Energy Efficiency

Energy Effciency
edited by
Jenny Palm
SCIYO
Energy Effciency
Edited by Jenny Palm
Published by Sciyo
Janeza Trdine 9, 51000 Rijeka, Croatia
Copyright © 2010 Sciyo
All chapters are Open Access articles distributed under the Creative Commons Non Commercial Share
Alike Attribution 3.0 license, which permits to copy, distribute, transmit, and adapt the work in any
medium, so long as the original work is properly cited. After this work has been published by Sciyo,
authors have the right to republish it, in whole or part, in any publication of which they are the author,
and to make other personal use of the work. Any republication, referencing or personal use of the work
must explicitly identify the original source.
Statements and opinions expressed in the chapters are these of the individual contributors and
not necessarily those of the editors or publisher. No responsibility is accepted for the accuracy of
information contained in the published articles. The publisher assumes no responsibility for any
damage or injury to persons or property arising out of the use of any materials, instructions, methods
or ideas contained in the book.

Publishing Process Manager Ana Nikolic
Technical Editor Sonja Mujacic
Cover Designer Martina Sirotic
Image Copyright Alfgar, 2010. Used under license from Shutterstock.com
First published September 2010
Printed in India
A free online edition of this book is available at www.sciyo.com
Additional hard copies can be obtained from [email protected]
Energy Effciency, Edited by Jenny Palm
p. cm.
ISBN 978-953-307-137-4
SCIYO.COM
WHERE KNOWLEDGE IS FREE
free online editions of Sciyo
Books, Journals and Videos can
be found at www.sciyo.com
Chapter 1
Chapter 2
Chapter 3
Chapter 4
Chapter 5
Chapter 6
Chapter 7
Chapter 8
Chapter 9
Preface VII
Energy Effciency Policy 1
Zoran Morvaj and Vesna Bukarica
Energy growth, complexity and effciency 27
Franco Ruzzenenti and Riccardo Basosi
Categorizing Barriers to Energy Effciency: An Interdisciplinary
Perspective 49
Patrik Thollander, Jenny Palm and Patrik Rohdin
Factors infuencing energy effciency in the German and Colombian
manufacturing industries 63
Clara Inés Pardo Martínez
Oxyfuel combustion in the steel industry: energy effciency and
decrease of co2 emissions 83
Author Name
Low-energy buildings – scientifc trends and developments 103
Dr. Patrik Rohdin, Dr. Wiktoria Glad and Dr. Jenny Palm
Energy transformed: building capacity in the engineering profession in
australia 125
Cheryl Desha and Karlson ‘Charlie’ Hargroves
The energy effciency of onboard hydrogen storage 143
Jens Oluf Jensen, Qingfeng Li and Niels J. Bjerrum
Energy effciency of Fuel Processor – PEM Fuel Cell systems 157
Lucia Salemme, Laura Menna and Marino Simeone
Contents
Global warming resulting from the use of fossil fuels is threatening the environment
and energy effciency is one of the most important ways to reduce this threat. Industry,
transport and buildings are all high energy-using sectors in the world and even in the most
technologically optimistic perspectives energy use is projected to increase in the next 50 years.
How and when energy is used determines society’s ability to create long-term sustainable
energy systems. This is why this book, focusing on energy effciency in these sectors and from
different perspectives, is sharp and also important for keeping a well-founded discussion on
the subject.
Transforming energy systems toward greater sustainability requires technological shifts
as well as transformations in behaviour, values, and routines to conserve energy. This
transformation can be facilitated by policy means and government initiatives as well as
technological improvements and innovations. This book combines engineering and social
science approaches to enhance our understanding of energy effciency and broaden our
perspective on policy making regarding energy effciency. The book will be an essential read
for anyone interested in how to contribute to the development of sustainable energy policies
and achieve improved energy effciency in industry, transport and the built environment.
The book is organised as follows. In the frst chapter Morvaj and Bukarica discuss how to
design, implement and evaluate energy effcient policy. This is followed by chapter 2 where
Basosi and Ruzzenenti highlight the rebound effect and problematise why the world sees a
growth in energy consumption despite the trend of higher effciency.
The following three chapters focus on industrial energy effciency. Thollander, Palm and
Rohdin discuss earlier studies on industrial barriers and how STS-perspective can contribute
to the barrier literature. Martinez compares factors that infuence energy effciency in German
and Colombian manufacturing. Such comparison is important to improve our understanding
of which factors are globally valid and which factors are more locally anchored. In chapter
5 von Schéele shows how specifc technologies become important for achieving increased
energy effciency in industrial processes.
Chapters 6 and 7 in different ways relate to development in the building sector. In chapter
6 Rohdin, Glad and Palm have done a literature review on methods and main results in
scientifc publications on low-energy buildings and low-energy architecture. In chapter 7
Desha and Hargroves discuss education of built professionals, such as architects, planners
and engineers, and the challenge and opportunities that exist for future professionals with
extensive knowledge about energy effciency in buildings.
Preface
VIII
The last two chapters both concern how different technologies can contribute to achieve
ambitious policy goals on energy effciency. In chapter 8 Jensen, Li and Bjerrum compare
different hydrogen storage techniques in terms of energy effciency and capacity available. In
the last chapter Simeone, Salemme and Menna present a comprehensive analysis of energy
effciency of fuel processor.
Sustainable development demands new strategies, solutions, and policy-making approaches.
This book discusses a wide spectrum of research on how to achieve ambitious policy goals on
energy effciency ranging from how energy effcient policy can be improved to how different
technologies can contribute to a more energy effcient future.
Editor
Jenny Palm
Tema T, Linköping University,
Sweden
Energy Effciency Policy 1
Energy Effciency Policy
Zoran Morvaj and Vesna Bukarica
x

Energy efficiency policy

Zoran Morvaj
1
and Vesna Bukarica
2

1
United Nations Development Programme (UNDP)
2
University of Zagreb Faculty of Electrical Engineering and Computing
Croatia

1. Introduction
Access to all forms of energy at affordable prices is an impetus for economic and social
development of the society. At the same time, energy sector is responsible for approximately
75 percent of total greenhouse gases emissions, which makes it the main provocative of
climate change. The convergence of international concerns about climate change and energy
security in the past decade has led to the increased awareness of policy-makers and general
public about energy issues and creation of new energy paradigm, the focus of which is
energy efficiency. Energy not used is arguably the best, the cheapest and the least
environmentally damaging source of energy supply and nowadays the concept of
"negawatts" in energy strategies worldwide is being introduced. However, energy efficiency
being typically demand side option is hard to implement due to the variety of stakeholders,
i.e. players in the energy efficiency market that need to be stimulated to adopt energy
efficiency as a way of doing business and ultimately a way of living - the change of mindset
is needed. As higher efficiency of energy use is indisputably a public interest, especially in
the light of the climate change combat, policy interventions are necessary to remove existing
market barriers hindering the fulfilment of potentials for cost-effective efficiency
improvements. Policy instruments to enhance energy efficiency improvements must
stimulate the transformation of the market towards higher efficiency, with the final aim of
achieving cleaner environment, better standard of living, more competitive industry and
improved security of energy supply. Moreover, they have to be designed according to the
real needs of the market (tailor-made), and have to have the flexibility and ability to respond
(adapt) to the changing market requirements in order to achieve goals in the optimal
manner.
Although there are excellent policies in place worldwide, with the European Union (EU)
being the indisputable energy efficiency and climate change combat leader, the results in
terms of reduced energy consumption are missing in the desired extent. Therefore, energy
efficiency policy making needs new, innovative approaches the main feature of which is
dynamics. Dynamic policy making means that it has to be learning, continuous, closed-loop
process which involves and balances policy design, implementation and evaluation. The
aim of this chapter is to explain these three main pillars of effective energy efficiency policy
making, focusing especially on implementation issues, which are usually highly neglected in
policy making process but are crucial for policy success.
1
Energy Effciency 2

2. Understanding energy efficiency policy making
2.1. Energy efficiency concept: avoid, reduce, monitor and manage
The basis for understanding the concept of energy efficiency is energy flow, from primary
energy contained in energy carriers to the useful energy consumed through various
activities of the society (Fig. 1).


Fig. 1. Energy flow - basis for understanding energy efficiency

Energy efficiency is all about tackling energy losses. As shown in Fig. 1, it boils down to the
very simple and understandable equation:

E
useful
= E
primary
-E
losses
(1)

Losses occur in processes of energy transformation, transmission, and distribution as well as
in the final uses of energy. While reducing losses in the first three activities is mainly a
matter of technology, the latest should be tackled by both technical and non-technical
measures. Often unnecessary uses of energy could be avoided by better organisation, better
energy management and changes in consumers’ behaviour and increasingly so by changing
lifestyle, which is the most difficult part. Energy efficiency has to be considered as a
continuous process that does not include only one-time actions to avoid excessive use of
energy and to minimise energy losses, but also includes monitoring and controlling energy
consumption with the aim of achieving continuous minimal energy consumption level.
Therefore, energy efficiency improvements rest on the following pillars (Morvaj & Bukarica,
2010):
 Avoiding excessive and unnecessary use of energy through regulation (e.g. building
codes and minimal standards) and policies that stimulate behavioural changes;
 Reducing energy losses by implementing energy efficiency improvement measures and
new technologies (e.g. waste heat recovery or use of LED lighting);
 Monitoring energy consumption in order to improve knowledge on energy
consumption patterns and their consequences (e.g. smart metering and real-time
pricing).
 Managing energy consumption by improving operational and maintenance practices.

To ensure continuity of energy efficiency improvements, energy consumption has to be
managed as any other activity. Actually, energy management can be denoted as a
framework for ensuring continuous avoidance of excessive energy use and reduction of
energy losses supported by a body of knowledge and adequate measuring and ICT
technology (Morvaj & Gvozdenac, 2008). It should not only consider techno-economic
features of energy consumption but should make energy efficiency an ongoing social
process. It also rests on the fact that energy has to be priced in a manner that more
accurately reflects its actual costs, which include, inter alia impacts on the environment,
health and geopolitics, and that consumers have to be made aware of these consequences of
energy use. These main pillars for achieving energy efficiency improvements have to be
taken into account in the policy making process - "avoiding" and stimulation of "reducing"
shall be a main driver in design of policy instruments, while for "monitoring" and
"managing" implementing capacities with appropriate capabilities and supporting
infrastructure shall be ensured.

2.2. Rationale behind energy efficiency: means not an end
Energy efficiency shall be regarded as a mean to achieve overall efficient resource allocation
(Dennis, 2006), rather then the goal in it self. As a consequence of improved energy
efficiency, other public policy goals will be achieved as well, the most important of which
are the goals of economic development and climate change mitigation.
In economic terms, and taking into account the fact that energy costs typically account to 15
to 20 percent of national gross domestic product, the significance of energy efficiency is
evident - reduced energy consumption lowers the costs for energy. For example, it is
estimated that the EU, although the world's most energy efficient region, still uses 20
percent more energy than it would be economically justified, which is the equivalent to
some of 390 Mtoe (European Commission, 2006) or the gross inland consumption of
Germany and Sweden together (Eurostat, 2009).
Furthermore, global consensus is emerging about consequences of inaction for mitigation of
an adaptation to climate change, and clear quantifiable targets (limiting CO
2
concentration
and temperature increase) within the given time frame (until 2012, than 2020 and finally
2050) need to be achieved if wish to avert a major disasters in the foreseeable future. For the
first time energy policy making is faced with such strict constraints, which require a
radically different approach in the whole cycle of policy making with special emphasis on
policy implementation. Energy efficiency is globally considered to be the most readily
available and rapid way to achieve desired greenhouse gases reductions in the short to
medium term. And taking into account the possible grave threats of climate change, the time
scale in energy policy has never been more important.
Let us briefly look at the evolution of energy policy making and the role of energy efficiency
(Fig. 2.). The standard energy policy making approach implied balancing of energy demand
and supply and slow evolution of policy goals, mixes and objectives as a response to various
external changes and drivers. The standard energy policy making was not faced with
serious constrains and specifically not time constraints for achieving certain results and
objectives. The time scales of energy policies were rather long, actions were gradually
undertaken (leading often to under investing in energy sector) and mainly left to the
decisions of energy companies, which led to the critical neglect of energy policy
implementing capacities at various levels of jurisdiction and in the society in general.
Energy Effciency Policy 3

2. Understanding energy efficiency policy making
2.1. Energy efficiency concept: avoid, reduce, monitor and manage
The basis for understanding the concept of energy efficiency is energy flow, from primary
energy contained in energy carriers to the useful energy consumed through various
activities of the society (Fig. 1).


Fig. 1. Energy flow - basis for understanding energy efficiency

Energy efficiency is all about tackling energy losses. As shown in Fig. 1, it boils down to the
very simple and understandable equation:

E
useful
= E
primary
-E
losses
(1)

Losses occur in processes of energy transformation, transmission, and distribution as well as
in the final uses of energy. While reducing losses in the first three activities is mainly a
matter of technology, the latest should be tackled by both technical and non-technical
measures. Often unnecessary uses of energy could be avoided by better organisation, better
energy management and changes in consumers’ behaviour and increasingly so by changing
lifestyle, which is the most difficult part. Energy efficiency has to be considered as a
continuous process that does not include only one-time actions to avoid excessive use of
energy and to minimise energy losses, but also includes monitoring and controlling energy
consumption with the aim of achieving continuous minimal energy consumption level.
Therefore, energy efficiency improvements rest on the following pillars (Morvaj & Bukarica,
2010):
 Avoiding excessive and unnecessary use of energy through regulation (e.g. building
codes and minimal standards) and policies that stimulate behavioural changes;
 Reducing energy losses by implementing energy efficiency improvement measures and
new technologies (e.g. waste heat recovery or use of LED lighting);
 Monitoring energy consumption in order to improve knowledge on energy
consumption patterns and their consequences (e.g. smart metering and real-time
pricing).
 Managing energy consumption by improving operational and maintenance practices.

To ensure continuity of energy efficiency improvements, energy consumption has to be
managed as any other activity. Actually, energy management can be denoted as a
framework for ensuring continuous avoidance of excessive energy use and reduction of
energy losses supported by a body of knowledge and adequate measuring and ICT
technology (Morvaj & Gvozdenac, 2008). It should not only consider techno-economic
features of energy consumption but should make energy efficiency an ongoing social
process. It also rests on the fact that energy has to be priced in a manner that more
accurately reflects its actual costs, which include, inter alia impacts on the environment,
health and geopolitics, and that consumers have to be made aware of these consequences of
energy use. These main pillars for achieving energy efficiency improvements have to be
taken into account in the policy making process - "avoiding" and stimulation of "reducing"
shall be a main driver in design of policy instruments, while for "monitoring" and
"managing" implementing capacities with appropriate capabilities and supporting
infrastructure shall be ensured.

2.2. Rationale behind energy efficiency: means not an end
Energy efficiency shall be regarded as a mean to achieve overall efficient resource allocation
(Dennis, 2006), rather then the goal in it self. As a consequence of improved energy
efficiency, other public policy goals will be achieved as well, the most important of which
are the goals of economic development and climate change mitigation.
In economic terms, and taking into account the fact that energy costs typically account to 15
to 20 percent of national gross domestic product, the significance of energy efficiency is
evident - reduced energy consumption lowers the costs for energy. For example, it is
estimated that the EU, although the world's most energy efficient region, still uses 20
percent more energy than it would be economically justified, which is the equivalent to
some of 390 Mtoe (European Commission, 2006) or the gross inland consumption of
Germany and Sweden together (Eurostat, 2009).
Furthermore, global consensus is emerging about consequences of inaction for mitigation of
an adaptation to climate change, and clear quantifiable targets (limiting CO
2
concentration
and temperature increase) within the given time frame (until 2012, than 2020 and finally
2050) need to be achieved if wish to avert a major disasters in the foreseeable future. For the
first time energy policy making is faced with such strict constraints, which require a
radically different approach in the whole cycle of policy making with special emphasis on
policy implementation. Energy efficiency is globally considered to be the most readily
available and rapid way to achieve desired greenhouse gases reductions in the short to
medium term. And taking into account the possible grave threats of climate change, the time
scale in energy policy has never been more important.
Let us briefly look at the evolution of energy policy making and the role of energy efficiency
(Fig. 2.). The standard energy policy making approach implied balancing of energy demand
and supply and slow evolution of policy goals, mixes and objectives as a response to various
external changes and drivers. The standard energy policy making was not faced with
serious constrains and specifically not time constraints for achieving certain results and
objectives. The time scales of energy policies were rather long, actions were gradually
undertaken (leading often to under investing in energy sector) and mainly left to the
decisions of energy companies, which led to the critical neglect of energy policy
implementing capacities at various levels of jurisdiction and in the society in general.
Energy Effciency 4

Nowadays, energy policy is entering a new constrained phase, with time as the main
constrain being imposed by the desire to combat climate change.



Fig. 2. Gradual changes of energy policy accents due to various drivers (Morvaj & Bukarica, 2010)

Energy efficiency solely can deliver the desired greenhouse gases reduction targets to the large
extent. To confirm the statement, the EU has been taken as an example. It is estimated that
fulfilling 20 percent target for energy efficiency improvements by 2020 would mean reducing
greenhouse gases emissions by 780 million tonnes, more than twice the EU reductions needed
under the Kyoto Protocol by 2012 (European Commission, 2006). Since the EU has committed to
reduce its greenhouse gases emissions by 20 percent compared to 1990 by 2020 and since the
EU's greenhouse gases emissions in 1990 amounted 5,564 million tonnes (European Environment
Agency, 2009), it is evident that 20 percent of energy efficiency improvement can deliver almost
three fourths of desired greenhouse gases reduction target. The power of energy efficiency as a
tool for climate change combat is therefore obvious.

2.3. Levels of energy efficiency policy: from enabling to implementing
Taking into account the role energy efficiency plays in reaching global goals of climate change
combat, it is understandable that there is a need for coordinated actions at all levels -
international, regional (e.g. European Union) and national to ensure enabling environment for
energy efficiency improvements by formulating appropriate policy instruments. However, the
real power to change is local. Policies have to be designed in a way that enables local
implementation in homes, public services and businesses. The interconnection between levels of
energy efficiency policy is illustrated in Fig. 3.

Fig. 3. Levels of energy efficiency policy

2.3.1. International aspect of energy efficiency policy
Due to its significance, energy efficiency is the topic of international agreements related to
climate change combat, environmental protection and security of energy supply. Money and
effort are put into promotion of energy efficiency by numerous international institutions, as
briefly demonstrated in Table 1.

International treaties and agreements on Climate Change and EE
Name of the document Year Main features
Energy Charter Treaty 1994 Legally-biding multilateral instrument, obliging parties,
inter alia, to reducing negative environmental impact of
energy cycle through improving energy efficiency
Energy Charter Protocol on
EE and Related
Environmental Aspects
(PEEREA)
1994 Recognises EE as considerable source of energy and obliges
parties to promote EE and to create framework which will
induce both producers and consumers to use energy in the
most efficient and environment friendly way as possible
Kyoto Protocol to United
Nations Framework
Convention on Climate
Change (UNFCCC)
1997 Obliges parties to reduce GHG in time period 2008-2012.
Defines flexible mechanisms that will ease the achievement
of targets at the least cost
International institutions/programmes for energy efficiency
Institution/Programme Year Main features
Global Environment
Facility
1991 -
2009
GEF is main financial mechanism of UNFCCC; GEF has
supported 131 EE projects with portfolio of approximately
850 million USD
World Bank Group 2005-
2009
Renewable energy and EE at the heart of WBG energy
agenda; in period 2005-2009 over 4 billion USD given for EE
projects world wide
United Nations
Development Programme,
United Nations Foundation
/ Energy as an important factor in reaching Millennium
Development Goals and reducing Poverty; Calls for
international “Efficiency First” agreement; Number of EE
projects financed world wide
International Energy
Agency
/ EE one of six broad focus areas of IEA's G8 Gleneagles
Programme - IEA submitted 25 policy recommendations to
the G8 for promoting EE that could reduce global CO2
emissions by 8.2 gigatonnes by 2030.
Table 1. International treaties and programmes for energy efficiency (Morvaj & Bukarica, 2010)

As seen from Table 1, international treaties and programmes are supported by various
financing tools, bilateral and international donors, but there is very little focus on how to
implement policy measures and instruments, hence the real results in terms of sustainable
and verifiable energy efficiency improvements and greenhouse gases reductions are
missing. It is absolutely crucial to shift the focus of international policies towards real-life
application, respecting in this process different local circumstances.
Namely, the drivers for energy efficiency and implementing environments differ
significantly on the global scene. Four "blocks" could be identified as shown in the Fig. 4.
The EU, followed by some other OECD countries, is certainly a forerunner in combating
climate change and in related energy efficiency activities. USA and BRIC countries are the
most vocal in defending their national interests and resisting any firm commitments for CO
2

reduction. Developing countries collectively represent a significant block in terms of
Energy Effciency Policy 5

Nowadays, energy policy is entering a new constrained phase, with time as the main
constrain being imposed by the desire to combat climate change.



Fig. 2. Gradual changes of energy policy accents due to various drivers (Morvaj & Bukarica, 2010)

Energy efficiency solely can deliver the desired greenhouse gases reduction targets to the large
extent. To confirm the statement, the EU has been taken as an example. It is estimated that
fulfilling 20 percent target for energy efficiency improvements by 2020 would mean reducing
greenhouse gases emissions by 780 million tonnes, more than twice the EU reductions needed
under the Kyoto Protocol by 2012 (European Commission, 2006). Since the EU has committed to
reduce its greenhouse gases emissions by 20 percent compared to 1990 by 2020 and since the
EU's greenhouse gases emissions in 1990 amounted 5,564 million tonnes (European Environment
Agency, 2009), it is evident that 20 percent of energy efficiency improvement can deliver almost
three fourths of desired greenhouse gases reduction target. The power of energy efficiency as a
tool for climate change combat is therefore obvious.

2.3. Levels of energy efficiency policy: from enabling to implementing
Taking into account the role energy efficiency plays in reaching global goals of climate change
combat, it is understandable that there is a need for coordinated actions at all levels -
international, regional (e.g. European Union) and national to ensure enabling environment for
energy efficiency improvements by formulating appropriate policy instruments. However, the
real power to change is local. Policies have to be designed in a way that enables local
implementation in homes, public services and businesses. The interconnection between levels of
energy efficiency policy is illustrated in Fig. 3.

Fig. 3. Levels of energy efficiency policy

2.3.1. International aspect of energy efficiency policy
Due to its significance, energy efficiency is the topic of international agreements related to
climate change combat, environmental protection and security of energy supply. Money and
effort are put into promotion of energy efficiency by numerous international institutions, as
briefly demonstrated in Table 1.

International treaties and agreements on Climate Change and EE
Name of the document Year Main features
Energy Charter Treaty 1994 Legally-biding multilateral instrument, obliging parties,
inter alia, to reducing negative environmental impact of
energy cycle through improving energy efficiency
Energy Charter Protocol on
EE and Related
Environmental Aspects
(PEEREA)
1994 Recognises EE as considerable source of energy and obliges
parties to promote EE and to create framework which will
induce both producers and consumers to use energy in the
most efficient and environment friendly way as possible
Kyoto Protocol to United
Nations Framework
Convention on Climate
Change (UNFCCC)
1997 Obliges parties to reduce GHG in time period 2008-2012.
Defines flexible mechanisms that will ease the achievement
of targets at the least cost
International institutions/programmes for energy efficiency
Institution/Programme Year Main features
Global Environment
Facility
1991 -
2009
GEF is main financial mechanism of UNFCCC; GEF has
supported 131 EE projects with portfolio of approximately
850 million USD
World Bank Group 2005-
2009
Renewable energy and EE at the heart of WBG energy
agenda; in period 2005-2009 over 4 billion USD given for EE
projects world wide
United Nations
Development Programme,
United Nations Foundation
/ Energy as an important factor in reaching Millennium
Development Goals and reducing Poverty; Calls for
international “Efficiency First” agreement; Number of EE
projects financed world wide
International Energy
Agency
/ EE one of six broad focus areas of IEA's G8 Gleneagles
Programme - IEA submitted 25 policy recommendations to
the G8 for promoting EE that could reduce global CO2
emissions by 8.2 gigatonnes by 2030.
Table 1. International treaties and programmes for energy efficiency (Morvaj & Bukarica, 2010)

As seen from Table 1, international treaties and programmes are supported by various
financing tools, bilateral and international donors, but there is very little focus on how to
implement policy measures and instruments, hence the real results in terms of sustainable
and verifiable energy efficiency improvements and greenhouse gases reductions are
missing. It is absolutely crucial to shift the focus of international policies towards real-life
application, respecting in this process different local circumstances.
Namely, the drivers for energy efficiency and implementing environments differ
significantly on the global scene. Four "blocks" could be identified as shown in the Fig. 4.
The EU, followed by some other OECD countries, is certainly a forerunner in combating
climate change and in related energy efficiency activities. USA and BRIC countries are the
most vocal in defending their national interests and resisting any firm commitments for CO
2

reduction. Developing countries collectively represent a significant block in terms of
Energy Effciency 6

greenhouse gases emissions. Energy efficiency is for them a win-win approach for reducing
the greenhouse gases emissions while also reducing costs of energy for their fragile
economies. Therefore, energy efficiency in developing countries should be addressed
immediately and incorporated in energy policies with strong supporting implementation
mechanisms.


Fig. 4. World differences in climate change and energy efficiency policies adoption (Morvaj
& Bukarica, 2010)

The efforts from the international level are extremely useful and necessary, but they are still
not enough, i.e. they are generic in their nature, hence are not able to deliver real results.
International policies, programmes and aids shall be brought down to the national and local
level in every "block", where conditions for policy implementation are different, requiring
thus tailor-made solutions in both policy instruments and implementing capacities.

2.3.2. Regional energy efficiency policy: case EU
The indisputable "energy efficiency forerunner" in the world is the European Union (EU).
The EU has strongly stressed its aim to achieve the "20-20-20" targets by 2020: to reduce
greenhouse gases emissions minimally 20 percent (with the intention to even achieve 30
percent greenhouse gases emission cut by 2030); to increase the proportion of renewable
energies in the energy mix by 20 percent and to reduce primary energy consumption by 20
percent. In order to achieve the energy efficiency improvement goals, the EU has introduced
a well thought of set of voluntary and some mandatory polices. The most important policy
and legislative documents related to energy efficiency in the EU are summarised in the
Table 2.

EU policy documents on EE
Name of the document Year Main features
EE in European Community –
Towards a Strategy for the
1998 Analyse available economical potential for
improvements in energy efficiency, identifies barriers

Rational Use of Energy (COM
(1998)) 246 final)
and gives proposals to remove those barriers. Estimates
that saving of 18% of 1995 energy consumption can be
achieved by 2010 (160 Mtoe).
Action Plan to Improve EE in
the European Community
(COM (2000) 247 final)
2000 Sets a target for energy intensity improvement by an
additional 1% per year compared to a business as usual
trend resulting in 100 Mtoe avoided energy consumption
by 2010.
Green Paper on EE or Doing
More with Less (COM (2005)
265 final)
2005 Expresses urging need to put energy saving policy
higher on the EU agenda and estimates that EU is using
20% more energy then economically justifiable and if
additional efforts are not made, this potential will not be
fulfilled by current policies.
Action Plan for Energy
Efficiency: Realising the
Potential (COM(2006) 545)
2006 Sets energy saving target of 20 percent by 2020 (390
Mtoe) and defines 6 priority policy measures (energy
performance standards; improving energy
transformation; focusing on transport; providing
financial incentives and ensuring correct energy pricing;
changing energy behaviour; fostering international
partnership).
Second Strategic Energy Review
- An EU Energy Security and
Solidarity Action Plan
(COM/2008/0781)
2008 Reinforces EE efforts to achieve 20% target - calls for
revision of directives on energy performance of
buildings, appliance labelling and eco-design, strongly
promotes Covenant of Mayors, use of cohesion policy
and funds and tax system to boost energy efficiency.
EU EE legislation (directives)
Directive 92/75/EEC on energy
labelling of household appliances
and implementing directives
1992 Prescribes obligatory EE labelling for 8 groups of
household appliances.
Directive 2002/91/EC on the
energy performance of buildings
(Proposal for a Directive on the
energy performance of buildings
(recast) [COM(2008)780])
2002
(reca
st
prop
osed
in
2008)
Calls for minimum energy requirements for new and
existing buildings, energy certification and regular
inspection of boilers and air conditioning systems.
Directive 2004/8/EC on the
promotion of cogeneration based
on a useful heat demand in the
internal energy market
2004 Facilitate the installation and operation of electrical
cogeneration plants.
Directive 2005/32/EC
establishing a framework for the
setting of eco-design
requirements for energy-using
products and implementing
directives
2005 Defines the principles, conditions and criteria for setting
environmental requirements for energy-using
appliances.
Directive 2006/32/EC on
Energy end-use Efficiency and
Energy Services
2006 Calls for establishment of indicative energy savings
target for the Member States, obligations on national
public authorities as regards energy savings and energy
efficient procurement, and measures to promote EE and
energy services.
Table 2. EU policy documents for energy efficiency (Morvaj & Bukarica, 2010)
Energy Effciency Policy 7

greenhouse gases emissions. Energy efficiency is for them a win-win approach for reducing
the greenhouse gases emissions while also reducing costs of energy for their fragile
economies. Therefore, energy efficiency in developing countries should be addressed
immediately and incorporated in energy policies with strong supporting implementation
mechanisms.


Fig. 4. World differences in climate change and energy efficiency policies adoption (Morvaj
& Bukarica, 2010)

The efforts from the international level are extremely useful and necessary, but they are still
not enough, i.e. they are generic in their nature, hence are not able to deliver real results.
International policies, programmes and aids shall be brought down to the national and local
level in every "block", where conditions for policy implementation are different, requiring
thus tailor-made solutions in both policy instruments and implementing capacities.

2.3.2. Regional energy efficiency policy: case EU
The indisputable "energy efficiency forerunner" in the world is the European Union (EU).
The EU has strongly stressed its aim to achieve the "20-20-20" targets by 2020: to reduce
greenhouse gases emissions minimally 20 percent (with the intention to even achieve 30
percent greenhouse gases emission cut by 2030); to increase the proportion of renewable
energies in the energy mix by 20 percent and to reduce primary energy consumption by 20
percent. In order to achieve the energy efficiency improvement goals, the EU has introduced
a well thought of set of voluntary and some mandatory polices. The most important policy
and legislative documents related to energy efficiency in the EU are summarised in the
Table 2.

EU policy documents on EE
Name of the document Year Main features
EE in European Community –
Towards a Strategy for the
1998 Analyse available economical potential for
improvements in energy efficiency, identifies barriers

Rational Use of Energy (COM
(1998)) 246 final)
and gives proposals to remove those barriers. Estimates
that saving of 18% of 1995 energy consumption can be
achieved by 2010 (160 Mtoe).
Action Plan to Improve EE in
the European Community
(COM (2000) 247 final)
2000 Sets a target for energy intensity improvement by an
additional 1% per year compared to a business as usual
trend resulting in 100 Mtoe avoided energy consumption
by 2010.
Green Paper on EE or Doing
More with Less (COM (2005)
265 final)
2005 Expresses urging need to put energy saving policy
higher on the EU agenda and estimates that EU is using
20% more energy then economically justifiable and if
additional efforts are not made, this potential will not be
fulfilled by current policies.
Action Plan for Energy
Efficiency: Realising the
Potential (COM(2006) 545)
2006 Sets energy saving target of 20 percent by 2020 (390
Mtoe) and defines 6 priority policy measures (energy
performance standards; improving energy
transformation; focusing on transport; providing
financial incentives and ensuring correct energy pricing;
changing energy behaviour; fostering international
partnership).
Second Strategic Energy Review
- An EU Energy Security and
Solidarity Action Plan
(COM/2008/0781)
2008 Reinforces EE efforts to achieve 20% target - calls for
revision of directives on energy performance of
buildings, appliance labelling and eco-design, strongly
promotes Covenant of Mayors, use of cohesion policy
and funds and tax system to boost energy efficiency.
EU EE legislation (directives)
Directive 92/75/EEC on energy
labelling of household appliances
and implementing directives
1992 Prescribes obligatory EE labelling for 8 groups of
household appliances.
Directive 2002/91/EC on the
energy performance of buildings
(Proposal for a Directive on the
energy performance of buildings
(recast) [COM(2008)780])
2002
(reca
st
prop
osed
in
2008)
Calls for minimum energy requirements for new and
existing buildings, energy certification and regular
inspection of boilers and air conditioning systems.
Directive 2004/8/EC on the
promotion of cogeneration based
on a useful heat demand in the
internal energy market
2004 Facilitate the installation and operation of electrical
cogeneration plants.
Directive 2005/32/EC
establishing a framework for the
setting of eco-design
requirements for energy-using
products and implementing
directives
2005 Defines the principles, conditions and criteria for setting
environmental requirements for energy-using
appliances.
Directive 2006/32/EC on
Energy end-use Efficiency and
Energy Services
2006 Calls for establishment of indicative energy savings
target for the Member States, obligations on national
public authorities as regards energy savings and energy
efficient procurement, and measures to promote EE and
energy services.
Table 2. EU policy documents for energy efficiency (Morvaj & Bukarica, 2010)
Energy Effciency 8

The analysis of these documents clearly shows the commitment and huge policy efforts to
boost energy efficiency improvements. Despite that, the EU is far from reaching its 20
percent energy efficiency improvement target by 2020. The results of the policy
implementation are missing in the desired extent, leaving the huge potential of "negawatts"
idle. With the current legislation and policy instruments in place, a reduction of only 8.5
percent will be achieved. Even taking into account additional measures in the pipeline, at
the best only 11 percent reductions will be achieved, as shown in the Fig. 5 (European
Commission, 2009). However, the EU policy only provides the framework national policies
have to cope with. It is, to the largest extent, the task of national policies to deliver actual
energy efficiency improvements. Obviously, they are failing to do so.


1990 1995 2000 2005 2010 2015 2020
1.500
1.600
1.700
1.800
1.900
2.000
2.100
2.200
Mtoe
445 Mtoe
- 8, 5%
-11 ,3%
-20%
No EE policy
PRIMES 2007 baseline
PRIMES 2009 baseline ( adopted, policies)
EE policy mix (PRIMES 2009 + additional measures)

20% EE target

(according to PRIMES 2007 baseline)



Fig. 5. Development and projection of Gross Inland Energy Consumption for EU by 2020
(European Commission, 2009)

2.3.2. National energy efficiency policy: (not) delivering targets
In national energy efficiency policy there is a symptomatic unbalance between efforts for
preparing polices, and preparations for policy implementation. The vast majority of policy
makers are focused on incorporating requirements of international policies and
requirements into national strategic and legislative frameworks, without thorough
consideration of national circumstances, i.e. without taking into account the level of energy
efficiency market maturity in a country. Moreover, there is a general lack of focus on policy
implementation and a sort of general expectation that implementation is straightforward,
will hopefully happened by itself, hence there is no need to put too much efforts into that.
Current national energy efficiency policies are persistently missing or underachieving the
desired results. There are number of reasons behind this policy failure, but the problem is
essentially threefold:
1. Policy makers do not fully tackle all stakeholders relevant for energy efficiency,
i.e. not all market players are tackled with appropriate policy instruments that
would remove market imperfections and enable sustainability. There is a need for
all-a-compassing, tailor-made policies, adaptive to specific changing market
conditions.

2. Policy making needs to appreciate specific implementing environment
conditions and time constraints for implementation, thus focusing on creating
sufficient and appropriate implementing capacities that are adequate for achieving
the targets. A model for developing implementing capacities shall be established.
3. Policies are not static, meaning that policy making is not on-time job. It requires
well established procedures for policy monitoring and evaluation that will reveal
what works and what does not work in the practice and provide inputs for policy
improved redesign.
Obviously, new approach in overall energy efficiency policy making is needed, the main
feature of which is dynamics.

2.4. Policy dynamics: key to effective energy efficiency policy making
For energy efficiency policy to be successful its creation has to be a learning process based
on both theoretical knowledge and empirical data. This learning process can be the most
appropriately described by the closed-loop process (Fig. 6) consisting of the following
stages:
 Policy design:
o Policy definition: objectives, targets, approaches for different target groups, legal
and regulatory frameworks;
o Policy instruments development: incentives, penalties, standards, technical
assistance, financing support;
 Policy implementation: institutional framework, stakeholders, human resources,
capacity and capability development, supporting infrastructure (ICT);
 Policy evaluation: monitoring of achieved results through energy statistics and energy
efficiency indicators, qualitative and quantitative evaluation of policy instruments'
impacts.
Design
of
EE Policy
Implementation
of
EE Policy
Evaluation
of
EE Policy
SUCCESSFULL
IMPLEMENTATION
of Energy Effciency
Improvement
Project
Target
Group
Energy Efficiency Market
REPETITION!
TRANSFORMATION!
M
A
R
K
E
T

A
S
S
E
S
M
E
N
T
Deciding on
Redefinition
of EE policy

Fig. 6. Dynamics of energy efficiency policy (Bukarica et al., 2007)

Energy Effciency Policy 9

The analysis of these documents clearly shows the commitment and huge policy efforts to
boost energy efficiency improvements. Despite that, the EU is far from reaching its 20
percent energy efficiency improvement target by 2020. The results of the policy
implementation are missing in the desired extent, leaving the huge potential of "negawatts"
idle. With the current legislation and policy instruments in place, a reduction of only 8.5
percent will be achieved. Even taking into account additional measures in the pipeline, at
the best only 11 percent reductions will be achieved, as shown in the Fig. 5 (European
Commission, 2009). However, the EU policy only provides the framework national policies
have to cope with. It is, to the largest extent, the task of national policies to deliver actual
energy efficiency improvements. Obviously, they are failing to do so.


1990 1995 2000 2005 2010 2015 2020
1.500
1.600
1.700
1.800
1.900
2.000
2.100
2.200
Mtoe
445 Mtoe
- 8, 5%
-11 ,3%
-20%
No EE policy
PRIMES 2007 baseline
PRIMES 2009 baseline ( adopted, policies)
EE policy mix (PRIMES 2009 + additional measures)

20% EE target

(according to PRIMES 2007 baseline)



Fig. 5. Development and projection of Gross Inland Energy Consumption for EU by 2020
(European Commission, 2009)

2.3.2. National energy efficiency policy: (not) delivering targets
In national energy efficiency policy there is a symptomatic unbalance between efforts for
preparing polices, and preparations for policy implementation. The vast majority of policy
makers are focused on incorporating requirements of international policies and
requirements into national strategic and legislative frameworks, without thorough
consideration of national circumstances, i.e. without taking into account the level of energy
efficiency market maturity in a country. Moreover, there is a general lack of focus on policy
implementation and a sort of general expectation that implementation is straightforward,
will hopefully happened by itself, hence there is no need to put too much efforts into that.
Current national energy efficiency policies are persistently missing or underachieving the
desired results. There are number of reasons behind this policy failure, but the problem is
essentially threefold:
1. Policy makers do not fully tackle all stakeholders relevant for energy efficiency,
i.e. not all market players are tackled with appropriate policy instruments that
would remove market imperfections and enable sustainability. There is a need for
all-a-compassing, tailor-made policies, adaptive to specific changing market
conditions.

2. Policy making needs to appreciate specific implementing environment
conditions and time constraints for implementation, thus focusing on creating
sufficient and appropriate implementing capacities that are adequate for achieving
the targets. A model for developing implementing capacities shall be established.
3. Policies are not static, meaning that policy making is not on-time job. It requires
well established procedures for policy monitoring and evaluation that will reveal
what works and what does not work in the practice and provide inputs for policy
improved redesign.
Obviously, new approach in overall energy efficiency policy making is needed, the main
feature of which is dynamics.

2.4. Policy dynamics: key to effective energy efficiency policy making
For energy efficiency policy to be successful its creation has to be a learning process based
on both theoretical knowledge and empirical data. This learning process can be the most
appropriately described by the closed-loop process (Fig. 6) consisting of the following
stages:
 Policy design:
o Policy definition: objectives, targets, approaches for different target groups, legal
and regulatory frameworks;
o Policy instruments development: incentives, penalties, standards, technical
assistance, financing support;
 Policy implementation: institutional framework, stakeholders, human resources,
capacity and capability development, supporting infrastructure (ICT);
 Policy evaluation: monitoring of achieved results through energy statistics and energy
efficiency indicators, qualitative and quantitative evaluation of policy instruments'
impacts.
Design
of
EE Policy
Implementation
of
EE Policy
Evaluation
of
EE Policy
SUCCESSFULL
IMPLEMENTATION
of Energy Effciency
Improvement
Project
Target
Group
Energy Efficiency Market
REPETITION!
TRANSFORMATION!
M
A
R
K
E
T

A
S
S
E
S
M
E
N
T
Deciding on
Redefinition
of EE policy

Fig. 6. Dynamics of energy efficiency policy (Bukarica et al., 2007)

Energy Effciency 10

Energy efficiency policy in its essence shall be a market transformation programme.
Market transformation programmes are strategic interventions that cause lasting changes in
the structure or function of markets for all energy-efficient products/services/practices
(Brinner & Martinot, 2005). The effective market transformation programme rests on the
following key pillars:
 mix of policy instruments created to remove market barriers identified throughout all
stages of the individual energy efficiency project development;
 policy interventions adaptive to market conditions ensuring sustainability of energy
efficiency improvements through replications of successfully implemented energy
efficiency projects;
 policy instruments tailored to enable all market players (government, private sector,
consumers, equipment producers, service providers, financing institutions, etc.) to find
their interest in improved energy efficiency;
 energy efficiency improvements achieved as the result of supply-demand interactions
based on competitive market forces.
Therefore, prior to the start of energy efficiency policy design the market assessment shall be
preformed. It shall reveal the maturity of the market. This is extremely important, as
different instruments have different effects and are therefore appropriate at different market
maturity levels, i.e. some measures could stimulate market introduction, whereas other
measures could accelerate commercialisation, or increase the overall penetration of energy-
efficient products and services (Brinner & Martinot, 2005). Market analysis is required to
identify market forces that have to be strengthened by incentives or diminished by
penalties. The policy instruments should be carefully designed and mixed in order to tackle
identified market barriers.
Conceptually, the typical energy efficiency policy cycle starts with strategic planning and
determination of targets leading to the design of specific instruments to tackle different
target groups, i.e. market players. The implementation of policy instruments follows and
one cycle is concluded with the evaluation of policy impacts. The results of the policy
evaluation process are then fed into the planning, design, and implementation processes,
and the cycle repeats itself (Vine, 2008). Every stage in this dynamic loop requires
methodical and systematic approach and will be given all due attention in the subsequent
sections.

3. Main postulates for defining effective energy efficiency policy
3.1. Understanding energy efficiency markets
The starting point in creation of any policy is to understand how market operates and how
well developed it really is. Unlike the economic theory that assumes perfect competition, the
real markets are imperfect due to various barriers preventing market forces to deliver desired
results. The task of any policy is to identify these barriers and to develop market-based
incentives and well-designed, forward-looking instruments for their removal (Dennis, 2006).
Policies usually define various instruments to support implementation of energy efficiency
measures in energy end-use sectors (households, services, industry, transport). Very often,
the proposed instruments are generic and designed without a proper appreciation of the
situation on the ground – an energy efficiency market place where energy efficiency
measures need to be adopted by consumers, supported by energy service providing

companies. Addressing end-users solely is not nearly sufficient to ensure self-sustainable
energy efficiency improvements. The concept of energy efficiency market shall be
introduced and understood for creating and implementing energy efficiency policy.
Energy efficiency market is not exactly one market but a conglomeration of various and very
diverse businesses acting in the field and having different interests in energy efficiency
realm. Energy efficiency market's supply side includes providers of energy efficient
equipment and services as well as institutions involved in financing and implementation of
energy efficiency projects (banks, investment funds, design engineers, constructors, etc.).
The demand side of energy efficiency market includes project sponsors with ideas for
energy efficiency improvements (end-users, i.e. building owners and renters, building
managers, public sector institutions and local authorities, industries).
The performance of energy efficiency market is evaluated according to the actual energy
efficiency improvements delivered, i.e. according to number of successfully implemented
energy efficiency projects. Basically, the energy efficiency market transformation depends
on the success of the project development process. Development of an energy efficiency
project goes through various stages, from the very initial idea, until the final and actual
implementation of the project that operates and yields results in terms of reduced energy
consumption and emissions (Fig. 7). Due to various market barriers, only few of a variety of
identified opportunities for energy efficiency improvements reach the stage of a bankable
project, becoming actually implemented; hence the narrowed pipeline presentation is
chosen.


Fig. 7. Understanding energy efficiency projects' development cycle and energy efficiency
markets (Bukarica et al., 2007)

3.2. Definition of policy instruments for market transformation
One of the main reasons for energy efficiency policy failure lies in the preference of policy
makers to use universal solutions in definition of energy efficiency policy and basically to
copy-paste policy instruments from others without considering the specificities of own
country's energy efficiency market. There are, of course, some general market barriers for
energy efficiency which require such universal solutions (Table 3), but they are not nearly
sufficient to provoke market transformation and to fulfil the final goal - creation of self-
sustainable energy efficiency market.


Energy Effciency Policy 11

Energy efficiency policy in its essence shall be a market transformation programme.
Market transformation programmes are strategic interventions that cause lasting changes in
the structure or function of markets for all energy-efficient products/services/practices
(Brinner & Martinot, 2005). The effective market transformation programme rests on the
following key pillars:
 mix of policy instruments created to remove market barriers identified throughout all
stages of the individual energy efficiency project development;
 policy interventions adaptive to market conditions ensuring sustainability of energy
efficiency improvements through replications of successfully implemented energy
efficiency projects;
 policy instruments tailored to enable all market players (government, private sector,
consumers, equipment producers, service providers, financing institutions, etc.) to find
their interest in improved energy efficiency;
 energy efficiency improvements achieved as the result of supply-demand interactions
based on competitive market forces.
Therefore, prior to the start of energy efficiency policy design the market assessment shall be
preformed. It shall reveal the maturity of the market. This is extremely important, as
different instruments have different effects and are therefore appropriate at different market
maturity levels, i.e. some measures could stimulate market introduction, whereas other
measures could accelerate commercialisation, or increase the overall penetration of energy-
efficient products and services (Brinner & Martinot, 2005). Market analysis is required to
identify market forces that have to be strengthened by incentives or diminished by
penalties. The policy instruments should be carefully designed and mixed in order to tackle
identified market barriers.
Conceptually, the typical energy efficiency policy cycle starts with strategic planning and
determination of targets leading to the design of specific instruments to tackle different
target groups, i.e. market players. The implementation of policy instruments follows and
one cycle is concluded with the evaluation of policy impacts. The results of the policy
evaluation process are then fed into the planning, design, and implementation processes,
and the cycle repeats itself (Vine, 2008). Every stage in this dynamic loop requires
methodical and systematic approach and will be given all due attention in the subsequent
sections.

3. Main postulates for defining effective energy efficiency policy
3.1. Understanding energy efficiency markets
The starting point in creation of any policy is to understand how market operates and how
well developed it really is. Unlike the economic theory that assumes perfect competition, the
real markets are imperfect due to various barriers preventing market forces to deliver desired
results. The task of any policy is to identify these barriers and to develop market-based
incentives and well-designed, forward-looking instruments for their removal (Dennis, 2006).
Policies usually define various instruments to support implementation of energy efficiency
measures in energy end-use sectors (households, services, industry, transport). Very often,
the proposed instruments are generic and designed without a proper appreciation of the
situation on the ground – an energy efficiency market place where energy efficiency
measures need to be adopted by consumers, supported by energy service providing

companies. Addressing end-users solely is not nearly sufficient to ensure self-sustainable
energy efficiency improvements. The concept of energy efficiency market shall be
introduced and understood for creating and implementing energy efficiency policy.
Energy efficiency market is not exactly one market but a conglomeration of various and very
diverse businesses acting in the field and having different interests in energy efficiency
realm. Energy efficiency market's supply side includes providers of energy efficient
equipment and services as well as institutions involved in financing and implementation of
energy efficiency projects (banks, investment funds, design engineers, constructors, etc.).
The demand side of energy efficiency market includes project sponsors with ideas for
energy efficiency improvements (end-users, i.e. building owners and renters, building
managers, public sector institutions and local authorities, industries).
The performance of energy efficiency market is evaluated according to the actual energy
efficiency improvements delivered, i.e. according to number of successfully implemented
energy efficiency projects. Basically, the energy efficiency market transformation depends
on the success of the project development process. Development of an energy efficiency
project goes through various stages, from the very initial idea, until the final and actual
implementation of the project that operates and yields results in terms of reduced energy
consumption and emissions (Fig. 7). Due to various market barriers, only few of a variety of
identified opportunities for energy efficiency improvements reach the stage of a bankable
project, becoming actually implemented; hence the narrowed pipeline presentation is
chosen.


Fig. 7. Understanding energy efficiency projects' development cycle and energy efficiency
markets (Bukarica et al., 2007)

3.2. Definition of policy instruments for market transformation
One of the main reasons for energy efficiency policy failure lies in the preference of policy
makers to use universal solutions in definition of energy efficiency policy and basically to
copy-paste policy instruments from others without considering the specificities of own
country's energy efficiency market. There are, of course, some general market barriers for
energy efficiency which require such universal solutions (Table 3), but they are not nearly
sufficient to provoke market transformation and to fulfil the final goal - creation of self-
sustainable energy efficiency market.


Energy Effciency 12

Primary
Barriers
Effects Solutions
Incomplete
(imperfect)
information
Affects both demand and supply side of
EE market leaving the demand
underdeveloped and supply side
disinterested
Dedicated promotional and
informational campaigns;
Energy labelling of appliance,
equipment, buildings and cars
EE as public
goods
Markets tend to undersupply public
goods
Stimulating Research and
Development of energy efficient
technologies;
Voluntary agreements with
manufacturing industries
Externalities Energy price does not reflect the adverse
environmental and human health effects
of energy consumption nor impacts of
political instabilities related to energy
supply;
Positive externalities of improved EE
should also be taken into account.
Correct energy pricing and energy
taxation;
Environmental fees (but usually
imposed to large consumers only);
Tax credits for EE investments ;
Minimal efficiency standards;
Utilising purchasing power (green
public procurement and consumers'
awareness)
Market power
(imperfect
market
structures)
Remains of monopoly in energy sectors
prevent development of truly competitive
energy markets and restructuring of
utilities to become energy service
companies;
Improper structures of energy prices
based on historical average costs and not
on short-run marginal costs

Transforming utilities to become
energy service companies;
Smart metering and real-time
pricing;
Smart appliances

Secondary
Barriers -
consequences
of primary
barriers
Effects Solutions
Lack of access
to capital
Makes it difficult or impossible to invest
in energy efficiency
EE (revolving) funds (as initial
driver of demand for energy
efficient solutions);
Transforming utilities to become
energy service companies
Mindset
(rather then
market)
barrier
Effects Solutions
Consumers'
behaviour
Optimal decisions will not be made
regardless sufficient information provided
due to bounded rationality
Energy and climate literacy (a top
educational priority in schools and
in the public discourse)
Table 3. General market barriers to energy efficiency and universal solutions (Morvaj &
Bukarica, 2010)

Instead of routine proposals of generic policy instruments, specific status of energy efficiency
market in a given jurisdiction has to be understood, and for every stage in the energy

efficiency project development process specific barriers must be identified and support policy
instruments designed to ensure project pipeline throughput (Bukarica et al. 2007). In other
words, policy instruments have to be tailor-made for specific market circumstances.
Energy efficiency market has a variety of players with different backgrounds and as such is
highly influenced by behavioural, socio-economic and psychological factors that govern
market players’ decisions. All these influences have to be taken into account when defining
policy instruments for energy efficiency improvement. As indicated in the Fig. 8, combination
of policy instruments has to be used to remove both supply and demand side barriers, i.e. both
supply and demand side have to be addressed simultaneously when markets are “stuck”. In
other words, producers/service providers have to be stimulated to produce/offer more
efficient products/services, while consumers have to be stimulated to by such
products/services. What this means is that if there is no demand for energy efficient
products/services suppliers are not interested in improving their performance by themselves
and vice verso, if there is no efficient products/services offered in the market, there is no
demand for them either. Policy instruments have to be designed to move this situation from
the deadlock and to fulfil the ultimate goal of market transformation - to achieve public
benefits from increased energy efficiency as accepted mode of behaviour (Bukarica et al., 2007).


Fig. 8. Defining energy efficiency policy instruments based on actual status of a specific
energy efficiency market (Morvaj & Bukarica, 2010) (Note: the scheme was developed during
market assessment and creation of energy efficiency policy in the Republic of Croatia)

Policy-makers have to understand that policy instruments are not equally relevant at all points
in time – the requirement for different instruments vary with maturity of the market and
timing of utilisation. Therefore, policies have to be adaptive to changing market conditions.
Energy Effciency Policy 13

Primary
Barriers
Effects Solutions
Incomplete
(imperfect)
information
Affects both demand and supply side of
EE market leaving the demand
underdeveloped and supply side
disinterested
Dedicated promotional and
informational campaigns;
Energy labelling of appliance,
equipment, buildings and cars
EE as public
goods
Markets tend to undersupply public
goods
Stimulating Research and
Development of energy efficient
technologies;
Voluntary agreements with
manufacturing industries
Externalities Energy price does not reflect the adverse
environmental and human health effects
of energy consumption nor impacts of
political instabilities related to energy
supply;
Positive externalities of improved EE
should also be taken into account.
Correct energy pricing and energy
taxation;
Environmental fees (but usually
imposed to large consumers only);
Tax credits for EE investments ;
Minimal efficiency standards;
Utilising purchasing power (green
public procurement and consumers'
awareness)
Market power
(imperfect
market
structures)
Remains of monopoly in energy sectors
prevent development of truly competitive
energy markets and restructuring of
utilities to become energy service
companies;
Improper structures of energy prices
based on historical average costs and not
on short-run marginal costs

Transforming utilities to become
energy service companies;
Smart metering and real-time
pricing;
Smart appliances

Secondary
Barriers -
consequences
of primary
barriers
Effects Solutions
Lack of access
to capital
Makes it difficult or impossible to invest
in energy efficiency
EE (revolving) funds (as initial
driver of demand for energy
efficient solutions);
Transforming utilities to become
energy service companies
Mindset
(rather then
market)
barrier
Effects Solutions
Consumers'
behaviour
Optimal decisions will not be made
regardless sufficient information provided
due to bounded rationality
Energy and climate literacy (a top
educational priority in schools and
in the public discourse)
Table 3. General market barriers to energy efficiency and universal solutions (Morvaj &
Bukarica, 2010)

Instead of routine proposals of generic policy instruments, specific status of energy efficiency
market in a given jurisdiction has to be understood, and for every stage in the energy

efficiency project development process specific barriers must be identified and support policy
instruments designed to ensure project pipeline throughput (Bukarica et al. 2007). In other
words, policy instruments have to be tailor-made for specific market circumstances.
Energy efficiency market has a variety of players with different backgrounds and as such is
highly influenced by behavioural, socio-economic and psychological factors that govern
market players’ decisions. All these influences have to be taken into account when defining
policy instruments for energy efficiency improvement. As indicated in the Fig. 8, combination
of policy instruments has to be used to remove both supply and demand side barriers, i.e. both
supply and demand side have to be addressed simultaneously when markets are “stuck”. In
other words, producers/service providers have to be stimulated to produce/offer more
efficient products/services, while consumers have to be stimulated to by such
products/services. What this means is that if there is no demand for energy efficient
products/services suppliers are not interested in improving their performance by themselves
and vice verso, if there is no efficient products/services offered in the market, there is no
demand for them either. Policy instruments have to be designed to move this situation from
the deadlock and to fulfil the ultimate goal of market transformation - to achieve public
benefits from increased energy efficiency as accepted mode of behaviour (Bukarica et al., 2007).


Fig. 8. Defining energy efficiency policy instruments based on actual status of a specific
energy efficiency market (Morvaj & Bukarica, 2010) (Note: the scheme was developed during
market assessment and creation of energy efficiency policy in the Republic of Croatia)

Policy-makers have to understand that policy instruments are not equally relevant at all points
in time – the requirement for different instruments vary with maturity of the market and
timing of utilisation. Therefore, policies have to be adaptive to changing market conditions.
Energy Effciency 14

Adaptive policy response means that utilisation of instruments and funding designated for
their implementation must correspond to the market demands. E.g. offering partial financial
guarantees to the banks will have very modest impact in markets where there is no demand
for energy efficiency projects and banks do not find the interest to offer specialised financial
products for the. As a general guideline, instruments for awareness raising and technical
assistance are more important in developing energy efficiency markets, while with its maturity
financial incentives become increasingly desired.

 Not all policy instruments are suitable for all markets:
o Understand the maturity level of country's energy efficiency market and tailor
policy instruments to overcome identified barriers;
o Use experiences of others, but do not copy-paste without taking into account
real market situation - what works in one country, does not have to work in
other;
o Every policy instrument has its right timing for implementation - take one step
at time to ensure smooth transformation of the market i.e. smooth transition
from one phase to another as shown in Fig. 7;
 Not all policy instruments are suitable for all market players - be specific in
determining target groups for a certain policy instrument (e.g. voluntary agreements
are not suitable for households consumers, while appliance labelling will have little
to do with large industry consumers);
 Not all policy instruments are suitable for all energy end-use sectors (households,
public services, private services, industry, and transport) - sectors' specificities shall
be taken into account;
 Sometimes it is useful to determine package of instruments (combinations of two or
more instruments, e.g. building code in combination with subsidies for
demonstrating achievement of higher standards or promotion campaign for cleaner
transport in combination with subsidies for purchasing hybrid cars) to increase
policy effectiveness and efficiency;
 Identify sectors that can be the best tackled by policy and that would have the
largest immediate and spill-over effects:
o Experience shows that putting policy focus on public sector is both easiest to
implement and it provides the largest spill-over effect to other sectors by
demonstrating effects of energy efficiency improvements, but it also has a
potential to transform the market in a short span of time due to large
purchasing power of the public sector;
o Buildings usually consume more then 40 percent of country's energy demand,
therefore this sector offers the largest potential for energy efficiency
improvements (especially existing building stock) that could be achieved
through advanced building codes and energy performance standards;
 Look for local best practices and make them national - often there are local
initiatives in a country that have great results and capability for replication;
 Be aware of your implementing capabilities - available budget and, even more
important, institutional capacities needed for implementation of policy instruments.

Box 1. "Quick-win" guidelines for designing successful energy efficiency policy instruments

4. Energy efficiency policy implementation
4.1. Understanding implementing environment
The immediate questions aimed at understanding the "implementing environment for
energy efficiency policies" are:
 Who has to do what? In other words, what are the roles and responsibilities of different
stakeholders.
 Were the implementation has to happen? The answer, although as simple as possible,
is often overlooked - policy needs to be implemented where energy is used everyday –
and this is at our places of work and at our homes.
It is very simple fact that all energy delivered is consumed directly by people or indirectly
through different institutional and business forms created by people (Fig. 9), during the
course of our professional and private life. Therefore, for implementation of energy
efficiency measures and a full policy uptake, the mobilisation and cooperation of all
stakeholders is needed. The international institutions and efforts form an umbrella of this
implementing environment, dictating the framework for policy creation and
implementation (as discussed in the section 2). At national level, four key groups of
stakeholders, i.e. vertical social structures can be identified (Fig. 9), all of which have their
specific roles in energy efficiency policy implementation and their activities (or lack thereof)
influence the energy efficiency market.
The primary role of the public sector institutions is to ensure national policy
implementation in all end-use sectors (households, services, industry and transport).
However, at the same time the public sector, same as businesses, are the realms where
policy is actually being implemented. Civil society organisations and media, on the other
hand, play the key role in providing information and promoting energy efficiency on the
wide scale, which will, in the long run, enable changing the consumers' mindset towards
more energy efficient behaviour.


Fig. 9. Main pillars of implementing environment for energy efficiency policy

Energy Effciency Policy 15

Adaptive policy response means that utilisation of instruments and funding designated for
their implementation must correspond to the market demands. E.g. offering partial financial
guarantees to the banks will have very modest impact in markets where there is no demand
for energy efficiency projects and banks do not find the interest to offer specialised financial
products for the. As a general guideline, instruments for awareness raising and technical
assistance are more important in developing energy efficiency markets, while with its maturity
financial incentives become increasingly desired.

 Not all policy instruments are suitable for all markets:
o Understand the maturity level of country's energy efficiency market and tailor
policy instruments to overcome identified barriers;
o Use experiences of others, but do not copy-paste without taking into account
real market situation - what works in one country, does not have to work in
other;
o Every policy instrument has its right timing for implementation - take one step
at time to ensure smooth transformation of the market i.e. smooth transition
from one phase to another as shown in Fig. 7;
 Not all policy instruments are suitable for all market players - be specific in
determining target groups for a certain policy instrument (e.g. voluntary agreements
are not suitable for households consumers, while appliance labelling will have little
to do with large industry consumers);
 Not all policy instruments are suitable for all energy end-use sectors (households,
public services, private services, industry, and transport) - sectors' specificities shall
be taken into account;
 Sometimes it is useful to determine package of instruments (combinations of two or
more instruments, e.g. building code in combination with subsidies for
demonstrating achievement of higher standards or promotion campaign for cleaner
transport in combination with subsidies for purchasing hybrid cars) to increase
policy effectiveness and efficiency;
 Identify sectors that can be the best tackled by policy and that would have the
largest immediate and spill-over effects:
o Experience shows that putting policy focus on public sector is both easiest to
implement and it provides the largest spill-over effect to other sectors by
demonstrating effects of energy efficiency improvements, but it also has a
potential to transform the market in a short span of time due to large
purchasing power of the public sector;
o Buildings usually consume more then 40 percent of country's energy demand,
therefore this sector offers the largest potential for energy efficiency
improvements (especially existing building stock) that could be achieved
through advanced building codes and energy performance standards;
 Look for local best practices and make them national - often there are local
initiatives in a country that have great results and capability for replication;
 Be aware of your implementing capabilities - available budget and, even more
important, institutional capacities needed for implementation of policy instruments.

Box 1. "Quick-win" guidelines for designing successful energy efficiency policy instruments

4. Energy efficiency policy implementation
4.1. Understanding implementing environment
The immediate questions aimed at understanding the "implementing environment for
energy efficiency policies" are:
 Who has to do what? In other words, what are the roles and responsibilities of different
stakeholders.
 Were the implementation has to happen? The answer, although as simple as possible,
is often overlooked - policy needs to be implemented where energy is used everyday –
and this is at our places of work and at our homes.
It is very simple fact that all energy delivered is consumed directly by people or indirectly
through different institutional and business forms created by people (Fig. 9), during the
course of our professional and private life. Therefore, for implementation of energy
efficiency measures and a full policy uptake, the mobilisation and cooperation of all
stakeholders is needed. The international institutions and efforts form an umbrella of this
implementing environment, dictating the framework for policy creation and
implementation (as discussed in the section 2). At national level, four key groups of
stakeholders, i.e. vertical social structures can be identified (Fig. 9), all of which have their
specific roles in energy efficiency policy implementation and their activities (or lack thereof)
influence the energy efficiency market.
The primary role of the public sector institutions is to ensure national policy
implementation in all end-use sectors (households, services, industry and transport).
However, at the same time the public sector, same as businesses, are the realms where
policy is actually being implemented. Civil society organisations and media, on the other
hand, play the key role in providing information and promoting energy efficiency on the
wide scale, which will, in the long run, enable changing the consumers' mindset towards
more energy efficient behaviour.


Fig. 9. Main pillars of implementing environment for energy efficiency policy

Energy Effciency 16

4.2. Roles and responsibilities of key stakeholders
Public institutions play, with no doubt, pivotal role in enabling and enhancing policy
implementation. However, the governments, i.e. competent ministries themselves rarely
have the capacities to deal with policy implementation issues. Therefore, in many countries
specialised national energy efficiency agencies are established as governmental
implementing bodies. They have a crucial role in initiating energy efficiency programmes,
coordination of activities and especially in monitoring and evaluation of policy
implementation.
To support this statement, a fact that nowadays more than 70 percent of European
population lives in cities has to be emphasised. Even more so, in 2009 for the first time in
history official statistics have reported that globally more than 50 percent of world
population lives in cities. Hence cities are obvious places where vigorous, continuous and
focused implementation of energy efficiency measures needs to be carried out by all key
stakeholders (see Fig. 3).
Being closest to places where energy is consumed and still having executive powers, local
authorities more than ever have a pivotal role to play at reducing energy consumption.
Actions that local authorities (and public sector in general) should undertake are twofold:
 Firstly, energy consumption in facilities and services in their jurisdiction should be
properly managed. This means that local authorities shall demonstrate their
commitment by implementing energy efficiency improvement measures in all buildings
in their jurisdiction (office buildings, schools and kindergartens, hospitals, etc.) as well
as in public services they provide (public lighting, transport, energy and water supply).
 Secondly, information must be made publicly available and cooperation with civil
society organisations, businesses and media has to be established to improve citizens’
awareness and facilitate change of energy related behaviour and attitude.
Building local capacities to perform these activities is the most important precondition for
successful policy implementation and delivering policy targets. Introduction of full-scale
energy management is instrumental there, which could be a backbone for evolution of
"smart cities" and sustainable urban development (Paskaleva, 2009).
In all business sectors, the climate change awareness and social responsibility are driving
companies to demonstrate their "greenness". The new "green" revolution in the corporative
world is led by the biggest - Google and Microsoft are going solar, Dell is committed to
neutralising carbon impact of its operations, Wal-Mart aims at completely renewable energy
supply, crating zero waste and selling products that sustain resources and the environment
(Stanislaw, 2008). However, while corporations do have money and human capacities to
turn their business towards more efficient and environmentally friendly solutions, small and
medium enterprises (SMEs) need role-models and support to improve their energy
efficiency, hence the overall business performance. The 2007 Observatory of EU SMEs
indicates that only 29 percent of SMEs have instituted some measures for preserving energy
and resources (46 percent in the case of large enterprises) and that only 4 percent of EU
SMEs have a comprehensive system in place for energy efficiency, which is much lover then
for large enterprises (19 percent) (European Commission, 2009). Again, energy management
is the solution.
And finally, policy makers together with civil society organisations, businesses and media
have to work together to ensure that energy and climate change literacy (Stanislaw, 2008)
becomes a top educational priority in schools and in the public discourse. In this task, civil

society organisations and media have particularly important role, since they formulate the
public opinion and are able to establish a new "green" ethic in rising generations.
Therefore, the solution for ensuring proper implementing environment for energy efficiency
policies lies in bringing together and mobilizing for action all stakeholders so that every
pillar of the society contributes fully according to their own means for achievement of
energy efficiency policy targets. Strong links, as demonstrated in Fig.10, between each and
every stakeholder shall be established, not only whilst implementing policy, but
immediately during the process of energy efficiency policy design. Either link is equally
important as the current practice has indicated that policy making lacking feedback from all
stakeholders results in weak and slow implementation. The Fig. 10 aims to illustrate the
need for stakeholders' interactions in various energy efficiency activities, and points that
such coordinated and collaborative approach will influence citizens and eventually
transform the market and society towards higher efficiency.


Fig. 10. Stakeholders' interactions in different energy efficiency activities

4.3 Building implementing capacities through Energy Management System
Implementing capacities can be successfully strengthen through the process known as
Energy Management System (EMS). It comprises a specific set of knowledge and skills
based on organizational structure incorporating the following elements:
 people with assigned responsibilities
 energy efficiency monitoring through calculation and analysis of:
o energy consumption indicators
o energy efficiency improvement targets
Energy Effciency Policy 17

4.2. Roles and responsibilities of key stakeholders
Public institutions play, with no doubt, pivotal role in enabling and enhancing policy
implementation. However, the governments, i.e. competent ministries themselves rarely
have the capacities to deal with policy implementation issues. Therefore, in many countries
specialised national energy efficiency agencies are established as governmental
implementing bodies. They have a crucial role in initiating energy efficiency programmes,
coordination of activities and especially in monitoring and evaluation of policy
implementation.
To support this statement, a fact that nowadays more than 70 percent of European
population lives in cities has to be emphasised. Even more so, in 2009 for the first time in
history official statistics have reported that globally more than 50 percent of world
population lives in cities. Hence cities are obvious places where vigorous, continuous and
focused implementation of energy efficiency measures needs to be carried out by all key
stakeholders (see Fig. 3).
Being closest to places where energy is consumed and still having executive powers, local
authorities more than ever have a pivotal role to play at reducing energy consumption.
Actions that local authorities (and public sector in general) should undertake are twofold:
 Firstly, energy consumption in facilities and services in their jurisdiction should be
properly managed. This means that local authorities shall demonstrate their
commitment by implementing energy efficiency improvement measures in all buildings
in their jurisdiction (office buildings, schools and kindergartens, hospitals, etc.) as well
as in public services they provide (public lighting, transport, energy and water supply).
 Secondly, information must be made publicly available and cooperation with civil
society organisations, businesses and media has to be established to improve citizens’
awareness and facilitate change of energy related behaviour and attitude.
Building local capacities to perform these activities is the most important precondition for
successful policy implementation and delivering policy targets. Introduction of full-scale
energy management is instrumental there, which could be a backbone for evolution of
"smart cities" and sustainable urban development (Paskaleva, 2009).
In all business sectors, the climate change awareness and social responsibility are driving
companies to demonstrate their "greenness". The new "green" revolution in the corporative
world is led by the biggest - Google and Microsoft are going solar, Dell is committed to
neutralising carbon impact of its operations, Wal-Mart aims at completely renewable energy
supply, crating zero waste and selling products that sustain resources and the environment
(Stanislaw, 2008). However, while corporations do have money and human capacities to
turn their business towards more efficient and environmentally friendly solutions, small and
medium enterprises (SMEs) need role-models and support to improve their energy
efficiency, hence the overall business performance. The 2007 Observatory of EU SMEs
indicates that only 29 percent of SMEs have instituted some measures for preserving energy
and resources (46 percent in the case of large enterprises) and that only 4 percent of EU
SMEs have a comprehensive system in place for energy efficiency, which is much lover then
for large enterprises (19 percent) (European Commission, 2009). Again, energy management
is the solution.
And finally, policy makers together with civil society organisations, businesses and media
have to work together to ensure that energy and climate change literacy (Stanislaw, 2008)
becomes a top educational priority in schools and in the public discourse. In this task, civil

society organisations and media have particularly important role, since they formulate the
public opinion and are able to establish a new "green" ethic in rising generations.
Therefore, the solution for ensuring proper implementing environment for energy efficiency
policies lies in bringing together and mobilizing for action all stakeholders so that every
pillar of the society contributes fully according to their own means for achievement of
energy efficiency policy targets. Strong links, as demonstrated in Fig.10, between each and
every stakeholder shall be established, not only whilst implementing policy, but
immediately during the process of energy efficiency policy design. Either link is equally
important as the current practice has indicated that policy making lacking feedback from all
stakeholders results in weak and slow implementation. The Fig. 10 aims to illustrate the
need for stakeholders' interactions in various energy efficiency activities, and points that
such coordinated and collaborative approach will influence citizens and eventually
transform the market and society towards higher efficiency.


Fig. 10. Stakeholders' interactions in different energy efficiency activities

4.3 Building implementing capacities through Energy Management System
Implementing capacities can be successfully strengthen through the process known as
Energy Management System (EMS). It comprises a specific set of knowledge and skills
based on organizational structure incorporating the following elements:
 people with assigned responsibilities
 energy efficiency monitoring through calculation and analysis of:
o energy consumption indicators
o energy efficiency improvement targets
Energy Effciency 18

 continuous measuring and improvement of efficiency.

Fig. 11. Concept of energy management system (Note: EMS is equally applicable in public
and business sector)

The process of introducing energy management starts from the decision of adopting an
energy management policy statement. It then leads to an energy management action plan
being adopted at the top management level. Measurable goals to be achieved are set within
the plan. The plan with defined goals is made public. This act ensures a constant support of
the top management and all employees to the implementation of energy management
project. This is followed by introduction of organizational infrastructure to deliver the plan.
A dedicated energy management team is appointed which assumes the obligation of overall
energy management on the level of a city or a company. Furthermore, every facility in the
structure of a company or in the ownership of a city has to have a person (usually technical
or maintenance) appointed as the one responsible for the local energy management. And
finally, all members of energy management team shall be adequately educated and trained
to perform their tasks. This way capacities and capabilities for implementation of energy
efficiency projects are ensured. Additionally, they need to be supported by appropriate ICT
tool for continuous collection, storage, monitoring and analysis of data on energy
consumption. Moreover, energy management team is also responsible for further
educational and promotional activities to change employees' behaviour and attitudes
towards energy consumption at the work place and for initiating green public procurement
activities to stimulate market transformation by utilizing public sector's huge purchase
power. And last, but not the least, energy management teams, especially those established
within pubic sector (i.e. local authorities) are reaching out to the citizens by publicly
announcing their activities and by providing advisory services. This comprehensive process
of energy management system introduction is shown in the Fig. 12. Although it shows the
process applied in the cities, it could be easily adjusted for business sector as well.
Once it is understood that policy implementation is happening locally, capacitating both
public and commercial business market players for implementing energy efficiency policy
through systematic introduction of energy management practices becomes the key to the
policy success.
Another look at the Fig. 8 reminds us that implemented projects are only vehicle that deliver
actual energy consumption reductions and they appear merely like a drops at the end of
pipeline that involves huge number of actors, actions, barriers and instruments to overcome
them. Without strong, focused, competent and effective capacities for implementing energy

efficiency policies it is unlikely to expect that projects would flow from the pipeline and that
the targets would be delivered.



Fig. 12. Energy management process in a city (Note: The scheme is applied in the cities of the
Republic of Croatia. The process is easily adjusted for business sector.) (Morvaj et al., 2008)

5. Evaluating energy efficiency policy: measurement and verification (M&V)
5.1. General issues on policy evaluation
In the energy efficiency policy cyclic loop policy evaluation has an essential position,
although it might not appear so. Namely, evaluation procedures are at the same time an
integral part of policy design phase as well as both parallel and consecutive activity to
policy implementation.
The first step in policy design shall be establishment of a plausible theory on how a policy
instrument (or a package of instruments) is expected to lead to energy efficiency
improvements (Blumstein, 2000). Based on well-reasoned assumptions (theory) policy
instruments mix shall be created. Well-reasoned means that strong believe exists that exactly
this instrument will lead to cost-beneficial improvements in energy efficiency market
performance. Policy makers should have as precise as possible conception of impacts that
policy instrument will deliver, prior to its implementation. This is referred to as ex-ante or
beforehand policy evaluation during which impacts (social, technological and financial) of
policy instruments are forecasted. Expected impact in terms of reduced energy consumption
and cost-effectiveness of the instruments are evaluated and compared to business-as-usual
scenario in which no instruments are applied. However, often policymakers do not have
enough experience and knowledge to confirm the established theory is right. Therefore,
policymaking has to be publicly open process involving all stakeholders and market actors
that could contribute to the overall understanding how the policy instrument is intended to
work.
Energy Effciency Policy 19

 continuous measuring and improvement of efficiency.

Fig. 11. Concept of energy management system (Note: EMS is equally applicable in public
and business sector)

The process of introducing energy management starts from the decision of adopting an
energy management policy statement. It then leads to an energy management action plan
being adopted at the top management level. Measurable goals to be achieved are set within
the plan. The plan with defined goals is made public. This act ensures a constant support of
the top management and all employees to the implementation of energy management
project. This is followed by introduction of organizational infrastructure to deliver the plan.
A dedicated energy management team is appointed which assumes the obligation of overall
energy management on the level of a city or a company. Furthermore, every facility in the
structure of a company or in the ownership of a city has to have a person (usually technical
or maintenance) appointed as the one responsible for the local energy management. And
finally, all members of energy management team shall be adequately educated and trained
to perform their tasks. This way capacities and capabilities for implementation of energy
efficiency projects are ensured. Additionally, they need to be supported by appropriate ICT
tool for continuous collection, storage, monitoring and analysis of data on energy
consumption. Moreover, energy management team is also responsible for further
educational and promotional activities to change employees' behaviour and attitudes
towards energy consumption at the work place and for initiating green public procurement
activities to stimulate market transformation by utilizing public sector's huge purchase
power. And last, but not the least, energy management teams, especially those established
within pubic sector (i.e. local authorities) are reaching out to the citizens by publicly
announcing their activities and by providing advisory services. This comprehensive process
of energy management system introduction is shown in the Fig. 12. Although it shows the
process applied in the cities, it could be easily adjusted for business sector as well.
Once it is understood that policy implementation is happening locally, capacitating both
public and commercial business market players for implementing energy efficiency policy
through systematic introduction of energy management practices becomes the key to the
policy success.
Another look at the Fig. 8 reminds us that implemented projects are only vehicle that deliver
actual energy consumption reductions and they appear merely like a drops at the end of
pipeline that involves huge number of actors, actions, barriers and instruments to overcome
them. Without strong, focused, competent and effective capacities for implementing energy

efficiency policies it is unlikely to expect that projects would flow from the pipeline and that
the targets would be delivered.



Fig. 12. Energy management process in a city (Note: The scheme is applied in the cities of the
Republic of Croatia. The process is easily adjusted for business sector.) (Morvaj et al., 2008)

5. Evaluating energy efficiency policy: measurement and verification (M&V)
5.1. General issues on policy evaluation
In the energy efficiency policy cyclic loop policy evaluation has an essential position,
although it might not appear so. Namely, evaluation procedures are at the same time an
integral part of policy design phase as well as both parallel and consecutive activity to
policy implementation.
The first step in policy design shall be establishment of a plausible theory on how a policy
instrument (or a package of instruments) is expected to lead to energy efficiency
improvements (Blumstein, 2000). Based on well-reasoned assumptions (theory) policy
instruments mix shall be created. Well-reasoned means that strong believe exists that exactly
this instrument will lead to cost-beneficial improvements in energy efficiency market
performance. Policy makers should have as precise as possible conception of impacts that
policy instrument will deliver, prior to its implementation. This is referred to as ex-ante or
beforehand policy evaluation during which impacts (social, technological and financial) of
policy instruments are forecasted. Expected impact in terms of reduced energy consumption
and cost-effectiveness of the instruments are evaluated and compared to business-as-usual
scenario in which no instruments are applied. However, often policymakers do not have
enough experience and knowledge to confirm the established theory is right. Therefore,
policymaking has to be publicly open process involving all stakeholders and market actors
that could contribute to the overall understanding how the policy instrument is intended to
work.
Energy Effciency 20

Unlike ex-ante evaluation of a policy, ex-post approach is applied after a certain time of the
policy instrument implementation, effects of which should be evaluated to answer two key
questions (Joosen& Harmelink, 2006):
 What was the contribution of policy instrument in the realisation of policy targets
(effectiveness of policy instruments)?
o Effectiveness of a policy instrument is measured as its net impact in the relation to
the policy target set in the design phase. Net impact is equal to the difference
between amount of energy used prior and after instrument is implemented. These
are net energy savings but also related net CO
2
emission reductions that can be
attributed to specific energy efficiency instrument taking free rider, spill over,
rebound effect and other possible effects into account. Net impact is determined
according to the previously defined baseline scenario.
 What was the cost effectiveness of policy instruments, and could targets have been
reached against lower costs?
o Cost effectiveness is the ratio between the additional costs caused by the
instrument for the end-user, the society as whole or the government, and the net
impact of the investigated instrument. Government costs are related to
implementation, administration, enforcement of regulations, monitoring and
evaluation, subsidies and tax relieves. In other words, cost effectiveness is used to
determine how well public money is used to achieve socially beneficial goals. For
end-users costs are determined by energy price, marginal investment and marginal
operation and maintenance costs of energy efficiency measure.
However, instruments of energy efficiency policies might have other effects as well, so the
third question it should be raised is:
 What other impacts did the policy achieved outside its main realm?
o Most usually mentioned side effects of energy efficiency policy are environmental
benefits and creation of new jobs, which are a positive effects in terms of ecological,
social and economic stability and progress. However, sometimes negative effects
are also possible to appear. E.g. CFLs are using far less energy and have longer life
time and in a world's combat against climate change they are now starting to
completely replace "old" incandescent light bulbs. However, CFLs do bring some
other hazards, like small amount of highly toxic mercury they contain. Policy
makers have to be aware of these relationships and often trade-offs have to be done
- in this case, the trade-off has to be done between efficiency and potential health
risk.
Answering these questions is referred as ex-post evaluation. It goes beyond evaluation of
final delivered energy savings and tries to reveal success and failure factor enhancing in that
way our knowledge about market performance. Enhanced knowledge gives the opportunity
to improve effectiveness of policy instruments and to redefine our policy. Here both
qualitative and quantitative assessments are needed and should be preferably supported by
empirical data about policy performance. The backbone is cause-impact relationship,
supplemented by indicators that measure the existence of cause-impact relationship, then
failure and success factors should be listed (qualitative) and relationships with other policy
instruments should be emphasised (other instruments can enhance or mitigate the impact of
analysed instrument). In evaluation process empirical data are also very important as they
are additional and often the only indicators of certain instruments impacts.

Both ex-ante and ex-post evaluation need to be supported with quantitative data, i.e. with
data on energy efficiency improvements actually realised by implementation of policy
instruments and energy efficiency improvement projects. The tools used for this purpose are
referred to as measurement and verification (M&V) of energy savings. M&V is absolutely
crucial part of any energy efficiency policy – it captures the overall improvement in energy
efficiency and assesses the impact of individual measures. M&V procedures include two
major methodological approaches: top-down and bottom-up. Both approaches must be
combined to appropriately and as exact as possible evaluate the success of national energy
efficiency policy and the magnitude of energy efficiency improvement measures’ impact.
Both approaches will be briefly explained hereafter, although it has to be emphasised that
the detailed elaboration of M&V principles goes far beyond the scope of this chapter.

5.2. Top-down M&V methods
A top-down calculation method means that the amount of energy savings is calculated
using the national or large-scale aggregated sectoral levels of energy saving as a starting
point. This is purely statistical approach, often referred to as “energy efficiency indicators”
because it gives an indication of developments.
Top-down methodology is based on collection of extensive data sets for not only energy
consumption but also for various factors influencing it, and on calculation and monitoring
of energy efficiency indicators. There are six types of indicators most commonly used. These
are as follows
1
.
1. Energy intensity – ratio between an energy consumption (measured in energy units:
toe, Joule) and an indicator of activity measured in monetary units (Gross Domestic
Product, value added). Energy intensities are the only indicators that can be used
every time energy efficiency is assessed at a high level of aggregation, where it is not
possible to characterize the activity with a technical or physical indicator, i.e. at the
level of the whole economy or of a sector.
2. Unit consumption or specific consumption – relates energy consumption to an
indicator of activity measured in physical terms (tons of steel, number of vehicle-km,
etc.) or to a consumption unit (vehicle, dwelling …).
3. Energy efficiency index (ODEX) – provides an overall assessment of energy
efficiency trends of a sector. They are calculated as a weighted average of detailed
sub-sectoral indicators (by end-use, transport mode...). A decrease means an energy
efficiency improvement. Such index is more relevant for grasping the reality of
energy efficiency changes than energy intensities.
4. Diffusion indicators – there are three types of such indicators: (i) market penetration
of renewables (number of solar water heaters, percentage of wood boilers for
heating, etc.); (ii) market penetration of efficient technologies (number of efficient
lamps sold, percentage of label A in new sales of electrical appliance, etc.); (iii)
diffusion of energy efficient practices (percentage of passenger transport by public
modes, by non motorised modes; percentage of transport of goods by rail, by
combined rail-road transport, percentage of efficient process in industry, etc.).
Diffusion indicators have been introduced to complement the existing energy

1
These indicators are developed within ODYSSEE project and are used Europe- wide. More can be
found at: http://www.odyssee-indicators.org/
Energy Effciency Policy 21

Unlike ex-ante evaluation of a policy, ex-post approach is applied after a certain time of the
policy instrument implementation, effects of which should be evaluated to answer two key
questions (Joosen& Harmelink, 2006):
 What was the contribution of policy instrument in the realisation of policy targets
(effectiveness of policy instruments)?
o Effectiveness of a policy instrument is measured as its net impact in the relation to
the policy target set in the design phase. Net impact is equal to the difference
between amount of energy used prior and after instrument is implemented. These
are net energy savings but also related net CO
2
emission reductions that can be
attributed to specific energy efficiency instrument taking free rider, spill over,
rebound effect and other possible effects into account. Net impact is determined
according to the previously defined baseline scenario.
 What was the cost effectiveness of policy instruments, and could targets have been
reached against lower costs?
o Cost effectiveness is the ratio between the additional costs caused by the
instrument for the end-user, the society as whole or the government, and the net
impact of the investigated instrument. Government costs are related to
implementation, administration, enforcement of regulations, monitoring and
evaluation, subsidies and tax relieves. In other words, cost effectiveness is used to
determine how well public money is used to achieve socially beneficial goals. For
end-users costs are determined by energy price, marginal investment and marginal
operation and maintenance costs of energy efficiency measure.
However, instruments of energy efficiency policies might have other effects as well, so the
third question it should be raised is:
 What other impacts did the policy achieved outside its main realm?
o Most usually mentioned side effects of energy efficiency policy are environmental
benefits and creation of new jobs, which are a positive effects in terms of ecological,
social and economic stability and progress. However, sometimes negative effects
are also possible to appear. E.g. CFLs are using far less energy and have longer life
time and in a world's combat against climate change they are now starting to
completely replace "old" incandescent light bulbs. However, CFLs do bring some
other hazards, like small amount of highly toxic mercury they contain. Policy
makers have to be aware of these relationships and often trade-offs have to be done
- in this case, the trade-off has to be done between efficiency and potential health
risk.
Answering these questions is referred as ex-post evaluation. It goes beyond evaluation of
final delivered energy savings and tries to reveal success and failure factor enhancing in that
way our knowledge about market performance. Enhanced knowledge gives the opportunity
to improve effectiveness of policy instruments and to redefine our policy. Here both
qualitative and quantitative assessments are needed and should be preferably supported by
empirical data about policy performance. The backbone is cause-impact relationship,
supplemented by indicators that measure the existence of cause-impact relationship, then
failure and success factors should be listed (qualitative) and relationships with other policy
instruments should be emphasised (other instruments can enhance or mitigate the impact of
analysed instrument). In evaluation process empirical data are also very important as they
are additional and often the only indicators of certain instruments impacts.

Both ex-ante and ex-post evaluation need to be supported with quantitative data, i.e. with
data on energy efficiency improvements actually realised by implementation of policy
instruments and energy efficiency improvement projects. The tools used for this purpose are
referred to as measurement and verification (M&V) of energy savings. M&V is absolutely
crucial part of any energy efficiency policy – it captures the overall improvement in energy
efficiency and assesses the impact of individual measures. M&V procedures include two
major methodological approaches: top-down and bottom-up. Both approaches must be
combined to appropriately and as exact as possible evaluate the success of national energy
efficiency policy and the magnitude of energy efficiency improvement measures’ impact.
Both approaches will be briefly explained hereafter, although it has to be emphasised that
the detailed elaboration of M&V principles goes far beyond the scope of this chapter.

5.2. Top-down M&V methods
A top-down calculation method means that the amount of energy savings is calculated
using the national or large-scale aggregated sectoral levels of energy saving as a starting
point. This is purely statistical approach, often referred to as “energy efficiency indicators”
because it gives an indication of developments.
Top-down methodology is based on collection of extensive data sets for not only energy
consumption but also for various factors influencing it, and on calculation and monitoring
of energy efficiency indicators. There are six types of indicators most commonly used. These
are as follows
1
.
1. Energy intensity – ratio between an energy consumption (measured in energy units:
toe, Joule) and an indicator of activity measured in monetary units (Gross Domestic
Product, value added). Energy intensities are the only indicators that can be used
every time energy efficiency is assessed at a high level of aggregation, where it is not
possible to characterize the activity with a technical or physical indicator, i.e. at the
level of the whole economy or of a sector.
2. Unit consumption or specific consumption – relates energy consumption to an
indicator of activity measured in physical terms (tons of steel, number of vehicle-km,
etc.) or to a consumption unit (vehicle, dwelling …).
3. Energy efficiency index (ODEX) – provides an overall assessment of energy
efficiency trends of a sector. They are calculated as a weighted average of detailed
sub-sectoral indicators (by end-use, transport mode...). A decrease means an energy
efficiency improvement. Such index is more relevant for grasping the reality of
energy efficiency changes than energy intensities.
4. Diffusion indicators – there are three types of such indicators: (i) market penetration
of renewables (number of solar water heaters, percentage of wood boilers for
heating, etc.); (ii) market penetration of efficient technologies (number of efficient
lamps sold, percentage of label A in new sales of electrical appliance, etc.); (iii)
diffusion of energy efficient practices (percentage of passenger transport by public
modes, by non motorised modes; percentage of transport of goods by rail, by
combined rail-road transport, percentage of efficient process in industry, etc.).
Diffusion indicators have been introduced to complement the existing energy

1
These indicators are developed within ODYSSEE project and are used Europe- wide. More can be
found at: http://www.odyssee-indicators.org/
Energy Effciency 22

efficiency indicators, as they are easier to monitor, often with a more rapid updating.
They aim at improving the interpretation of trends observed on the energy efficiency
indicators.
5. Adjusted energy efficiency indicators – account for differences existing among
countries in the climate, in economic structures or in technologies. Comparisons of
energy efficiency performance across countries are only meaningful if they are based
on such indicators. External factors that might influence energy consumption
include: (a) weather conditions, such as degree days; (b) occupancy levels; (c)
opening hours for non-domestic buildings; (d) installed equipment intensity (plant
throughput); product mix; (e) plant throughput, level of production, volume or
added value, including changes in GDP level; (f) schedules for installation and
vehicles; (g) relationship with other units. Some of these factors are relevant for
correction of aggregated indicators, while some are to be used for the individual
facilities in which energy efficiency measures are implemented.
6. Target indicators – aim at providing reference values to show possible target of
energy efficiency improvements or energy efficiency potentials for a given country.
They are somehow similar to benchmark value but defined at a macro level, which
implies a careful interpretation of differences. The target is defined as the distance to
the average of the 3 best countries; this distance shows what gain can be achieved.
The main advantages of the usage of top-down methods is their simplicity, lower costs and
reliance on the existing systems of energy statistics needed for development of a country's
energy balance. On the other hand, these indicators do not consider individual energy
efficiency measures and their impact nor do they show cause and effect relationships
between measures and their resulting energy savings. Developing such indicators requires
huge amount of data (not only energy statistics, but whole set of macro and microeconomic
data that are influencing energy consumption in all end-use sectors is needed), and data
availability and reliability are often questionable in practice, sometimes leading to the huge
need for modelling and expert judgement to overcome the lack of data. Nevertheless,
energy efficiency indicators are inevitable part of energy efficiency evaluation process (both
ex-ante and ex-post) as they are the only means to benchmark own performance against the
performance of others, to reveal the potentials and help determine policy targets, to quantify
the success/failure of the policy instruments and to track down the progress made in
achieving the defined targets.

5.3. Bottom-up M&V methods
A bottom-up M&V method means that energy consumption reductions obtained through
the implementation of a specific energy efficiency improvement measure are measured in
kilowatt-hours (kWh), in Joules (J) or in kilogram oil equivalent (kgoe) and added to energy
savings results from other specific energy efficiency improvement measures to obtain an
overall impact. The bottom-up M&V methods are oriented towards evaluation of individual
measures and are rarely used solely to perform evaluation of overall energy efficiency
policy impacts. However, they should be used whenever possible to provide more details on
performance of energy efficiency improvement measures. Bottom-up methods include
mathematical models (formulas) that are specific for every measure, so only the principle of
their definition will be briefly explained hereafter.

M&V approach boils down to the fact that the absence of energy use can be only determined
by comparing measurements of energy use made before (baseline) and after (post-retrofit)
implementation of energy efficiency measure or expressed in a simple equation:

Energy Savings = Baseline Energy Use - Post-Retrofit Energy Use ± Adjustments (2)

The baseline conditions can change after the energy efficiency measures are installed and
the term "Adjustments" (can be positive or negative) in equation (2) aiming at bringing
energy use in the two time periods (before and after) to the same set of conditions.
Conditions commonly affecting energy use are weather, occupancy, plant throughput, and
equipment operations required by these conditions. These factors must be taken into
account and analysed after measure is undertaken and adjustments have to be made in
order to ensure correct comparisons of the state pre- and post-retrofit. This kind of M&V
scheme (often referred to as ex-post) may be very costly but they guarantee the detections of
real savings. The costs are related to the actual measurement, i.e. to the measurement
equipment. To avoid a large increase in the M&V costs, only the largest or unpredictable
measures should be analysed through this methodology.
Individual energy efficiency projects might also be evaluated using well reasoned
estimations of individual energy efficiency improvement measures impacts. This approach
(ex-ante) means that certain type of energy efficiency measure is awarded with a certain
amount of energy savings prior to its actual realisation. This approach has significantly
lower costs and is especially appropriate for replicable measures, for which one can agree on
a reasonable estimate. There are also some "hybrid" solutions that combine ex-ante and ex-
post approaches in bottom-up M&V. This hybrid approach is often referred to as
parameterised ex-ante method. It applies to measures for which energy savings are known but
they may differ depending on a number of restricted factors (e.g. availability factor or
number of working hours). The set up of a hybrid approach can be more accurate than a
pure ex-ante methodology, without a substantial increase of the M&V costs.

5.4. Establishing evaluation procedures supported by M&V
The success of national energy efficiency policy has to be constantly monitored and its
impact evaluated. Findings of evaluation process shall be used to redesign policies and
enable their higher effectiveness. Regardless to its importance, policy evaluation is often
highly neglected. Policy documents are often adopted by governments and parliaments and
afterwards there is no interest for impacts they have produced. Therefore, setting up the
fully operable system for evaluation of energy efficiency is a complex process, which
requires structural and practice changes among main stakeholders in policy making.
Additionally, it has to be supported by M&V procedures, which require comprehensive data
collection and analysis systems to develop energy efficiency indicators that will quantify
policy effects.

6. Conclusion
Evidently, energy efficiency policy making is not one-time job. It is a continuous, dynamic
process that should create enabling conditions for energy efficiency market as complex
Energy Effciency Policy 23

efficiency indicators, as they are easier to monitor, often with a more rapid updating.
They aim at improving the interpretation of trends observed on the energy efficiency
indicators.
5. Adjusted energy efficiency indicators – account for differences existing among
countries in the climate, in economic structures or in technologies. Comparisons of
energy efficiency performance across countries are only meaningful if they are based
on such indicators. External factors that might influence energy consumption
include: (a) weather conditions, such as degree days; (b) occupancy levels; (c)
opening hours for non-domestic buildings; (d) installed equipment intensity (plant
throughput); product mix; (e) plant throughput, level of production, volume or
added value, including changes in GDP level; (f) schedules for installation and
vehicles; (g) relationship with other units. Some of these factors are relevant for
correction of aggregated indicators, while some are to be used for the individual
facilities in which energy efficiency measures are implemented.
6. Target indicators – aim at providing reference values to show possible target of
energy efficiency improvements or energy efficiency potentials for a given country.
They are somehow similar to benchmark value but defined at a macro level, which
implies a careful interpretation of differences. The target is defined as the distance to
the average of the 3 best countries; this distance shows what gain can be achieved.
The main advantages of the usage of top-down methods is their simplicity, lower costs and
reliance on the existing systems of energy statistics needed for development of a country's
energy balance. On the other hand, these indicators do not consider individual energy
efficiency measures and their impact nor do they show cause and effect relationships
between measures and their resulting energy savings. Developing such indicators requires
huge amount of data (not only energy statistics, but whole set of macro and microeconomic
data that are influencing energy consumption in all end-use sectors is needed), and data
availability and reliability are often questionable in practice, sometimes leading to the huge
need for modelling and expert judgement to overcome the lack of data. Nevertheless,
energy efficiency indicators are inevitable part of energy efficiency evaluation process (both
ex-ante and ex-post) as they are the only means to benchmark own performance against the
performance of others, to reveal the potentials and help determine policy targets, to quantify
the success/failure of the policy instruments and to track down the progress made in
achieving the defined targets.

5.3. Bottom-up M&V methods
A bottom-up M&V method means that energy consumption reductions obtained through
the implementation of a specific energy efficiency improvement measure are measured in
kilowatt-hours (kWh), in Joules (J) or in kilogram oil equivalent (kgoe) and added to energy
savings results from other specific energy efficiency improvement measures to obtain an
overall impact. The bottom-up M&V methods are oriented towards evaluation of individual
measures and are rarely used solely to perform evaluation of overall energy efficiency
policy impacts. However, they should be used whenever possible to provide more details on
performance of energy efficiency improvement measures. Bottom-up methods include
mathematical models (formulas) that are specific for every measure, so only the principle of
their definition will be briefly explained hereafter.

M&V approach boils down to the fact that the absence of energy use can be only determined
by comparing measurements of energy use made before (baseline) and after (post-retrofit)
implementation of energy efficiency measure or expressed in a simple equation:

Energy Savings = Baseline Energy Use - Post-Retrofit Energy Use ± Adjustments (2)

The baseline conditions can change after the energy efficiency measures are installed and
the term "Adjustments" (can be positive or negative) in equation (2) aiming at bringing
energy use in the two time periods (before and after) to the same set of conditions.
Conditions commonly affecting energy use are weather, occupancy, plant throughput, and
equipment operations required by these conditions. These factors must be taken into
account and analysed after measure is undertaken and adjustments have to be made in
order to ensure correct comparisons of the state pre- and post-retrofit. This kind of M&V
scheme (often referred to as ex-post) may be very costly but they guarantee the detections of
real savings. The costs are related to the actual measurement, i.e. to the measurement
equipment. To avoid a large increase in the M&V costs, only the largest or unpredictable
measures should be analysed through this methodology.
Individual energy efficiency projects might also be evaluated using well reasoned
estimations of individual energy efficiency improvement measures impacts. This approach
(ex-ante) means that certain type of energy efficiency measure is awarded with a certain
amount of energy savings prior to its actual realisation. This approach has significantly
lower costs and is especially appropriate for replicable measures, for which one can agree on
a reasonable estimate. There are also some "hybrid" solutions that combine ex-ante and ex-
post approaches in bottom-up M&V. This hybrid approach is often referred to as
parameterised ex-ante method. It applies to measures for which energy savings are known but
they may differ depending on a number of restricted factors (e.g. availability factor or
number of working hours). The set up of a hybrid approach can be more accurate than a
pure ex-ante methodology, without a substantial increase of the M&V costs.

5.4. Establishing evaluation procedures supported by M&V
The success of national energy efficiency policy has to be constantly monitored and its
impact evaluated. Findings of evaluation process shall be used to redesign policies and
enable their higher effectiveness. Regardless to its importance, policy evaluation is often
highly neglected. Policy documents are often adopted by governments and parliaments and
afterwards there is no interest for impacts they have produced. Therefore, setting up the
fully operable system for evaluation of energy efficiency is a complex process, which
requires structural and practice changes among main stakeholders in policy making.
Additionally, it has to be supported by M&V procedures, which require comprehensive data
collection and analysis systems to develop energy efficiency indicators that will quantify
policy effects.

6. Conclusion
Evidently, energy efficiency policy making is not one-time job. It is a continuous, dynamic
process that should create enabling conditions for energy efficiency market as complex
Energy Effciency 24

system of supply-demand interactions undergoing evolutionary change and direct that
change toward efficiency, environmental benefits and social well-being. However, there are
number of barriers preventing optimal functioning of energy efficiency market, which
should determine the choice of policy instruments. Policy instruments have to be flexible
and able to respond (adapt) to the market requirements in order to achieve goals in the
optimal manner, i.e. to the least cost for the society. Due to fast changing market conditions,
Policy instruments can no longer be documents once produced and then intact for several
years. Continuous policy evaluation process has to become a usual. Future research work to
support policy making shall be exactly directed towards elaboration of methodology that
will be able to qualitatively and quantitatively evaluate effectiveness and cost-effectiveness
of policy instruments and enable selection of optimal policy instruments mix depending on
current development stage of the energy efficiency market.
Evaluation procedures will advance and deepen our knowledge on success or failure factors
of energy efficiency policy. The analysis of current situation shows that policies world-wide
tend to fail in delivering desired targets in terms of energy consumption reduction. The
main reason lies in the lack of understanding and focus on implementing adequate
capacities, which are far too underdeveloped, insufficient and inappropriate for ambitious
goals that have to be achieved. It has to be understood that policy implementation will not
just happen by it self, and that capacities and capabilities in all society structures are needed.
Embracing full-scale energy management systems in both public service and business sector
can make the difference. Additionally, with the positive pressure from civil society
organisations and media, understanding the interdependences of energy and climate change
issues will improve, gradually changing the society's mindset towards higher efficiency, and
eventually towards the change of lifestyle.

7. References
Morvaj, Z. & Bukarica, V. (2010). Immediate challenge of combating climate change:
effective implementation of energy efficiency policies, paper accepted for 21
st
World
Energy Congress, 12-16 September, Montreal, 2010
Morvaj, Z. & Gvozdenac, D.(2008). Applied Industrial Energy and Environmental Management,
John Wiley and Sons - IEEE press, ISBN: 978-0-470-69742-9, UK
Dennis, K. (2006). The Compatibility of Economic Theory and Proactive Energy Efficiency
Policy. The Electricity Journal, Vol. 19, Issue 7, (August/September 2006) 58-73,
ISSN: 1040-6190
European Commission. (2006). Action Plan for Energy Efficiency COM(2006)545 final, Brussels
Eurostat. (2009). Energy, transport and environment indicators, Office for Official Publications
of the European Communities, ISBN 978-92-79-09835-2, Luxembourg
European Environment Agency. (2009).Annual European Community greenhouse gas inventory
1990–2007 and inventory report 2009, Office for Official Publications of the European
Communities, ISBN 978-92-9167-980-5, Copenhagen
European Commission. (2009). Draft Communication from the Commission to the Council and the
European Parliament: 7 Measures for 2 Million New EU Jobs: Low Carbon Eco Efficient &
Cleaner Economy for European Citizens, Brussels

Bukarica, V.; Morvaj, Z. & Tomšić, Ž. (2007). Evaluation of Energy Efficiency Policy
Instruments Effectiveness – Case Study Croatia, Proceedings of IASTED International
conference “Power and Energy Systems 2007”, ISBN: 978-0-88986-689-8, Palma de
Mallorca, August, 2007, The International Association of Science and Technology
for Development
Briner, S. & Martinot, E. (2005). Promoting energy-efficient products: GEF experience and
lessons for market transformation in developing countries. Energy Policy, 33 (2005)
1765-1779, ISSN: 0301-4215
Vine, E. (2008). Strategies and policies for improving energy efficiency programs: Closing
the loop between evaluation and implementation. Energy Policy, 36 (2008) 3872–
3881, ISSN: 0301-4215
Bulmstein, C.; Goldstone, S. & Lutzenhiser, L. (2000). A theory-based approach to market
transformation, Energy Policy, 28 (2000) 137-144, ISSN: 0301-4215
Paskaleva, K. (2009). Enabling the smart city: The progress of e-city governance in Europe.
International Journal of Innovation and Regional Development, 1 (January 2009) 405–
422(18), ISSN 1753-0660
Stanislaw, J.A. (2008). Climate Changes Everything: The Dawn of the Green Economy, Delloite
Development LCC, USA
Morvaj, Z. et al. (2008). Energy management in cities: learning through change, Proceedings of
11
th
EURA conference, Learning Cities in a Knowledge based Societies, 9-11 October 2008,
Milan
Joosen, S. & Harmelink, M. (2006). Guidelines for the ex-post evaluation of 20 energy efficiency
instruments applied across Europe, publication published within AID-EE project
supported by Intelligent Energy Europe programme.
Energy Effciency Policy 25

system of supply-demand interactions undergoing evolutionary change and direct that
change toward efficiency, environmental benefits and social well-being. However, there are
number of barriers preventing optimal functioning of energy efficiency market, which
should determine the choice of policy instruments. Policy instruments have to be flexible
and able to respond (adapt) to the market requirements in order to achieve goals in the
optimal manner, i.e. to the least cost for the society. Due to fast changing market conditions,
Policy instruments can no longer be documents once produced and then intact for several
years. Continuous policy evaluation process has to become a usual. Future research work to
support policy making shall be exactly directed towards elaboration of methodology that
will be able to qualitatively and quantitatively evaluate effectiveness and cost-effectiveness
of policy instruments and enable selection of optimal policy instruments mix depending on
current development stage of the energy efficiency market.
Evaluation procedures will advance and deepen our knowledge on success or failure factors
of energy efficiency policy. The analysis of current situation shows that policies world-wide
tend to fail in delivering desired targets in terms of energy consumption reduction. The
main reason lies in the lack of understanding and focus on implementing adequate
capacities, which are far too underdeveloped, insufficient and inappropriate for ambitious
goals that have to be achieved. It has to be understood that policy implementation will not
just happen by it self, and that capacities and capabilities in all society structures are needed.
Embracing full-scale energy management systems in both public service and business sector
can make the difference. Additionally, with the positive pressure from civil society
organisations and media, understanding the interdependences of energy and climate change
issues will improve, gradually changing the society's mindset towards higher efficiency, and
eventually towards the change of lifestyle.

7. References
Morvaj, Z. & Bukarica, V. (2010). Immediate challenge of combating climate change:
effective implementation of energy efficiency policies, paper accepted for 21
st
World
Energy Congress, 12-16 September, Montreal, 2010
Morvaj, Z. & Gvozdenac, D.(2008). Applied Industrial Energy and Environmental Management,
John Wiley and Sons - IEEE press, ISBN: 978-0-470-69742-9, UK
Dennis, K. (2006). The Compatibility of Economic Theory and Proactive Energy Efficiency
Policy. The Electricity Journal, Vol. 19, Issue 7, (August/September 2006) 58-73,
ISSN: 1040-6190
European Commission. (2006). Action Plan for Energy Efficiency COM(2006)545 final, Brussels
Eurostat. (2009). Energy, transport and environment indicators, Office for Official Publications
of the European Communities, ISBN 978-92-79-09835-2, Luxembourg
European Environment Agency. (2009).Annual European Community greenhouse gas inventory
1990–2007 and inventory report 2009, Office for Official Publications of the European
Communities, ISBN 978-92-9167-980-5, Copenhagen
European Commission. (2009). Draft Communication from the Commission to the Council and the
European Parliament: 7 Measures for 2 Million New EU Jobs: Low Carbon Eco Efficient &
Cleaner Economy for European Citizens, Brussels

Bukarica, V.; Morvaj, Z. & Tomšić, Ž. (2007). Evaluation of Energy Efficiency Policy
Instruments Effectiveness – Case Study Croatia, Proceedings of IASTED International
conference “Power and Energy Systems 2007”, ISBN: 978-0-88986-689-8, Palma de
Mallorca, August, 2007, The International Association of Science and Technology
for Development
Briner, S. & Martinot, E. (2005). Promoting energy-efficient products: GEF experience and
lessons for market transformation in developing countries. Energy Policy, 33 (2005)
1765-1779, ISSN: 0301-4215
Vine, E. (2008). Strategies and policies for improving energy efficiency programs: Closing
the loop between evaluation and implementation. Energy Policy, 36 (2008) 3872–
3881, ISSN: 0301-4215
Bulmstein, C.; Goldstone, S. & Lutzenhiser, L. (2000). A theory-based approach to market
transformation, Energy Policy, 28 (2000) 137-144, ISSN: 0301-4215
Paskaleva, K. (2009). Enabling the smart city: The progress of e-city governance in Europe.
International Journal of Innovation and Regional Development, 1 (January 2009) 405–
422(18), ISSN 1753-0660
Stanislaw, J.A. (2008). Climate Changes Everything: The Dawn of the Green Economy, Delloite
Development LCC, USA
Morvaj, Z. et al. (2008). Energy management in cities: learning through change, Proceedings of
11
th
EURA conference, Learning Cities in a Knowledge based Societies, 9-11 October 2008,
Milan
Joosen, S. & Harmelink, M. (2006). Guidelines for the ex-post evaluation of 20 energy efficiency
instruments applied across Europe, publication published within AID-EE project
supported by Intelligent Energy Europe programme.
Energy Effciency 26
Energy growth, complexity and effciency 27
Energy growth, complexity and effciency
Franco Ruzzenenti and Riccardo Basosi
x

Energy growth, complexity and efficiency

Franco Ruzzenenti* and Riccardo Basosi*°
*Center for the Studies of Complex Systems, University of Siena
°Department of Chemistry, University of Siena
Italy

1. Introduction
Over the last two centuries, the human capacity to harness energy or transform heat into
work, has dramatically improved. Since the first steam engine appeared in Great Britain, the
first order thermodynamic efficiency (the rate of useful work over the heat released by the
energy source) has soared from a mere 1 % to the 40 % of present engines, up to the 70% of
the most recent power plants. Despite this efficiency revolution, energy consumption per
capita has always increased (Banks, 2007).
The economy and society have undeniably faced an expanding frontier, and both household
and global energy intensities have commonly been linked to economic growth and social
progress. The rising issue of energy conservation has prompted us to consider energy
efficiency as more than merely a characteristic of economic growth, but also as a cause
(Ayres and Warr, 2004). We thus wonder if it is possible to increase efficiency, reduce global
energy consumption, and foster economic development within an energy decreasing
pattern, by separating efficiency and energy growth. In other words, by reducing efficiency
positive feed-backs on the system’s energy level (Alcott, 2008).
In 1865, the economist Stanley Jevons was the first to point out the existence of a circular
causal process linking energy efficiency, energy use, and the economic system. Jevons was
convinced that efficiency was a driving force of energy growth and highlighted the risk
associated with an energy conservation policy thoroughly committed to efficiency
1
.
Recently the Jevon’s paradox has been approached in the field of Economics and termed
“rebound effect”. It has been the subject of articles, research, as well as a great deal of
controversy over the last two decades (Schipper, 2000). Although many economists are still
sceptical as to its actual relevance, most of them have agreed on the existence and
importance of such an effect. Some are deeply concerned (Khazzoum, 1980, Brookes 1990,

1
“It is very commonly urged, that the failing supply of coal will be met by new modes of using it
efficiently and economically. The amount of useful work got out of coal may be made to increase
manifold, while the amount of coal consumed is stationary or diminishing. We have thus, it is
supposed, the means of completely neutralizing the eveils of scarce and costly fuel. But the
economy of coal in manufacturing is a different matter. It is a wholly confusion of ideas to suppose that
the economical use of fuel is equivalent to a diminished consumption. The very contrary is the truth (Jevons,
1965).”
2
Energy Effciency 28
Saunders, 2000, Herring, 2006) about the overall net effect and its capacity to counterbalance
the gains due to efficiency. Others, however, still believe in the net benefit of energy policies
focused on developing energy efficiency, although they admit the burden of having to pay a
loss of savings (Shipper and Haas, 1998; Washida, 2004; Grepperud & Rasmussen, 2003).
The most accurate and simple definition of rebound effect is: a measure of the difference
between projected and actual savings due to increased efficiency (Sorrell and
Dimitropoulos, 2007).
Three different kinds of rebound effects are now widely used and accepted(Greening and
Greene,1997):
1. Direct effects: those directly linked to consumer behaviour in response to the more
advantageous cost of the service provided. They depend on changes in the final
energy use of appliances, devices or vehicles (i.e. if my car is more efficient, I drive
longer).
2. Indirect effects: those related to shifts in purchasing choices of customers, either
dependent on income effects or substitution effects, which have an ultimate impact
on other energy services (i.e. new generation engines are economical, then I buy a
bigger car or I spend the money saved for an air conditioner).
3. General equilibrium effects: changes in market demands as well as in relative costs
of productive inputs that ultimately have a deep impact in the productive
structure, possibly affecting the employment of energy as a productive factor (i.e.
the well known substitution of capital to labour, subsequent to a rise of labour
costs, is otherwise an increase of the energy intensity of the system. Labour cost
may increases relative to a subsidiary process that employs more energy to run).
The above classification displays the circular feedback process’s (increasing) time lag
scheme, beginning with a quick response, the altered use of energy devices due to changes
in energy costs, followed by a slower mechanism, changes in purchasing choices, and
finally, the long term restructuring process affecting economic factors. While direct and
indirect effects have found considerable attention in the literature, general equilibrium
effects remain relatively unexplored due to the uneasiness of their time scale and the variety
of involved variables (Binswanger, 2001)
2
.

2. The economic approach to the rebound effect
However paradoxical the rebound effect may seem, it can be explained by classic economic
theory. Energy is a derived demand because it is not the actual good purchased, but a
means by which a good or a service is enjoyed. Thus, technology that is able to reduce the
amount of energy employed by good or service lowers the cost of that item. It is said that
efficiency improvements reduce the implicit price of energy services and, according to the
basic theory of market demands, the amount of goods consumed rises when prices decrease.
Happy with this explanation, economic theory focused on measuring and forecasting the
rebound effect. Both econometric models and neoclassical forcasting models have been

2
“Third, changes in the prices of firms’ outputs and changes in the demand for inputs caused by
income and substitution effects will propagate throughout the economy and result in adjustments
of supply and demand in all sectors, resulting in general equilibrium effects. By taking care of the
income effect, we also include the indirect rebound effect in our analysis, but we still neglect
general equilibrium effects (Binswanger, 2001).
developed that exhibit sound results, except for the third kind of effect, that unfortunately
presents many features unfit for these models (Saunders, 1992; Greening and Greene, 1997;
Binswanger, 2001; Sorrell, 2009).
Forecasting models are mainly based on Cobb-Douglas production functions, with three
factors of production (capital, labour, energy), and which derive market demands for these
factors. Since the first attempts, calculations confirmed the existence of the effect under the
assumption of constant energy prices (Saunders, 1992). Econometric based research also
verified the relevance of the rebound effect and further provided valid measures of the
effect in a variety of economic sectors. Such measures mainly utilize the relative elasticities
of demand curves. Demand curves are built on statistical regressions in prices and
quantities of goods, while elasticity is a measure of the sensitivity in demand to the
variation of a good’s price. Although these models may be accurate, they are all single good
or service designed and are consequently viable only for the detection of direct effects.
Other models based on substitution elasticities between goods or factors as well as income
elasticities have addressed indirect effects (Greening and Greene, 1997). Such contributions
brought the level of detection to a whole sector of an economy or to a variety of aspects
related to the process of substitution highlighted in the rebound effect like the role of time-
saving technologies and their impact on energy intensities (Bentzen, 2004; Binswanger,
2001). Nevertheless, very few attempts have been made to evaluate general equilibrium
effects, a task which entails the recognition of the main connecting variables of an economy,
spread over a long period of time. These contributions, however, fail to describe and explain
major structural changes in the productive systems that cause discontinuity in the economic
relations among variables. All these models are, in fact, based on a stationary framework,
and therefore neglect evolutionary changes that heighten the developing pattern of an
economy (Dimitropoulos, 2007).
As a result of being the first who introduced the paradox behind the development of
efficiency, Jevons’ work has to be considered a landmark in this matter, for he was able to
trace a line that goes beyond the mere economical, or the implicit price mechanism,
explanation. He thought that any technological improvement rendering the energy source
more economical would stimulate the demand for energy. Furthermore Jevons had some
advanced and valuable intuitions about the role of energy sources in the economic
development, as well as about the dynamic between technology, energy and the economy
that were too often neglected by modern economists. His contributions are summarized as
follows:
1. Fuel efficiency affects market size and shape, and not just a process of substitution
among factors. He noticed that both time scale and space scale of travels changed
with engine technologies making new markets or new places reachable
3
.
2. Features of energy sources other than efficiency are relevant for economic purposes
like energy intensity and time disposal (power). He argued that what made steam

3
Such structural changes are unfit for common, wide spread modeling approaches. Is noteworthy
that when Jevons was developing his analysis, consumer theory was far to come and main sectors
were those of steal, mining and machinery industries. Economy was chiefly engaged in building his
back bone and changes at any rate were basically structural. His view of economic processes was
consequentially affected by that turmoil and can be considered, to a certain extent, evolutionary.
Shipper has raised the attention on structural changes, which are, according to his opinion, hardly
detectable but very important in energy demand long term pattern (Shipper and Grubb, 2000).
Energy growth, complexity and effciency 29
Saunders, 2000, Herring, 2006) about the overall net effect and its capacity to counterbalance
the gains due to efficiency. Others, however, still believe in the net benefit of energy policies
focused on developing energy efficiency, although they admit the burden of having to pay a
loss of savings (Shipper and Haas, 1998; Washida, 2004; Grepperud & Rasmussen, 2003).
The most accurate and simple definition of rebound effect is: a measure of the difference
between projected and actual savings due to increased efficiency (Sorrell and
Dimitropoulos, 2007).
Three different kinds of rebound effects are now widely used and accepted(Greening and
Greene,1997):
1. Direct effects: those directly linked to consumer behaviour in response to the more
advantageous cost of the service provided. They depend on changes in the final
energy use of appliances, devices or vehicles (i.e. if my car is more efficient, I drive
longer).
2. Indirect effects: those related to shifts in purchasing choices of customers, either
dependent on income effects or substitution effects, which have an ultimate impact
on other energy services (i.e. new generation engines are economical, then I buy a
bigger car or I spend the money saved for an air conditioner).
3. General equilibrium effects: changes in market demands as well as in relative costs
of productive inputs that ultimately have a deep impact in the productive
structure, possibly affecting the employment of energy as a productive factor (i.e.
the well known substitution of capital to labour, subsequent to a rise of labour
costs, is otherwise an increase of the energy intensity of the system. Labour cost
may increases relative to a subsidiary process that employs more energy to run).
The above classification displays the circular feedback process’s (increasing) time lag
scheme, beginning with a quick response, the altered use of energy devices due to changes
in energy costs, followed by a slower mechanism, changes in purchasing choices, and
finally, the long term restructuring process affecting economic factors. While direct and
indirect effects have found considerable attention in the literature, general equilibrium
effects remain relatively unexplored due to the uneasiness of their time scale and the variety
of involved variables (Binswanger, 2001)
2
.

2. The economic approach to the rebound effect
However paradoxical the rebound effect may seem, it can be explained by classic economic
theory. Energy is a derived demand because it is not the actual good purchased, but a
means by which a good or a service is enjoyed. Thus, technology that is able to reduce the
amount of energy employed by good or service lowers the cost of that item. It is said that
efficiency improvements reduce the implicit price of energy services and, according to the
basic theory of market demands, the amount of goods consumed rises when prices decrease.
Happy with this explanation, economic theory focused on measuring and forecasting the
rebound effect. Both econometric models and neoclassical forcasting models have been

2
“Third, changes in the prices of firms’ outputs and changes in the demand for inputs caused by
income and substitution effects will propagate throughout the economy and result in adjustments
of supply and demand in all sectors, resulting in general equilibrium effects. By taking care of the
income effect, we also include the indirect rebound effect in our analysis, but we still neglect
general equilibrium effects (Binswanger, 2001).
developed that exhibit sound results, except for the third kind of effect, that unfortunately
presents many features unfit for these models (Saunders, 1992; Greening and Greene, 1997;
Binswanger, 2001; Sorrell, 2009).
Forecasting models are mainly based on Cobb-Douglas production functions, with three
factors of production (capital, labour, energy), and which derive market demands for these
factors. Since the first attempts, calculations confirmed the existence of the effect under the
assumption of constant energy prices (Saunders, 1992). Econometric based research also
verified the relevance of the rebound effect and further provided valid measures of the
effect in a variety of economic sectors. Such measures mainly utilize the relative elasticities
of demand curves. Demand curves are built on statistical regressions in prices and
quantities of goods, while elasticity is a measure of the sensitivity in demand to the
variation of a good’s price. Although these models may be accurate, they are all single good
or service designed and are consequently viable only for the detection of direct effects.
Other models based on substitution elasticities between goods or factors as well as income
elasticities have addressed indirect effects (Greening and Greene, 1997). Such contributions
brought the level of detection to a whole sector of an economy or to a variety of aspects
related to the process of substitution highlighted in the rebound effect like the role of time-
saving technologies and their impact on energy intensities (Bentzen, 2004; Binswanger,
2001). Nevertheless, very few attempts have been made to evaluate general equilibrium
effects, a task which entails the recognition of the main connecting variables of an economy,
spread over a long period of time. These contributions, however, fail to describe and explain
major structural changes in the productive systems that cause discontinuity in the economic
relations among variables. All these models are, in fact, based on a stationary framework,
and therefore neglect evolutionary changes that heighten the developing pattern of an
economy (Dimitropoulos, 2007).
As a result of being the first who introduced the paradox behind the development of
efficiency, Jevons’ work has to be considered a landmark in this matter, for he was able to
trace a line that goes beyond the mere economical, or the implicit price mechanism,
explanation. He thought that any technological improvement rendering the energy source
more economical would stimulate the demand for energy. Furthermore Jevons had some
advanced and valuable intuitions about the role of energy sources in the economic
development, as well as about the dynamic between technology, energy and the economy
that were too often neglected by modern economists. His contributions are summarized as
follows:
1. Fuel efficiency affects market size and shape, and not just a process of substitution
among factors. He noticed that both time scale and space scale of travels changed
with engine technologies making new markets or new places reachable
3
.
2. Features of energy sources other than efficiency are relevant for economic purposes
like energy intensity and time disposal (power). He argued that what made steam

3
Such structural changes are unfit for common, wide spread modeling approaches. Is noteworthy
that when Jevons was developing his analysis, consumer theory was far to come and main sectors
were those of steal, mining and machinery industries. Economy was chiefly engaged in building his
back bone and changes at any rate were basically structural. His view of economic processes was
consequentially affected by that turmoil and can be considered, to a certain extent, evolutionary.
Shipper has raised the attention on structural changes, which are, according to his opinion, hardly
detectable but very important in energy demand long term pattern (Shipper and Grubb, 2000).
Energy Effciency 30
vessels more economical was neither fuel efficiency (wind power is more efficient)
nor unit costs (wind vessels are almost costless), but instead the availability and
disposal of coal as an energy source which had an incomparable positive impact on
the capital return cycle.
3. A sink or a flux of free energy becomes an energy source when there is an
exploiting technology and an economic need forward. He argues that from the
beginning onward, a developing process of energy sources has a fundamental role
as an economic driving force and not vice versa. In other words, when economic
needs are compelling, technology development is significantly accelerated and as a
result, feeds back to the whole economic system.
4. Prosperity is dependent on economical energy sources, and economic development
is mainly shaped by energy sources and its quantity
4
. However pessimistic we may
consider this statement, Jevons meant to call for an economical austerity in order to
prevent society form a hard landing due to the running out of low cost coal
5
. He
claimed that was more recommendable a stationary economy together with social
progress.
What we can therefore gain from his teachings is that there is an inner tendency of an
economy to render energy sources more economical and that this is the true driving force of
economic development
6
.
Thus, for Jevons, societal development—civilization—is “the economy of power” or the
constant strain on humanity of harnessing energy in a productive way, and its “history is a
history of successive steps of economy (energy efficiency, n.d.r.).” The incremental process

4
“We may observe, in the first place, that almost all the arts practiced in England before the middle
of the eighteenth century were of continental origin. England, until lately, was young and inferior in
the arts. Secondly, we may observe that by far the grater part of arts and inventions we have of late
contributed, spring from our command of coal, or at any rate depend upon its profuse
consumption” (Jevons, 1965).
5
A misleading, wide spread, opinion is that Jevons skepticism was misjudged and the rising age of
oil gave proof of it; but he clearly foresaw the drawbacks of such a solution: “Petroleum has, of late
years, become the matter of a most extensive trade, and has been found admirably adapted for use
in marine steam-engine boilers. It is undoubtedly superior to coal for many purposes, and is
capable of replacing it. But then, What is Petroleum but Essence of Coal, distilled from it by terrestrial
or artificial heat? Its natural supply is far more limited and uncertain than of coal, and an artificial
supply can only be had by the distillation of some kind of coal at considerable cost. To extend the
use of petroleum, then, is only a new way of pushing the consumption of coal. It is more likely to be
an aggravation of the drain then a remedy.”
6
“The steam-engine is the motive power of this country, and its history is a history of successive
steps of economy. But every such improvement of the engine, when effected, does but accelerate
anew the consumption of coal. Every branch of manufacture receives a fresh impulse-hand labour is
still further replaced by mechanical labour, and greatly extended works can be undertaken which
were not commercially possible by the use of the more costly steam-power. But no one must
suppose that coal thus saved is spared –it is only saved from one use to be employed in others, and
the profits gained soon lead to extended employment in many new forms. The several branches of
industry are closely interdependent, and the progress of any one leads to the progress of nearly all.
And if economy in the past has been the main source of our progress and growing consumption of
coal, the same effect will follow from the same cause in the future.”

of energy efficiency drives more and more energy into the system, but how does it occur?
Jevons, in the following passage, provides insight into such a controversial question:
Again, the quantity consumed by each individual is a composite quantity, increased either
by multiplying the scale of former applications of coal, or finding new applications. We
cannot, indeed, always be doubling the length of our railways, the magnitude of our ships,
and bridges, and factories. In every kind of enterprise we shall no doubt meet a natural limit
of convenience, or commercial practicability, as we do in the cultivation of land. I do not
mean a fixed and impassible limit, but as it were an elastic limit, which we may push
against a little further, but ever with increasing difficulty. But the new applications of coal
are of an unlimited character (Jevons, 1965).

3. Complexity and Efficiency
Jevons believed that the natural tendency of economy is to expand linearly, “multiplying the
scale of former applications,” up to a limit and then, to overcome such limits, the system
works within itself to develop “new applications”. Sketched roughly, the scheme here is:
growth-saturation-innovation-growth.
Jevons found an unsuspected counterpart in a famous biologist, Alfred Lotka, who was
interested in the relation between energy and evolution. Indeed there are several analogies
among their theories. Lotka too believed in the need for looking synoptically at the
biological system in order to understand the energetics of evolution. Lotka also shares
Jevons’ cyclic view of processes, which, in the case of energy “transformers,” he understood
to be formed by an alternation growth-limit to growth- evolution- growth
7
. According to
Lotka, the reason why this process was doomed to an ever growing amount of energy flow
boiled down to the cross action of selection-evolution on the one hand and the
thermodynamics law on the other. In his opinion, evolution is the result of a stochastic
process and a selective pressure, and moreover, “the life contest is primarily competition for
available free energy.” Thus, selection rewards those species adapted to thrive on a
particular substrate, and the growth of such species will divert an increasing quantity of free
energy into the biological system. Those species' growth will proceed until the free energy
available for that transformation process is completely exploited. The dual action of case
and selection will then favor new transformers more efficient in employing the free energy
still available. The developmental stages of ecological succession mirror this evolutionary
energetic pattern. In the first stage of ecological succession, plant pioneering species
dominate, growing rapidly, but inefficiently disposing of resources. In the climax stage,

7
“But in detail the engine is infinitely complex, and the main cycle contains within its self a maze of
subsidiary cycles. And, since the parts of the engine are all interrelated, it may happen that the
output of the great wheel is limited, or at least hampered, by the performance of one or more of the
wheels within the wheel. For it must be remembered that the output of each transformer is
determined both by its mass and by its rate of revolution. Hence if the working substance, or any
ingredient of the working substance of any of the subsidiary transformers, reaches its limits, a limit
may at the same time be set for the performance of the great transformer as a whole. Conversely, if
any one of the subsidiary transformers develops new activity, either by acquiring new resources of
working substance, or by accelerating its rate of revolution, the output of the entire system may be
reflexly stimulated
Energy growth, complexity and effciency 31
vessels more economical was neither fuel efficiency (wind power is more efficient)
nor unit costs (wind vessels are almost costless), but instead the availability and
disposal of coal as an energy source which had an incomparable positive impact on
the capital return cycle.
3. A sink or a flux of free energy becomes an energy source when there is an
exploiting technology and an economic need forward. He argues that from the
beginning onward, a developing process of energy sources has a fundamental role
as an economic driving force and not vice versa. In other words, when economic
needs are compelling, technology development is significantly accelerated and as a
result, feeds back to the whole economic system.
4. Prosperity is dependent on economical energy sources, and economic development
is mainly shaped by energy sources and its quantity
4
. However pessimistic we may
consider this statement, Jevons meant to call for an economical austerity in order to
prevent society form a hard landing due to the running out of low cost coal
5
. He
claimed that was more recommendable a stationary economy together with social
progress.
What we can therefore gain from his teachings is that there is an inner tendency of an
economy to render energy sources more economical and that this is the true driving force of
economic development
6
.
Thus, for Jevons, societal development—civilization—is “the economy of power” or the
constant strain on humanity of harnessing energy in a productive way, and its “history is a
history of successive steps of economy (energy efficiency, n.d.r.).” The incremental process

4
“We may observe, in the first place, that almost all the arts practiced in England before the middle
of the eighteenth century were of continental origin. England, until lately, was young and inferior in
the arts. Secondly, we may observe that by far the grater part of arts and inventions we have of late
contributed, spring from our command of coal, or at any rate depend upon its profuse
consumption” (Jevons, 1965).
5
A misleading, wide spread, opinion is that Jevons skepticism was misjudged and the rising age of
oil gave proof of it; but he clearly foresaw the drawbacks of such a solution: “Petroleum has, of late
years, become the matter of a most extensive trade, and has been found admirably adapted for use
in marine steam-engine boilers. It is undoubtedly superior to coal for many purposes, and is
capable of replacing it. But then, What is Petroleum but Essence of Coal, distilled from it by terrestrial
or artificial heat? Its natural supply is far more limited and uncertain than of coal, and an artificial
supply can only be had by the distillation of some kind of coal at considerable cost. To extend the
use of petroleum, then, is only a new way of pushing the consumption of coal. It is more likely to be
an aggravation of the drain then a remedy.”
6
“The steam-engine is the motive power of this country, and its history is a history of successive
steps of economy. But every such improvement of the engine, when effected, does but accelerate
anew the consumption of coal. Every branch of manufacture receives a fresh impulse-hand labour is
still further replaced by mechanical labour, and greatly extended works can be undertaken which
were not commercially possible by the use of the more costly steam-power. But no one must
suppose that coal thus saved is spared –it is only saved from one use to be employed in others, and
the profits gained soon lead to extended employment in many new forms. The several branches of
industry are closely interdependent, and the progress of any one leads to the progress of nearly all.
And if economy in the past has been the main source of our progress and growing consumption of
coal, the same effect will follow from the same cause in the future.”

of energy efficiency drives more and more energy into the system, but how does it occur?
Jevons, in the following passage, provides insight into such a controversial question:
Again, the quantity consumed by each individual is a composite quantity, increased either
by multiplying the scale of former applications of coal, or finding new applications. We
cannot, indeed, always be doubling the length of our railways, the magnitude of our ships,
and bridges, and factories. In every kind of enterprise we shall no doubt meet a natural limit
of convenience, or commercial practicability, as we do in the cultivation of land. I do not
mean a fixed and impassible limit, but as it were an elastic limit, which we may push
against a little further, but ever with increasing difficulty. But the new applications of coal
are of an unlimited character (Jevons, 1965).

3. Complexity and Efficiency
Jevons believed that the natural tendency of economy is to expand linearly, “multiplying the
scale of former applications,” up to a limit and then, to overcome such limits, the system
works within itself to develop “new applications”. Sketched roughly, the scheme here is:
growth-saturation-innovation-growth.
Jevons found an unsuspected counterpart in a famous biologist, Alfred Lotka, who was
interested in the relation between energy and evolution. Indeed there are several analogies
among their theories. Lotka too believed in the need for looking synoptically at the
biological system in order to understand the energetics of evolution. Lotka also shares
Jevons’ cyclic view of processes, which, in the case of energy “transformers,” he understood
to be formed by an alternation growth-limit to growth- evolution- growth
7
. According to
Lotka, the reason why this process was doomed to an ever growing amount of energy flow
boiled down to the cross action of selection-evolution on the one hand and the
thermodynamics law on the other. In his opinion, evolution is the result of a stochastic
process and a selective pressure, and moreover, “the life contest is primarily competition for
available free energy.” Thus, selection rewards those species adapted to thrive on a
particular substrate, and the growth of such species will divert an increasing quantity of free
energy into the biological system. Those species' growth will proceed until the free energy
available for that transformation process is completely exploited. The dual action of case
and selection will then favor new transformers more efficient in employing the free energy
still available. The developmental stages of ecological succession mirror this evolutionary
energetic pattern. In the first stage of ecological succession, plant pioneering species
dominate, growing rapidly, but inefficiently disposing of resources. In the climax stage,

7
“But in detail the engine is infinitely complex, and the main cycle contains within its self a maze of
subsidiary cycles. And, since the parts of the engine are all interrelated, it may happen that the
output of the great wheel is limited, or at least hampered, by the performance of one or more of the
wheels within the wheel. For it must be remembered that the output of each transformer is
determined both by its mass and by its rate of revolution. Hence if the working substance, or any
ingredient of the working substance of any of the subsidiary transformers, reaches its limits, a limit
may at the same time be set for the performance of the great transformer as a whole. Conversely, if
any one of the subsidiary transformers develops new activity, either by acquiring new resources of
working substance, or by accelerating its rate of revolution, the output of the entire system may be
reflexly stimulated
Energy Effciency 32
however, the most efficient species in converting resources prevail (Odum, 1997). The
following passage stresses this key concept:
This at least seems probable, that so long as there is abundant surplus of available energy
running “to waste” over the sides of the mill wheel, so to speak, so long will a marked
advantage be gained by any species that may develop talents to utilize this “lost portion of
the stream”. Such a species will therefore, other things equal, tend to grow in extent
(numbers) and its growth will further increase the flux of energy through the system. It is to
be observed that in this argument the principle of the survival of the fittest yields us
information beyond that attainable by the reasoning of thermodynamics. As to the other
aspect of the matter, the problem of economy in husbanding resources will not rise to its full
importance until the available resources are more completely tapped than they are today.
Every indication is that man will learn to utilize some of the sunlight that now goes to waste
(Lotka, 1956).
Economy and biology are both evolutionary systems and both can be approached from
thermodynamics. By contrast, not all analogies are suitable. Whilst less efficient transformers
like bacteria persist together with more evolved vertebrates, hence biosphere makes
manifest the entire evolutionary path, economy dismisses obsolete technologies (we don’t
see any more steam motive engines around). So, if we abandon inefficient technologies, why
isn’t the net effect over consumptions negative? In other words why, if we employ more
efficient devices, energy use doesn’t drop? History has so far proved that more efficiency
results in more energy consumption. Where does this paradox come from?
Is this paradox due to the counteractive effect of population or affluence growth over
efficiency or is efficiency evolution the driving factor of economic growth? We will here
attempt to show how the causality chain initiate with an efficiency improvement and that
growth comes after. Growth featured by those changes affecting the economic system
comparable to “new applications of unlimited character” mentioned by Jevons or an
“acceleration to the revolution rate of the world engine” envisioned by Lotka.
What it is being argued here is that all those changes, or among them, those affecting the
structure or delivering brand new technologies into the system, may be regarded as a leap
of complexity occurring to the system. Complexity, in the acceptation of organizational
complexity, if it was observed as a feature of whatsoever of a system, has always displayed
a high energy density rate. This means that growing complexity implies growing energy
consumption. That is to say, a more complex system consumes more (more connections,
more variety, more hierarchical levels). It is therefore possible that the energy saved by new
and more efficient processes is absorbed or perhaps a better word, dissipated, by a more
complex system. Energy savings resulting from increased efficiency would then be offset by
an organization restructuring process within the system.

4. Evolutionary Pattern
We have advanced the hypothesis of the existence of a common, recursive pattern in
evolutionary systems. This pattern underlies a broad, complex thermodynamic process
involving the entire system and arises from forces embedded within the system. We have
described this pattern as the following circular process: growth-saturation-complexity leap-
growth and can be depicted it as a circular process.

Fig. 1. Evolutionary Pattern

The growth stage relies on the presence of inner forces that drive the system to expand
while seeking survival and reproduction. These forces are species (the genome) in the
domain of biology, and firms (the capital) in the economy. Although it is clear how these
autocatalytic processes cause the system’s expansion, it is less clear how, coupled with
efficiency improvements, they can divert more energy into the system or in the words of
Lotka, “maximize the energy flow.” It must be kept in mind that neither Lotka nor Jevons
claims that the overflow of energy is the actual aim of system components. It is rather a
result of their interaction with each other and with the environment. Lotka, for example,
believes that two main thermodynamic strategies are adopted by organisms in order to
adapt to the environment: maximizing output (power maximum) and minimizing input
(efficiency maximum). The former is developed by species thriving in resource abundance
and the latter by organisms struggling in scarcity conditions. According to Lotka, by
pursuing unexploited free-energy more energy is driven through the system thus
maximizing global output. The dichotomy between efficiency and power is therefore quite
apparent
8
.
And there is indeed something well founded in this revelation, which is rooted in
thermodynamics. The antagonism between efficiency and power is less evident from a
thermodynamics perspective, meaning that if other factors are left unchanged, an efficiency
improvement always leads to empowerment. The misunderstanding and thereby the
paradox of efficiency comes from two major misconceptions, which can be outlined as
follows:
 Thermodynamic efficiency, from the Carnot Engine onward, concerns the
conversion of heat into work, not just the mere transformation of one form of
energy to another.
 Efficiency, as a rate between output and input or benefits and costs, pertains to a
static analysis despite the fact that the conversion process actually takes place in
time and therefore costs and benefits also depend on the time elapsed.

8
There is a simplification of Lotka’s vision of the energetics of evolution that states that two
strategies would top evolutionary thermodynamics: one that maximizes work over time (power) in
the case of resource abundance and another that minimizes energy consumed per for amount of
work delivered (efficiency) in the case of scarcity. These two strategies have been summarized in
the “maximum power principle,” despite Lotka himself being reluctant to adopt any lofty and
ambitious term like “principle” for his thinking. Moreover, in this formulation, scarcity and
abundance are unrelated whatsoever to magnitude, while Lotka clearly stresses what scarcity must
be compared to: the ability of a transformer to get hold of free energy and its growing rate. What
are indisputably scarce or plenty are nutrients, row materials or water, which eventually affect
energy efficiency.
Energy growth, complexity and effciency 33
however, the most efficient species in converting resources prevail (Odum, 1997). The
following passage stresses this key concept:
This at least seems probable, that so long as there is abundant surplus of available energy
running “to waste” over the sides of the mill wheel, so to speak, so long will a marked
advantage be gained by any species that may develop talents to utilize this “lost portion of
the stream”. Such a species will therefore, other things equal, tend to grow in extent
(numbers) and its growth will further increase the flux of energy through the system. It is to
be observed that in this argument the principle of the survival of the fittest yields us
information beyond that attainable by the reasoning of thermodynamics. As to the other
aspect of the matter, the problem of economy in husbanding resources will not rise to its full
importance until the available resources are more completely tapped than they are today.
Every indication is that man will learn to utilize some of the sunlight that now goes to waste
(Lotka, 1956).
Economy and biology are both evolutionary systems and both can be approached from
thermodynamics. By contrast, not all analogies are suitable. Whilst less efficient transformers
like bacteria persist together with more evolved vertebrates, hence biosphere makes
manifest the entire evolutionary path, economy dismisses obsolete technologies (we don’t
see any more steam motive engines around). So, if we abandon inefficient technologies, why
isn’t the net effect over consumptions negative? In other words why, if we employ more
efficient devices, energy use doesn’t drop? History has so far proved that more efficiency
results in more energy consumption. Where does this paradox come from?
Is this paradox due to the counteractive effect of population or affluence growth over
efficiency or is efficiency evolution the driving factor of economic growth? We will here
attempt to show how the causality chain initiate with an efficiency improvement and that
growth comes after. Growth featured by those changes affecting the economic system
comparable to “new applications of unlimited character” mentioned by Jevons or an
“acceleration to the revolution rate of the world engine” envisioned by Lotka.
What it is being argued here is that all those changes, or among them, those affecting the
structure or delivering brand new technologies into the system, may be regarded as a leap
of complexity occurring to the system. Complexity, in the acceptation of organizational
complexity, if it was observed as a feature of whatsoever of a system, has always displayed
a high energy density rate. This means that growing complexity implies growing energy
consumption. That is to say, a more complex system consumes more (more connections,
more variety, more hierarchical levels). It is therefore possible that the energy saved by new
and more efficient processes is absorbed or perhaps a better word, dissipated, by a more
complex system. Energy savings resulting from increased efficiency would then be offset by
an organization restructuring process within the system.

4. Evolutionary Pattern
We have advanced the hypothesis of the existence of a common, recursive pattern in
evolutionary systems. This pattern underlies a broad, complex thermodynamic process
involving the entire system and arises from forces embedded within the system. We have
described this pattern as the following circular process: growth-saturation-complexity leap-
growth and can be depicted it as a circular process.

Fig. 1. Evolutionary Pattern

The growth stage relies on the presence of inner forces that drive the system to expand
while seeking survival and reproduction. These forces are species (the genome) in the
domain of biology, and firms (the capital) in the economy. Although it is clear how these
autocatalytic processes cause the system’s expansion, it is less clear how, coupled with
efficiency improvements, they can divert more energy into the system or in the words of
Lotka, “maximize the energy flow.” It must be kept in mind that neither Lotka nor Jevons
claims that the overflow of energy is the actual aim of system components. It is rather a
result of their interaction with each other and with the environment. Lotka, for example,
believes that two main thermodynamic strategies are adopted by organisms in order to
adapt to the environment: maximizing output (power maximum) and minimizing input
(efficiency maximum). The former is developed by species thriving in resource abundance
and the latter by organisms struggling in scarcity conditions. According to Lotka, by
pursuing unexploited free-energy more energy is driven through the system thus
maximizing global output. The dichotomy between efficiency and power is therefore quite
apparent
8
.
And there is indeed something well founded in this revelation, which is rooted in
thermodynamics. The antagonism between efficiency and power is less evident from a
thermodynamics perspective, meaning that if other factors are left unchanged, an efficiency
improvement always leads to empowerment. The misunderstanding and thereby the
paradox of efficiency comes from two major misconceptions, which can be outlined as
follows:
 Thermodynamic efficiency, from the Carnot Engine onward, concerns the
conversion of heat into work, not just the mere transformation of one form of
energy to another.
 Efficiency, as a rate between output and input or benefits and costs, pertains to a
static analysis despite the fact that the conversion process actually takes place in
time and therefore costs and benefits also depend on the time elapsed.

8
There is a simplification of Lotka’s vision of the energetics of evolution that states that two
strategies would top evolutionary thermodynamics: one that maximizes work over time (power) in
the case of resource abundance and another that minimizes energy consumed per for amount of
work delivered (efficiency) in the case of scarcity. These two strategies have been summarized in
the “maximum power principle,” despite Lotka himself being reluctant to adopt any lofty and
ambitious term like “principle” for his thinking. Moreover, in this formulation, scarcity and
abundance are unrelated whatsoever to magnitude, while Lotka clearly stresses what scarcity must
be compared to: the ability of a transformer to get hold of free energy and its growing rate. What
are indisputably scarce or plenty are nutrients, row materials or water, which eventually affect
energy efficiency.
Energy Effciency 34
The first statement assumes the custom of considering conversion rates, such as the
transformation of chemical energy into heat, as thermodynamic efficiencies. As previously
noted, most of the controversies surrounding the rebound effect in the residential sector
arise from the misleading concept of efficiency. The rate of transformation of chemical
energy into heat in e.g. a bomb calorimeter is a calorie while out of the laboratory, it is a
thermal efficiency, and should not be considered a thermodynamic efficiency because no
work is involved
9
. The theoretical apparatus we have so far employed is therefore
inapplicable. Only work needs an entropy change into the (work) reservoir in order to be
dissipated while a heat sink is of unlimited disposal to the environment. In other words, the
system’s structure needs to change in order to dissipate (more) mechanical work, but not the
same can be said for heat. This kind of efficiency, known as thermal efficiency, has much
more to do with squandering. When a process becomes more thermodynamically efficient,
more work is extracted from the same amount of energy (heat) and when it becomes more
thermally efficient, less heat for our purpose is wasted from a heat source.

4.1 The Time Variable Determines the Efficiency Level
In the second statement, the attention is focused on a theoretical aspect that needs a formal
treatment to be understood. It is indeed very difficult to intuitively sense that, in physics
terms, a system that improves its efficiency also enhances its power. It is even more difficult
to see how this can be true if a trade off exists between power maximization and efficiency
optimization. A system that maximizes its efficiency actually minimizes its power and vice
versa. Thus, if we improve the efficiency, we increase the power. Nevertheless, if we seek
the best efficiency, we have to set the minimum power output. Is this a paradox? In a sense,
yes, but only if our analysis is oblivious to the passage of time.
We have formulated two assertions in apparent contradiction. The first is that when
thermodynamic efficiency improves, power increases. This direct relationship is evident by
observing the definitions of efficiency and power:


h
W
η=
Q
,
W
P =
Δt


(1)
As long as the specific consumption—the rate at which the energy source is depleted—
remains constant, the power increases. It is noteworthy that this relationship strictly relates
to the capacity of the system to draw from a particular source. The capacity depends on the
specific consumption:

∂Q
h
∂t
(2)
The specific consumption is the rate of depletion of the energy source or the amount of
input (fuel) per the unit of time. It reflects the capacity of the system to convey energy

9
Thermodynamic efficiency concerns the transformation of heat into work. Other non-thermodynamic
efficiencies are, for example, heat transport and heat regulation or the cinematic chain.
Nevertheless, any kind of efficiency can contribute to the overall thermodynamic efficiency, when a
work output is obtained out of heat.
throughout the process.The second assertion that there exists a trade-off between efficiency
and power needs more mathematics to be explained. It will be illustrated by means of a
Carnot Cycle, revisited with the addition of the time variable. In the Carnot Cycle, to
achieve the maximum efficiency, the isothermal expansion and compression (Figure 2), need
to occur at an infinitely slow speed in order to maintain an infinitesimal temperature
gradient between the working substance (T
hw,
T
cw
) and the heat reservoirs (T
h
, T
c
). Under
these circumstances, the power of the machine approaches zero since it takes infinite time to
produce a finite amount of work. To speed up the process, we need to increase the gradient
since the heat transfer rate is proportional to it. To thereby get more than an infinitesimal
amount of power from a Carnot Engine, we have to keep the temperature of its working
substance below that of the hot reservoir and above that of the cold reservoir.


Fig. 2. Carnot Cycle

The more we increase the two gradients, the closer the extreme temperatures of the working
substance. Ultimately, the two isothermal stages take place with no change in the
temperature of the working substance. Heat flows directly from the hot source to the cold
sink and no work is done. Hence the power output is zero and the engine has zero efficiency
as well. In this model, we consider a Carnot Engine with a working substance absorbing
heat from the hot source at T
hw
and releasing heat to the cold source at T
cw
. Under most
circumstances, the rates of heat transfer will be proportional to the temperature gradients.
We assume the constant of proportionality (K –meaning that heat absorption/release occurs
in the same conditions) and the same ∆t for the expansion and the compression
10
. We also
assume that the two adiabatic transformations remain unaltered. We now have the
following equations describing the once isothermal processes:

 
h
h hw
1
Q
= k T T
Δt

(3)

10
These assumptions can be abandoned without changing the results of the model, see Curzon and
Ahlborn (Curzon and Ahlborn, 1975).
Energy growth, complexity and effciency 35
The first statement assumes the custom of considering conversion rates, such as the
transformation of chemical energy into heat, as thermodynamic efficiencies. As previously
noted, most of the controversies surrounding the rebound effect in the residential sector
arise from the misleading concept of efficiency. The rate of transformation of chemical
energy into heat in e.g. a bomb calorimeter is a calorie while out of the laboratory, it is a
thermal efficiency, and should not be considered a thermodynamic efficiency because no
work is involved
9
. The theoretical apparatus we have so far employed is therefore
inapplicable. Only work needs an entropy change into the (work) reservoir in order to be
dissipated while a heat sink is of unlimited disposal to the environment. In other words, the
system’s structure needs to change in order to dissipate (more) mechanical work, but not the
same can be said for heat. This kind of efficiency, known as thermal efficiency, has much
more to do with squandering. When a process becomes more thermodynamically efficient,
more work is extracted from the same amount of energy (heat) and when it becomes more
thermally efficient, less heat for our purpose is wasted from a heat source.

4.1 The Time Variable Determines the Efficiency Level
In the second statement, the attention is focused on a theoretical aspect that needs a formal
treatment to be understood. It is indeed very difficult to intuitively sense that, in physics
terms, a system that improves its efficiency also enhances its power. It is even more difficult
to see how this can be true if a trade off exists between power maximization and efficiency
optimization. A system that maximizes its efficiency actually minimizes its power and vice
versa. Thus, if we improve the efficiency, we increase the power. Nevertheless, if we seek
the best efficiency, we have to set the minimum power output. Is this a paradox? In a sense,
yes, but only if our analysis is oblivious to the passage of time.
We have formulated two assertions in apparent contradiction. The first is that when
thermodynamic efficiency improves, power increases. This direct relationship is evident by
observing the definitions of efficiency and power:


h
W
η=
Q
,
W
P =
Δt


(1)
As long as the specific consumption—the rate at which the energy source is depleted—
remains constant, the power increases. It is noteworthy that this relationship strictly relates
to the capacity of the system to draw from a particular source. The capacity depends on the
specific consumption:

∂Q
h
∂t
(2)
The specific consumption is the rate of depletion of the energy source or the amount of
input (fuel) per the unit of time. It reflects the capacity of the system to convey energy

9
Thermodynamic efficiency concerns the transformation of heat into work. Other non-thermodynamic
efficiencies are, for example, heat transport and heat regulation or the cinematic chain.
Nevertheless, any kind of efficiency can contribute to the overall thermodynamic efficiency, when a
work output is obtained out of heat.
throughout the process.The second assertion that there exists a trade-off between efficiency
and power needs more mathematics to be explained. It will be illustrated by means of a
Carnot Cycle, revisited with the addition of the time variable. In the Carnot Cycle, to
achieve the maximum efficiency, the isothermal expansion and compression (Figure 2), need
to occur at an infinitely slow speed in order to maintain an infinitesimal temperature
gradient between the working substance (T
hw,
T
cw
) and the heat reservoirs (T
h
, T
c
). Under
these circumstances, the power of the machine approaches zero since it takes infinite time to
produce a finite amount of work. To speed up the process, we need to increase the gradient
since the heat transfer rate is proportional to it. To thereby get more than an infinitesimal
amount of power from a Carnot Engine, we have to keep the temperature of its working
substance below that of the hot reservoir and above that of the cold reservoir.


Fig. 2. Carnot Cycle

The more we increase the two gradients, the closer the extreme temperatures of the working
substance. Ultimately, the two isothermal stages take place with no change in the
temperature of the working substance. Heat flows directly from the hot source to the cold
sink and no work is done. Hence the power output is zero and the engine has zero efficiency
as well. In this model, we consider a Carnot Engine with a working substance absorbing
heat from the hot source at T
hw
and releasing heat to the cold source at T
cw
. Under most
circumstances, the rates of heat transfer will be proportional to the temperature gradients.
We assume the constant of proportionality (K –meaning that heat absorption/release occurs
in the same conditions) and the same ∆t for the expansion and the compression
10
. We also
assume that the two adiabatic transformations remain unaltered. We now have the
following equations describing the once isothermal processes:

 
h
h hw
1
Q
= k T T
Δt

(3)

10
These assumptions can be abandoned without changing the results of the model, see Curzon and
Ahlborn (Curzon and Ahlborn, 1975).
Energy Effciency 36
c
cw c
2
Q
= k(T T )
Δt
÷
(4)
T
h
= temperature of the hot source, T
c
=temperature of the cold source, T
hw
=max temperature
of the working fluid, T
cw
=min temperature of the working fluid
Since the remaining two processes are adiabatic, they follow the relation (5):

Q
h
T
hw
=
Q
c
T
cw

(5)
The power of the system will be defined in equation (6):

P=
W
2Δt

(6)
W=Q
h
− Q
c
,
Δt
1
=Δt
2

(7)
The maximization of the power, as a function of T
hw
, the hotter working temperature, will
give the following result for the optimum power output:

( )
1
2
hw
h h cw
T = T + T T
(8)

at a corresponding efficiency of
1
c
h
T
η=
T
÷
(9)

It will be useful to do a variables’ substitution to depict the trade off so we now fix
x=T
cw
/T
hw
. According to this model, the efficiency-power trade off can be sketched as
function of x and whereby Carnot efficiency will be represented by curve (10) and power
output curve (11):

η=1− x

(10)
4
c
c h h
T k
P = T +T T x
x
| |
÷ ÷
|
\ .

(11)
The two curves can be plot in a graph, assuming T
h
and T
c
of 300 and 25 degree Celsius;
and fixing k at 0.05 (Fig.3). To reach the maximum theoretical efficiency (η for the isothermal
transformation) the system must approach thermal equilibrium and therefore maximum
slowness. Since it arises from power maximization, the optimal output will be somewhere
between theoretical maximum efficiency and zero efficiency and it will only be determined
by the sources’ temperatures (T
h
and T
c
). So for every boundary condition in a Carnot
Cycle, there is a single optimal value of output. Even if we abandon most of the abstract
assumptions about the Carnot Cycle thus introducing further irreversibility, the peak of the
curve will probably shift, but the trade off is unavoidable. We have to set the engine at
either maximum efficiency or maximum power. “However, when the cost of building an
engine is much greater than the cost of fuel (as is often the case), it is desirable to optimize
the engine for maximum power output, not maximum efficiency (Schroeder, 2000).”


Fig. 3. Power-efficiency trade off

The power maximization will lead to sub-optimal efficiency (with respect to Carnot
efficiency) which depends on sources’ temperatures with the explicit relation ( 9) while
Carnot efficiency is:
η
Carnot
= 1−
T
c
T
h

(12)

It is noteworthy that such an efficiency level seems to be much closer to the running
efficiency of most of energy converting sources than the Carnot efficiency (Table 1).

4.2 Efficiency improvement and power enhancement
We can further assume that efficiency improvements also apply to engine parts, in addition
to working temperatures
11
. Any technical improvement concerning the material employed

11
If we consider sources’ temperature changes, we return to the dominion of Carnot efficiency while if
we take into account working temperatures, we resort to the efficiency-power trade off sketched by the
model.
Energy growth, complexity and effciency 37
c
cw c
2
Q
= k(T T )
Δt
÷
(4)
T
h
= temperature of the hot source, T
c
=temperature of the cold source, T
hw
=max temperature
of the working fluid, T
cw
=min temperature of the working fluid
Since the remaining two processes are adiabatic, they follow the relation (5):

Q
h
T
hw
=
Q
c
T
cw

(5)
The power of the system will be defined in equation (6):

P=
W
2Δt

(6)
W=Q
h
− Q
c
,
Δt
1
=Δt
2

(7)
The maximization of the power, as a function of T
hw
, the hotter working temperature, will
give the following result for the optimum power output:

( )
1
2
hw
h h cw
T = T + T T
(8)

at a corresponding efficiency of
1
c
h
T
η=
T
÷
(9)

It will be useful to do a variables’ substitution to depict the trade off so we now fix
x=T
cw
/T
hw
. According to this model, the efficiency-power trade off can be sketched as
function of x and whereby Carnot efficiency will be represented by curve (10) and power
output curve (11):

η=1− x

(10)
4
c
c h h
T k
P = T +T T x
x
| |
÷ ÷
|
\ .

(11)
The two curves can be plot in a graph, assuming T
h
and T
c
of 300 and 25 degree Celsius;
and fixing k at 0.05 (Fig.3). To reach the maximum theoretical efficiency (η for the isothermal
transformation) the system must approach thermal equilibrium and therefore maximum
slowness. Since it arises from power maximization, the optimal output will be somewhere
between theoretical maximum efficiency and zero efficiency and it will only be determined
by the sources’ temperatures (T
h
and T
c
). So for every boundary condition in a Carnot
Cycle, there is a single optimal value of output. Even if we abandon most of the abstract
assumptions about the Carnot Cycle thus introducing further irreversibility, the peak of the
curve will probably shift, but the trade off is unavoidable. We have to set the engine at
either maximum efficiency or maximum power. “However, when the cost of building an
engine is much greater than the cost of fuel (as is often the case), it is desirable to optimize
the engine for maximum power output, not maximum efficiency (Schroeder, 2000).”


Fig. 3. Power-efficiency trade off

The power maximization will lead to sub-optimal efficiency (with respect to Carnot
efficiency) which depends on sources’ temperatures with the explicit relation ( 9) while
Carnot efficiency is:
η
Carnot
= 1−
T
c
T
h

(12)

It is noteworthy that such an efficiency level seems to be much closer to the running
efficiency of most of energy converting sources than the Carnot efficiency (Table 1).

4.2 Efficiency improvement and power enhancement
We can further assume that efficiency improvements also apply to engine parts, in addition
to working temperatures
11
. Any technical improvement concerning the material employed

11
If we consider sources’ temperature changes, we return to the dominion of Carnot efficiency while if
we take into account working temperatures, we resort to the efficiency-power trade off sketched by the
model.
Energy Effciency 38
or the reduction of friction would lead to a higher K and a better (faster) heat transfer across
the machinery
12
. Since K does not affect the output regulation (the maximum value is not
dependent on K), this will in turn, increase the rate of Q
h
and the power. According to the
value of the maximum power, it is clear that any increase in K (given T
h
and T
c
) will
augment the power by a factor of 1/4, shifting the peak upward. More efficiency will
therefore lead to higher power.
Suppose we want to increase efficiency as much as possible, leading the control parameter
K. We may push further K, in order to increase the heat transfer rate and get an higher
efficiency, but we will end up moving away from the theoretical maximum efficiency level,
toward an higher power output, as it is shown in the animation
13
.
Through this model, we have shown how the efficiency-power paradox is apparent and we
have also described the thermodynamic conditions of the efficiency-power trade off. We can
thus draw Lotka’s conceptual framework of power maximization versus efficiency
optimization in the context of the economy of power. It is, as we have already highlighted,
an economic optimization that leads to maximum output
14
. Whenever the cost of fuel is
relatively less constraining than the cost of machinery, power will be maximized.
Nevertheless, every efficiency improvement involving technological development will
probably lead to a more complex engine (or process) and therefore, will, on the one hand,
reduce the relative price of energy, but on the other, raise the cost of the apparatus. This will
ultimately amplify the bifurcation and positively feed back to the optimum power level
15
.

12
The paradoxical effect of increasing both efficiency and power can be easily understood if we think
energy as space integral and work as the time integral of force. A process that reduces energy input in
less time, increases power, as integrations over the same function are not independent. That is to say: if
we use less energy per unit of space, and unit of time, in the same amount space we will save energy,
yet in the same time lag we may use more energy!
13
Animation at: http//sciyo.com
14
Concepts of the like of “costs” and “economic optimization” should not be intended in a strict way.
Broadly speaking, costs are to be meant as thermodynamic cost.
15
The idea of sub-optimal efficiency level output was investigated in the filed of biological systems. As
early as the 1955, Hodum and Pinkerton (Hodum H.T., Pinkerton R., 1955) published an article in
which, adopting Lotka approach an vision for life’s energetics, tried to demonstrate that “natural
systems tend to operate at that efficiency which produces a maximum power output”. Such efficiency
was lower then the maximum attainable and, according to them, was exactly of 50%. “In natural
systems there is a general tendency to sacrifice efficiency for power output”. The idea of the 50% set
point was based on the finding that most of energy converting systems were featured by coupled
antagonists processes. “The essence of biochemical workings of an organism is the coupling of an
exergonic catabolism to an endergonic anabolism that results in growth, reproduction and
maintenance”. Although this paradigm may account partially or even totally, for the derivation of the
50% value of efficiency, it was a striking intuition. It is remarkable, for example, that in the former
model, whereas it is not so evident, there are two counteractive processes: the heat absorption an then
heat release. The heat disposal affect the power output as much as the heat intake, as we know
empirically, from electric power plants. Thus, this simple thermodynamic model resembles, by this
point of view, the “living systems” of Odum theory. Conversely, as already Odum didn’t fail to
mention later on (Odum, 1983), the article of Cuzon and Ahlborn of 1975, on which this model is based,
gave a sound evidence and a formal basis, to the postulate of the “maximum power principle”
Power Source Tc Th η (Carnot) η* (model) η (observed)
Coal Fired Steam Plant 25 565 64.10% 40% 36.00%
Nuclear Reactor 25 300 48.00% 28% 30%
Geothermal Steam Plant 80 250 32.30% 17,5% 16%
Table 1 Source: Cuzon and Ahlborn, 1975.

4.3 The Case of Trucks
The truck industry and therefore, the road freight transport sector, gives a useful example of
empowerment brought about by the efficiency improvement of the engine and vehicle
technologies. From the late 1970’s onward, efficiency rose steadily as an effect of technology
research that tried to overcome the effects of soaring energy costs. Initially, such
improvements were employed to reduce consumptions, but later technology development
partially addressed power enhancement. Energy efficiency, as measured by fuel economy -
distance travelled, at constant speed, for unit of fuel consumed, increased since late 1970’s to
late 1990’s of about 30%. However, if we rescale fuel economy to the power shift of engines
(adjusted fuel economy), we can observe a major change in efficiency (Ruzzenenti and
Basosi, 2009a). This is also evident from the comparison of two trends of fuel economy and
adjusted fuel economy (fuel economy divided by the engines’ power) for a sample of 97
different European heavy-duty trucks. Initially the two metrics are coupled and show how
efficiency was employed to reduce fuel consumption; we can see a dramatic drop in both
fuel economy and adjusted fuel economy. Later trends display a sharp bifurcation, from
mid 1990’s onward, that explains how efficiency was then employed to enhance power and
reduce consumptions (Figure 3)

Fig. 4. Efficiency and power bifurcation in European Truck Industry (Ruzzenenti and Basosi,
2009a)
Energy growth, complexity and effciency 39
or the reduction of friction would lead to a higher K and a better (faster) heat transfer across
the machinery
12
. Since K does not affect the output regulation (the maximum value is not
dependent on K), this will in turn, increase the rate of Q
h
and the power. According to the
value of the maximum power, it is clear that any increase in K (given T
h
and T
c
) will
augment the power by a factor of 1/4, shifting the peak upward. More efficiency will
therefore lead to higher power.
Suppose we want to increase efficiency as much as possible, leading the control parameter
K. We may push further K, in order to increase the heat transfer rate and get an higher
efficiency, but we will end up moving away from the theoretical maximum efficiency level,
toward an higher power output, as it is shown in the animation
13
.
Through this model, we have shown how the efficiency-power paradox is apparent and we
have also described the thermodynamic conditions of the efficiency-power trade off. We can
thus draw Lotka’s conceptual framework of power maximization versus efficiency
optimization in the context of the economy of power. It is, as we have already highlighted,
an economic optimization that leads to maximum output
14
. Whenever the cost of fuel is
relatively less constraining than the cost of machinery, power will be maximized.
Nevertheless, every efficiency improvement involving technological development will
probably lead to a more complex engine (or process) and therefore, will, on the one hand,
reduce the relative price of energy, but on the other, raise the cost of the apparatus. This will
ultimately amplify the bifurcation and positively feed back to the optimum power level
15
.

12
The paradoxical effect of increasing both efficiency and power can be easily understood if we think
energy as space integral and work as the time integral of force. A process that reduces energy input in
less time, increases power, as integrations over the same function are not independent. That is to say: if
we use less energy per unit of space, and unit of time, in the same amount space we will save energy,
yet in the same time lag we may use more energy!
13
Animation at: http//sciyo.com
14
Concepts of the like of “costs” and “economic optimization” should not be intended in a strict way.
Broadly speaking, costs are to be meant as thermodynamic cost.
15
The idea of sub-optimal efficiency level output was investigated in the filed of biological systems. As
early as the 1955, Hodum and Pinkerton (Hodum H.T., Pinkerton R., 1955) published an article in
which, adopting Lotka approach an vision for life’s energetics, tried to demonstrate that “natural
systems tend to operate at that efficiency which produces a maximum power output”. Such efficiency
was lower then the maximum attainable and, according to them, was exactly of 50%. “In natural
systems there is a general tendency to sacrifice efficiency for power output”. The idea of the 50% set
point was based on the finding that most of energy converting systems were featured by coupled
antagonists processes. “The essence of biochemical workings of an organism is the coupling of an
exergonic catabolism to an endergonic anabolism that results in growth, reproduction and
maintenance”. Although this paradigm may account partially or even totally, for the derivation of the
50% value of efficiency, it was a striking intuition. It is remarkable, for example, that in the former
model, whereas it is not so evident, there are two counteractive processes: the heat absorption an then
heat release. The heat disposal affect the power output as much as the heat intake, as we know
empirically, from electric power plants. Thus, this simple thermodynamic model resembles, by this
point of view, the “living systems” of Odum theory. Conversely, as already Odum didn’t fail to
mention later on (Odum, 1983), the article of Cuzon and Ahlborn of 1975, on which this model is based,
gave a sound evidence and a formal basis, to the postulate of the “maximum power principle”
Power Source Tc Th η (Carnot) η* (model) η (observed)
Coal Fired Steam Plant 25 565 64.10% 40% 36.00%
Nuclear Reactor 25 300 48.00% 28% 30%
Geothermal Steam Plant 80 250 32.30% 17,5% 16%
Table 1 Source: Cuzon and Ahlborn, 1975.

4.3 The Case of Trucks
The truck industry and therefore, the road freight transport sector, gives a useful example of
empowerment brought about by the efficiency improvement of the engine and vehicle
technologies. From the late 1970’s onward, efficiency rose steadily as an effect of technology
research that tried to overcome the effects of soaring energy costs. Initially, such
improvements were employed to reduce consumptions, but later technology development
partially addressed power enhancement. Energy efficiency, as measured by fuel economy -
distance travelled, at constant speed, for unit of fuel consumed, increased since late 1970’s to
late 1990’s of about 30%. However, if we rescale fuel economy to the power shift of engines
(adjusted fuel economy), we can observe a major change in efficiency (Ruzzenenti and
Basosi, 2009a). This is also evident from the comparison of two trends of fuel economy and
adjusted fuel economy (fuel economy divided by the engines’ power) for a sample of 97
different European heavy-duty trucks. Initially the two metrics are coupled and show how
efficiency was employed to reduce fuel consumption; we can see a dramatic drop in both
fuel economy and adjusted fuel economy. Later trends display a sharp bifurcation, from
mid 1990’s onward, that explains how efficiency was then employed to enhance power and
reduce consumptions (Figure 3)

Fig. 4. Efficiency and power bifurcation in European Truck Industry (Ruzzenenti and Basosi,
2009a)
Energy Effciency 40
5 Structural Complexity
The underlying hypothesis of this work is that higher complexity counterbalances, on a
global scale, the effects of higher efficiency on a process scale. It is our general
understanding of evolution that selection operates by reward complexity. More complex, in
the context of biology is often used as a synonym for fittest in terms of the competition for
resources. Technological advances also develop from less to more complex devices. The
meaning of complexity has never been questioned, for it has been evident in the semantic of
nature or progress since earliest observations. A eukaryotic cell is more complex than a
prokaryotic one and a Ferrari F1 is more complex than a Ford T. Under this perspective,
complexity is countable, if not measurable, by the number of different components, parts or
organs. If we abandon the conviction that progress always evolves toward higher
complexity, that is to say, if we relinquish the belief of an immanent trend in nature; or if
we are dealing with systems that differ in structure rather than in number of components,
how can we apply such well established knowledge?
It is beyond the goals of this analysis to establish what complexity is or how it should be
approached. The scientific community has been unable to establish or agree on a universal
definition or paradigm of complexity and any attempt to univocally measure complexity is
therefore doomed to failure
16
. It will be here assumed that a more complex system has a
higher energy density or in other words, consumes more energy per unit of mass and time. There
is a great deal of evidence, from biological records to cosmological entities, of such a
relationship and therefore we believe it can be considered a reasonable assumption.
Indeed, this strongly recursive pattern in nature –linking energy density and compolexity,
caused many scientists to think that energy could itself be considered a measure of
complexity

(Odum H.T., 1996; Odum E.P., 1997; Chaisson, 2001). Let’s assume a more
complex system consumes more (per unit of mass and time) and the complexity we refer to,
it is a structural or morphological complexity, as we are dealing with systems with undefined
boundaries and innumerable components, like the productive and transport systems. The
two main assumptions regarding complexity that we are concerned with are:
1. A more complex system consumes more energy per unit of mass and unit of time
(higher energy density rate)
2. Structural complexity primarily concerns the components’ organization
17
rather
than the components’ variety or number


16
“By ‘complexity’, we refer to the term intuitively as used in ordinary discourse, a definition culled
from many sources: ‘a state of intricacy, complication, variety, or involvement, as in the
interconnected parts of a structure-a quality of having many interacting, different components.’In
this work we shall come to identify complexity in two operational ways: as a measure of the
information needed to describe a system’s structure and function, or as a measure of the rate of
energy flowing through a system of given mass. No attempt is made here to be rigorous with the
words ‘order’, ‘organization’, ‘complexity’, and the like; this is not a work of classical philology or
linguistic gymnastics. Indeed, no two researchers seem able to agree on a precise, technical
definition of such a specious word as complexity, which may be context-dependent in any case
(Chaisson, 2001)”.
17
For organization we refer to any system’s components acting or arranged in a cooperative, systematic
fashion.
Further remarks attain the duality efficiency/complexity. We should bear in mind that
while we are referring to energy efficiency improvements, we are dealing with a process-
scale analysis, whilst the leap in complexity concerns the global-scale analysis. These
phenomena are at two different hierarchical levels:
1. Energy efficiency concerns energy converting processes and is therefore at the
components level of the system
2. Complexity (structural) concerns the organization of the system and is thus at the
global level of the system
We try to hereafter relate energy efficiency enhancement to complexity change. To
accomplish this, we have to detect changes in system organization that move in the
direction of higher complexity. Yet, changes in which system?
Since we have been dealing in the case study with truck efficiency, it would make sense to
refer to the freight road transport system, but that would be misleading because the goods
transport sector is merely a sub-component of the whole productive system. Transport
service is just a derivative demand, which means, in economics terminology, that someone
wants a good to be moved from one place to another. The shipment is the means, not the
goal. Our analysis therefore has to address the productive system in order to detect long
term changes in transport demand. Transport demand is derived from the needs of the
productive system. The transport system and the productive system, under the scope of
present analysis, are two parts of a whole.

5.1 Complexity leap: structural analysis
The main feature of the shift from a fordian to a post-fordian system concerns the location of
the productive chain. Formerly, the productive chain was set entirely in one site, to which
raw materials were delivered and from which products were shipped. From the 1970’s
onward, big firms began disassembling the production chain and redistributing it over
several scattered structures, belonging to the same company, or, more generally, belonging
to other international firms or local producers system. As a matter of fact, the productive
chain changed shape thereafter and it changed in such a fashion that the complexity of the
structure increased. It can be shown, by means of graph theory, that the post-fordian
structure increased in connectivity and path-cycles diversity across its nodes (Ruzzenenti
and Basosi, 2008b). Hence the post-fordian structure presents a higher degree of freedom
and thus relates to a more complex system. A system with a higher degree of freedom is a
more complex system in the sense that, as for any physical system, it has increased
multiplicity or number of different available states. In other words, a more complex system
has more ways to arrange the components, in this case goods or raw materials, and
therefore to dissipate energy.
According to the hypothesis here advocated, complexity increases when the system can
rearrange its components in such a manner that the number and the path length, or the
speed of interactions, will be augmented within the same boundaries
18
. That is to say,
complexity growth consists of an intensive rather than an extensive change, affecting the

18
It noteworthy that, in a network, the number and the path length must be considered
intensive features as they can grow without affecting the extension of the network, which is determined
eventually only by the number of nodes (components).
Energy growth, complexity and effciency 41
5 Structural Complexity
The underlying hypothesis of this work is that higher complexity counterbalances, on a
global scale, the effects of higher efficiency on a process scale. It is our general
understanding of evolution that selection operates by reward complexity. More complex, in
the context of biology is often used as a synonym for fittest in terms of the competition for
resources. Technological advances also develop from less to more complex devices. The
meaning of complexity has never been questioned, for it has been evident in the semantic of
nature or progress since earliest observations. A eukaryotic cell is more complex than a
prokaryotic one and a Ferrari F1 is more complex than a Ford T. Under this perspective,
complexity is countable, if not measurable, by the number of different components, parts or
organs. If we abandon the conviction that progress always evolves toward higher
complexity, that is to say, if we relinquish the belief of an immanent trend in nature; or if
we are dealing with systems that differ in structure rather than in number of components,
how can we apply such well established knowledge?
It is beyond the goals of this analysis to establish what complexity is or how it should be
approached. The scientific community has been unable to establish or agree on a universal
definition or paradigm of complexity and any attempt to univocally measure complexity is
therefore doomed to failure
16
. It will be here assumed that a more complex system has a
higher energy density or in other words, consumes more energy per unit of mass and time. There
is a great deal of evidence, from biological records to cosmological entities, of such a
relationship and therefore we believe it can be considered a reasonable assumption.
Indeed, this strongly recursive pattern in nature –linking energy density and compolexity,
caused many scientists to think that energy could itself be considered a measure of
complexity

(Odum H.T., 1996; Odum E.P., 1997; Chaisson, 2001). Let’s assume a more
complex system consumes more (per unit of mass and time) and the complexity we refer to,
it is a structural or morphological complexity, as we are dealing with systems with undefined
boundaries and innumerable components, like the productive and transport systems. The
two main assumptions regarding complexity that we are concerned with are:
1. A more complex system consumes more energy per unit of mass and unit of time
(higher energy density rate)
2. Structural complexity primarily concerns the components’ organization
17
rather
than the components’ variety or number


16
“By ‘complexity’, we refer to the term intuitively as used in ordinary discourse, a definition culled
from many sources: ‘a state of intricacy, complication, variety, or involvement, as in the
interconnected parts of a structure-a quality of having many interacting, different components.’In
this work we shall come to identify complexity in two operational ways: as a measure of the
information needed to describe a system’s structure and function, or as a measure of the rate of
energy flowing through a system of given mass. No attempt is made here to be rigorous with the
words ‘order’, ‘organization’, ‘complexity’, and the like; this is not a work of classical philology or
linguistic gymnastics. Indeed, no two researchers seem able to agree on a precise, technical
definition of such a specious word as complexity, which may be context-dependent in any case
(Chaisson, 2001)”.
17
For organization we refer to any system’s components acting or arranged in a cooperative, systematic
fashion.
Further remarks attain the duality efficiency/complexity. We should bear in mind that
while we are referring to energy efficiency improvements, we are dealing with a process-
scale analysis, whilst the leap in complexity concerns the global-scale analysis. These
phenomena are at two different hierarchical levels:
1. Energy efficiency concerns energy converting processes and is therefore at the
components level of the system
2. Complexity (structural) concerns the organization of the system and is thus at the
global level of the system
We try to hereafter relate energy efficiency enhancement to complexity change. To
accomplish this, we have to detect changes in system organization that move in the
direction of higher complexity. Yet, changes in which system?
Since we have been dealing in the case study with truck efficiency, it would make sense to
refer to the freight road transport system, but that would be misleading because the goods
transport sector is merely a sub-component of the whole productive system. Transport
service is just a derivative demand, which means, in economics terminology, that someone
wants a good to be moved from one place to another. The shipment is the means, not the
goal. Our analysis therefore has to address the productive system in order to detect long
term changes in transport demand. Transport demand is derived from the needs of the
productive system. The transport system and the productive system, under the scope of
present analysis, are two parts of a whole.

5.1 Complexity leap: structural analysis
The main feature of the shift from a fordian to a post-fordian system concerns the location of
the productive chain. Formerly, the productive chain was set entirely in one site, to which
raw materials were delivered and from which products were shipped. From the 1970’s
onward, big firms began disassembling the production chain and redistributing it over
several scattered structures, belonging to the same company, or, more generally, belonging
to other international firms or local producers system. As a matter of fact, the productive
chain changed shape thereafter and it changed in such a fashion that the complexity of the
structure increased. It can be shown, by means of graph theory, that the post-fordian
structure increased in connectivity and path-cycles diversity across its nodes (Ruzzenenti
and Basosi, 2008b). Hence the post-fordian structure presents a higher degree of freedom
and thus relates to a more complex system. A system with a higher degree of freedom is a
more complex system in the sense that, as for any physical system, it has increased
multiplicity or number of different available states. In other words, a more complex system
has more ways to arrange the components, in this case goods or raw materials, and
therefore to dissipate energy.
According to the hypothesis here advocated, complexity increases when the system can
rearrange its components in such a manner that the number and the path length, or the
speed of interactions, will be augmented within the same boundaries
18
. That is to say,
complexity growth consists of an intensive rather than an extensive change, affecting the

18
It noteworthy that, in a network, the number and the path length must be considered
intensive features as they can grow without affecting the extension of the network, which is determined
eventually only by the number of nodes (components).
Energy Effciency 42
internal structure of the system, which may be expressed by a new arrangement of system
components
19
. This hypothesis expresses view of complexity based on the concept of
geometry
20
.
In our opinion the network structure development, that eventually results in complexity
growth at any system level, is the outcome of forces (energy influx driven by autocatalytic
processes) in the context of hindering boundary conditions. It is the simple growth (in extent
and in number of components) the normal behavior. That is to say, without hindering
boundary conditions, the system expands its structure, qualitatively unaltered (spatial
growth). The system develops in a primary and spatial manner initially, then, when
saturation is reached, in a secondary and geometrical (structural) one. It is such geometrical
development that enables the system to increase its degree of freedom and to host more
energy (or energy density rate) within the same constraints. When this complexity change
emerges, the incoming structure, albeit already available to system components, becomes
now more probable. The boundary conditions ultimately determine the likelihood of the
new structure. It is therefore the role played by saturation in system’s growth that must be
addressed in order to understand the surge of complexity leap.
After the first oil crisis, worldwide industrial production dramatically decreased. There are
many clues, indeed, that industrial production at that time reached a saturation point.
Statistics show that between the early 1970’s and 1990’s, a revolution occurred in economic
and societal structure that might be considered the end of the industrial age (IEA, 1997).
Until that time, linear growth lasted for about 20 years and consisted of a shift in the active
population (which was itself growing) from agriculture to industry. It was the nature of the
“economic boom”, the linear “growth in extent and numbers” (Lotka, 1956). The birth rate
thereafter inverted its trend (also in relation to the average income) and population
employed in industry reached a maximum and started decreasing (Ruzzenenti and Basosi,
2009b). Industry received a dramatic set back and consequently began to explore new
strategies to reduce labour costs and regenerate production. The structural change we have
been describing thus far—the globalization and outsourcing revolution—took between 10
and 15 years to become established and influential. However, after the 1990’s, the growth
trend in the industrial sector resumed and the economy retrieved.

5.2 Degree of freedom reduction/increase
When analyzing the structural complexity change resulting from globalization, of
paramount importance is the shift from a uni-located, national productive chain to one that
is pluri-located and international. For those firms relying on external resources to pursue
their productive needs, production became less costly, but more subdued due to
uncontrollable factors. Part of its activity, formerly controlled managerially and internally,

19
An example in cells is represented by the internal skeleton of microtubules that increase the
speed of molecules across the cell compared to a transportation system based on simple diffusion. In
ecosystems, furthermore, there are food chains and predator-prey dynamics that represent another
“transportation network” over which matter flows faster.
20
According to Lotka, geometry is a prominent feature of thermodynamics of living systems and
thus, of a sort completely different from those normally addressed by equilibrium thermodynamics.
Whereas the latter mainly deals with “structure-less systems”, of the like of chemical coefficients, the
former must deals with the “geometrical features” of the system (Lotka, 1956).
was then focused on the free market. This shift reduced the stability of the firm and reduced
its degree of freedom (choices of allocation). After globalization, firms could explore labour
costs according to various national legislations and average incomes. The same occurred for
financial and fiscal conditions or the proximity to productive districts. The system (entire
market) could thereby reduce production costs by selecting where to set plants or rely on
suppliers. It is in this sense that globalization produced the rise of new spatial gradients in
the productive system. The whole system thereby increased its degree of freedom. We face
therefore the counteractive interplay of degree of freedom, on two hierarchical levels,
triggered by a saturation stage. To better stress how this interplay of degree of freedom,
working in opposite directions, can be caused by saturation, it is best to approach physical
systems for analogies.
For example, if we increase the pressure of a gas in a specific volume, we reduce its degree
of freedom and it consequently can become a liquid, at certain temperatures. At the same
time, when a liquid changes its motion regime, as in Benárd cells, from a pure, random
dissipative system (Figure 5A) to a global dissipative one (Figure 5B), which displays
features several magnitudes larger than molecules, a superstructure can arise that was
previously available, but very unlikely.


Fig. 5. Degree reduction/increase in dissipative structures (Ruzzenenti and Basosi, 2009b)

Gravity and viscosity constraints make such a structure, beyond a certain level of energy
input, possible. The Benárd cells phenomenon is indeed possible when the gradient
temperature and water level thickness are known, but not when the vessel permits the fluid
to dissipate heat in random motions. In other words, the boundary conditions together with
the pressure imposed upon the system by an increasing energy flux, changes the macrostate
(energy density) of the system by modifying its microstates (the molecular motion). The
random motion of molecules reflects one gradient, the temperature, which is not spatial
(geometrical), while the superstructure is exposed to the spatial gradient. That is to say,
while the first gradient is defined by one variable, the latter is described by three variables
and probabilities consequently change. Dissipation into one variable is therefore more
probable than onto three variables, unless boundary conditions render the former
impossible. In Benárd cells, such conditions are exemplified in Van der Waals forces, the
low heat capacity of water, and restrained vessel thickness (Chandrasekhar, 1961; Prigogine
and Stengers, 1984; Swenson, 1997). The connectivity recasts the same trade off in a network
system’s conceptual framework. A network system grows in complication as long as a new
Energy growth, complexity and effciency 43
internal structure of the system, which may be expressed by a new arrangement of system
components
19
. This hypothesis expresses view of complexity based on the concept of
geometry
20
.
In our opinion the network structure development, that eventually results in complexity
growth at any system level, is the outcome of forces (energy influx driven by autocatalytic
processes) in the context of hindering boundary conditions. It is the simple growth (in extent
and in number of components) the normal behavior. That is to say, without hindering
boundary conditions, the system expands its structure, qualitatively unaltered (spatial
growth). The system develops in a primary and spatial manner initially, then, when
saturation is reached, in a secondary and geometrical (structural) one. It is such geometrical
development that enables the system to increase its degree of freedom and to host more
energy (or energy density rate) within the same constraints. When this complexity change
emerges, the incoming structure, albeit already available to system components, becomes
now more probable. The boundary conditions ultimately determine the likelihood of the
new structure. It is therefore the role played by saturation in system’s growth that must be
addressed in order to understand the surge of complexity leap.
After the first oil crisis, worldwide industrial production dramatically decreased. There are
many clues, indeed, that industrial production at that time reached a saturation point.
Statistics show that between the early 1970’s and 1990’s, a revolution occurred in economic
and societal structure that might be considered the end of the industrial age (IEA, 1997).
Until that time, linear growth lasted for about 20 years and consisted of a shift in the active
population (which was itself growing) from agriculture to industry. It was the nature of the
“economic boom”, the linear “growth in extent and numbers” (Lotka, 1956). The birth rate
thereafter inverted its trend (also in relation to the average income) and population
employed in industry reached a maximum and started decreasing (Ruzzenenti and Basosi,
2009b). Industry received a dramatic set back and consequently began to explore new
strategies to reduce labour costs and regenerate production. The structural change we have
been describing thus far—the globalization and outsourcing revolution—took between 10
and 15 years to become established and influential. However, after the 1990’s, the growth
trend in the industrial sector resumed and the economy retrieved.

5.2 Degree of freedom reduction/increase
When analyzing the structural complexity change resulting from globalization, of
paramount importance is the shift from a uni-located, national productive chain to one that
is pluri-located and international. For those firms relying on external resources to pursue
their productive needs, production became less costly, but more subdued due to
uncontrollable factors. Part of its activity, formerly controlled managerially and internally,

19
An example in cells is represented by the internal skeleton of microtubules that increase the
speed of molecules across the cell compared to a transportation system based on simple diffusion. In
ecosystems, furthermore, there are food chains and predator-prey dynamics that represent another
“transportation network” over which matter flows faster.
20
According to Lotka, geometry is a prominent feature of thermodynamics of living systems and
thus, of a sort completely different from those normally addressed by equilibrium thermodynamics.
Whereas the latter mainly deals with “structure-less systems”, of the like of chemical coefficients, the
former must deals with the “geometrical features” of the system (Lotka, 1956).
was then focused on the free market. This shift reduced the stability of the firm and reduced
its degree of freedom (choices of allocation). After globalization, firms could explore labour
costs according to various national legislations and average incomes. The same occurred for
financial and fiscal conditions or the proximity to productive districts. The system (entire
market) could thereby reduce production costs by selecting where to set plants or rely on
suppliers. It is in this sense that globalization produced the rise of new spatial gradients in
the productive system. The whole system thereby increased its degree of freedom. We face
therefore the counteractive interplay of degree of freedom, on two hierarchical levels,
triggered by a saturation stage. To better stress how this interplay of degree of freedom,
working in opposite directions, can be caused by saturation, it is best to approach physical
systems for analogies.
For example, if we increase the pressure of a gas in a specific volume, we reduce its degree
of freedom and it consequently can become a liquid, at certain temperatures. At the same
time, when a liquid changes its motion regime, as in Benárd cells, from a pure, random
dissipative system (Figure 5A) to a global dissipative one (Figure 5B), which displays
features several magnitudes larger than molecules, a superstructure can arise that was
previously available, but very unlikely.


Fig. 5. Degree reduction/increase in dissipative structures (Ruzzenenti and Basosi, 2009b)

Gravity and viscosity constraints make such a structure, beyond a certain level of energy
input, possible. The Benárd cells phenomenon is indeed possible when the gradient
temperature and water level thickness are known, but not when the vessel permits the fluid
to dissipate heat in random motions. In other words, the boundary conditions together with
the pressure imposed upon the system by an increasing energy flux, changes the macrostate
(energy density) of the system by modifying its microstates (the molecular motion). The
random motion of molecules reflects one gradient, the temperature, which is not spatial
(geometrical), while the superstructure is exposed to the spatial gradient. That is to say,
while the first gradient is defined by one variable, the latter is described by three variables
and probabilities consequently change. Dissipation into one variable is therefore more
probable than onto three variables, unless boundary conditions render the former
impossible. In Benárd cells, such conditions are exemplified in Van der Waals forces, the
low heat capacity of water, and restrained vessel thickness (Chandrasekhar, 1961; Prigogine
and Stengers, 1984; Swenson, 1997). The connectivity recasts the same trade off in a network
system’s conceptual framework. A network system grows in complication as long as a new
Energy Effciency 44
component is connected on the same hierarchical level and it grows in complexity when a
new component is introduced on a higher hierarchy (Allen and Starr, 1982). The emergence
of a new hierarchy involves coherent behavior for lower level components to the same
extent as molecules in Benárd cells, and most importantly, the onset of a new spatial
gradient

for the higher component, which must now recognize system boundaries.

On a
molecular scale, cells in the body behave like a network. From the stand point of the
organism, however, they act as a whole unit. Indeed “free” cells in substrates are mainly
exposed to chemical gradients (temperature, pressure and gravitational gradients as well),
while “embedded” cells in tissues that form organs are described by spatial, three
dimensional, gradients.

5.3 Spatial symmetry rupture
We believe economic systems (and macroscopic complex systems in general) can exhibit a
similar evolutionary pattern: a space symmetry rupture emerges from compelling boundary
conditions and increasing energy inflow. In the case of the productive structure’s evolution
it can be shown that space was isotropic
21
in the former state (fordian) and non isotropic in
the latter (post-fordian): a spatial symmetry breaking occurred (Figure 6). What made this
spatial gradient rise was a reduction in firms’ degree of freedom production settings,
coupled with an energy efficiency leap. More energy was thus available to the system amid
a condition of hindering forces applied to its boundaries. Two counteractive forces are
beneath a symmetry rupture. In this case the symmetry rupture put a space gradient upon
the system, with which it induces its variables (components) to organize themselves.
Globalization and outsourcing set production plants in a new, oriented space that was
formerly homogeneous.
We would like now to clarify the reason why it has been used the word rupture has been
used in place of breaking to describe the symmetry change. The concept of “symmetry
breaking” applies to the temporal scale, whereas here space symmetry has been considered.
That is to say, the time-symmetry concept concerns a sudden change in the developmental
path of the system; nevertheless this change affects the system itself, rather than the space of
the system. In Prigogine’s paradigm a dynamic system is considered and it is thus described
by a dynamic function, whereas, in the symmetry rupture a phase transition rather than a
dynamical, however non-linear, change is described
22
.


21
In other words, there is just one way to go from the periphery to the centre, regardless of the number
of nodes considered, while there are many ways to connect the same number of points in the path.
Furthermore, the number of different ways increases with the number of points. This does not mean
that, in a scattered productive chain, factories (points), are connected randomly, but instead means that
there are multiple ways for a chain to develop its pattern and just one for a centralized system.
22
„We see therefore, that the appearance of a periodic reaction is a time-symmetry breaking process
exactly as ferromagnetism is a space-symmetry breaking one. [….] To understand at least qualitatively
this result let us consider the analogy with phase transitions. When we cool down a paramagnetic
substance, we come to the so-called Curie point below which the system behaves like a Ferro magnet.
Above the Curie point, all directions play the same role. Below, there is a privileged direction
corresponding to the direction of magnetization “(Prigogine, 1977).


Fig. 6. Spatial symmetry and productive chain (Ruzzenenti and Basosi, 2009b)

It is noteworthy that Prigogine used the concept of space-symmetry breaking as a metaphor
to introduce the new concept of time-symmetry breaking. Now, we want to retrieve the
concept of space symmetry breaking (symmetry rupture in the jargon so far adopted) as we
think it is fundamental to understand how evolution may concern the space of the system,
rather than the system itself. Furthermore, it should also be noticed that the concept of
space-symmetry breaking includes the concept of time-symmetry breaking and not vice
versa.

5.4 Time scale and Spatial scale
Complex systems display a spatial gradient which is sometimes many orders of magnitude
larger than gradients involving the scale of components. This important feature of
complexity was first envisaged by Prigogine. Parameters describing dissipative structures,
like Bénard cells, are macroscopic compared to parameters describing structures at
thermodynamic equilibrium. Indeed, while crystals are described by interactions of the
order of 10
-10
meter, convective cells display a size of the order of the 10
-2
meter (Prigogine
and Stengers, 1984). The same can be said for the characteristic time of phenomena. Time
scale varies greatly for the above mentioned systems: the vibration period of molecules is of
the order of 10
-15
seconds whereas convective motions have a period in the order of seconds.
It is noteworthy that this scale effect, consequential to the hierarchical leap, seems to be a
common feature of all complex systems, spanning from simple, non-living dissipative
structures to greatly complex human-made and biological systems. In the case study here
presented the magnification of time scale is clearly established in the nature of decision
process that characterize the view-point of firms. As long as the entire chain was engulfed in
the same firm, if not in the same production plant, any decision inherent to the volume of
Energy growth, complexity and effciency 45
component is connected on the same hierarchical level and it grows in complexity when a
new component is introduced on a higher hierarchy (Allen and Starr, 1982). The emergence
of a new hierarchy involves coherent behavior for lower level components to the same
extent as molecules in Benárd cells, and most importantly, the onset of a new spatial
gradient

for the higher component, which must now recognize system boundaries.

On a
molecular scale, cells in the body behave like a network. From the stand point of the
organism, however, they act as a whole unit. Indeed “free” cells in substrates are mainly
exposed to chemical gradients (temperature, pressure and gravitational gradients as well),
while “embedded” cells in tissues that form organs are described by spatial, three
dimensional, gradients.

5.3 Spatial symmetry rupture
We believe economic systems (and macroscopic complex systems in general) can exhibit a
similar evolutionary pattern: a space symmetry rupture emerges from compelling boundary
conditions and increasing energy inflow. In the case of the productive structure’s evolution
it can be shown that space was isotropic
21
in the former state (fordian) and non isotropic in
the latter (post-fordian): a spatial symmetry breaking occurred (Figure 6). What made this
spatial gradient rise was a reduction in firms’ degree of freedom production settings,
coupled with an energy efficiency leap. More energy was thus available to the system amid
a condition of hindering forces applied to its boundaries. Two counteractive forces are
beneath a symmetry rupture. In this case the symmetry rupture put a space gradient upon
the system, with which it induces its variables (components) to organize themselves.
Globalization and outsourcing set production plants in a new, oriented space that was
formerly homogeneous.
We would like now to clarify the reason why it has been used the word rupture has been
used in place of breaking to describe the symmetry change. The concept of “symmetry
breaking” applies to the temporal scale, whereas here space symmetry has been considered.
That is to say, the time-symmetry concept concerns a sudden change in the developmental
path of the system; nevertheless this change affects the system itself, rather than the space of
the system. In Prigogine’s paradigm a dynamic system is considered and it is thus described
by a dynamic function, whereas, in the symmetry rupture a phase transition rather than a
dynamical, however non-linear, change is described
22
.


21
In other words, there is just one way to go from the periphery to the centre, regardless of the number
of nodes considered, while there are many ways to connect the same number of points in the path.
Furthermore, the number of different ways increases with the number of points. This does not mean
that, in a scattered productive chain, factories (points), are connected randomly, but instead means that
there are multiple ways for a chain to develop its pattern and just one for a centralized system.
22
„We see therefore, that the appearance of a periodic reaction is a time-symmetry breaking process
exactly as ferromagnetism is a space-symmetry breaking one. [….] To understand at least qualitatively
this result let us consider the analogy with phase transitions. When we cool down a paramagnetic
substance, we come to the so-called Curie point below which the system behaves like a Ferro magnet.
Above the Curie point, all directions play the same role. Below, there is a privileged direction
corresponding to the direction of magnetization “(Prigogine, 1977).


Fig. 6. Spatial symmetry and productive chain (Ruzzenenti and Basosi, 2009b)

It is noteworthy that Prigogine used the concept of space-symmetry breaking as a metaphor
to introduce the new concept of time-symmetry breaking. Now, we want to retrieve the
concept of space symmetry breaking (symmetry rupture in the jargon so far adopted) as we
think it is fundamental to understand how evolution may concern the space of the system,
rather than the system itself. Furthermore, it should also be noticed that the concept of
space-symmetry breaking includes the concept of time-symmetry breaking and not vice
versa.

5.4 Time scale and Spatial scale
Complex systems display a spatial gradient which is sometimes many orders of magnitude
larger than gradients involving the scale of components. This important feature of
complexity was first envisaged by Prigogine. Parameters describing dissipative structures,
like Bénard cells, are macroscopic compared to parameters describing structures at
thermodynamic equilibrium. Indeed, while crystals are described by interactions of the
order of 10
-10
meter, convective cells display a size of the order of the 10
-2
meter (Prigogine
and Stengers, 1984). The same can be said for the characteristic time of phenomena. Time
scale varies greatly for the above mentioned systems: the vibration period of molecules is of
the order of 10
-15
seconds whereas convective motions have a period in the order of seconds.
It is noteworthy that this scale effect, consequential to the hierarchical leap, seems to be a
common feature of all complex systems, spanning from simple, non-living dissipative
structures to greatly complex human-made and biological systems. In the case study here
presented the magnification of time scale is clearly established in the nature of decision
process that characterize the view-point of firms. As long as the entire chain was engulfed in
the same firm, if not in the same production plant, any decision inherent to the volume of
Energy Effciency 46
the production was almost readily attainable. Outsourcing, conversely, brought the
productive chain outside the firm, making the setting of the plant an endogenous variable
and the volume of the production an exogenous variable (or at least a less controllable one).
Nevertheless, decisions having to do with production’s settlement develop themselves over
a much larger range of time.
Obviously, the time lag scales up due to the spatial extent of interactions, which increases
many orders of magnitude throughout hierarchy as components’ size remains the same.
This is why the time scale is commensurate to the spatial extent of the system. Nevertheless
it seems that magnification of the time scale is affected by space in a fashion that is not
entirely reducible to an extensive factor. Time scale as it grows displays a cyclic
phenomenon which seems to relate to the symmetry property of the space. In mechanics a
cyclic system displays properties of invariance that are proportional to the symmetry
properties of the space: to every local symmetry there corresponds a conservation law. A
conservation law states that some quantity describing a system remains constant throughout
its motion; expressed mathematically, the rate of change of its derivative with respect to
time is zero.
A system that is cyclic exhibits symmetry as if the space was homogeneous. Therefore by
means of cyclic patterns, symmetry in space is re-established and growth can develop again
in a continuous way. It is needless to emphasize that cycles are a prominent feature of
complex systems, regardless of the nature or the scale that is involved.

6. Conclusions
In the first part of the chapter the rebound effect -the growing energy use coupled with an
efficiency enhancement, was employed to introduce the broader question that relates energy
efficiency to energy density rate. It was shown that the paradox partially derives from
misconception of energy efficiency and power. It must be firstly conceptually set apart
thermodynamic efficiency from other forms of efficiency. It must than bore in mind that
thermodynamic efficiency is strictly connected to power, in two ways. On the one hand
there is power-efficiency trade-off and evolutionary systems tend to maximize power rather
than efficiency. On the other hand, an efficiency enhancement normally leads to a power
shift, as a side effect. In the second part of the chapter, we approached the question of the
interdependence between efficiency, complexity and energy density, to illustrate how the
causality chain can be reversed: efficiency leads to an higher energy density rate and
eventually, to a complexity leap. A complexity leap that is underlined by a change in the
space of the system.
As in a phase transition, space symmetry rupture seems to be an important aspect of
complexity change. Symmetry rupture, introducing a new gradient in the system space,
allows variables (components) to organize themselves. This new arrangement, on the scale
of variables, reduces their degree of freedom, on the scale of the whole system, increases it.
The change deeply concerns the structure, therefore the geometry of the system. Between the
new and the old structure a topological change occurs. In our opinion, the topology of the
system has to be addressed with graph theory. Yet, the transition between the two phases is
still an open question and more research is needed. A formal analysis of it should start from
recent acquisitions in the field of network theory (Ruzzenenti, Garlaschelli and Basosi, 2010).
However, as we tried to illustrate in this chapter, the new arising structure will be more
complex and more energy intensive. Higher energy density rate will be an outcome of a
transition that will maximize links and frequency of interactions. Such a transition lays
behind, in our opinion, the so called “rebound effect” (Jevons paradox) and explains why
energy efficiency has always led to energy growth. Energy conservation policies should
therefore contemplate, together with strategies prompted at fostering energy efficiency,
measures directed at balancing the long term positive effect over energy consumptions due
to a structural changes in economy.

7. References
Alcott B., 2008. The sufficiency strategy: Would rich-world frugality lower environmental
impact? Ecological Economics, 6 4, 7 7 0 – 7 8 6
Allen T. H. F., and T. B. Starr. 1982. Hierarchy: perspectives for ecological complexity.
University of Chicago Press, Chicago, IL.
Ayres R. U., Warr B., 2005. Accounting for growth: the role of physical work. Structural
Change and Economic Dynamics ,16, 181–209
Banks F. E., 2007. The political Economy of World Energy: An Introductory Textbook.
Singapore and New York, World Scientific.
Bentzen J., 2004. Estimating the rebound effect in US manufacturing energy consumption,
Energy Economics, 26, 123-134.
Binswanger M., 2001. Technological progress and sustainable development: what about the
rebound effect? Ecological Economics 36 (2001) 119 – 132
Brookes, Leonard, 1990. The greenhouse effect: the fallacies in the energy efficiency solution.
Energy Policy 18 (2), 199–201.
Chaisson E., 2001. Cosmic Evolution-The Rise of Complexity in Nature, Harvard University
Press, Cambridge, Massachusetts, London, U.K., 2001.
Chandrasekhar, S. 1961. Hydrodynamic and hydromagnetic stability. Oxford, Clarendon.
Curzon F., Ahlborn B., 1975. Efficiency of a Carnot Engine at Maximum Power Output.
American Journal of Physics 41, 22-24 (1975).
Dimitropoulos J., 2007. Energy productivity improvements and the rebound effect: An
overview of the state of knowledge. Energy Policy 35, 6354–6363.
Geening, Lorna A., Greene, David L., Difiglio, Carmen, 2000. Energy efficiency and
consumption—the rebound effect—a survey. Energy Policy 28 (6/7), 389 – 401.
Grepperud S., Rasmussen I. 2003. General equilibrium assessment of rebound effects,
Energy Economics, 2003. 189–203.
Herring H., Energy efficiency—a critical view. Energy 31 (2006) 10–20.
IEA, Indicators of energy use and efficiency, 1997.
IEA Energy Statistics Statistics on the Web: http://www.iea.org/statist/index.htm
Jevons, W. Stanley, The coal question – An Inquiry Concerning the Progress of the Nation,
and the Probable Exhaustion of our Coal-mines, M.A., ll.D., F.R.S., Augustus M.
Kelley Publisher, New York, 1965
Khazzoom, J.D., 1980, “Economic Implications of Mandated Efficiency in Standards for
Houshold Appliances.” The Energy Journal, Vol.1, No.4, pp21-40.
Kummel, R., 1989. Energy as a factor of production and entropy as a pollution, Ecological
Economics, 1 (1989) 161-180.
Energy growth, complexity and effciency 47
the production was almost readily attainable. Outsourcing, conversely, brought the
productive chain outside the firm, making the setting of the plant an endogenous variable
and the volume of the production an exogenous variable (or at least a less controllable one).
Nevertheless, decisions having to do with production’s settlement develop themselves over
a much larger range of time.
Obviously, the time lag scales up due to the spatial extent of interactions, which increases
many orders of magnitude throughout hierarchy as components’ size remains the same.
This is why the time scale is commensurate to the spatial extent of the system. Nevertheless
it seems that magnification of the time scale is affected by space in a fashion that is not
entirely reducible to an extensive factor. Time scale as it grows displays a cyclic
phenomenon which seems to relate to the symmetry property of the space. In mechanics a
cyclic system displays properties of invariance that are proportional to the symmetry
properties of the space: to every local symmetry there corresponds a conservation law. A
conservation law states that some quantity describing a system remains constant throughout
its motion; expressed mathematically, the rate of change of its derivative with respect to
time is zero.
A system that is cyclic exhibits symmetry as if the space was homogeneous. Therefore by
means of cyclic patterns, symmetry in space is re-established and growth can develop again
in a continuous way. It is needless to emphasize that cycles are a prominent feature of
complex systems, regardless of the nature or the scale that is involved.

6. Conclusions
In the first part of the chapter the rebound effect -the growing energy use coupled with an
efficiency enhancement, was employed to introduce the broader question that relates energy
efficiency to energy density rate. It was shown that the paradox partially derives from
misconception of energy efficiency and power. It must be firstly conceptually set apart
thermodynamic efficiency from other forms of efficiency. It must than bore in mind that
thermodynamic efficiency is strictly connected to power, in two ways. On the one hand
there is power-efficiency trade-off and evolutionary systems tend to maximize power rather
than efficiency. On the other hand, an efficiency enhancement normally leads to a power
shift, as a side effect. In the second part of the chapter, we approached the question of the
interdependence between efficiency, complexity and energy density, to illustrate how the
causality chain can be reversed: efficiency leads to an higher energy density rate and
eventually, to a complexity leap. A complexity leap that is underlined by a change in the
space of the system.
As in a phase transition, space symmetry rupture seems to be an important aspect of
complexity change. Symmetry rupture, introducing a new gradient in the system space,
allows variables (components) to organize themselves. This new arrangement, on the scale
of variables, reduces their degree of freedom, on the scale of the whole system, increases it.
The change deeply concerns the structure, therefore the geometry of the system. Between the
new and the old structure a topological change occurs. In our opinion, the topology of the
system has to be addressed with graph theory. Yet, the transition between the two phases is
still an open question and more research is needed. A formal analysis of it should start from
recent acquisitions in the field of network theory (Ruzzenenti, Garlaschelli and Basosi, 2010).
However, as we tried to illustrate in this chapter, the new arising structure will be more
complex and more energy intensive. Higher energy density rate will be an outcome of a
transition that will maximize links and frequency of interactions. Such a transition lays
behind, in our opinion, the so called “rebound effect” (Jevons paradox) and explains why
energy efficiency has always led to energy growth. Energy conservation policies should
therefore contemplate, together with strategies prompted at fostering energy efficiency,
measures directed at balancing the long term positive effect over energy consumptions due
to a structural changes in economy.

7. References
Alcott B., 2008. The sufficiency strategy: Would rich-world frugality lower environmental
impact? Ecological Economics, 6 4, 7 7 0 – 7 8 6
Allen T. H. F., and T. B. Starr. 1982. Hierarchy: perspectives for ecological complexity.
University of Chicago Press, Chicago, IL.
Ayres R. U., Warr B., 2005. Accounting for growth: the role of physical work. Structural
Change and Economic Dynamics ,16, 181–209
Banks F. E., 2007. The political Economy of World Energy: An Introductory Textbook.
Singapore and New York, World Scientific.
Bentzen J., 2004. Estimating the rebound effect in US manufacturing energy consumption,
Energy Economics, 26, 123-134.
Binswanger M., 2001. Technological progress and sustainable development: what about the
rebound effect? Ecological Economics 36 (2001) 119 – 132
Brookes, Leonard, 1990. The greenhouse effect: the fallacies in the energy efficiency solution.
Energy Policy 18 (2), 199–201.
Chaisson E., 2001. Cosmic Evolution-The Rise of Complexity in Nature, Harvard University
Press, Cambridge, Massachusetts, London, U.K., 2001.
Chandrasekhar, S. 1961. Hydrodynamic and hydromagnetic stability. Oxford, Clarendon.
Curzon F., Ahlborn B., 1975. Efficiency of a Carnot Engine at Maximum Power Output.
American Journal of Physics 41, 22-24 (1975).
Dimitropoulos J., 2007. Energy productivity improvements and the rebound effect: An
overview of the state of knowledge. Energy Policy 35, 6354–6363.
Geening, Lorna A., Greene, David L., Difiglio, Carmen, 2000. Energy efficiency and
consumption—the rebound effect—a survey. Energy Policy 28 (6/7), 389 – 401.
Grepperud S., Rasmussen I. 2003. General equilibrium assessment of rebound effects,
Energy Economics, 2003. 189–203.
Herring H., Energy efficiency—a critical view. Energy 31 (2006) 10–20.
IEA, Indicators of energy use and efficiency, 1997.
IEA Energy Statistics Statistics on the Web: http://www.iea.org/statist/index.htm
Jevons, W. Stanley, The coal question – An Inquiry Concerning the Progress of the Nation,
and the Probable Exhaustion of our Coal-mines, M.A., ll.D., F.R.S., Augustus M.
Kelley Publisher, New York, 1965
Khazzoom, J.D., 1980, “Economic Implications of Mandated Efficiency in Standards for
Houshold Appliances.” The Energy Journal, Vol.1, No.4, pp21-40.
Kummel, R., 1989. Energy as a factor of production and entropy as a pollution, Ecological
Economics, 1 (1989) 161-180.
Energy Effciency 48
Lotka A., 1956. Elements of Mathematical Biology, Dover Publications, Inc, New York. (first
publication: elements of physical biology, The Williams and Wilkins Co., Inc, 1924).
Odum E.P., 1997. Ecology: A Bridge Between Science and Society, Sinauer Associates, Inc.,
Publishers. Sunderland, Massachusetts 01375 U.S.A.
Odum H.T., 1955. The speed regulation: the optimum efficiency for maximum power output
in physical and biological systems. American Scientist. 43 (1955), 331-343.
Odum H.T., 1983. Maximum Power and Efficiency: a rebutal, Ecological Modelling, 20, 71-
82.
Odum H.T., 1996. Environmental Accounting. Emergy and Environmental Decision
Making. John Wiley & Sons, Inc., New York, USA.
Prigogine I., G. Nicolis, A. Babloyantz, 1972, Thermodynamics of evolution. Physics Today
25 (11).
Prigogine Y., 1977. Time, Structure and Fluctuations, Nobel Lecture.
Ruzzenenti, F., Basosi, R., 2008a The role of the power/efficiency misconception in the rebound
effect’s size debate: Does efficiency actually lead to a power enhancement? Energy Policy
36-9, September 2008, 3626-3632.
Ruzzenenti F., Basosi R., 2008b The rebound effect: An evolutionary perspective. Ecological
Economics, 67 (2008) 526 – 537.
Ruzzenenti F., Basosi R., 2009b Complexity change and space symmetry rupture, Ecological
Modelling, 220(2009)1880–1885.
Ruzzenenti F., Basosi R., 2009a Evaluation of the energy efficiency evolution in the European road
freight transport sector. Energy Policy, 37 (2009) 4079–4085.
Ruzzenenti, F., Garlaschelli D., Basosi R., Complex Networks and Symmetry: a Review with
Applications to the Evolution of World Trade. Article in Press, pre-print:
arXiv:1006.3923v1 [q-fin.GN]
Saunders, Harry D., 2000. A view from the macro side: rebound, backfire and Khazzoom–
Brookes. Energy Policy 28 (6/7), 439–449.
Schroeder D., 2000. Thermal Physics, an introduction to. Addison Wesley Longman.
Schipper L., Haas R., 1998. Residential energy demand in OECD-countries and the role of
irreversible efficiency improvements. Energy Economics 20 (1998) 421-442.
Schipper L., Grubb M., 2000 On the rebound? Feedback between energy intensities and
energy uses in IEA countries. Energy Policy 28, 367 } 388
Sorrell, S., Dimitropoulos, J., 2007. UKERC Review of Evidence for the Rebound Effect:
Technical Report 5—Energy Productivity and Economic Growth Studies. UK
Energy Research Centre, London.
Sorrell S., 2009. Jevons’ Paradox revisited: The evidence for backfire from improved energy
efficiency. Energy Policy, 37 (2009) 1456-1469.
Swenson R., 1997. Autokatakinetics, evolution, and the law of maximum entropy
production: a principled foundation toward the study of human ecology, in: Freese,
L., Advances in human Ecology, vol.&. JAI Press, Greewich, CT, pp. 1-47; 1989.
Emergent attractors and the law of maximum entropy production: foundations to a
theory of general evolution. Syst. Res. 6, 187-197.
Washida T., 2004. Economy-wide Model of Rebound Effect for Environmental Efficiency ,
proceedings of International Workshop on Sustainable Consumption, University of
Leeds, March 5-6, 2004.

Categorizing Barriers to Energy Effciency: An Interdisciplinary Perspective 49
Categorizing Barriers to Energy Effciency: An Interdisciplinary
Perspective
Patrik Thollander, Jenny Palm and Patrik Rohdin
x

Categorizing Barriers to Energy Efficiency:
An Interdisciplinary Perspective

Patrik Thollander
1
, Jenny Palm
2
and Patrik Rohdin
1

1
Energy systems, Linköping University
2
Tema T, Linköping University
Sweden

1. Introduction
This chapter presents theoretical perspectives on barriers to energy efficiency identified in
different scientific disciplines, and briefly describes each barrier and its mode of operation. In an
attempt to categorize barriers to energy efficiency, the chapter addresses socio-technical regimes,
leading to a novel interdisciplinary categorization of barriers to energy efficiency in three
categories: the technological system, the technological regime, and the socio-technical regime.
The threat of climate change resulting from the use of fossil fuels is posing a threat to the
environment, and energy efficiency is one of the most important means of reducing this
threat (IPCC, 2007). Despite this, there are a number of publications stating the existence of a
“gap” between potential cost-effective energy efficiency measures and measures actually
implemented—the so called “energy efficiency gap” or “energy paradox” (York et al., 1978;
Blumstein et al., 1980; Stern and Aronsson, 1984; Hirst and Brown, 1990; Gruber and Brand,
1991; Stern, 1992; DeCanio, 1993; Jaffe and Stavins 1994; Sanstad and Howarth, 1994; Weber,
1997; Ostertag, 1999; Sorrell et al., 2000; Brown, 2001; de Groot et al., 2001; Schleich, 2004;
Sorrell et al., 2004; Schleich and Gruber, 2008). This “energy efficiency gap” or “energy
paradox” exists due to barriers to energy efficiency. A barrier may be defined as a
postulated mechanism that inhibits investments in technologies that are both energy-
efficient and economically efficient (Sorrell et al., 2004).
Barriers are explanations for the reluctance to adopt cost-effective energy efficiency
measures derived from mainstream economics, organizational economics, and
organizational and behavioural theories. There are also institutional or structural barriers to
energy efficiency that do not directly affect the “gap”, even though it does affect the overall
level of energy efficiency. Barriers may be divided into three broad categories: Economic,
Organizational and Behavioural. Inspired by an extensive review of the existing literature on
barriers to energy efficiency, Sorrell et al. (2000) compiled a barrier framework categorized
into different barriers (see Table 1; the barriers are explained in greater detail in the
following sections of this chapter). It should be noted that the above classification of barriers
is not unambiguous; one type of real-world phenomena may be explained by several of the
theoretically derived barriers presented (Weber, 1997).
3
Energy Effciency 50

Jaffe and Stavins (1994) outlined a number of different levels of “energy efficiency potential”, or
“energy efficiency gaps” (see Figure 1). The figure states that the actual potential level of energy
efficiency depends on which view is applied—while the technologist’s potential is real in a sense,
the economist’s potential is actually real for that person or organization, with the difference
between the two levels depending on which theoretical perspective is being applied.

Theoretical Barriers Comment
Imperfect information
(Howarth and Andersson, 1993)
Lack of information may lead to cost-effective energy efficiency
measures opportunities being missed.
Adverse selection
(Sanstad and Howarth, 1994)
(Jaffe and Stavins, 1994)
If suppliers know more about the energy performance of goods than
purchasers, the purchasers may select goods on the basis of visible
aspects such as price.
Principal-agent
relationships
(Jaffe and Stavins, 1994)
Strict monitoring and control by the principal, since he or she cannot
see what the agent is doing, may result in energy efficiency measures
being ignored.
Split incentives
(Jaffe and Stavins, 1994)
(Hirst and Brown, 1990)
If a person or department cannot gain benefits from energy efficiency
investment it is likely that implementation will be of less interest.
Hidden costs
(Jaffe and Stavins, 1994)
(Ostertag, 1999)
Examples of hidden costs are overhead costs, cost of collecting and
analyzing information, production disruptions, inconvenience etc..
Access to capital
(Hirst and Brown, 1990)
(Jaffe and Stavins, 1994)
Limited access to capital may prevent energy efficiency measures
from being implemented.
Risk
(Hirst and Brown, 1990)
Risk aversion may be the reason why energy efficiency measures are
constrained by short pay-back criteria.
Heterogeneity
(Jaffe and Stavins, 1994)
A technology or measure may be cost-effective in general, but not in
all cases.
Form of information
(Stern and Aronsson, 1984)
Research has shown that the form of information is critical.
Information should be specific, vivid, simple, and personal to
increase its chances of being accepted.
Credibility and trust
(Stern and Aronsson, 1984)
The information source should be credible and trustworthy in order to
successfully deliver information regarding energy efficiency
measures. If these factors are lacking this will result in inefficient
choices.
Values
(Stern, 1992)
Efficiency improvements are most likely to be successful if there are
individuals with real ambition, preferably represented by a key
individual within top management.
Inertia
(Stern and Aronsson, 1984)
Individuals who are opponents to change within an organization may
result in overlooking energy efficiency measures that are cost-
effective.
Bounded rationality
(Sanstad and Howarth, 1994)
Instead of being based on perfect information, decisions are made by
rule of thumb.
Power
(Sorrell et al., 2000)
Low status of energy management may lead to lower priority of
energy issues within organizations.
Culture
(Sorrell et al., 2000)
Organizations may encourage energy efficiency investments by
developing a culture characterized by environmental values.


Table 1. Classification of barriers to energy efficiency (inspired by Sorrell et al., 2000).

1.1 Economic barriers – market failures
One important category with regard to barriers is the group of barriers that may be seen as
market failures violating the underlying axioms of mainstream economic theory. According

to mainstream economic theory, a market failure may justify public policy intervention.
However, the mere existence of a market failure may not in and of itself be sufficient to
justify intervention. As Brown (2001) writes:
“The existence of market failures and barriers that inhibit socially optimal levels of investment in
energy efficiency is the primary reason for considering public policy interventions. In many
instances, feasible, low cost policies can be implemented that either eliminate or compensate for
market imperfections and barriers, enabling markets to operate more efficiently to the benefit of
society. In other instances, policies may not be feasible; they may not fully eliminate the targeted
barrier or imperfection; or they may do so at costs that exceed the benefits.” (Brown, 2001).
The elimination of a market failure barrier may thus only be put into operation if the
benefits arising from an intervention exceed the cost of implementation.
Increasing
energy
efficiency
Hypothetical potential
Eliminate market
failures in energy
markets
Effect of market
Technologist's barriers that
economic cannot be
potential eliminated at
acceptable cost
Eliminate high
discount rates due True
to uncertainty, social
overcome inertia, optimum
ignore
heterogeneity Additional
efficiency
Economist's justified by
economic environmental
potential externalities
Narrow
Eliminate social
market failures optimum
in the market for
energy efficient Eliminate those
technologies market failures
whose elimination
can pass a
benefit/cost test
Baseline or business as usual energy efficiency level

Fig. 1. Different levels of energy efficiency potential (Jaffe and Stavins, 1994).

An often cited market failure barrier is imperfect information. Other market failure barriers
include asymmetric information, a special form of imperfect information where split incentives,
Categorizing Barriers to Energy Effciency: An Interdisciplinary Perspective 51

Jaffe and Stavins (1994) outlined a number of different levels of “energy efficiency potential”, or
“energy efficiency gaps” (see Figure 1). The figure states that the actual potential level of energy
efficiency depends on which view is applied—while the technologist’s potential is real in a sense,
the economist’s potential is actually real for that person or organization, with the difference
between the two levels depending on which theoretical perspective is being applied.

Theoretical Barriers Comment
Imperfect information
(Howarth and Andersson, 1993)
Lack of information may lead to cost-effective energy efficiency
measures opportunities being missed.
Adverse selection
(Sanstad and Howarth, 1994)
(Jaffe and Stavins, 1994)
If suppliers know more about the energy performance of goods than
purchasers, the purchasers may select goods on the basis of visible
aspects such as price.
Principal-agent
relationships
(Jaffe and Stavins, 1994)
Strict monitoring and control by the principal, since he or she cannot
see what the agent is doing, may result in energy efficiency measures
being ignored.
Split incentives
(Jaffe and Stavins, 1994)
(Hirst and Brown, 1990)
If a person or department cannot gain benefits from energy efficiency
investment it is likely that implementation will be of less interest.
Hidden costs
(Jaffe and Stavins, 1994)
(Ostertag, 1999)
Examples of hidden costs are overhead costs, cost of collecting and
analyzing information, production disruptions, inconvenience etc..
Access to capital
(Hirst and Brown, 1990)
(Jaffe and Stavins, 1994)
Limited access to capital may prevent energy efficiency measures
from being implemented.
Risk
(Hirst and Brown, 1990)
Risk aversion may be the reason why energy efficiency measures are
constrained by short pay-back criteria.
Heterogeneity
(Jaffe and Stavins, 1994)
A technology or measure may be cost-effective in general, but not in
all cases.
Form of information
(Stern and Aronsson, 1984)
Research has shown that the form of information is critical.
Information should be specific, vivid, simple, and personal to
increase its chances of being accepted.
Credibility and trust
(Stern and Aronsson, 1984)
The information source should be credible and trustworthy in order to
successfully deliver information regarding energy efficiency
measures. If these factors are lacking this will result in inefficient
choices.
Values
(Stern, 1992)
Efficiency improvements are most likely to be successful if there are
individuals with real ambition, preferably represented by a key
individual within top management.
Inertia
(Stern and Aronsson, 1984)
Individuals who are opponents to change within an organization may
result in overlooking energy efficiency measures that are cost-
effective.
Bounded rationality
(Sanstad and Howarth, 1994)
Instead of being based on perfect information, decisions are made by
rule of thumb.
Power
(Sorrell et al., 2000)
Low status of energy management may lead to lower priority of
energy issues within organizations.
Culture
(Sorrell et al., 2000)
Organizations may encourage energy efficiency investments by
developing a culture characterized by environmental values.


Table 1. Classification of barriers to energy efficiency (inspired by Sorrell et al., 2000).

1.1 Economic barriers – market failures
One important category with regard to barriers is the group of barriers that may be seen as
market failures violating the underlying axioms of mainstream economic theory. According

to mainstream economic theory, a market failure may justify public policy intervention.
However, the mere existence of a market failure may not in and of itself be sufficient to
justify intervention. As Brown (2001) writes:
“The existence of market failures and barriers that inhibit socially optimal levels of investment in
energy efficiency is the primary reason for considering public policy interventions. In many
instances, feasible, low cost policies can be implemented that either eliminate or compensate for
market imperfections and barriers, enabling markets to operate more efficiently to the benefit of
society. In other instances, policies may not be feasible; they may not fully eliminate the targeted
barrier or imperfection; or they may do so at costs that exceed the benefits.” (Brown, 2001).
The elimination of a market failure barrier may thus only be put into operation if the
benefits arising from an intervention exceed the cost of implementation.
Increasing
energy
efficiency
Hypothetical potential
Eliminate market
failures in energy
markets
Effect of market
Technologist's barriers that
economic cannot be
potential eliminated at
acceptable cost
Eliminate high
discount rates due True
to uncertainty, social
overcome inertia, optimum
ignore
heterogeneity Additional
efficiency
Economist's justified by
economic environmental
potential externalities
Narrow
Eliminate social
market failures optimum
in the market for
energy efficient Eliminate those
technologies market failures
whose elimination
can pass a
benefit/cost test
Baseline or business as usual energy efficiency level

Fig. 1. Different levels of energy efficiency potential (Jaffe and Stavins, 1994).

An often cited market failure barrier is imperfect information. Other market failure barriers
include asymmetric information, a special form of imperfect information where split incentives,
Energy Effciency 52

adverse selection, and principal-agent relationships may also be categorized. These market
failure barriers are presented below.

1.1.1 Imperfect information
A large body of research states that consumers are often poorly informed about market
conditions, technology characteristics and their own energy use. The lack of adequate
information about potential energy-efficient technologies inhibits investments in energy
efficiency measures (Sanstad and Howarth, 1994). Insufficient information is one form of
imperfect information, such as when the energy performance of energy-efficient
technologies is not made available to agents. Another form of imperfect information is the
cost of information, meaning that there are costs associated with searching and acquiring
information about the energy performance of an energy-efficient technology. Yet another
form is the accuracy of information, meaning that the information provider may not always
be transparent about the product being offered. Imperfect information is likely to be most
serious when the product is purchased infrequently, performance characteristics are difficult
to evaluate either before or soon after purchase, and the rate of technology change is rapid
relative to the purchase intervals (Sorrell et al., 2000), which is the case for many energy
efficiency measures. Issues related to imperfect information may be countered with different
forms of information campaigns.

1.1.2 Adverse selection
Adverse selection means that producers of energy-efficient equipment are, in general, more
informed about the characteristics and performance of equipment than prospective buyers.
In other words, the information between the two parties engaged in the transaction is
asymmetric. Since asymmetric information is extremely common in real world markets,
inefficient outcomes may be the rule rather than the exception (Sanstad and Howarth, 1994).

1.1.3 Principal-agent relationship
The principal-agent relationship arises due to a lack of trust between two parties at different
levels within an organization or transaction. The owner of a company, who may not be as
well-informed about the site-specific criteria for energy efficiency investments, may demand
short payback rates/high hurdle rates on energy efficiency investments due to his or her
distrust in the executive’s ability to convey such investments—leading to the neglect of cost-
effective energy efficiency investments (DeCanio 1993; Jaffe and Stavins, 1994).

1.1.4 Split incentives
A split incentive may occur when the potential adopter of an investment is not the party
that pays the energy bill. If so, information about available cost-effective energy efficiency
measures in the hands of the potential adopter may not be sufficient; adoption will only
occur if the adopter can recover the investment from the party that enjoys the energy
savings (Jaffe and Stavins, 1994). This is often referred to as the landlord-tenant relationship
For example, if a mid-level executive pays the energy bill for his or her division based on
number of employees, this decreases interest in the organization’s overall in-house energy
program to lower energy costs (including investments in energy efficiency technologies),

since there is “nothing in it” for him or her. This is a restriction to adopting energy-efficient
technologies, in particular those with higher initial costs but lower life cycle costs than
conventional technologies (Hirst and Brown, 1990). The lack of sub-metering within
multidivisional organizations may also be classified as a split incentive.

1.2 Economic barriers: non-market failures
Apart from market failure barriers, there are a number of barriers that explain the “gap” but
which cannot be categorized as market failures, but are rather non-market failure barriers or
market barriers. A market barrier, according to Jaffe and Stavins (1994), may be defined as
any factor that may account for the “gap”, while Brown (2001) defines market barriers as
obstacles that are not based on market failures but which nonetheless contribute to the slow diffusion
and adoption of energy-efficient measures. Barriers that may be categorized as market barriers
are, for example, hidden costs, limited access to capital, risk, and heterogeneity. These
barriers are presented below.

1.2.1 Hidden costs
Hidden costs are often used as an explanatory variable for the “gap” (DeCanio, 1998). In
short, the argument is that there are high costs associated with information-seeking, meeting
with sellers, writing contracts and other such activities; if these costs are higher than the
actual profit from implementation, they inhibit investment. Accordingly, cost-effective
measures are not cost-effective when such costs associated with the investment are included.
A study by Hein and Blok (1994) found that hidden costs in large energy-intensive
industrial firms ranged from three to eight percent of total investment costs. In smaller, non-
energy-intensive firms, such costs are thus likely to be even higher. Hidden costs are a
frequently used argument against the existence of an energy efficiency gap; it is argued that
engineering-economic models are not able to see the full cost of an energy efficiency
measure (Sorrell et al., 2000).

1.2.2 Limited access to capital
Technologies that are energy-efficient are often more expensive to purchase than alternative
technologies (Almeida, 1998). Moreover, obtaining additional capital in order to invest in
energy-efficient technology may be problematic. Apart from low liquidity, limited access to
capital may also arise due to restrictions on lending money (Hirst and Brown, 1990).
Sometimes such restrictions may be self-imposed.

1.2.3 Risk
Even though, for example, managers know what the capital cost is for an energy efficiency
investment, there can be uncertainty about the long-term savings in operating costs; this
means the investment poses a risk. Such concerns have been found to be very important to
decision-makers (Hirst and Brown, 1990).
Stern and Aronson (1984) also identify risk as a barrier to energy efficiency, since accurate
estimates of the net costs of implementing energy efficiency measures depend on future
economic conditions in general, and on future energy prices and availability in particular.
Energy prices have fluctuated as long as there has been a market for energy, leading to
Categorizing Barriers to Energy Effciency: An Interdisciplinary Perspective 53

adverse selection, and principal-agent relationships may also be categorized. These market
failure barriers are presented below.

1.1.1 Imperfect information
A large body of research states that consumers are often poorly informed about market
conditions, technology characteristics and their own energy use. The lack of adequate
information about potential energy-efficient technologies inhibits investments in energy
efficiency measures (Sanstad and Howarth, 1994). Insufficient information is one form of
imperfect information, such as when the energy performance of energy-efficient
technologies is not made available to agents. Another form of imperfect information is the
cost of information, meaning that there are costs associated with searching and acquiring
information about the energy performance of an energy-efficient technology. Yet another
form is the accuracy of information, meaning that the information provider may not always
be transparent about the product being offered. Imperfect information is likely to be most
serious when the product is purchased infrequently, performance characteristics are difficult
to evaluate either before or soon after purchase, and the rate of technology change is rapid
relative to the purchase intervals (Sorrell et al., 2000), which is the case for many energy
efficiency measures. Issues related to imperfect information may be countered with different
forms of information campaigns.

1.1.2 Adverse selection
Adverse selection means that producers of energy-efficient equipment are, in general, more
informed about the characteristics and performance of equipment than prospective buyers.
In other words, the information between the two parties engaged in the transaction is
asymmetric. Since asymmetric information is extremely common in real world markets,
inefficient outcomes may be the rule rather than the exception (Sanstad and Howarth, 1994).

1.1.3 Principal-agent relationship
The principal-agent relationship arises due to a lack of trust between two parties at different
levels within an organization or transaction. The owner of a company, who may not be as
well-informed about the site-specific criteria for energy efficiency investments, may demand
short payback rates/high hurdle rates on energy efficiency investments due to his or her
distrust in the executive’s ability to convey such investments—leading to the neglect of cost-
effective energy efficiency investments (DeCanio 1993; Jaffe and Stavins, 1994).

1.1.4 Split incentives
A split incentive may occur when the potential adopter of an investment is not the party
that pays the energy bill. If so, information about available cost-effective energy efficiency
measures in the hands of the potential adopter may not be sufficient; adoption will only
occur if the adopter can recover the investment from the party that enjoys the energy
savings (Jaffe and Stavins, 1994). This is often referred to as the landlord-tenant relationship
For example, if a mid-level executive pays the energy bill for his or her division based on
number of employees, this decreases interest in the organization’s overall in-house energy
program to lower energy costs (including investments in energy efficiency technologies),

since there is “nothing in it” for him or her. This is a restriction to adopting energy-efficient
technologies, in particular those with higher initial costs but lower life cycle costs than
conventional technologies (Hirst and Brown, 1990). The lack of sub-metering within
multidivisional organizations may also be classified as a split incentive.

1.2 Economic barriers: non-market failures
Apart from market failure barriers, there are a number of barriers that explain the “gap” but
which cannot be categorized as market failures, but are rather non-market failure barriers or
market barriers. A market barrier, according to Jaffe and Stavins (1994), may be defined as
any factor that may account for the “gap”, while Brown (2001) defines market barriers as
obstacles that are not based on market failures but which nonetheless contribute to the slow diffusion
and adoption of energy-efficient measures. Barriers that may be categorized as market barriers
are, for example, hidden costs, limited access to capital, risk, and heterogeneity. These
barriers are presented below.

1.2.1 Hidden costs
Hidden costs are often used as an explanatory variable for the “gap” (DeCanio, 1998). In
short, the argument is that there are high costs associated with information-seeking, meeting
with sellers, writing contracts and other such activities; if these costs are higher than the
actual profit from implementation, they inhibit investment. Accordingly, cost-effective
measures are not cost-effective when such costs associated with the investment are included.
A study by Hein and Blok (1994) found that hidden costs in large energy-intensive
industrial firms ranged from three to eight percent of total investment costs. In smaller, non-
energy-intensive firms, such costs are thus likely to be even higher. Hidden costs are a
frequently used argument against the existence of an energy efficiency gap; it is argued that
engineering-economic models are not able to see the full cost of an energy efficiency
measure (Sorrell et al., 2000).

1.2.2 Limited access to capital
Technologies that are energy-efficient are often more expensive to purchase than alternative
technologies (Almeida, 1998). Moreover, obtaining additional capital in order to invest in
energy-efficient technology may be problematic. Apart from low liquidity, limited access to
capital may also arise due to restrictions on lending money (Hirst and Brown, 1990).
Sometimes such restrictions may be self-imposed.

1.2.3 Risk
Even though, for example, managers know what the capital cost is for an energy efficiency
investment, there can be uncertainty about the long-term savings in operating costs; this
means the investment poses a risk. Such concerns have been found to be very important to
decision-makers (Hirst and Brown, 1990).
Stern and Aronson (1984) also identify risk as a barrier to energy efficiency, since accurate
estimates of the net costs of implementing energy efficiency measures depend on future
economic conditions in general, and on future energy prices and availability in particular.
Energy prices have fluctuated as long as there has been a market for energy, leading to
Energy Effciency 54

perceptions of uncertainty about future prices. How are consumers to make “rational” choices
about the purchase of new energy-using systems such as cars, heating equipment, new buildings, and
motors when the basis for estimating long-term operating costs is so uncertain? ... Uncertainty about
fuel prices is a barrier to investment in both the manufacture and purchase of energy-efficient systems
(Hirst and Brown, 1990). Studies among small and medium-sized enterprises have found
that some may not even be able to reduce uncertainty to a calculated risk due to a lack of
time and money to calculate the required estimates (Stern and Aronson, 1984).

1.2.4 Heterogeneity
The heterogeneity barrier is associated with the fact that even if a given technology is cost-
effective on average, it will most likely not be so for some individuals or firms.
Heterogeneity particularly impacts production processes of companies that often specialize
in one type of goods, and where a potential energy efficiency measure may be difficult to
implement in another company. Even though similar goods are produced, small differences
in the products, such as different size and shape, can inhibit the implementation of the
measure in another firm (Jaffe and Stavins, 1994). Heterogeneity may be an explanatory
variable for the “gap” when constructing (economic) models of a population of companies,
but is less likely to hold if site-specific information exists regarding a cost-effective energy
efficiency measure resulting from, for example, an energy audit.

1.3 Behavioural barriers
Apart from the explanations for the “gap” outlined above, there are also a number of
barriers derived from behavioural sciences that explain the “gap”, such as the form of
information, credibility and trust, values, inertia, and bounded rationality. These barriers
are presented below.

1.3.1 Form of information
One barrier to energy efficiency is the form of information, meaning that information does
not always receive as much attention as anticipated, since people are (often) not active
information-seekers but rather selective about attending to and assimilating information.
Research points out some characteristics in the way information is assimilated; some people,
for example, are more likely to remember information if it is specific and presented in a
vivid and personalized manner, and comes from a person who is similar to the receiver
(Stern and Aronson, 1984; Palm, 2009, 2010).

1.3.2 Credibility and trust
Another factor that may inhibit adoption is the receiver’s perceived credibility of and trust
in the information provider. Energy users cannot always easily gain accurate information
about the ultimate comparative cost of different investment options; they will rely on the
most credible available information. The following example from the household sector may
illustrate this. Pamphlets describing how to save energy in home air conditioning systems
were sent out to 1,000 households in New York. Fifty percent of the households received the
information in a mailing from the local electricity utility, and the other half received it from
the state regulatory agency for utilities. The following month, households that had received

the pamphlet from the state agency used about eight percent less electricity than the
households that had received the same pamphlet from the local electricity utility (Stern and
Aronson, 1984). The effective spread of information thus depends on a trustworthy
information provider. As regards the industry, intermediaries such as sector organizations
or consultants may play an important role, as these entities or individuals often tend to be
regarded as trustworthy (Ramirez et al., 2005; Stern and Aronson, 1984).

1.3.3 Values
Values such as helping others, concern for the environment and a moral commitment to use
energy more efficiently are influencing individuals and groups of individuals to adopt
energy efficiency measures. However, studies of households indicate that norms only have a
strong impact on cost-free energy efficiency and energy conservation measures (Stern and
Aronson, 1984). A study by Aronson and O’Leary (1983) on showering in a university
building showed that the number of students taking short, energy-saving showers increased
from six percent when a sign encouraging short showers was put up, to 19 percent when an
intrusive sign was used, to 49 percent when the researchers used a student to set an example
for others by always turning off the water and soaping up whenever someone came into the
facility, and to 67 percent when two students serving as examples were used (Aronsson and
O’Leary, 1983). Consequently, a lack of values related to energy efficiency may inhibit
measures from being undertaken.

1.3.4 Inertia
In short, inertia means that individuals and organizations are, in part, creatures of habit and
established routines, which may make it difficult to create changes to such behaviours and
habits. This is stated as an explanatory variable to the “gap”. People work to reduce
uncertainty and change in their environments, and avoid or ignore problems (Stern and
Aronson, 1984). Also, people who have recently made an important decision often seek to
justify that decision afterwards—convincing themselves and others that the decision was
correct. This description of inertia may partially explain the failure of many energy users to
take economically justifiable actions to save energy; energy efficiency also often begins with
small commitments that later lead to greater ones (Stern and Aronson, 1984).

1.3.5 Bounded rationality
Another explanation for why cost-effective energy efficiency measures are not undertaken is
bounded rationality (Simon, 1957). Most types of market failures are concerned with
problems in the economic environment that impede economic efficiency even when
assuming fully rational agents—that is, utility-maximizing consumers and profit-
maximizing firms (Palm and Thollander, 2010). In the case of energy efficiency-related
decisions, this hypothesis formally requires decision-makers to solve what may be
extremely complex optimization problems in order to obtain the lowest-cost provision of
energy services (Sanstad and Howarth, 1994). Studies of organizational decision-making
identify two major features of organizations that affect the linkage of a simple rational view
to their actions. First, the organization is not a single actor but rather consists of many actors
with different, sometimes conflicting, objectives. The interests of one employee or
department may, for example, be in conflict with those of others. Second, according to
Categorizing Barriers to Energy Effciency: An Interdisciplinary Perspective 55

perceptions of uncertainty about future prices. How are consumers to make “rational” choices
about the purchase of new energy-using systems such as cars, heating equipment, new buildings, and
motors when the basis for estimating long-term operating costs is so uncertain? ... Uncertainty about
fuel prices is a barrier to investment in both the manufacture and purchase of energy-efficient systems
(Hirst and Brown, 1990). Studies among small and medium-sized enterprises have found
that some may not even be able to reduce uncertainty to a calculated risk due to a lack of
time and money to calculate the required estimates (Stern and Aronson, 1984).

1.2.4 Heterogeneity
The heterogeneity barrier is associated with the fact that even if a given technology is cost-
effective on average, it will most likely not be so for some individuals or firms.
Heterogeneity particularly impacts production processes of companies that often specialize
in one type of goods, and where a potential energy efficiency measure may be difficult to
implement in another company. Even though similar goods are produced, small differences
in the products, such as different size and shape, can inhibit the implementation of the
measure in another firm (Jaffe and Stavins, 1994). Heterogeneity may be an explanatory
variable for the “gap” when constructing (economic) models of a population of companies,
but is less likely to hold if site-specific information exists regarding a cost-effective energy
efficiency measure resulting from, for example, an energy audit.

1.3 Behavioural barriers
Apart from the explanations for the “gap” outlined above, there are also a number of
barriers derived from behavioural sciences that explain the “gap”, such as the form of
information, credibility and trust, values, inertia, and bounded rationality. These barriers
are presented below.

1.3.1 Form of information
One barrier to energy efficiency is the form of information, meaning that information does
not always receive as much attention as anticipated, since people are (often) not active
information-seekers but rather selective about attending to and assimilating information.
Research points out some characteristics in the way information is assimilated; some people,
for example, are more likely to remember information if it is specific and presented in a
vivid and personalized manner, and comes from a person who is similar to the receiver
(Stern and Aronson, 1984; Palm, 2009, 2010).

1.3.2 Credibility and trust
Another factor that may inhibit adoption is the receiver’s perceived credibility of and trust
in the information provider. Energy users cannot always easily gain accurate information
about the ultimate comparative cost of different investment options; they will rely on the
most credible available information. The following example from the household sector may
illustrate this. Pamphlets describing how to save energy in home air conditioning systems
were sent out to 1,000 households in New York. Fifty percent of the households received the
information in a mailing from the local electricity utility, and the other half received it from
the state regulatory agency for utilities. The following month, households that had received

the pamphlet from the state agency used about eight percent less electricity than the
households that had received the same pamphlet from the local electricity utility (Stern and
Aronson, 1984). The effective spread of information thus depends on a trustworthy
information provider. As regards the industry, intermediaries such as sector organizations
or consultants may play an important role, as these entities or individuals often tend to be
regarded as trustworthy (Ramirez et al., 2005; Stern and Aronson, 1984).

1.3.3 Values
Values such as helping others, concern for the environment and a moral commitment to use
energy more efficiently are influencing individuals and groups of individuals to adopt
energy efficiency measures. However, studies of households indicate that norms only have a
strong impact on cost-free energy efficiency and energy conservation measures (Stern and
Aronson, 1984). A study by Aronson and O’Leary (1983) on showering in a university
building showed that the number of students taking short, energy-saving showers increased
from six percent when a sign encouraging short showers was put up, to 19 percent when an
intrusive sign was used, to 49 percent when the researchers used a student to set an example
for others by always turning off the water and soaping up whenever someone came into the
facility, and to 67 percent when two students serving as examples were used (Aronsson and
O’Leary, 1983). Consequently, a lack of values related to energy efficiency may inhibit
measures from being undertaken.

1.3.4 Inertia
In short, inertia means that individuals and organizations are, in part, creatures of habit and
established routines, which may make it difficult to create changes to such behaviours and
habits. This is stated as an explanatory variable to the “gap”. People work to reduce
uncertainty and change in their environments, and avoid or ignore problems (Stern and
Aronson, 1984). Also, people who have recently made an important decision often seek to
justify that decision afterwards—convincing themselves and others that the decision was
correct. This description of inertia may partially explain the failure of many energy users to
take economically justifiable actions to save energy; energy efficiency also often begins with
small commitments that later lead to greater ones (Stern and Aronson, 1984).

1.3.5 Bounded rationality
Another explanation for why cost-effective energy efficiency measures are not undertaken is
bounded rationality (Simon, 1957). Most types of market failures are concerned with
problems in the economic environment that impede economic efficiency even when
assuming fully rational agents—that is, utility-maximizing consumers and profit-
maximizing firms (Palm and Thollander, 2010). In the case of energy efficiency-related
decisions, this hypothesis formally requires decision-makers to solve what may be
extremely complex optimization problems in order to obtain the lowest-cost provision of
energy services (Sanstad and Howarth, 1994). Studies of organizational decision-making
identify two major features of organizations that affect the linkage of a simple rational view
to their actions. First, the organization is not a single actor but rather consists of many actors
with different, sometimes conflicting, objectives. The interests of one employee or
department may, for example, be in conflict with those of others. Second, according to
Energy Effciency 56

Sanstad and Howarth (1994), organizations (just like individuals) to some extent do not act
on the basis of complete information but rather make decisions by rule of thumb (Stern and
Aronson, 1984).

1.4 Organizational barriers
Apart from economic and behavioural barriers, there are also barriers such as power and
culture that emerge from organizational theory. These barriers are presented below.

1.4.1 Power
Lack of power among energy efficiency decision-makers (e.g., the energy controllers), is
often put forth as an explanatory variable for the “gap”. The low importance of energy
management within organizations leads to constraints when striving to implement energy
efficiency measures (Sorrell et al., 2000).

1.4.2 Culture
Culture is closely connected to the values of the individuals forming the culture. An
organization’s culture may be seen as the sum of each individual’s values, where the
executives’ values or the values of other workers who have influence within the
organization may have more impact on the organization’s culture than “lower status”
workers (Sorrell et al., 2000).

1.5 Different ways of categorizing barriers to energy efficiency
A review of research on barriers to energy efficiency reveals that a number of different
means of categorizing barriers exists.
A barrier model specifies three features: the objective obstacle, the subject hindered, and the
action hindered. The methodological question of how to determine a barrier model is: what
is an obstacle to whom reaching what in energy conservation (Weber, 1997)?
 What is an obstacle (persons, patterns of behaviour, attitudes, preferences, social
norms, habits, needs, organizations, cultural patterns, technical standards,
regulations, economic interests, financial incentives, etc.)
 ... is an obstacle to whom (consumers, tenants, workers, clerks, managers, voters,
politicians, local administration, parties, trade unions, households, firms, non-
governmental organizations)
 ... reaching what (buying more efficient equipment, retro-fitting, decreasing an
energy tax, establishing a public traffic network, improving operating practices,
etc.)
Different ways of categorizing barriers to energy efficiency have been developed. Sorrell et
al. (2000) distinguish three main categories: market failures, organizational failures and non-
failures, while Weber (1997) classifies the barriers as institutional, economic, organizational
and behavioral barriers. Hirst and Brown (1990) made yet another distinction of barriers to
energy efficiency, which divides the barriers into two broad categories: structural barriers
and behavioral barriers.

In the following section we will discuss another way of understanding technological
development and changes in organizations, namely transition theory and socio-technical
regimes.

2. Socio-technical regimes
At this stage it is useful to introduce Geels et.al.’s evolutionary model for socio-technical
change, which focuses on the dynamics in changing artifacts, technologies, regimes and
overall society. The model relies on the work of science and technology studies (STS), which
argues that technological and social change are interrelated.
In this model, radical novelties are developed in special spaces or technological niches,
where they are sheltered from mainstream competition (Schot and Geels, 2008). These can
be small market niches or technological niches where resources are provided by public
subsidies. Niches need protection because new technologies initially have low
price/performance ratios. Since small networks of actors protect the niches, when initiating
new technology building social networks is a vital activity (Verbong and Geels, 2007).
Niches form the micro level at which radical novelties emerge. The meso level is the regime
level, and includes routines, knowledge, defining problems and so on embedded in
institutions and infrastructures (Shove 2003). The macro level is the socio-technical
landscape, which is the environment that changes slowly. Verbong and Geels (2007)
describe the relationship between the three levels as a “nested hierarchy”. New technologies
have problems breaking through because of deep-rooted, established regimes. Transition
only takes place when all three levels link up and reinforce each other.
Geels (2004) has developed Nelson and winter’s “technological regimes” and discusses
socio-technical regimes. Technological regimes refer to cognitive routines that are shared in
a community of engineers and that guides research and development activities. The
technological regime is the rule-set embedded in “engineering practices, production process
technologies, product characteristics, skills and procedures, ways of handling relevant
artefacts and persons, ways of defining problems; all of them embedded in institutions and
infrastructures”. It highlights the fact that engineers act in a social context of social
structures, regulations and norms (Geels and Kemp, 2007, pp 443). Technological regimes
are broadened to include socio-technical regimes by including the institutional and market
aspects needed to make the technical regime work. A socio-technical regime is characterized
by the set of rules that guide technical design, as well as the rules that shape market
development such as user preferences and rules for regulating these markets (Schot and
Geels, 2007). The use of socio-technical regimes also implies the existence of different
regimes and the existence of a connection and mutual dependency between them. In a
company, different social groups can be distinguished by their own special features. Actors
within these groups then share a set of rules, or a regime. Because different groups share
different rules, it is possible to distinguish different regimes, such as technological regimes,
science regimes, and financial regimes and so on. They share aims, values, problems,
agendas, professional journals, etc. However, rules are not just linked within regimes but
also between regimes, and regimes influence each other; this is why socio-technical regimes
are a better concept for explaining this (Geels, 2004). When regimes are widened to socio-
technical regimes, they include interaction with other social groups, besides engineers and
firms, in society such as users, policy-makers and social groups. Regimes not only refer to
Categorizing Barriers to Energy Effciency: An Interdisciplinary Perspective 57

Sanstad and Howarth (1994), organizations (just like individuals) to some extent do not act
on the basis of complete information but rather make decisions by rule of thumb (Stern and
Aronson, 1984).

1.4 Organizational barriers
Apart from economic and behavioural barriers, there are also barriers such as power and
culture that emerge from organizational theory. These barriers are presented below.

1.4.1 Power
Lack of power among energy efficiency decision-makers (e.g., the energy controllers), is
often put forth as an explanatory variable for the “gap”. The low importance of energy
management within organizations leads to constraints when striving to implement energy
efficiency measures (Sorrell et al., 2000).

1.4.2 Culture
Culture is closely connected to the values of the individuals forming the culture. An
organization’s culture may be seen as the sum of each individual’s values, where the
executives’ values or the values of other workers who have influence within the
organization may have more impact on the organization’s culture than “lower status”
workers (Sorrell et al., 2000).

1.5 Different ways of categorizing barriers to energy efficiency
A review of research on barriers to energy efficiency reveals that a number of different
means of categorizing barriers exists.
A barrier model specifies three features: the objective obstacle, the subject hindered, and the
action hindered. The methodological question of how to determine a barrier model is: what
is an obstacle to whom reaching what in energy conservation (Weber, 1997)?
 What is an obstacle (persons, patterns of behaviour, attitudes, preferences, social
norms, habits, needs, organizations, cultural patterns, technical standards,
regulations, economic interests, financial incentives, etc.)
 ... is an obstacle to whom (consumers, tenants, workers, clerks, managers, voters,
politicians, local administration, parties, trade unions, households, firms, non-
governmental organizations)
 ... reaching what (buying more efficient equipment, retro-fitting, decreasing an
energy tax, establishing a public traffic network, improving operating practices,
etc.)
Different ways of categorizing barriers to energy efficiency have been developed. Sorrell et
al. (2000) distinguish three main categories: market failures, organizational failures and non-
failures, while Weber (1997) classifies the barriers as institutional, economic, organizational
and behavioral barriers. Hirst and Brown (1990) made yet another distinction of barriers to
energy efficiency, which divides the barriers into two broad categories: structural barriers
and behavioral barriers.

In the following section we will discuss another way of understanding technological
development and changes in organizations, namely transition theory and socio-technical
regimes.

2. Socio-technical regimes
At this stage it is useful to introduce Geels et.al.’s evolutionary model for socio-technical
change, which focuses on the dynamics in changing artifacts, technologies, regimes and
overall society. The model relies on the work of science and technology studies (STS), which
argues that technological and social change are interrelated.
In this model, radical novelties are developed in special spaces or technological niches,
where they are sheltered from mainstream competition (Schot and Geels, 2008). These can
be small market niches or technological niches where resources are provided by public
subsidies. Niches need protection because new technologies initially have low
price/performance ratios. Since small networks of actors protect the niches, when initiating
new technology building social networks is a vital activity (Verbong and Geels, 2007).
Niches form the micro level at which radical novelties emerge. The meso level is the regime
level, and includes routines, knowledge, defining problems and so on embedded in
institutions and infrastructures (Shove 2003). The macro level is the socio-technical
landscape, which is the environment that changes slowly. Verbong and Geels (2007)
describe the relationship between the three levels as a “nested hierarchy”. New technologies
have problems breaking through because of deep-rooted, established regimes. Transition
only takes place when all three levels link up and reinforce each other.
Geels (2004) has developed Nelson and winter’s “technological regimes” and discusses
socio-technical regimes. Technological regimes refer to cognitive routines that are shared in
a community of engineers and that guides research and development activities. The
technological regime is the rule-set embedded in “engineering practices, production process
technologies, product characteristics, skills and procedures, ways of handling relevant
artefacts and persons, ways of defining problems; all of them embedded in institutions and
infrastructures”. It highlights the fact that engineers act in a social context of social
structures, regulations and norms (Geels and Kemp, 2007, pp 443). Technological regimes
are broadened to include socio-technical regimes by including the institutional and market
aspects needed to make the technical regime work. A socio-technical regime is characterized
by the set of rules that guide technical design, as well as the rules that shape market
development such as user preferences and rules for regulating these markets (Schot and
Geels, 2007). The use of socio-technical regimes also implies the existence of different
regimes and the existence of a connection and mutual dependency between them. In a
company, different social groups can be distinguished by their own special features. Actors
within these groups then share a set of rules, or a regime. Because different groups share
different rules, it is possible to distinguish different regimes, such as technological regimes,
science regimes, and financial regimes and so on. They share aims, values, problems,
agendas, professional journals, etc. However, rules are not just linked within regimes but
also between regimes, and regimes influence each other; this is why socio-technical regimes
are a better concept for explaining this (Geels, 2004). When regimes are widened to socio-
technical regimes, they include interaction with other social groups, besides engineers and
firms, in society such as users, policy-makers and social groups. Regimes not only refer to
Energy Effciency 58

cognitive routines and belief systems, but also to regulative rules and normative roles. From
this perspective, different regimes are relatively autonomous, but also interdependent. A
socio-technical regime thus binds producers, users and regulators together.
As mentioned above, the socio-technical regime forms the meso level, which accounts for
the stability of existing large-scale systems such as energy systems. The macro level is
formed by the socio-technical landscape, and cannot be under direct influence of niche and
regime actors. Changes at the landscape level occur slowly. Niche actors hope that novelties
will eventually be used in the regime. Niche actors can contribute to changes in the practices
and routines of existing regime actors. Sometimes niches can also replace the existing
regime. It is not easy, however, to replace an established regime, not least because of lock-in
effects wherein new technology often needs to fit into existing system solutions (Schot and
Geels, 2008).
Socio-technical regimes highlight the fact that actors are embedded in structures that shape
their preferences, aims and strategies. But from this perspective, actors also have agency and
perform conscious and strategic actions. The model confirms Gidden’s duality of structure,
and when that structure produces and mediates action. Actors can then act upon and
restructure these systems (Geels, 2004). Regimes then implement and (re)produce rules in
social activities that take place in local practices. By implementing shared rule systems, the
regime actors generate patterns of activity that are similar across different local practices.
There may be variation, however, between local practices due to the fact that there are
differences between group members, so regimes can have somewhat different strategies,
resources, problems and aims. Strategies, aims and the like are also not very flexible within a
regime, and undergo only incremental change over time (Geels, 2004). In addition, incremental
innovation still occurs in stable regimes and is important because these changes can
accumulate and result in major performance improvements over time (Geels and Kemp, 2007).
A dominant regime can be forced to restructure and invest in new technical directions. For
example, changes in the socio-technical landscape can put pressure on the regime. Climate
change has forced the energy and transport sector to find new technical strategies. Internal
technical problems, change in user preferences and negative externalities such as health
risks may also trigger actors to act. Competitive games between firms are another example
of developments that can open up a regime (Geels, 2004).
If we cross-pollinate barriers theories with ideas from transition theories and socio-technical
regimes, we have a new categorization of barriers and, therefore, a new way of reflecting on
and discussing efficiency gaps. This will be discussed in the following section.

3. Conclusions: A proposed structure for empirical studies on barriers to
energy efficiency
How we define a problem determines whether we can solve it; this is elementary knowledge
in all of the sciences. Clear definitions are the foundation for all innovative thoughts, which
is why it is important to discuss how barriers to energy efficiency can be categorized in
potentially different ways. In an attempt to categorize barriers to energy efficiency, the 15
theoretical barriers are divided into three different categories, depending on each barrier’s
system complexity (see table 2). In the first category—the technical system—the results are
quite restricted to technology and its associated costs. In the second category—the
technological regime—the results are influenced by human factors but nevertheless coupled

to the technology in question. In the third category—the socio-technical regime—the results
are heavily influenced by human factors, and less influenced by the technology in question.

Classification
Theoretical Barriers
Access to capital
(Hirst and Brown, 1990)
Heterogeneity
(Jaffe and Stavins, 1994)
Hidden costs
(Ostertag, 1999)
Risk
(Hirst and Brown, 1990)
Imperfect information
(Howarth and Andersson, 1993)
Adverse selection
(Sanstad and Howarth, 1994)
Split incentives
(Jaffe and Stavins, 1994)
Form of information
(Stern and Aronsson, 1984)
Credibility and trust
(Stern and Aronsson, 1984)
Principal-agent relationship
(Jaffe and Stavins, 1994)
Values
(Stern, 1992)
Inertia
(Stern and Aronsson, 1984)
Bounded rationality
(Sanstad and Howarth, 1994)
Power
(Sorrell et al., 2000)
Culture
(Sorrell et al., 2000)
The technical system
The technological regime
The socio-technical regime

Table 2. Proposed classification of barriers to energy efficiency.

Re-defining how we should categorize barriers could open up new ways of looking at the
problem, which in turn might lead to other suggestions for addressing the energy efficiency
gap. Energy efficiency problems are multi-faceted and should be approached accordingly. If
a barrier is identified as belonging to a technological regime or a socio-technical regime, it
should be approached differently and addressed via different policy means. If a barrier is
seen as belonging to a technological regime, then more information on existing energy
efficient measures could be a possible solution. If a barrier is more related to a socio-
technical perspective on barriers, then aspects such as corporate culture and established
Categorizing Barriers to Energy Effciency: An Interdisciplinary Perspective 59

cognitive routines and belief systems, but also to regulative rules and normative roles. From
this perspective, different regimes are relatively autonomous, but also interdependent. A
socio-technical regime thus binds producers, users and regulators together.
As mentioned above, the socio-technical regime forms the meso level, which accounts for
the stability of existing large-scale systems such as energy systems. The macro level is
formed by the socio-technical landscape, and cannot be under direct influence of niche and
regime actors. Changes at the landscape level occur slowly. Niche actors hope that novelties
will eventually be used in the regime. Niche actors can contribute to changes in the practices
and routines of existing regime actors. Sometimes niches can also replace the existing
regime. It is not easy, however, to replace an established regime, not least because of lock-in
effects wherein new technology often needs to fit into existing system solutions (Schot and
Geels, 2008).
Socio-technical regimes highlight the fact that actors are embedded in structures that shape
their preferences, aims and strategies. But from this perspective, actors also have agency and
perform conscious and strategic actions. The model confirms Gidden’s duality of structure,
and when that structure produces and mediates action. Actors can then act upon and
restructure these systems (Geels, 2004). Regimes then implement and (re)produce rules in
social activities that take place in local practices. By implementing shared rule systems, the
regime actors generate patterns of activity that are similar across different local practices.
There may be variation, however, between local practices due to the fact that there are
differences between group members, so regimes can have somewhat different strategies,
resources, problems and aims. Strategies, aims and the like are also not very flexible within a
regime, and undergo only incremental change over time (Geels, 2004). In addition, incremental
innovation still occurs in stable regimes and is important because these changes can
accumulate and result in major performance improvements over time (Geels and Kemp, 2007).
A dominant regime can be forced to restructure and invest in new technical directions. For
example, changes in the socio-technical landscape can put pressure on the regime. Climate
change has forced the energy and transport sector to find new technical strategies. Internal
technical problems, change in user preferences and negative externalities such as health
risks may also trigger actors to act. Competitive games between firms are another example
of developments that can open up a regime (Geels, 2004).
If we cross-pollinate barriers theories with ideas from transition theories and socio-technical
regimes, we have a new categorization of barriers and, therefore, a new way of reflecting on
and discussing efficiency gaps. This will be discussed in the following section.

3. Conclusions: A proposed structure for empirical studies on barriers to
energy efficiency
How we define a problem determines whether we can solve it; this is elementary knowledge
in all of the sciences. Clear definitions are the foundation for all innovative thoughts, which
is why it is important to discuss how barriers to energy efficiency can be categorized in
potentially different ways. In an attempt to categorize barriers to energy efficiency, the 15
theoretical barriers are divided into three different categories, depending on each barrier’s
system complexity (see table 2). In the first category—the technical system—the results are
quite restricted to technology and its associated costs. In the second category—the
technological regime—the results are influenced by human factors but nevertheless coupled

to the technology in question. In the third category—the socio-technical regime—the results
are heavily influenced by human factors, and less influenced by the technology in question.

Classification
Theoretical Barriers
Access to capital
(Hirst and Brown, 1990)
Heterogeneity
(Jaffe and Stavins, 1994)
Hidden costs
(Ostertag, 1999)
Risk
(Hirst and Brown, 1990)
Imperfect information
(Howarth and Andersson, 1993)
Adverse selection
(Sanstad and Howarth, 1994)
Split incentives
(Jaffe and Stavins, 1994)
Form of information
(Stern and Aronsson, 1984)
Credibility and trust
(Stern and Aronsson, 1984)
Principal-agent relationship
(Jaffe and Stavins, 1994)
Values
(Stern, 1992)
Inertia
(Stern and Aronsson, 1984)
Bounded rationality
(Sanstad and Howarth, 1994)
Power
(Sorrell et al., 2000)
Culture
(Sorrell et al., 2000)
The technical system
The technological regime
The socio-technical regime

Table 2. Proposed classification of barriers to energy efficiency.

Re-defining how we should categorize barriers could open up new ways of looking at the
problem, which in turn might lead to other suggestions for addressing the energy efficiency
gap. Energy efficiency problems are multi-faceted and should be approached accordingly. If
a barrier is identified as belonging to a technological regime or a socio-technical regime, it
should be approached differently and addressed via different policy means. If a barrier is
seen as belonging to a technological regime, then more information on existing energy
efficient measures could be a possible solution. If a barrier is more related to a socio-
technical perspective on barriers, then aspects such as corporate culture and established
Energy Effciency 60

internal values should be problematized and highlighted. In other words, how we perceive
and define these barriers will lead to different solutions for overcoming the barriers and,
ultimately, to different policy recommendations.
Finding solutions to the energy efficiency gap is vital for solving the climate change
problem. To define and redefine the empirically identified barriers is therefore important for
challenging existing solutions and developing new, creative ways of approaching
companies and other actors. Employing this categorization of barriers would lead to a
greater focus on social practices in companies and existing routines in decision-making and
industrial processes.

4. References
Almeida, E. L. (1998). Energy efficiency and the limits of market forces: The example of the
electric motor market in France. Energy Policy, 26, 8, 643–653, ISSN 0301-4215.
Aronson, E., O’Leary, M. (1983). The relative effectiveness of models and prompts on energy
conservation: field experiment in a shower room. Journal of Environmental Systems,
12, 3, 219-224, ISSN 0047-2433.
Blumstein, C., Krieg, B., Schipper, L., York, C.M. (1980). Overcoming social and institutional
barriers to energy conservation. Energy, 5, 355-371, ISSN 0144-2600.
Brown, M.A. (2001). Market failures and barriers as a basis for clean energy policies., Energy
Policy, 29, 14, 1197-1207, ISSN 0301-4215.
de Groot, H., Verhoef, E., Nijkamp, P. (2001). Energy saving by firms: decision-making,
barriers and policies. Energy Economics, 23, 6, 717-740, ISSN 0140-9833.
DeCanio, S. (1998). The efficiency paradox: bureaucratic and organizational barriers to
profitable energy-saving investments. Energy Policy, 26, 5, 441-458, ISSN 0301-4215.
DeCanio, S. (1993). Barriers within firms to energy efficient investments. Energy Policy, 9, 1,
906-914, ISSN 0301-4215.
Geels, F. (2004) From Sectoral systems of innovation to socio-technical systems. Insights
about dynamics and change from sociology and institutional theory. Research policy,
33, 897-920, ISSN 0048-7333.
Geels, F and Kemp, R. (2007). Dynamics in socio-technical systems: Typology of change
processes and contrasting case studies. Technology in Society, 29, 441-455, ISSN 0160-
791x.
Gruber, E., Brand, M. (1991). Promoting energy conservation in small and medium-sized
companies. Energy Policy, 19, 3, 279-287, ISSN 0301-4215.
Hein, L., Blok, K. (1995). Transaction costs of energy efficiency improvement. In Proceedings
of the 1995 ECEEE summer study, Panel 2, 1-8.
Hirst, E., Brown, M., A.( 1990). Closing the efficiency gap: barriers to the efficient use of
energy. Resources, Conservation and Recycling, 3, 4, 267-281, ISSN 0921-3449.
Howarth, R., Andersson, B. (1993). Market barriers to energy efficiency. Energy Economics,
15, 4), 262-272, ISSN 0140-9833.
Jaffe, A.B., Stavins, R.N. (1994). The energy efficiency gap: what does it mean? Energy Policy,
22, 10, 60-71, ISSN 0301-4215.

Ostertag, K. (1999). Transaction Costs of Raising Energy Efficiency. In: Proceedings of the
2007 IEA international Workshop on Technologies to Reduce Greenhouse gas
Emissions: Engineering-Economic Analyses of Conserved Energy and Carbon.
Washington DC, 5-7 May 1999.
Palm, J. (2009). Placing barriers to industrial energy efficiency in a social context: a
discussion of lifestyle categorisation. Energy Efficiency, 2, 3, 263-270, ISSN 1570-646x.
Palm, J. (2010). The public-private divide in household bahavior. How far into the home can
energy guidance reach? Energy Policy, 38, 6, 2858-2864, ISSN 0301-4215.
Palm, J. and Thollander, P. (2010). An interdisciplinary perspective on industrial energy
efficiency. Applied Energy 87, 10, 3255-3261, ISSN 0306-2619.
Ramirez, C.A., Patel, M., Blok, K. (2005). The non-energy intensive manufacturing sector. An
energy analysis relating to the Netherlands. Energy, 30, 5, 749-767, ISSN 0144-2600.
Rohdin, P., Thollander, P. (2006). Barriers to and driving forces for energy efficiency in the
non-energy-intensive manufacturing industry in Sweden, Energy 31, 12, 1836-1844,
ISSN 0144-2600.
Rohdin, P., Thollander, P., Solding, P., 2007. Barriers to and drivers for energy efficiency in
the Swedish foundry industry. Energy Policy doi: 10.1016 35, 1, 672-677,
ISSN 0301-4215.
Sanstad, A., Howarth, R.,(1994). ‘Normal’ markets, market imperfections and energy
efficiency. Energy Policy, 10, 811-818, ISSN 0301-4215.
Schleich, J., Gruber, E. (2008). Beyond case studies: Barriers to energy efficiency in commerce
and the services sector. Energy Economics, 30, 2, 449-464, ISSN 0140-9833.
Schleich, J. (2004). Do energy audits help reduce barriers to energy efficiency? An empirical
analysis for Germany. International Journal of Energy Technology and Policy, 2, 3, 226-
239, ISSN 1472-8923.
Schot, J and Geels, F. (2007). Niches in evolutionary theories of technical change. A critical
survey of the literature. Journal of Evolutionary Economics, 17, 605-622, ISSN 0936-
9937.
Schot, J and Geels, F. (2008) Strategic niche management and sustainable innovation
journeys: theory, findings, research agenda and policy. Technology Analysis &
Strategig Management, 20, 5, 537-554, ISSN 0953-7325.
Shove, E. (2003). Users, Technologies and Expectations of Comfort, Cleanliness and
Convenience. Innovation, 16, 2, 193-205, ISSN 1469-8412.
Simon, H.A. (1957). Models of Man. Wiley, London.
Sorrell S., O'Malley, E., Schleich, J., Scott, S. (2004). The Economics of Energy Efficiency -
Barriers to Cost-Effective Investment, Edward Elgar, Cheltenham.
Sorrell, S., Schleich, J., Scott, S., O’Malley, E., Trace, F., Boede, E., Ostertag, K. Radgen, P.
(2000). Reducing Barriers to Energy Efficiency in Public and Private Organizations.
Retrieved October 8, 2007, from the SPRU’s (Science and Technology Policy
Research) Retrieved October 8, 2007, from: http://www.sussex. ac.uk/Units/spru
/publications/reports/ barriers/final.html.
Stern, P.C. (1992). What Psychology Knows About Energy Conservation. American
Psychologist, 47, 10, 1224-1232, ISSN 0003-066x.
Stern, P.C., Aronson, E. (1984, Eds). Energy Use: The Human Dimension, W.H Freeman,
0716716216, New York.
Categorizing Barriers to Energy Effciency: An Interdisciplinary Perspective 61

internal values should be problematized and highlighted. In other words, how we perceive
and define these barriers will lead to different solutions for overcoming the barriers and,
ultimately, to different policy recommendations.
Finding solutions to the energy efficiency gap is vital for solving the climate change
problem. To define and redefine the empirically identified barriers is therefore important for
challenging existing solutions and developing new, creative ways of approaching
companies and other actors. Employing this categorization of barriers would lead to a
greater focus on social practices in companies and existing routines in decision-making and
industrial processes.

4. References
Almeida, E. L. (1998). Energy efficiency and the limits of market forces: The example of the
electric motor market in France. Energy Policy, 26, 8, 643–653, ISSN 0301-4215.
Aronson, E., O’Leary, M. (1983). The relative effectiveness of models and prompts on energy
conservation: field experiment in a shower room. Journal of Environmental Systems,
12, 3, 219-224, ISSN 0047-2433.
Blumstein, C., Krieg, B., Schipper, L., York, C.M. (1980). Overcoming social and institutional
barriers to energy conservation. Energy, 5, 355-371, ISSN 0144-2600.
Brown, M.A. (2001). Market failures and barriers as a basis for clean energy policies., Energy
Policy, 29, 14, 1197-1207, ISSN 0301-4215.
de Groot, H., Verhoef, E., Nijkamp, P. (2001). Energy saving by firms: decision-making,
barriers and policies. Energy Economics, 23, 6, 717-740, ISSN 0140-9833.
DeCanio, S. (1998). The efficiency paradox: bureaucratic and organizational barriers to
profitable energy-saving investments. Energy Policy, 26, 5, 441-458, ISSN 0301-4215.
DeCanio, S. (1993). Barriers within firms to energy efficient investments. Energy Policy, 9, 1,
906-914, ISSN 0301-4215.
Geels, F. (2004) From Sectoral systems of innovation to socio-technical systems. Insights
about dynamics and change from sociology and institutional theory. Research policy,
33, 897-920, ISSN 0048-7333.
Geels, F and Kemp, R. (2007). Dynamics in socio-technical systems: Typology of change
processes and contrasting case studies. Technology in Society, 29, 441-455, ISSN 0160-
791x.
Gruber, E., Brand, M. (1991). Promoting energy conservation in small and medium-sized
companies. Energy Policy, 19, 3, 279-287, ISSN 0301-4215.
Hein, L., Blok, K. (1995). Transaction costs of energy efficiency improvement. In Proceedings
of the 1995 ECEEE summer study, Panel 2, 1-8.
Hirst, E., Brown, M., A.( 1990). Closing the efficiency gap: barriers to the efficient use of
energy. Resources, Conservation and Recycling, 3, 4, 267-281, ISSN 0921-3449.
Howarth, R., Andersson, B. (1993). Market barriers to energy efficiency. Energy Economics,
15, 4), 262-272, ISSN 0140-9833.
Jaffe, A.B., Stavins, R.N. (1994). The energy efficiency gap: what does it mean? Energy Policy,
22, 10, 60-71, ISSN 0301-4215.

Ostertag, K. (1999). Transaction Costs of Raising Energy Efficiency. In: Proceedings of the
2007 IEA international Workshop on Technologies to Reduce Greenhouse gas
Emissions: Engineering-Economic Analyses of Conserved Energy and Carbon.
Washington DC, 5-7 May 1999.
Palm, J. (2009). Placing barriers to industrial energy efficiency in a social context: a
discussion of lifestyle categorisation. Energy Efficiency, 2, 3, 263-270, ISSN 1570-646x.
Palm, J. (2010). The public-private divide in household bahavior. How far into the home can
energy guidance reach? Energy Policy, 38, 6, 2858-2864, ISSN 0301-4215.
Palm, J. and Thollander, P. (2010). An interdisciplinary perspective on industrial energy
efficiency. Applied Energy 87, 10, 3255-3261, ISSN 0306-2619.
Ramirez, C.A., Patel, M., Blok, K. (2005). The non-energy intensive manufacturing sector. An
energy analysis relating to the Netherlands. Energy, 30, 5, 749-767, ISSN 0144-2600.
Rohdin, P., Thollander, P. (2006). Barriers to and driving forces for energy efficiency in the
non-energy-intensive manufacturing industry in Sweden, Energy 31, 12, 1836-1844,
ISSN 0144-2600.
Rohdin, P., Thollander, P., Solding, P., 2007. Barriers to and drivers for energy efficiency in
the Swedish foundry industry. Energy Policy doi: 10.1016 35, 1, 672-677,
ISSN 0301-4215.
Sanstad, A., Howarth, R.,(1994). ‘Normal’ markets, market imperfections and energy
efficiency. Energy Policy, 10, 811-818, ISSN 0301-4215.
Schleich, J., Gruber, E. (2008). Beyond case studies: Barriers to energy efficiency in commerce
and the services sector. Energy Economics, 30, 2, 449-464, ISSN 0140-9833.
Schleich, J. (2004). Do energy audits help reduce barriers to energy efficiency? An empirical
analysis for Germany. International Journal of Energy Technology and Policy, 2, 3, 226-
239, ISSN 1472-8923.
Schot, J and Geels, F. (2007). Niches in evolutionary theories of technical change. A critical
survey of the literature. Journal of Evolutionary Economics, 17, 605-622, ISSN 0936-
9937.
Schot, J and Geels, F. (2008) Strategic niche management and sustainable innovation
journeys: theory, findings, research agenda and policy. Technology Analysis &
Strategig Management, 20, 5, 537-554, ISSN 0953-7325.
Shove, E. (2003). Users, Technologies and Expectations of Comfort, Cleanliness and
Convenience. Innovation, 16, 2, 193-205, ISSN 1469-8412.
Simon, H.A. (1957). Models of Man. Wiley, London.
Sorrell S., O'Malley, E., Schleich, J., Scott, S. (2004). The Economics of Energy Efficiency -
Barriers to Cost-Effective Investment, Edward Elgar, Cheltenham.
Sorrell, S., Schleich, J., Scott, S., O’Malley, E., Trace, F., Boede, E., Ostertag, K. Radgen, P.
(2000). Reducing Barriers to Energy Efficiency in Public and Private Organizations.
Retrieved October 8, 2007, from the SPRU’s (Science and Technology Policy
Research) Retrieved October 8, 2007, from: http://www.sussex. ac.uk/Units/spru
/publications/reports/ barriers/final.html.
Stern, P.C. (1992). What Psychology Knows About Energy Conservation. American
Psychologist, 47, 10, 1224-1232, ISSN 0003-066x.
Stern, P.C., Aronson, E. (1984, Eds). Energy Use: The Human Dimension, W.H Freeman,
0716716216, New York.
Energy Effciency 62

Thollander, P., Ottosson, M., 2008. An energy-efficient Swedish pulp and paper industry –
exploring barriers to and driving forces for cost-effective energy efficiency
investments. Energy Efficiency 1, 1, 21-34, ISSN 1570-646x.
Thollander, P., Rohdin, P., Danestig, M., 2007. Energy policies for increased industrial
energy efficiency: Evaluation of a local energy programme for manufacturing
SMEs. Energy Policy 35, 11, 5774-5783, ISSN 0301-4215.
Verbong, G and Geels, F. (2007). The ongoing energy transition: Lessons from a socio-
technical multi-level analysis of the Dutch electricity system (1960-2004). Energy
Policy, 35, 1025-1037, ISSN 0301-4215.
Weber, L. (1997). Some reflections on barriers to the efficient use of energy. Energy Policy, 25,
10, 833-835, ISSN 0301-4215.
York, C.M., Blumstein, C., Krieg, B., Schipper, L. (1978). Bibliography in institutional barriers to
energy conservation. Lawrence Berkeley Laboratory and University of California,
Berkeley.
Factors infuencing energy effciency in the German and Colombian manufacturing industries 63
Factors infuencing energy effciency in the German and Colombian
manufacturing industries
Clara Inés Pardo Martínez
X

Factors influencing energy efficiency
in the German and Colombian
manufacturing industries

Clara Inés Pardo Martínez
University of Wuppertal, Wuppertal Institute and University of La Salle
Germany and Colombia

1. Introduction
Energy is a basic factor for industrial production, and the level of electricity consumption is used
to measure the progress and economic development of nations. Globally, growing population,
industrialisation and rising living standards have substantially increased dependence on energy.
As a result, the development of conventional energy resources, the search for new or renewable
energy sources, energy conservation (using less energy), and energy efficiency (same service or
output, less energy) have become unavoidable topics within politics.
Generally, an ideal policy cycle sees a given policy formulated, implemented, monitored and
evaluated to verify its effectiveness and fulfilment of the proposed objectives and in accordance
with the results of this evaluation, the policy is then kept, reformulated or abolished. In this
cycle—and above all, in industrial energy politics—it is important that the policy makers
recognise the influence of economic, technical and political factors and have an understanding of
the mechanisms that determine energy efficiency performance such that the instruments and
strategy they formulate become successful.
Strategies and instruments developers drafting an energy policy need to understand the
behaviour of the manufacturing industry with respect to energy consumption in order to (i)
motivate, (ii) target energy actions that will be adopted, and (iii) develop energy saving and
energy efficiency actions and technologies that will be of interest (Kant, 1995 and Thollander et
al., 2007). The quantity and quality of energy conservation support or energy efficiency programs
will depend on perceived interest and as well as the need for energy conservation changes.
There are limited studies and information currently available on the perception of approach to
energy efficiency in companies. Therefore, this study seeks to analyse the factors and strategies
that address energy efficiency in the manufacturing industries. This information may be useful
for energy policy and program development as well as pollution prevention and energy
efficiency strategies. The research questions that guide this chapter are:
 What is the role of energy consumption and energy efficiency in business strategies in
the manufacturing industries?
 What are the variables of political factors that may have more influence on energy
efficiency performance?
4
Energy Effciency 64

 What are the strategies and instruments that may generate better results to improve
energy efficiency in the manufacturing industries?
These questions were investigated in this study by means of the opinions and expectations of the
main stakeholders (associations and representative firms in Germany and Colombia) through a
questionnaire and analysis of literature.
This chapter is structured as follows. In section 2, examines energy efficiency policy in both
countries. Section 3 shows the methodology used in this study. Section 4 analyses changes in
energy efficiency in German and Colombian manufacturing industries. Results and discussion
appear in section 5 while the section 6 shows different strategies and recommendations for an
effective energy efficiency policy in the Colombian manufacturing industry. The main
conclusions of the study are presented in section 7.

2. General characteristics of energy efficiency policy in Germany and
Colombia
2.1 The German energy efficiency policy
The German energy policy is based in the commitment to the “3 Es”: energy security, economic
efficiency and environmental sustainability. In this context, Germany emphasises environment
and climate change objectives, and energy efficiency assumes increased importance in the
country’s overall energy policy. Moreover, in the last decade, the key German energy policies
have been based on the expansion of the use of renewable energy and the establishment of new
energy efficiency targets and an energy research program (IEA, 2007).
From the mid-1990s, the dominant instruments employed to improve energy efficiency in the
German manufacturing industries were voluntary agreements. Since its introduction in 2004,
however, the emissions trading system has become the most important policy measure in the
manufacturing industrial sector, and it has also provided a key incentive to raise energy
efficiency (Eichhammer, et al., 2006).
Regarding cross-cutting measures to improve energy efficiency in Germany, the main policy is
the Ecological Tax Reform, i.e., the introduction of a so-called Eco Tax on oil, gas and electricity
1
.
Additionally, the Renewable Energy Sources Act provides digressive compensation rates for new
installations for all renewable energies
2
.
The German energy efficiency policies for the manufacturing industries have worked mainly
with the following strategies:
 Voluntary agreements: the improvements in the efficiency of on-site electricity generation,
particularly combined heat and power (CHP).
 Eco-tax: Germany's red-green coalition government introduced a set of ecotaxes on 1 April
1999 designed to make energy and resource consumption more expensive while lowering
the cost of labour. Taxes on petrol and diesel, electricity, heating oil and natural gas had

1
The tax was introduced in two stages: a first tax increase from 1 April 1999 and a further
four-step increase in taxation from 2000 to 2003. There are tax reductions for some
consumers, chiefly within the manufacturing industry, agriculture and the railways. The
revenue from this tax is used for a reduction of the non-wage labour costs and the
promotion of renewable energies (Eichhammer, et al. 2006).
2
The rates are adapted to the efficiency potential of the different branches. This will provide
a strong incentive to reduce costs and increase efficiency (Eichhammer, et al. 2006).

been increased in five stages, and the bulk of the tax revenue generated used to reduce
pension insurance contributions.
 Emission trading system means to achieve ecological and economic success. It means
assuring the ecological integrity of the instrument, competition neutrality and low
transaction costs. In other words, the emission trading system makes use of market-based
mechanisms to encourage the reduction of greenhouse gas emissions in a cost-effective and
economically-efficient manner, while maintaining the environmental integrity of the
system.
 Specific Regulations such as: the Energy Performance of Buildings that seek to promote the
energy performance of buildings taking into account outdoor climatic and local conditions
as well as indoor climate requirements and cost-effectiveness, and the Minimum Energy
Performance Standards for appliances or equipments and mandatory labels that are used to
increase the energy efficiency of individual technologies.
 German CHP Law supports of cost efficient technology to reduce CO
2
emissions. This law
contains the definition of CHP electricity and heat; support mechanism for high efficiency
CHP, and mechanise to supervise reporting of CHP electricity production in CHP plants.
 Renewable Energy Sources Act creates a feed-in tariff system which requires utilities to
purchase a predetermined amount of renewable energy at a fixed price. The policy
provides economic security for investors and manufacturers and is responsible for the bulk
of Germany’s dynamic scale-up of renewable electricity capacity and equipment
production.
 Grants and loans: the Kreditanstalt für Wiederaufbau (KfW) Umweltprogramm
(Environment Program) that provides capital for investment in environmental protection
activities and the low-interest loans to SMEs that can be used to supplement the European
Recovery Programme’s Environment and Energy Saving Program.
 Technology specific rebates are programs used to promote energy management and new
energy-efficient technologies.
 Public information and advice: the sub-project under the Initiative Energieeffizienz (Energy
Efficiency Initiative) campaign, DENA, the German Energy Agency.

2.2 The Colombian energy efficiency policy
In 1991, with the introduction of the new Constitution, Colombia adopted the principles of
sustainable development as a guide to economic development and assigned to
municipalities the duty to regulate especially the industry and energy intensive activities.
The deregulation of the Colombian electricity system
3
began in the same period, as did the
restructuration of the public environmental management system
4
. These elements have
characterised the development of energy policies in this country, where the emphasis has

3
The Colombian electricity industry is characterized by a large hydroelectricity component,
close to 70%, and is considered to be one of the most open markets in the developing world,
and the market evolution with this model has been satisfactory in terms of investment,
competition, efficiency and reduction in electricity losses (Larsen et al., 2004).
4
The Colombian environmental administration characterizes to be decentralized,
democratic, participatory, fiscally solvent, and socially legitimate with measures as a system
of pollution taxes, require environmental impact assessments for large construction projects,
and institutionalize legal remedies against polluters (Blackman et al., 2006).
Factors infuencing energy effciency in the German and Colombian manufacturing industries 65

 What are the strategies and instruments that may generate better results to improve
energy efficiency in the manufacturing industries?
These questions were investigated in this study by means of the opinions and expectations of the
main stakeholders (associations and representative firms in Germany and Colombia) through a
questionnaire and analysis of literature.
This chapter is structured as follows. In section 2, examines energy efficiency policy in both
countries. Section 3 shows the methodology used in this study. Section 4 analyses changes in
energy efficiency in German and Colombian manufacturing industries. Results and discussion
appear in section 5 while the section 6 shows different strategies and recommendations for an
effective energy efficiency policy in the Colombian manufacturing industry. The main
conclusions of the study are presented in section 7.

2. General characteristics of energy efficiency policy in Germany and
Colombia
2.1 The German energy efficiency policy
The German energy policy is based in the commitment to the “3 Es”: energy security, economic
efficiency and environmental sustainability. In this context, Germany emphasises environment
and climate change objectives, and energy efficiency assumes increased importance in the
country’s overall energy policy. Moreover, in the last decade, the key German energy policies
have been based on the expansion of the use of renewable energy and the establishment of new
energy efficiency targets and an energy research program (IEA, 2007).
From the mid-1990s, the dominant instruments employed to improve energy efficiency in the
German manufacturing industries were voluntary agreements. Since its introduction in 2004,
however, the emissions trading system has become the most important policy measure in the
manufacturing industrial sector, and it has also provided a key incentive to raise energy
efficiency (Eichhammer, et al., 2006).
Regarding cross-cutting measures to improve energy efficiency in Germany, the main policy is
the Ecological Tax Reform, i.e., the introduction of a so-called Eco Tax on oil, gas and electricity
1
.
Additionally, the Renewable Energy Sources Act provides digressive compensation rates for new
installations for all renewable energies
2
.
The German energy efficiency policies for the manufacturing industries have worked mainly
with the following strategies:
 Voluntary agreements: the improvements in the efficiency of on-site electricity generation,
particularly combined heat and power (CHP).
 Eco-tax: Germany's red-green coalition government introduced a set of ecotaxes on 1 April
1999 designed to make energy and resource consumption more expensive while lowering
the cost of labour. Taxes on petrol and diesel, electricity, heating oil and natural gas had

1
The tax was introduced in two stages: a first tax increase from 1 April 1999 and a further
four-step increase in taxation from 2000 to 2003. There are tax reductions for some
consumers, chiefly within the manufacturing industry, agriculture and the railways. The
revenue from this tax is used for a reduction of the non-wage labour costs and the
promotion of renewable energies (Eichhammer, et al. 2006).
2
The rates are adapted to the efficiency potential of the different branches. This will provide
a strong incentive to reduce costs and increase efficiency (Eichhammer, et al. 2006).

been increased in five stages, and the bulk of the tax revenue generated used to reduce
pension insurance contributions.
 Emission trading system means to achieve ecological and economic success. It means
assuring the ecological integrity of the instrument, competition neutrality and low
transaction costs. In other words, the emission trading system makes use of market-based
mechanisms to encourage the reduction of greenhouse gas emissions in a cost-effective and
economically-efficient manner, while maintaining the environmental integrity of the
system.
 Specific Regulations such as: the Energy Performance of Buildings that seek to promote the
energy performance of buildings taking into account outdoor climatic and local conditions
as well as indoor climate requirements and cost-effectiveness, and the Minimum Energy
Performance Standards for appliances or equipments and mandatory labels that are used to
increase the energy efficiency of individual technologies.
 German CHP Law supports of cost efficient technology to reduce CO
2
emissions. This law
contains the definition of CHP electricity and heat; support mechanism for high efficiency
CHP, and mechanise to supervise reporting of CHP electricity production in CHP plants.
 Renewable Energy Sources Act creates a feed-in tariff system which requires utilities to
purchase a predetermined amount of renewable energy at a fixed price. The policy
provides economic security for investors and manufacturers and is responsible for the bulk
of Germany’s dynamic scale-up of renewable electricity capacity and equipment
production.
 Grants and loans: the Kreditanstalt für Wiederaufbau (KfW) Umweltprogramm
(Environment Program) that provides capital for investment in environmental protection
activities and the low-interest loans to SMEs that can be used to supplement the European
Recovery Programme’s Environment and Energy Saving Program.
 Technology specific rebates are programs used to promote energy management and new
energy-efficient technologies.
 Public information and advice: the sub-project under the Initiative Energieeffizienz (Energy
Efficiency Initiative) campaign, DENA, the German Energy Agency.

2.2 The Colombian energy efficiency policy
In 1991, with the introduction of the new Constitution, Colombia adopted the principles of
sustainable development as a guide to economic development and assigned to
municipalities the duty to regulate especially the industry and energy intensive activities.
The deregulation of the Colombian electricity system
3
began in the same period, as did the
restructuration of the public environmental management system
4
. These elements have
characterised the development of energy policies in this country, where the emphasis has

3
The Colombian electricity industry is characterized by a large hydroelectricity component,
close to 70%, and is considered to be one of the most open markets in the developing world,
and the market evolution with this model has been satisfactory in terms of investment,
competition, efficiency and reduction in electricity losses (Larsen et al., 2004).
4
The Colombian environmental administration characterizes to be decentralized,
democratic, participatory, fiscally solvent, and socially legitimate with measures as a system
of pollution taxes, require environmental impact assessments for large construction projects,
and institutionalize legal remedies against polluters (Blackman et al., 2006).
Energy Effciency 66

been on the formulation of projects and regulations concerning energy efficiency in the
manufacturing industrial sector. Moreover, additional instruments for environmental
management involve agreements with industry or other relevant organisations. In 1997, the
National Environmental Council approved the National Policy of Clean Production. The key
objectives of this consensus-based energy policy were to increase the environmental
efficiency and quality of energy resources and to develop environmental guides (guias
ambientales) detailing options for improving energy efficiency performance in specific
sectors. Other strategies used to increase energy efficiency in the manufacturing industries
included the establishment of the energy excellence program (Merito URE), the conversion
of urban factories from coal or diesel to natural gas and the development of strategies
planning for energy efficiency and renewable energy. Currently, the government is
developing two legislation projects to improve energy efficiency: Cogeneration Law and the
design of the Colombian program of normalisation, accreditation, certification, and labelling
of final use of energy equipment.
Hence, Colombian energy policies are based almost entirely on direct regulation. Apart from
some small exemptions to VAT taxes for environmental investments, the principal use of
economic incentives in energy policies involves the pricing of fuels and agreements with
specific manufacturing industrial sector that have high potentials to improve energy
efficiency or to carry out changes in technology and renewable energy.

3. Methodology
Changes in energy efficiency were monitored by examining energy use by unit of activity
and the application of two indicators of energy efficiency. The first indicator (EI
i
) Measures
energy use per euro of gross production (equation 1); and the second indicator (CEI
i
)
Carbon emission intensity the generation of greenhouses gases (in terms of CO
2
emissions)
per euro of gross production by each sector i of German and Colombian manufacturing
industry (equation 2).
��
�
�
�
�
�
�
(1)
��
�
� ������ ��������� ��������� ��� � ⁄ �
�
�
� ������ ����������� �� ��� ������� ������������� �������� � ��� �� � ���
�
�
� ���������� �� ������� ������ � ���

���
�
�
��
�
�
�
(3)
���
�
� ��
�
��������� ��������� ���
�
��������� �� ⁄
��
�
� ��
�
��������� ��������

To identify the factors and variables that influencing energy efficiency in the German and
Colombian manufacturing industries, we summarises the opinions and expectations of the
main stakeholders (associations and representative firms in Germany and Colombia)
through a questionnaire and existing scientific studies.
The questions were designed to identify factors and variables that determine energy
efficiency in the manufacturing industries. It included three sections, each with a unique

objective. The first section was designed to establish general information about energy
consumption, structure of energy source and energy efficiency.
The second section was designed to assess and rank the importance of different factors and
variables in the achievement of improved energy efficiency performance. Questions were
asked on issues relating to economic, technical and political factors with their respective
variables.
The third section was designed to assess external factors and instruments that would cause
or encourage improvements in energy efficiency performance, and what kinds of internal
measures or actions would tend to increase energy efficiency performance in the industry.

4. Changes in energy efficiency in German and Colombian manufacturing
industries
Energy consumption in the manufacturing industries increased by 2.3% in Germany and
5.5% in Colombia during the sample period (figure 1). The manufacturing industries with
the largest increases in energy consumption in this period were paper and tobacco in
Germany, and the automotive industry and cement industry in Colombia, whereas the
largest decrease in Germany was by the cement industry and in Colombia the machinery
industry.


Fig. 1. Energy consumption developments in German and Colombian manufacturing
industries

Figure 2 shows developments in average energy intensity for the German and Colombian
manufacturing industries between 1998 and 2005. In Germany, this indicator decreased 11%
and in the Colombian case decreased 10%. In both countries, several energy intensive
sectors have driven the decreases in these indicators for the whole manufacturing sector (in
the case of Germany, the chemical industry and basic metal, and in Colombia, basic metal
and some sectors of the glass industry).

0.94
0.96
0.98
1
1.02
1.04
1.06
1.08
1998 1999 2000 2001 2002 2003 2004 2005
I
n
d
e
x
 
1
9
9
8
 
=
 
1
Energy consumption
Germany Colombia
Factors infuencing energy effciency in the German and Colombian manufacturing industries 67

been on the formulation of projects and regulations concerning energy efficiency in the
manufacturing industrial sector. Moreover, additional instruments for environmental
management involve agreements with industry or other relevant organisations. In 1997, the
National Environmental Council approved the National Policy of Clean Production. The key
objectives of this consensus-based energy policy were to increase the environmental
efficiency and quality of energy resources and to develop environmental guides (guias
ambientales) detailing options for improving energy efficiency performance in specific
sectors. Other strategies used to increase energy efficiency in the manufacturing industries
included the establishment of the energy excellence program (Merito URE), the conversion
of urban factories from coal or diesel to natural gas and the development of strategies
planning for energy efficiency and renewable energy. Currently, the government is
developing two legislation projects to improve energy efficiency: Cogeneration Law and the
design of the Colombian program of normalisation, accreditation, certification, and labelling
of final use of energy equipment.
Hence, Colombian energy policies are based almost entirely on direct regulation. Apart from
some small exemptions to VAT taxes for environmental investments, the principal use of
economic incentives in energy policies involves the pricing of fuels and agreements with
specific manufacturing industrial sector that have high potentials to improve energy
efficiency or to carry out changes in technology and renewable energy.

3. Methodology
Changes in energy efficiency were monitored by examining energy use by unit of activity
and the application of two indicators of energy efficiency. The first indicator (EI
i
) Measures
energy use per euro of gross production (equation 1); and the second indicator (CEI
i
)
Carbon emission intensity the generation of greenhouses gases (in terms of CO
2
emissions)
per euro of gross production by each sector i of German and Colombian manufacturing
industry (equation 2).
��
�
�
�
�
�
�
(1)
��
�
� ������ ��������� ��������� ��� � ⁄ �
�
�
� ������ ����������� �� ��� ������� ������������� �������� � ��� �� � ���
�
�
� ���������� �� ������� ������ � ���

���
�
�
��
�
�
�
(3)
���
�
� ��
�
��������� ��������� ���
�
��������� �� ⁄
��
�
� ��
�
��������� ��������

To identify the factors and variables that influencing energy efficiency in the German and
Colombian manufacturing industries, we summarises the opinions and expectations of the
main stakeholders (associations and representative firms in Germany and Colombia)
through a questionnaire and existing scientific studies.
The questions were designed to identify factors and variables that determine energy
efficiency in the manufacturing industries. It included three sections, each with a unique

objective. The first section was designed to establish general information about energy
consumption, structure of energy source and energy efficiency.
The second section was designed to assess and rank the importance of different factors and
variables in the achievement of improved energy efficiency performance. Questions were
asked on issues relating to economic, technical and political factors with their respective
variables.
The third section was designed to assess external factors and instruments that would cause
or encourage improvements in energy efficiency performance, and what kinds of internal
measures or actions would tend to increase energy efficiency performance in the industry.

4. Changes in energy efficiency in German and Colombian manufacturing
industries
Energy consumption in the manufacturing industries increased by 2.3% in Germany and
5.5% in Colombia during the sample period (figure 1). The manufacturing industries with
the largest increases in energy consumption in this period were paper and tobacco in
Germany, and the automotive industry and cement industry in Colombia, whereas the
largest decrease in Germany was by the cement industry and in Colombia the machinery
industry.


Fig. 1. Energy consumption developments in German and Colombian manufacturing
industries

Figure 2 shows developments in average energy intensity for the German and Colombian
manufacturing industries between 1998 and 2005. In Germany, this indicator decreased 11%
and in the Colombian case decreased 10%. In both countries, several energy intensive
sectors have driven the decreases in these indicators for the whole manufacturing sector (in
the case of Germany, the chemical industry and basic metal, and in Colombia, basic metal
and some sectors of the glass industry).

0.94
0.96
0.98
1
1.02
1.04
1.06
1.08
1998 1999 2000 2001 2002 2003 2004 2005
I
n
d
e
x
 
1
9
9
8
 
=
 
1
Energy consumption
Germany Colombia
Energy Effciency 68


Fig. 2. Energy intensity developments for the German and Colombian manufacturing
industries, 1998-2005

The indicator (CEIi) assessed in terms of generation of greenhouse gas emissions,
specifically tonnes of CO
2
per gross production. In Germany, the manufacturing industries
this indicator decreased 10%. The Colombian manufacturing industries decreased 13% this
indicator (see figure 3).


Fig. 3. CO
2
emissions intensity developments for the German and Colombian manufacturing
industries, 1998-2005

In Colombia, this indicator in general are still very high in comparison to the German
manufacturing industries, and thus there are plenty of opportunities for the Colombian
manufacturing industries to further lower this indicator and achieve better and cleaner
production figures by improved use of energy resources and a better selection of fuels. By
achieving these goals, Colombia will be able to meet international environmental
requirements and thus will assure its permanence in the market.

0.85
0.9
0.95
1
1.05
1.1
1.15
1.2
1998 1999 2000 2001 2002 2003 2004 2005
I
n
d
e
x
 
1
9
9
8
 
=
 
1
Energy intensity
Germany Colombia
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
1998 1999 2000 2001 2002 2003 2004 2005
I
n
d
e
x
 
1
9
9
8
 
=
 
1
CO
2
emissions  intensity
Germany Colombia

5. Results and discussion
The opinions and expectations of the main stakeholders as primary data are the following:
In the German case, two associations and twelve companies, and in the Colombian case,
four associations and 26 companies. (see figure 4).



Fig. 4. Breakdown of the primary data from the German and Colombian associations and
companies

5.1 Features of energy consumption, energy efficiency and energy source in German
and Colombian industries
The results of primary data show that in the German and Colombian cases more than 50% of
companies or associations consulted have made studies on energy efficiency and that within
of these companies and associations, the majority has analysed and assessed energy
efficiency performance and its advantages and disadvantages and included the topic of
energy efficiency within their business plans and strategies.
The results also show that the majority of firms and associations know their energy
consumption. However, in both countries, the assessment of energy intensity in the
companies and associations is a fairly new topic. Moreover, from 2000 to 2008, the
assessment of energy consumption and energy intensity has become more prevalent,
indicating, possibly, that within the German and Colombian manufacturing industries, the
energy topic is becoming more important in the production system and management. This
trend would coincide with the increase in certifications of environmental management
systems by the countries’ in the German case 65% and in the Colombian case 30% by year
during this period (ISO, 2007). Hence, energy management is a key program to improve
sustainability and environmental performance.
In both countries, the main energy sources for the firms consulted are electricity and natural
gas. Energy costs for the firms were between 0.5% and 3% in the German case and between
0.5% and 5% in the Colombian case.
Industries
86%
Asociations
14%
Germany
Textil
industry
37%
Food
industry
27%
Automotive
industry
20%
Germany
Industries
87%
Asociations
13%
Colombia
Textil
industry
37%
Food
industry
27%
Automotive
industry
20%
Other
sectors
17%
Colombia
Factors infuencing energy effciency in the German and Colombian manufacturing industries 69


Fig. 2. Energy intensity developments for the German and Colombian manufacturing
industries, 1998-2005

The indicator (CEIi) assessed in terms of generation of greenhouse gas emissions,
specifically tonnes of CO
2
per gross production. In Germany, the manufacturing industries
this indicator decreased 10%. The Colombian manufacturing industries decreased 13% this
indicator (see figure 3).


Fig. 3. CO
2
emissions intensity developments for the German and Colombian manufacturing
industries, 1998-2005

In Colombia, this indicator in general are still very high in comparison to the German
manufacturing industries, and thus there are plenty of opportunities for the Colombian
manufacturing industries to further lower this indicator and achieve better and cleaner
production figures by improved use of energy resources and a better selection of fuels. By
achieving these goals, Colombia will be able to meet international environmental
requirements and thus will assure its permanence in the market.

0.85
0.9
0.95
1
1.05
1.1
1.15
1.2
1998 1999 2000 2001 2002 2003 2004 2005
I
n
d
e
x
 
1
9
9
8
 
=
 
1
Energy intensity
Germany Colombia
0.85
0.90
0.95
1.00
1.05
1.10
1.15
1.20
1998 1999 2000 2001 2002 2003 2004 2005
I
n
d
e
x
 
1
9
9
8
 
=
 
1
CO
2
emissions  intensity
Germany Colombia

5. Results and discussion
The opinions and expectations of the main stakeholders as primary data are the following:
In the German case, two associations and twelve companies, and in the Colombian case,
four associations and 26 companies. (see figure 4).



Fig. 4. Breakdown of the primary data from the German and Colombian associations and
companies

5.1 Features of energy consumption, energy efficiency and energy source in German
and Colombian industries
The results of primary data show that in the German and Colombian cases more than 50% of
companies or associations consulted have made studies on energy efficiency and that within
of these companies and associations, the majority has analysed and assessed energy
efficiency performance and its advantages and disadvantages and included the topic of
energy efficiency within their business plans and strategies.
The results also show that the majority of firms and associations know their energy
consumption. However, in both countries, the assessment of energy intensity in the
companies and associations is a fairly new topic. Moreover, from 2000 to 2008, the
assessment of energy consumption and energy intensity has become more prevalent,
indicating, possibly, that within the German and Colombian manufacturing industries, the
energy topic is becoming more important in the production system and management. This
trend would coincide with the increase in certifications of environmental management
systems by the countries’ in the German case 65% and in the Colombian case 30% by year
during this period (ISO, 2007). Hence, energy management is a key program to improve
sustainability and environmental performance.
In both countries, the main energy sources for the firms consulted are electricity and natural
gas. Energy costs for the firms were between 0.5% and 3% in the German case and between
0.5% and 5% in the Colombian case.
Industries
86%
Asociations
14%
Germany
Textil
industry
37%
Food
industry
27%
Automotive
industry
20%
Germany
Industries
87%
Asociations
13%
Colombia
Textil
industry
37%
Food
industry
27%
Automotive
industry
20%
Other
sectors
17%
Colombia
Energy Effciency 70

The results in both countries indicate that energy management in the manufacturing
industries is important for business strategy and that the quantification and assessment of
energy consumption and energy efficiency are input indicators to improve upon in
optimisation processes working towards sustainability.

5.2 Factors influencing energy efficiency
In the German case, 43% of firms and associations consider production technology factors
very important, and 71% feel that economic and political factors are important in the
improvement of energy efficiency performance. In the Colombian case, economic (69%) and
production technology factors (62%) are very important factors in achieving improvement of
energy efficiency, whereas the political factor is irrelevant (42%) for firms and associations
(see figure 5).
These results indicate that in the German case, the firms and associations consider that
economic, technical as well as political factors influence energy efficiency, whereas in the
Colombian manufacturing industries improvements in energy efficiency are only closely
related with economic and production technology factors, mainly because energy efficiency
policies are limited and are focalised mainly in support and recommendations of the better
technologies.


Fig. 5. Factors influencing energy efficiency in German and Colombian industries

 Variables in economic factors influencing energy efficiency
Energy consumption in the manufacturing industrial sector is influenced by the behaviour
of several economic variables—e.g., high energy prices or constrained energy supply
motivate industrial facilities to try to secure the amount of energy required for operations at
the lowest possible price (McKane et al., 2008); structural changes in the manufacturing
industries cause shifts in final energy use and energy intensities; and the plant capacity
utilisation provides an indication of how efficiently plants and equipment are utilised and
consequently, could measure the efficiency of energy use.
In the German case, the variables of the economic factor that have the most influence on
energy efficiency are improvement in structural operations and maintenance costs and
investments in new technologies, equipment or specific activities of energy management
14%
43%
14%
71%
43%
71%
14%
14%
14%
0%
20%
40%
60%
80%
100%
Political  Technical  Economic
Germany
Very important Important Irrevelant
33%
62%
69%
25%
31%
31%
42%
8%
0%
20%
40%
60%
80%
100%
Political  Technical  Economic 
Colombia
Very important Important Irrevelant

investments. Improvements in plant capacity utilisation and levels of production have less
importance. On the other hand, in the Colombian case, all variables of the economic factor
are important, but the most relevant are improvement in plant capacity utilisation and
improvement in levels of production (see figure 6).
These results indicate that manufacturing industries of Germany consider that energy
efficiency improvements have higher dependence of investments and production methods,
whereas manufacturing industries of Colombia relate energy efficiency improvements with
capacity and levels of production. This means that in Germany, improving energy efficiency
is important as an investment strategy, whereas in Colombia, energy efficiency is a
secondary result from production strategy. This finding concurs with Tholander et al.,
(2007) who identified the non-priority of energy efficiency investments and lack of access to
capital—especially in small and medium enterprises—as main barriers to increased energy
efficiency in the manufacturing industries of developing countries in contrast with the
situation in developed countries. Moreover, manufacturing industries in developing
countries likely prefers traditional investments like expansion of industrial plants or power
generation. Furthermore, energy efficiency projects without large capital investments are
often perceived as riskier and / or are too small to attract multilateral financial institution
lending (UNIDO, 2007).


Fig. 6. Variables in the economic factors influencing energy efficiency in German and
Colombian industries.

SO&MC: Improvement in the structure of operation and maintenance costs. Inv.: Investments in
new technologies, equipments or specific activities of energy management. PCU: Improvement in
plant capacity utilisation. LP: Improvement in levels of production.

 Variables in production technology factor influencing energy efficiency
The need for improvement of energy efficiency is just one of the drivers for technology
development in industry. Moreover, the potential technical energy savings are available
based on proven technologies, best practices and use of new energy sources (IEA, 2007).
The manufacturing industries of both countries consider the most important technical
variable in improving energy efficiency to be changes in process, operations and machinery.
However, for German industries, changes in the structure of energy sources and
29%
57%
14%
57%
71%
29%
57%
29%
14%
29%
14%
0%
50%
100%
SO&MC   Inv.  PCU LP
Germany. Economic factor
Very important Important Not too  important Irrevelant
31%
54% 54%
69%
46%
31%
38%
23%
23% 8%
8% 8%
8%
0%
50%
100%
SO&MC   Inv.  PCU LP
Colombia. Economic factor
Very important Important Not too  important Irrevelant
Factors infuencing energy effciency in the German and Colombian manufacturing industries 71

The results in both countries indicate that energy management in the manufacturing
industries is important for business strategy and that the quantification and assessment of
energy consumption and energy efficiency are input indicators to improve upon in
optimisation processes working towards sustainability.

5.2 Factors influencing energy efficiency
In the German case, 43% of firms and associations consider production technology factors
very important, and 71% feel that economic and political factors are important in the
improvement of energy efficiency performance. In the Colombian case, economic (69%) and
production technology factors (62%) are very important factors in achieving improvement of
energy efficiency, whereas the political factor is irrelevant (42%) for firms and associations
(see figure 5).
These results indicate that in the German case, the firms and associations consider that
economic, technical as well as political factors influence energy efficiency, whereas in the
Colombian manufacturing industries improvements in energy efficiency are only closely
related with economic and production technology factors, mainly because energy efficiency
policies are limited and are focalised mainly in support and recommendations of the better
technologies.


Fig. 5. Factors influencing energy efficiency in German and Colombian industries

 Variables in economic factors influencing energy efficiency
Energy consumption in the manufacturing industrial sector is influenced by the behaviour
of several economic variables—e.g., high energy prices or constrained energy supply
motivate industrial facilities to try to secure the amount of energy required for operations at
the lowest possible price (McKane et al., 2008); structural changes in the manufacturing
industries cause shifts in final energy use and energy intensities; and the plant capacity
utilisation provides an indication of how efficiently plants and equipment are utilised and
consequently, could measure the efficiency of energy use.
In the German case, the variables of the economic factor that have the most influence on
energy efficiency are improvement in structural operations and maintenance costs and
investments in new technologies, equipment or specific activities of energy management
14%
43%
14%
71%
43%
71%
14%
14%
14%
0%
20%
40%
60%
80%
100%
Political  Technical  Economic
Germany
Very important Important Irrevelant
33%
62%
69%
25%
31%
31%
42%
8%
0%
20%
40%
60%
80%
100%
Political  Technical  Economic 
Colombia
Very important Important Irrevelant

investments. Improvements in plant capacity utilisation and levels of production have less
importance. On the other hand, in the Colombian case, all variables of the economic factor
are important, but the most relevant are improvement in plant capacity utilisation and
improvement in levels of production (see figure 6).
These results indicate that manufacturing industries of Germany consider that energy
efficiency improvements have higher dependence of investments and production methods,
whereas manufacturing industries of Colombia relate energy efficiency improvements with
capacity and levels of production. This means that in Germany, improving energy efficiency
is important as an investment strategy, whereas in Colombia, energy efficiency is a
secondary result from production strategy. This finding concurs with Tholander et al.,
(2007) who identified the non-priority of energy efficiency investments and lack of access to
capital—especially in small and medium enterprises—as main barriers to increased energy
efficiency in the manufacturing industries of developing countries in contrast with the
situation in developed countries. Moreover, manufacturing industries in developing
countries likely prefers traditional investments like expansion of industrial plants or power
generation. Furthermore, energy efficiency projects without large capital investments are
often perceived as riskier and / or are too small to attract multilateral financial institution
lending (UNIDO, 2007).


Fig. 6. Variables in the economic factors influencing energy efficiency in German and
Colombian industries.

SO&MC: Improvement in the structure of operation and maintenance costs. Inv.: Investments in
new technologies, equipments or specific activities of energy management. PCU: Improvement in
plant capacity utilisation. LP: Improvement in levels of production.

 Variables in production technology factor influencing energy efficiency
The need for improvement of energy efficiency is just one of the drivers for technology
development in industry. Moreover, the potential technical energy savings are available
based on proven technologies, best practices and use of new energy sources (IEA, 2007).
The manufacturing industries of both countries consider the most important technical
variable in improving energy efficiency to be changes in process, operations and machinery.
However, for German industries, changes in the structure of energy sources and
29%
57%
14%
57%
71%
29%
57%
29%
14%
29%
14%
0%
50%
100%
SO&MC   Inv.  PCU LP
Germany. Economic factor
Very important Important Not too  important Irrevelant
31%
54% 54%
69%
46%
31%
38%
23%
23% 8%
8% 8%
8%
0%
50%
100%
SO&MC   Inv.  PCU LP
Colombia. Economic factor
Very important Important Not too  important Irrevelant
Energy Effciency 72

consumption patterns are also important, while in the Colombian case, in the emphasis is on
improved employment behaviour (see figure 7). These results concur with empirical
analysis where energy sources emerging as an important variable that influences energy
efficiency and in the case of automotive industry and food industry changes of raw
materials have been a key variable to improve energy efficiency.


Fig. 7. Variables in the production technology factor influencing energy efficiency in
German and Colombian industries.

IPO: Increase processes outsourcing. CRM: Changes of raw materials. IR&D: Increase in the
resources of R&D. CCP: Changes of consumption patterns. CSES: Changes in the structure of
energy sources. IEB: Improvements in employment behaviour. CPOM: Changes in the process,
operations and machinery.

These results show that the manufacturing industries of both countries feel that the best way
to improve energy efficiency is by changes in process, operations and machinery (Germany
71% and Colombia 62%) generally these processes in the organizations begin with an
internal analysis of the production process and machinery to determine opportunities to
decrease energy consumption and increase energy efficiency. Moreover, in the Colombian
case, it’s also important the analysis of employment behaviour because behaviour change
erodes the energy savings due to the technical energy efficiency improvements, especially in
developing countries (IEA, 2005).
Hence, the results confirm that Germany has achieved important developments in energy
efficient-technology and significant improvement in energy efficiency performance in the
manufacturing industries. According to the Federal Ministry of Economics and Technology,
Germany in recent years has achieved a decrease in its energy consumption even though the
gross domestic product has more than doubled and German researchers and companies
have submitted many global patent applications in the development of energy efficient
industrial cross application technologies.

 Variables in political factors influencing energy efficiency
Market forces and other factors determine energy efficiency in the manufacturing industries.
However, these factors can be influenced by an effective energy policy that encourages cost
29% 29%
57%
43%
14%
29% 29%
29%
14%
29%
43%
71%
43%
29%
57%
14%
14% 43%
29%
14%
14%
14%
14%
0%
20%
40%
60%
80%
100%
IPO CRM IR&D CCP CSES IEB CPOM
Germany. Technical factor
Very important Important Not too  important Irrevelant
8%
15%
8% 8%
46% 46%
62%
8%
38%
54%
69%
31%
38%
23%
77%
31%
23%
23%
15%
15%
15%
8%
15%
8%
8%
0%
20%
40%
60%
80%
100%
IPO CRM IR&D CCP CSES IEB CPOM
Colombia. Technical  factor
Very important Important Not too  important Irrevelant

effective energy efficiency through the application of different types of policy instruments
that include information, regulation and economic instruments.
Figure 8 shows the results of variables in the political factors affecting energy efficiency in
German and Colombian industries. In the German case, the most important variables of the
political factor are to encourage the application of energy management in the organizations,
mandatory standards (such as the efficiency of electric motors and the efficiency of
industrial boilers), and soft loans—especially for cogeneration (CHP). These results concur
with Eichhammer, et al. (2006), who showed that only some measures are seen as a high-
impact (the first voluntary agreement with German industry from 1995 and the second
financial measures (CHP Act, KfW Umweltprogramm)), whereas the impact of the
Ecological Tax Reform has been estimated as medium, and other measures have been
assessed as low-impact.
However, according to studies of Ecofis et al., (2206) voluntary agreements to save energy
are adequate in these circumstances when dealing with a small number of actors with which
you need to negotiate or a strongly organized sector and / or when there is much relatively
cheap energy saving potential. The characteristics that could determine the success of this
instrument are the following: the target group motivated to participate, there are penalties
in case of non- compliance, there is a good monitoring system, and adequate supporting
instruments such as audits, energy monitoring systems, financial incentives and
demonstrations projects.


Fig. 8. Variable in the political factors influencing energy efficiency in German and
Colombian industries

Eco-tax: Eco-tax.VA: Voluntary audits. IC: Information campaigns. MS: Mandatory standards (the
efficiency of electric motors and the efficiency of industrial boilers). G/S: Grants / subsidies. CDM:
Emission trading / Clean Development Mechanism. EM: to encourage the application of energy
management SL: Soft Loans for Energy Efficiency, Renewable energy and CHP.

In the Colombian firms, the most important variables are soft loans (for Energy Efficiency,
Renewable energy and cogeneration (CHP)), to encourage energy management and the
emissions trading / Clean Development Mechanism—indicating that in this country, a
barrier to improved energy efficiency is the limited amount of resources available to change
14%
43%
14% 14%
43% 43%
29% 57%
71%
43%
43%
14%
57%
29% 29%
29%
14%
14%
29%
29%
14%
29%
14%
14%
14%
29%
14%
0%
20%
40%
60%
80%
100%
Eco‐Tax VA IC MS G/S CDM EM SL
Germany. Political factor
Very important Important Not too  important Irrevelant
23% 23%
8%
46% 46%
23%
54%
62%
38% 38% 69%
38% 38%
69%
38%
31%
31% 31%
8%
8%
15%
8% 8%
8%
8%
0%
20%
40%
60%
80%
100%
Eco‐Tax VA IC MS G/S CDM EM SL
Colombia. Political  factor
Very important Important Not too  important Irrevelant
Factors infuencing energy effciency in the German and Colombian manufacturing industries 73

consumption patterns are also important, while in the Colombian case, in the emphasis is on
improved employment behaviour (see figure 7). These results concur with empirical
analysis where energy sources emerging as an important variable that influences energy
efficiency and in the case of automotive industry and food industry changes of raw
materials have been a key variable to improve energy efficiency.


Fig. 7. Variables in the production technology factor influencing energy efficiency in
German and Colombian industries.

IPO: Increase processes outsourcing. CRM: Changes of raw materials. IR&D: Increase in the
resources of R&D. CCP: Changes of consumption patterns. CSES: Changes in the structure of
energy sources. IEB: Improvements in employment behaviour. CPOM: Changes in the process,
operations and machinery.

These results show that the manufacturing industries of both countries feel that the best way
to improve energy efficiency is by changes in process, operations and machinery (Germany
71% and Colombia 62%) generally these processes in the organizations begin with an
internal analysis of the production process and machinery to determine opportunities to
decrease energy consumption and increase energy efficiency. Moreover, in the Colombian
case, it’s also important the analysis of employment behaviour because behaviour change
erodes the energy savings due to the technical energy efficiency improvements, especially in
developing countries (IEA, 2005).
Hence, the results confirm that Germany has achieved important developments in energy
efficient-technology and significant improvement in energy efficiency performance in the
manufacturing industries. According to the Federal Ministry of Economics and Technology,
Germany in recent years has achieved a decrease in its energy consumption even though the
gross domestic product has more than doubled and German researchers and companies
have submitted many global patent applications in the development of energy efficient
industrial cross application technologies.

 Variables in political factors influencing energy efficiency
Market forces and other factors determine energy efficiency in the manufacturing industries.
However, these factors can be influenced by an effective energy policy that encourages cost
29% 29%
57%
43%
14%
29% 29%
29%
14%
29%
43%
71%
43%
29%
57%
14%
14% 43%
29%
14%
14%
14%
14%
0%
20%
40%
60%
80%
100%
IPO CRM IR&D CCP CSES IEB CPOM
Germany. Technical factor
Very important Important Not too  important Irrevelant
8%
15%
8% 8%
46% 46%
62%
8%
38%
54%
69%
31%
38%
23%
77%
31%
23%
23%
15%
15%
15%
8%
15%
8%
8%
0%
20%
40%
60%
80%
100%
IPO CRM IR&D CCP CSES IEB CPOM
Colombia. Technical  factor
Very important Important Not too  important Irrevelant

effective energy efficiency through the application of different types of policy instruments
that include information, regulation and economic instruments.
Figure 8 shows the results of variables in the political factors affecting energy efficiency in
German and Colombian industries. In the German case, the most important variables of the
political factor are to encourage the application of energy management in the organizations,
mandatory standards (such as the efficiency of electric motors and the efficiency of
industrial boilers), and soft loans—especially for cogeneration (CHP). These results concur
with Eichhammer, et al. (2006), who showed that only some measures are seen as a high-
impact (the first voluntary agreement with German industry from 1995 and the second
financial measures (CHP Act, KfW Umweltprogramm)), whereas the impact of the
Ecological Tax Reform has been estimated as medium, and other measures have been
assessed as low-impact.
However, according to studies of Ecofis et al., (2206) voluntary agreements to save energy
are adequate in these circumstances when dealing with a small number of actors with which
you need to negotiate or a strongly organized sector and / or when there is much relatively
cheap energy saving potential. The characteristics that could determine the success of this
instrument are the following: the target group motivated to participate, there are penalties
in case of non- compliance, there is a good monitoring system, and adequate supporting
instruments such as audits, energy monitoring systems, financial incentives and
demonstrations projects.


Fig. 8. Variable in the political factors influencing energy efficiency in German and
Colombian industries

Eco-tax: Eco-tax.VA: Voluntary audits. IC: Information campaigns. MS: Mandatory standards (the
efficiency of electric motors and the efficiency of industrial boilers). G/S: Grants / subsidies. CDM:
Emission trading / Clean Development Mechanism. EM: to encourage the application of energy
management SL: Soft Loans for Energy Efficiency, Renewable energy and CHP.

In the Colombian firms, the most important variables are soft loans (for Energy Efficiency,
Renewable energy and cogeneration (CHP)), to encourage energy management and the
emissions trading / Clean Development Mechanism—indicating that in this country, a
barrier to improved energy efficiency is the limited amount of resources available to change
14%
43%
14% 14%
43% 43%
29% 57%
71%
43%
43%
14%
57%
29% 29%
29%
14%
14%
29%
29%
14%
29%
14%
14%
14%
29%
14%
0%
20%
40%
60%
80%
100%
Eco‐Tax VA IC MS G/S CDM EM SL
Germany. Political factor
Very important Important Not too  important Irrevelant
23% 23%
8%
46% 46%
23%
54%
62%
38% 38% 69%
38% 38%
69%
38%
31%
31% 31%
8%
8%
15%
8% 8%
8%
8%
0%
20%
40%
60%
80%
100%
Eco‐Tax VA IC MS G/S CDM EM SL
Colombia. Political  factor
Very important Important Not too  important Irrevelant
Energy Effciency 74

technology and to achieve improved energy efficiency, a conclusion which concurs with the
studies of Kant, 1995; Tanaka, 2008 and Gillingham et al., 2009.

5.3 Instruments influence interest to improve energy efficiency performance
Figure 9 shows that instruments and measures would cause or encourage the German and
Colombian manufacturing industries to improve energy efficiency performance. In both
countries, the main instruments are changes in upstream sector (energy prices) and
institutional regulations, whereas labelling to have a lower impact.


Fig. 9. Percentage of respondents who felt that specific measures and instruments could
improve energy efficiency performance

CUS: Changes in upstream sector (energy prices). IR: Institutional regulations (Regulatory
standards, - Fiscal policy, State aid for R&D). VA: Voluntary agreements. Lab: Labelling (e.g.
industrial motors, EMAS, ISO 14001).

The results are clear in the German case, where a series of energy-conservation instruments
have been implemented to include: the replacement of traditional gas- or oil-fired boilers
with condensing gas-fired boilers, the gradual replacement of traditional fuels with more
expensive bio-fuel, and the consecutive emergence of integrated gasification combined cycle
(CGC) and combined heat and power (CHP) systems. As a result, the energy intensity of
Germany has decreased 20% from 1990 to 2003, with an annual decrease rate of 1.75%.
Moreover, during the last decade, the energy policy of Germany has been strongly
influenced by environmental issues, and the German government has consecutively
introduced various acts related to renewable energy and energy efficiency. During 1999, to
stimulate energy conservation, energy efficiency, and the application of renewable energy
technologies, the German government introduced the Eco-tax, which subsequently became
the Renewable Energy Act, which targets a short-term goal of doubling renewable power
generation by 2010, together with an intermediate-term goal of increasing renewable power
generation capacity to 20% of total power generation capacity by 2020 (Blesl et al., 2007).

71% 71%
43%
29%
29% 29%
57%
71%
0%
20%
40%
60%
80%
100%
CUS IR VA Lab
Germany
Yes  No
92%
77%
69%
31%
8%
23%
31%
69%
0%
20%
40%
60%
80%
100%
CUS IR Lab VA
Colombia
Yes No

5.4 Internal measures and actions the manufacturing industries would consider to
increase energy efficiency performance
Figure 10 shows the kinds of internal measures and actions the manufacturing industries
would consider to increase energy efficiency performance. In the German case, the most
important internal measures in order of importance are energy management systems,
energy efficiency investments, and changes in machinery and equipment. In the Colombian
case, the most important internal measures in order of importance are energy efficiency
investments, changes in machinery and equipment, and optimisation of production capacity
and production level.


Fig. 10. Kinds of internal measures and actions the manufacturing industries would consider
to increase energy efficiency performance

EMS: Energy management systems. EEI: Energy efficiency investment (e.g. changes in machinery,
equipments and technology). CM&E: Changes in machinery and equipment. TA: Training activities.
VA: Voluntary audit. TC: Major product/process related technological changes, whether or not
introduced as part of public/private national and the R&D programmes. OCP: Optimization of
production capacity and production level. CIB: Conversion of industrial business (in terms of both
products and processes).

These results show that in both countries, the manufacturing industrial sector has an interest
in increasing their investments to improve energy efficiency through changes in machinery
and equipment—demonstrating that the manufacturing industrial sector considers
improvements in energy efficiency to be closely related with technological change. This
result coincides with opportunities to improve industrial energy efficiency through new
technologies such as the use of high-efficiency motor-driven systems, the optimisation of
compressed air systems and the potential that exists based on currently available
improvements. In fact, the possibility of implementing new and emerging technologies with
potential savings of as much as 35 percent in energy costs is creating entirely new lines of
business (IAC, 2007).
Finally, the results of this study suggest that policy strategies in the manufacturing
industries have to utilise legal and fiscal instruments to generate supporting framework
conditions as well as targeted programs in the fields of R&D, technological change, market
transformation, information, education, dissemination of best practice, etc. Moreover, policy
100%
86% 86%
71% 71% 71%
57%
29%
14% 14%
29% 29% 29%
43%
71%
0%
20%
40%
60%
80%
100%
EMS EEI CM&E TA VA TC OCP CIB
Germany
Yes  No
85% 85%
77%
62% 62%
54% 54%
38%
15% 15%
23%
38% 38%
46% 46%
62%
0%
20%
40%
60%
80%
100%
EEI CM&E OCP EMS TA VA CIB TC
Colombia
Yes No
Factors infuencing energy effciency in the German and Colombian manufacturing industries 75

technology and to achieve improved energy efficiency, a conclusion which concurs with the
studies of Kant, 1995; Tanaka, 2008 and Gillingham et al., 2009.

5.3 Instruments influence interest to improve energy efficiency performance
Figure 9 shows that instruments and measures would cause or encourage the German and
Colombian manufacturing industries to improve energy efficiency performance. In both
countries, the main instruments are changes in upstream sector (energy prices) and
institutional regulations, whereas labelling to have a lower impact.


Fig. 9. Percentage of respondents who felt that specific measures and instruments could
improve energy efficiency performance

CUS: Changes in upstream sector (energy prices). IR: Institutional regulations (Regulatory
standards, - Fiscal policy, State aid for R&D). VA: Voluntary agreements. Lab: Labelling (e.g.
industrial motors, EMAS, ISO 14001).

The results are clear in the German case, where a series of energy-conservation instruments
have been implemented to include: the replacement of traditional gas- or oil-fired boilers
with condensing gas-fired boilers, the gradual replacement of traditional fuels with more
expensive bio-fuel, and the consecutive emergence of integrated gasification combined cycle
(CGC) and combined heat and power (CHP) systems. As a result, the energy intensity of
Germany has decreased 20% from 1990 to 2003, with an annual decrease rate of 1.75%.
Moreover, during the last decade, the energy policy of Germany has been strongly
influenced by environmental issues, and the German government has consecutively
introduced various acts related to renewable energy and energy efficiency. During 1999, to
stimulate energy conservation, energy efficiency, and the application of renewable energy
technologies, the German government introduced the Eco-tax, which subsequently became
the Renewable Energy Act, which targets a short-term goal of doubling renewable power
generation by 2010, together with an intermediate-term goal of increasing renewable power
generation capacity to 20% of total power generation capacity by 2020 (Blesl et al., 2007).

71% 71%
43%
29%
29% 29%
57%
71%
0%
20%
40%
60%
80%
100%
CUS IR VA Lab
Germany
Yes  No
92%
77%
69%
31%
8%
23%
31%
69%
0%
20%
40%
60%
80%
100%
CUS IR Lab VA
Colombia
Yes No

5.4 Internal measures and actions the manufacturing industries would consider to
increase energy efficiency performance
Figure 10 shows the kinds of internal measures and actions the manufacturing industries
would consider to increase energy efficiency performance. In the German case, the most
important internal measures in order of importance are energy management systems,
energy efficiency investments, and changes in machinery and equipment. In the Colombian
case, the most important internal measures in order of importance are energy efficiency
investments, changes in machinery and equipment, and optimisation of production capacity
and production level.


Fig. 10. Kinds of internal measures and actions the manufacturing industries would consider
to increase energy efficiency performance

EMS: Energy management systems. EEI: Energy efficiency investment (e.g. changes in machinery,
equipments and technology). CM&E: Changes in machinery and equipment. TA: Training activities.
VA: Voluntary audit. TC: Major product/process related technological changes, whether or not
introduced as part of public/private national and the R&D programmes. OCP: Optimization of
production capacity and production level. CIB: Conversion of industrial business (in terms of both
products and processes).

These results show that in both countries, the manufacturing industrial sector has an interest
in increasing their investments to improve energy efficiency through changes in machinery
and equipment—demonstrating that the manufacturing industrial sector considers
improvements in energy efficiency to be closely related with technological change. This
result coincides with opportunities to improve industrial energy efficiency through new
technologies such as the use of high-efficiency motor-driven systems, the optimisation of
compressed air systems and the potential that exists based on currently available
improvements. In fact, the possibility of implementing new and emerging technologies with
potential savings of as much as 35 percent in energy costs is creating entirely new lines of
business (IAC, 2007).
Finally, the results of this study suggest that policy strategies in the manufacturing
industries have to utilise legal and fiscal instruments to generate supporting framework
conditions as well as targeted programs in the fields of R&D, technological change, market
transformation, information, education, dissemination of best practice, etc. Moreover, policy
100%
86% 86%
71% 71% 71%
57%
29%
14% 14%
29% 29% 29%
43%
71%
0%
20%
40%
60%
80%
100%
EMS EEI CM&E TA VA TC OCP CIB
Germany
Yes  No
85% 85%
77%
62% 62%
54% 54%
38%
15% 15%
23%
38% 38%
46% 46%
62%
0%
20%
40%
60%
80%
100%
EEI CM&E OCP EMS TA VA CIB TC
Colombia
Yes No
Energy Effciency 76

will always have to live with unavoidably sub-optimal solutions, while growing knowledge
and changing frameworks will constantly impose the need to search for better solutions and
new opportunities. In this context, energy policy strategies represent not only (static)
problems of policy choice but—above all—dynamic search and learning processes aimed at
designing effective policy measures.

6. Recommendations for the formulation of energy-efficiency policies in the
Colombian manufacturing industrial sector
According to our results and the literature, it is important that there be a formulation of an
adequate package of policies and measures that are addressed to guarantee effective and
efficient impact to improve energy-efficiency performance and reducing greenhouse
emissions in the Colombian manufacturing industries. The following strategies and
instruments in policy settings are recommended in order to achieve improvements in
energy efficiency in a cost-effective manner:

a. Policy support. Policy support should aim at making energy efficiency easy (“Make it
easy!”), realisable (“Make it possible!”), and beneficial (“Make it rewarding!”) for
stakeholders, thereby contributing to the development of the market for energy-efficient
technologies and services. Due to the implementation of the support programmes, it also
becomes clear that energy efficiency is politically intended and crucial (“Make it a policy!”).
A pre-planned, target-group-specific, differentiated mix of policy instruments and measures
is necessary, with integrated measures that are directly addressed to stakeholders. In such a
way, the specific situations, incentives, barriers and obstacles of different stakeholders
should be addressed by specific policy mixes (Thomas and Irrek, 2007).

b. Integral approach. The most effective way to improve industrial energy efficiency is
through an integrated approach, where a number of policies and programmes are combined
to create a strong overall industrial energy-efficiency policy that addresses a variety of needs
in Colombian manufacturing sectors. There should thus be an adoption of a policy of
energy-efficiency sector targets and related programmes in which individual manufacturing
industrial sectors committed to specific improvements in energy intensity over a given time
period in exchange for governmental support in the form of financial incentives,
information programmes, demonstration programmes, and training programmes,
significant energy savings could be realised.

c. Energy efficiency strategies. National energy efficiency strategies in Colombia could
accelerate the implementation of energy efficiency in the manufacturing industries. National
energy-efficiency strategies should be useful because during their development,
implementation and evaluation, they can help to achieve the following: make the vision for
energy efficiency explicit; focus attention on the important issues; identify gaps in current
work programmes; identify necessary tasks and resources and allocate implementation and
monitoring responsibility.

d. Energy data. The Colombian government through the statistical office and energy agency
(UPME) must improve the availability of high-quality energy efficiency data because

without accurate energy time series data, it is difficult to target and develop appropriate
energy efficiency policies in the manufacturing industries. Moreover, for developing
sectoral energy efficiency benchmarks and best practices, action plans should: assess energy
consumption by end-use in manufacturing industrial sector; identify the economy's energy-
saving potentials and establish objectives and adequate methods for evaluating the success
of the plan.

e. Mandatory standards. For the Colombian manufacturing industrial sector, the most
important technical variable to improve energy efficiency is change in processes, operations,
machinery and equipment. For this reason, the Colombian government should consider
adopting mandatory minimum energy performance standards for machinery and
equipment (e.g., the efficiency of industrial motors and the efficiency of industrial boilers) in
line with international best practices. Moreover, it should examine barriers to the
optimisation of energy efficiency through technology systems and design and implement
comprehensive policy portfolios aimed at overcoming such barriers.

f. Energy management. Among Colombian firms, one of the most important political
variables is the encouragement of the application of energy management
5.
The Colombian
government should thus consider providing effective assistance in the development of
energy management (EM) capability through the development and maintenance of EM
tools, training, certification and quality assurance. Moreover, it should encourage or require
major industrial energy users to implement comprehensive energy management procedures
and practices that could include, according to IEA, 2008:

 The development and adoption of a formal energy management policy. The process
and implementation of this policy should be reported and overseen at the company
board level and reported in company reports. Within this policy, companies would
need to demonstrate that effective organisational structures have been put in place to
ensure the following: that decisions regarding the procurement of energy-using
equipment are taken with the full knowledge of the equipment's expected life-cycle
costs and that procurement managers have an effective incentive to minimise the life-
cycle costs of their acquisitions.
 The appointment of full-time qualified energy managers at both the enterprise- and
plant-specific levels as appropriate.
 The establishment of a scheme to measure, monitor, evaluate and report industrial
energy consumption and efficiency at the individual company sector and national
levels. As a part of this effort, appropriate energy performance benchmarks should be
developed, monitored and reported at levels deemed suitable for each sector.

g. Small and Medium-sized Enterprises (SMEs). The size of company variable was
significant for Colombian industry. The Colombian government should thus consider

5
There are significant cost-effective energy savings to be realised in industry through the
more widespread adoption of best practices in energy management (EM). EM addresses the
way in which an industrial plant or facility is managed to identify and exploit cost-effective
energy savings opportunities (IEA, 2008).
Factors infuencing energy effciency in the German and Colombian manufacturing industries 77

will always have to live with unavoidably sub-optimal solutions, while growing knowledge
and changing frameworks will constantly impose the need to search for better solutions and
new opportunities. In this context, energy policy strategies represent not only (static)
problems of policy choice but—above all—dynamic search and learning processes aimed at
designing effective policy measures.

6. Recommendations for the formulation of energy-efficiency policies in the
Colombian manufacturing industrial sector
According to our results and the literature, it is important that there be a formulation of an
adequate package of policies and measures that are addressed to guarantee effective and
efficient impact to improve energy-efficiency performance and reducing greenhouse
emissions in the Colombian manufacturing industries. The following strategies and
instruments in policy settings are recommended in order to achieve improvements in
energy efficiency in a cost-effective manner:

a. Policy support. Policy support should aim at making energy efficiency easy (“Make it
easy!”), realisable (“Make it possible!”), and beneficial (“Make it rewarding!”) for
stakeholders, thereby contributing to the development of the market for energy-efficient
technologies and services. Due to the implementation of the support programmes, it also
becomes clear that energy efficiency is politically intended and crucial (“Make it a policy!”).
A pre-planned, target-group-specific, differentiated mix of policy instruments and measures
is necessary, with integrated measures that are directly addressed to stakeholders. In such a
way, the specific situations, incentives, barriers and obstacles of different stakeholders
should be addressed by specific policy mixes (Thomas and Irrek, 2007).

b. Integral approach. The most effective way to improve industrial energy efficiency is
through an integrated approach, where a number of policies and programmes are combined
to create a strong overall industrial energy-efficiency policy that addresses a variety of needs
in Colombian manufacturing sectors. There should thus be an adoption of a policy of
energy-efficiency sector targets and related programmes in which individual manufacturing
industrial sectors committed to specific improvements in energy intensity over a given time
period in exchange for governmental support in the form of financial incentives,
information programmes, demonstration programmes, and training programmes,
significant energy savings could be realised.

c. Energy efficiency strategies. National energy efficiency strategies in Colombia could
accelerate the implementation of energy efficiency in the manufacturing industries. National
energy-efficiency strategies should be useful because during their development,
implementation and evaluation, they can help to achieve the following: make the vision for
energy efficiency explicit; focus attention on the important issues; identify gaps in current
work programmes; identify necessary tasks and resources and allocate implementation and
monitoring responsibility.

d. Energy data. The Colombian government through the statistical office and energy agency
(UPME) must improve the availability of high-quality energy efficiency data because

without accurate energy time series data, it is difficult to target and develop appropriate
energy efficiency policies in the manufacturing industries. Moreover, for developing
sectoral energy efficiency benchmarks and best practices, action plans should: assess energy
consumption by end-use in manufacturing industrial sector; identify the economy's energy-
saving potentials and establish objectives and adequate methods for evaluating the success
of the plan.

e. Mandatory standards. For the Colombian manufacturing industrial sector, the most
important technical variable to improve energy efficiency is change in processes, operations,
machinery and equipment. For this reason, the Colombian government should consider
adopting mandatory minimum energy performance standards for machinery and
equipment (e.g., the efficiency of industrial motors and the efficiency of industrial boilers) in
line with international best practices. Moreover, it should examine barriers to the
optimisation of energy efficiency through technology systems and design and implement
comprehensive policy portfolios aimed at overcoming such barriers.

f. Energy management. Among Colombian firms, one of the most important political
variables is the encouragement of the application of energy management
5.
The Colombian
government should thus consider providing effective assistance in the development of
energy management (EM) capability through the development and maintenance of EM
tools, training, certification and quality assurance. Moreover, it should encourage or require
major industrial energy users to implement comprehensive energy management procedures
and practices that could include, according to IEA, 2008:

 The development and adoption of a formal energy management policy. The process
and implementation of this policy should be reported and overseen at the company
board level and reported in company reports. Within this policy, companies would
need to demonstrate that effective organisational structures have been put in place to
ensure the following: that decisions regarding the procurement of energy-using
equipment are taken with the full knowledge of the equipment's expected life-cycle
costs and that procurement managers have an effective incentive to minimise the life-
cycle costs of their acquisitions.
 The appointment of full-time qualified energy managers at both the enterprise- and
plant-specific levels as appropriate.
 The establishment of a scheme to measure, monitor, evaluate and report industrial
energy consumption and efficiency at the individual company sector and national
levels. As a part of this effort, appropriate energy performance benchmarks should be
developed, monitored and reported at levels deemed suitable for each sector.

g. Small and Medium-sized Enterprises (SMEs). The size of company variable was
significant for Colombian industry. The Colombian government should thus consider

5
There are significant cost-effective energy savings to be realised in industry through the
more widespread adoption of best practices in energy management (EM). EM addresses the
way in which an industrial plant or facility is managed to identify and exploit cost-effective
energy savings opportunities (IEA, 2008).
Energy Effciency 78

developing and implementing a package of policies and measures to promote energy
efficiency among SMEs. This package should include: a system for ensuring that energy
audits, carried out by qualified engineers, are widely promoted and easily accessible for all
SMEs; the provision of high-quality and relevant information on energy-efficiency best
practices; the provision of energy performance benchmarking information that ideally
would be structured to allow international and national economy comparisons; and
appropriate incentives to adopt capital acquisition and procurement procedures with the
lowest life-cycle costs.

h. Investments. For the Colombian manufacturing industrial sector, the results indicate that
energy efficiency investments are a key variable to improve energy efficiency. However,
among the many impediments to the adoption of cost-effective energy efficiency
investments is the “finance barrier” (Tholander et al., 2007 and IEA, 2008). The Colombian
government should facilitate the manufacturing industrial sector’s and stakeholders’
involvement in energy efficiency investments by: I) adopting and publicising to the
manufacturing industrial sector a common energy-efficiency savings verification and
measurement protocol in order to reduce existing uncertainties in quantifying the benefits of
energy efficiency investments and stimulate increased private sector involvement; II)
reviewing their current subsidies and fiscal incentive programmes to create more favourable
grounds for private energy-efficiency investments; III) collaborating with the private
financial sector to establish public-private tools to facilitate energy-efficiency financing; IV)
promoting risk-mitigation instruments such as securitisation or public-private partnerships;
V) putting in place institutional frameworks to ensure regular co-operation and exchanges
on energy efficiency issues between the public sector and financial institutions and VI)
design an energy tax programme to provide an incentive to industry to improve energy
management at firms’ facilities through both behavioural changes and investments in
energy-efficient equipment.

i. Taxes and tariff structure. This study demonstrated that energy costs and taxes are
important for improving energy efficiency. The Colombian government should design a
package of taxes and a tariff structure that include the following: I) the reduction of
subsidies or using energy to balance the effect of subsidies, providing the energy consumer
with a more realistic indication of the actual costs associated with certain forms of energy; II)
the use of taxes to more accurately reflect the environmental costs, or “externalities”,
associated with energy consumption; III) the imposition of taxes and fees associated with
energy use resulting from energy consumption on users with the goal of creating incentives
to reduce wasteful energy consumption practices or creating public programmes and funds
for encouraging energy efficiency and IV) having the price system ensure that all individual
agents are confronted with the full costs that their decisions impose on others; this means
addressing externalities and market failures through a greater use of taxes, charges and
tradable permits and correcting policy failures through reforms of support programmes that
are environmentally harmful and economically inefficient and have undesirable social
effects.

j. Control, monitoring and evaluation. Developing effective energy-efficiency policies
requires a good understanding of how energy is used as well as the various factors that

drive or restrain demand. Such an understanding requires accurate data on energy end-use
and the associated activities. The Colombian government should thus ensure that
instruments of energy efficiency policies are adequately monitored, enforced and evaluated
so as to ensure maximum compliance and that their energy-efficiency policies are supported
by adequate end-use information by substantially increasing their effort to collect energy
end-use data across all sectors and relating to all energy types.

k. Technology transfer and cooperation. In the Colombian manufacturing industrial sector,
this analysis demonstrated that the technology level is still moderate and that this technical
factor is a key strategy to improve energy efficiency. The Colombian government should
thus promote technology transfer through an appropriate enabling framework in order to
enhance international cooperation for the scaling up of sustainable energy solutions. The
transfer of technology requires a careful balancing act that includes both fair treatment for
innovators and energy policies that stimulate global diffusion of energy technology to
address energy efficiency.

7. Conclusions
In this chapter analysed the energy efficiency in German and Colombian manufacturing
industries in the time period 1998-2005 using economic indicators. We found that the
industrial sectors of both countries during the sample period increased their energy
consumption by 2.3% in Germany and 5.5% in Colombia and also decreased the energy
intensity (11% and 10% respectively). Therefore, German and Colombian manufacturing
industries improved energy efficiency and decreased CO
2
emissions demonstrating that the
trend of manufacturing industry is “make more with less energy consumption and clean
production”.
Based on the primary data from German and Colombian industrial associations and
representative firms in each country, the economic, technical and political factors were
studied with respect to impact on energy efficiency. The results in both countries indicate
that energy management for the manufacturing industrial sector is important within
business strategy and that the quantification and assessment of energy consumption and
energy efficiency are input indicators to be used in improvement and optimisation processes
within sustainability development.
The results also show that in German industry, economic, technical and political factors
influence energy efficiency, whereas in the Colombian case, improvements in energy
efficiency are closely related with economical and production technology factors.
In the German case, the results showed the following: (I) the variables in the economic factor
with the most influence on energy efficiency are the structural operations and maintenance
costs and investments, whereas plant capacity utilisation and levels of production have
lower importance. (II) The most important technical variables to improve energy efficiency
are changes in the processes, operations and machinery, changes in the structure of energy
sources, and changes of consumption patterns. (III) The most important variables in the
political factor are to encourage the application of energy management, mandatory
standards (such as the efficiency of electric motors and the efficiency of industrial boilers),
and soft loans especially for cogeneration (CHP). (IV) The most important internal measures
Factors infuencing energy effciency in the German and Colombian manufacturing industries 79

developing and implementing a package of policies and measures to promote energy
efficiency among SMEs. This package should include: a system for ensuring that energy
audits, carried out by qualified engineers, are widely promoted and easily accessible for all
SMEs; the provision of high-quality and relevant information on energy-efficiency best
practices; the provision of energy performance benchmarking information that ideally
would be structured to allow international and national economy comparisons; and
appropriate incentives to adopt capital acquisition and procurement procedures with the
lowest life-cycle costs.

h. Investments. For the Colombian manufacturing industrial sector, the results indicate that
energy efficiency investments are a key variable to improve energy efficiency. However,
among the many impediments to the adoption of cost-effective energy efficiency
investments is the “finance barrier” (Tholander et al., 2007 and IEA, 2008). The Colombian
government should facilitate the manufacturing industrial sector’s and stakeholders’
involvement in energy efficiency investments by: I) adopting and publicising to the
manufacturing industrial sector a common energy-efficiency savings verification and
measurement protocol in order to reduce existing uncertainties in quantifying the benefits of
energy efficiency investments and stimulate increased private sector involvement; II)
reviewing their current subsidies and fiscal incentive programmes to create more favourable
grounds for private energy-efficiency investments; III) collaborating with the private
financial sector to establish public-private tools to facilitate energy-efficiency financing; IV)
promoting risk-mitigation instruments such as securitisation or public-private partnerships;
V) putting in place institutional frameworks to ensure regular co-operation and exchanges
on energy efficiency issues between the public sector and financial institutions and VI)
design an energy tax programme to provide an incentive to industry to improve energy
management at firms’ facilities through both behavioural changes and investments in
energy-efficient equipment.

i. Taxes and tariff structure. This study demonstrated that energy costs and taxes are
important for improving energy efficiency. The Colombian government should design a
package of taxes and a tariff structure that include the following: I) the reduction of
subsidies or using energy to balance the effect of subsidies, providing the energy consumer
with a more realistic indication of the actual costs associated with certain forms of energy; II)
the use of taxes to more accurately reflect the environmental costs, or “externalities”,
associated with energy consumption; III) the imposition of taxes and fees associated with
energy use resulting from energy consumption on users with the goal of creating incentives
to reduce wasteful energy consumption practices or creating public programmes and funds
for encouraging energy efficiency and IV) having the price system ensure that all individual
agents are confronted with the full costs that their decisions impose on others; this means
addressing externalities and market failures through a greater use of taxes, charges and
tradable permits and correcting policy failures through reforms of support programmes that
are environmentally harmful and economically inefficient and have undesirable social
effects.

j. Control, monitoring and evaluation. Developing effective energy-efficiency policies
requires a good understanding of how energy is used as well as the various factors that

drive or restrain demand. Such an understanding requires accurate data on energy end-use
and the associated activities. The Colombian government should thus ensure that
instruments of energy efficiency policies are adequately monitored, enforced and evaluated
so as to ensure maximum compliance and that their energy-efficiency policies are supported
by adequate end-use information by substantially increasing their effort to collect energy
end-use data across all sectors and relating to all energy types.

k. Technology transfer and cooperation. In the Colombian manufacturing industrial sector,
this analysis demonstrated that the technology level is still moderate and that this technical
factor is a key strategy to improve energy efficiency. The Colombian government should
thus promote technology transfer through an appropriate enabling framework in order to
enhance international cooperation for the scaling up of sustainable energy solutions. The
transfer of technology requires a careful balancing act that includes both fair treatment for
innovators and energy policies that stimulate global diffusion of energy technology to
address energy efficiency.

7. Conclusions
In this chapter analysed the energy efficiency in German and Colombian manufacturing
industries in the time period 1998-2005 using economic indicators. We found that the
industrial sectors of both countries during the sample period increased their energy
consumption by 2.3% in Germany and 5.5% in Colombia and also decreased the energy
intensity (11% and 10% respectively). Therefore, German and Colombian manufacturing
industries improved energy efficiency and decreased CO
2
emissions demonstrating that the
trend of manufacturing industry is “make more with less energy consumption and clean
production”.
Based on the primary data from German and Colombian industrial associations and
representative firms in each country, the economic, technical and political factors were
studied with respect to impact on energy efficiency. The results in both countries indicate
that energy management for the manufacturing industrial sector is important within
business strategy and that the quantification and assessment of energy consumption and
energy efficiency are input indicators to be used in improvement and optimisation processes
within sustainability development.
The results also show that in German industry, economic, technical and political factors
influence energy efficiency, whereas in the Colombian case, improvements in energy
efficiency are closely related with economical and production technology factors.
In the German case, the results showed the following: (I) the variables in the economic factor
with the most influence on energy efficiency are the structural operations and maintenance
costs and investments, whereas plant capacity utilisation and levels of production have
lower importance. (II) The most important technical variables to improve energy efficiency
are changes in the processes, operations and machinery, changes in the structure of energy
sources, and changes of consumption patterns. (III) The most important variables in the
political factor are to encourage the application of energy management, mandatory
standards (such as the efficiency of electric motors and the efficiency of industrial boilers),
and soft loans especially for cogeneration (CHP). (IV) The most important internal measures
Energy Effciency 80

to improve energy efficiency are energy management systems, energy efficiency investment,
and changes in machinery and equipment.
In the Colombian case, the results showed the following: (I) All variables for the economic
factor are important, but the most relevant are plant capacity utilisation and levels of
production. (II) The most important technical variables to improve energy efficiency are
changes in the processes, operations and machinery, and improvements in employment
behaviour. (III) The most important variables of the political factor are soft loans (for Energy
Efficiency, Renewable energy and cogeneration (CHP)), to encourage the application of
energy management and emissions trading / Clean Development Mechanism. (IV) The
most important internal measures for increasing energy efficiency are energy efficiency
investments, changes in machinery and equipment and optimisation of production capacity
and production level.
Moreover, the results suggest that policy strategies in the Colombian manufacturing
industrial sector have to combine the following strategies: integral approach, energy data,
mandatory standards, energy management, the promotion of energy efficiency in small and
medium-sized enterprises, investments, a tax program, an adequate tariff structure, control
and evaluation, technology transfer and cooperation.

Acknowledgments
The author would like to thank Professor Dr Werner Bönte, Dr Wolfang Irrek and Dr
Alexander Cotte Poveda for the helpful suggestions and comments. The author is grateful
for the support provided by the Wuppertal Institute, Deutscher Akademischer Austausch
Dients and the University of La Salle. Any remaining errors are the responsibility of the
author.

8. References
Blesl M., Das A., Fahl U., Remme U. (2007). Role of energy efficiency standards in reducing
CO2 emissions in Germany: An assessment with TIMES. Energy Policy 35, 772-785.
Blackman A, Morgenstern R., Montealegre L., Murcia L., and García J. (2006). Review of the
efficiency and effectiveness of Colombia’s environmental policies. An RFF Report.
Kant A. (1995). Strategies and Instruments to promote energy efficiency in developing
countries. Project working paper 5. Effectiveness of industrial energy conservation
programmes in IEA countries ECN-C-94-113.
ECOFYS, Wuppertal Institut, Lund University. (2006). Guidelines for the monitoring,
evaluation and design of energy efficiency policies. How policy theory can guide
monitoring and evaluation efforts and support the design of SMART policies.
www.aid-ee.org
Eichhammer, W., Schlomann, B., Kling N. (2006). Energy Efficiency Policies and Measures in
Germany 2006. Monitoring of Energy Efficiency in EU 15 and Norway (ODYSSEE-
MURE). Fraunhofer Institute for Systems and Innovation Research (Fraunhofer ISI).
Gillingham K., Newell R., Palmer K. (2009). Energy efficiency economics and policy.
Working Paper 15031. http://www.nber.org/papers/w15031
International Energy Agency (IEA). (2008). Energy efficiency policy recommendations. In
support of the G8 Plan of Action. http://www.iea.org/G8/2008/G8_EE_
recommendations.pdf

International Standard Organisation (ISO). (2007). The ISO Survey of Certifications 2006.
www.iso.org
Inter Academy Council (IAC). (2007). Lighting the way. Toward a sustainable energy future.
www.interacademycouncil.net
International Energy Agency (IEA). (2005). The experience with energy efficiency policies
and programmes in IEA countries. Learning from the critics. IEA Information
paper.
International Energy Agency (IEA). (2007). Tracking Industrial Energy Efficiency and CO2
Emissions. In support of the G8 Plan of Action. Energy Indicators.
Larsen E., Dyner. I, Bedoya L., Franco C. (2004). Lessons from deregulation in Colombia:
successes, failures and the way ahead. Energy policy 32, 1767-1780.
McKane A., Price L., Rue S. (2008). Policies for Promoting Industrial Energy Efficiency in
Developing Countries and Transition Economies. United Nations Industrial
Development Organization.
Tanaka K. (2008). Assessment of energy efficiency performance measures in industry and
their application for policy. Energy policy (2008), doi:10.1016/j.enpol.2008.03.032.
Thomas S., Irrek W. (2007). Wie 20 Prozent Endenergieeinsparung möglich werden können.
Worschläge des Wuppertal Instituts zum deutschen Energieeffizinez-Aktionsplan
und zu Maßnahmen im Industriebereich. VIK Mitteilungen 3/07, 16-18.
Thollander P., Danestig M., Rohdin P. (2007). Energy policies for increased industrial energy
efficiency: Evaluation of a local energy programme for manufacturing SMEs.
Energy policy 35, 5774–5783.
United Nations Industrial Development Organization (UNIDO). (2007). Policies for
promoting industrial energy efficiency in developing countries and transition
economies. Commission for Sustainable Development (CSD-15).
Wuppertal Institute, 2008. Greenhouse Gas Mitigation in Industry in Developing Countries.
Final Report. On behalf of the Deutsche Gesellschaft für Technishe
Zusammenarbeit (GTZ). http://www.wupperinst.org/en/projects/proj/index.
Factors infuencing energy effciency in the German and Colombian manufacturing industries 81

to improve energy efficiency are energy management systems, energy efficiency investment,
and changes in machinery and equipment.
In the Colombian case, the results showed the following: (I) All variables for the economic
factor are important, but the most relevant are plant capacity utilisation and levels of
production. (II) The most important technical variables to improve energy efficiency are
changes in the processes, operations and machinery, and improvements in employment
behaviour. (III) The most important variables of the political factor are soft loans (for Energy
Efficiency, Renewable energy and cogeneration (CHP)), to encourage the application of
energy management and emissions trading / Clean Development Mechanism. (IV) The
most important internal measures for increasing energy efficiency are energy efficiency
investments, changes in machinery and equipment and optimisation of production capacity
and production level.
Moreover, the results suggest that policy strategies in the Colombian manufacturing
industrial sector have to combine the following strategies: integral approach, energy data,
mandatory standards, energy management, the promotion of energy efficiency in small and
medium-sized enterprises, investments, a tax program, an adequate tariff structure, control
and evaluation, technology transfer and cooperation.

Acknowledgments
The author would like to thank Professor Dr Werner Bönte, Dr Wolfang Irrek and Dr
Alexander Cotte Poveda for the helpful suggestions and comments. The author is grateful
for the support provided by the Wuppertal Institute, Deutscher Akademischer Austausch
Dients and the University of La Salle. Any remaining errors are the responsibility of the
author.

8. References
Blesl M., Das A., Fahl U., Remme U. (2007). Role of energy efficiency standards in reducing
CO2 emissions in Germany: An assessment with TIMES. Energy Policy 35, 772-785.
Blackman A, Morgenstern R., Montealegre L., Murcia L., and García J. (2006). Review of the
efficiency and effectiveness of Colombia’s environmental policies. An RFF Report.
Kant A. (1995). Strategies and Instruments to promote energy efficiency in developing
countries. Project working paper 5. Effectiveness of industrial energy conservation
programmes in IEA countries ECN-C-94-113.
ECOFYS, Wuppertal Institut, Lund University. (2006). Guidelines for the monitoring,
evaluation and design of energy efficiency policies. How policy theory can guide
monitoring and evaluation efforts and support the design of SMART policies.
www.aid-ee.org
Eichhammer, W., Schlomann, B., Kling N. (2006). Energy Efficiency Policies and Measures in
Germany 2006. Monitoring of Energy Efficiency in EU 15 and Norway (ODYSSEE-
MURE). Fraunhofer Institute for Systems and Innovation Research (Fraunhofer ISI).
Gillingham K., Newell R., Palmer K. (2009). Energy efficiency economics and policy.
Working Paper 15031. http://www.nber.org/papers/w15031
International Energy Agency (IEA). (2008). Energy efficiency policy recommendations. In
support of the G8 Plan of Action. http://www.iea.org/G8/2008/G8_EE_
recommendations.pdf

International Standard Organisation (ISO). (2007). The ISO Survey of Certifications 2006.
www.iso.org
Inter Academy Council (IAC). (2007). Lighting the way. Toward a sustainable energy future.
www.interacademycouncil.net
International Energy Agency (IEA). (2005). The experience with energy efficiency policies
and programmes in IEA countries. Learning from the critics. IEA Information
paper.
International Energy Agency (IEA). (2007). Tracking Industrial Energy Efficiency and CO2
Emissions. In support of the G8 Plan of Action. Energy Indicators.
Larsen E., Dyner. I, Bedoya L., Franco C. (2004). Lessons from deregulation in Colombia:
successes, failures and the way ahead. Energy policy 32, 1767-1780.
McKane A., Price L., Rue S. (2008). Policies for Promoting Industrial Energy Efficiency in
Developing Countries and Transition Economies. United Nations Industrial
Development Organization.
Tanaka K. (2008). Assessment of energy efficiency performance measures in industry and
their application for policy. Energy policy (2008), doi:10.1016/j.enpol.2008.03.032.
Thomas S., Irrek W. (2007). Wie 20 Prozent Endenergieeinsparung möglich werden können.
Worschläge des Wuppertal Instituts zum deutschen Energieeffizinez-Aktionsplan
und zu Maßnahmen im Industriebereich. VIK Mitteilungen 3/07, 16-18.
Thollander P., Danestig M., Rohdin P. (2007). Energy policies for increased industrial energy
efficiency: Evaluation of a local energy programme for manufacturing SMEs.
Energy policy 35, 5774–5783.
United Nations Industrial Development Organization (UNIDO). (2007). Policies for
promoting industrial energy efficiency in developing countries and transition
economies. Commission for Sustainable Development (CSD-15).
Wuppertal Institute, 2008. Greenhouse Gas Mitigation in Industry in Developing Countries.
Final Report. On behalf of the Deutsche Gesellschaft für Technishe
Zusammenarbeit (GTZ). http://www.wupperinst.org/en/projects/proj/index.
Energy Effciency 82
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 83
Oxyfuel combustion in the steel industry: energy effciency and decrease
of co2 emissions
Author Name
X

Oxyfuel combustion in the steel
industry: energy efficiency and
decrease of CO
2
emissions

Joachim von Schéele
The Linde Group
Germany

1. Introduction
The use of oxygen technologies within the steel industry has become increasingly important.
During the last decades increased throughput capacity and lowered average cost have been
the driving forces, however, today the positive impact on energy savings and reduced
emissions have come into the focal point, a fact that seems to be even further pronounced in
the future. This chapter describes how the oxygen technologies contribute to increased
energy efficiency in the melting and heating processes, how it reduces the fuel consumption
and CO
2
emission, and how in-plant generated low calorific gases can be effectively used to
further improve the overall energy efficiency of a steel production plant, reduce costs and
environmental impact.
The main production routes for steel are the integrated steel mill and the mini-mill. The
integrated steel mill uses iron ore as main source for iron, and includes processes like ore
sintering, coke-making, blast furnace iron-making and basic oxygen steel-making. The main
piece of equipment at a mini-mill is the electric arc furnace where steel scrap, its main raw
material, is melted. Both routes include subsequent casting and downstream heating and
rolling (or forging) operations.
Dependent on production route and status, a steel mill need 700 to 4,000 kWh to produce 1
tonne of finished product. This corresponds to a CO
2
emission of about 0.35 to 2.2 tonne per
tonne of steel produced. However, there are great opportunities to increase the efficiency,
using oxygen technologies make a substantial positive impact. Relating to how the oxygen is
introduced, we basically distinguish between injection of oxygen (normally through a lance)
and oxyfuel combustion (applying a burner), however, the end result is the same: oxyfuel
combustion. The main processes where oxygen technologies can be applied are: electric arc
furnace for scrap melting, blast furnace iron-making, preheating of different vessels (ladles,
etc.), and in the downstream reheating and heat treatment.
It is a well-known fact that only three things are needed to start and maintain combustion:
oxygen, fuel, and sufficient energy for ignition. The combustion process itself would be
most efficient if fuel and oxygen can meet without any restrictions. However, in practice it is
not simply a question of efficient combustion, the heat transfer efficiency is also extremely
important. Nevertheless, it has been clearly demonstrated that if oxygen (and not air) is
5
Energy Effciency 84
used to combust a fuel, all the heat transfer mechanisms (convection, conduction and
radiation) can be promoted at the same time. Air contains 21% oxygen and 79% ballast. In a
combustion process, this ballast, practically all nitrogen, has to be heated, without taking
part in the process. By using oxygen instead of air we get the beneficial oxyfuel combustion.
New demands and challenges from the industry have been met by a continuous
development work. As a result, in parallel to the conventional oxyfuel – for example widely
used to boost melting in electric arc furnaces – there are today established very interesting
technologies. Among those, the most important ones seem to be flameless combustion and
direct flame impingement. These new technologies not only fulfil the existing needs with
astonishing results, they also open up for completely new areas of application.
Flameless oxyfuel is today applied in drying and preheating of ladles and converters, for
heating in reheat furnaces and annealing lines, and for melting when avoiding oxidation. It
provides excellent temperature uniformity and reduced NO
X
emissions. Additionally, it can
be applied in, for example, preheating of air in the blast furnace hot stoves.
The use of direct flame impingement has so far been limited to boosting of strip annealing
and galvanizing lines, but its opportunities are almost uncountable. For example, there are
ideas about applying this technology to substantially shorten process routes by omitting
process steps, or using it in the iron-making step.
In reheating, today’s best air-fuel solutions need at least 1.3 GJ (360 kWh) for heating a tonne
of steel to the right temperature for rolling or forging; employing oxyfuel the comparable
figure is below 1 GJ, a saving of 25%. For continuous heating operations it is also possible to
economically operate the furnace at a higher temperature at the entry (loading) side of the
furnace. This will even further increase the possible throughput in any furnace unit. Oxyfuel
combustion allows all installation pipes and flow trains to be compact without any need for
recuperative or regenerative heat recovery solutions. Combustion air-blowers and related
low frequency noise problems are avoided.
Oxyfuel solutions deliver a unique combination of advantages in reheat and annealing.
Thanks to improved thermal efficiency (about 80% compared with 40-60% for air-fuel), the
heating rate and productivity are increased and less fuel is required to heat the product to
the desired temperature, at the same time saving on CO
2
and NO
X
emissions. In summary
the results include:
 Throughput capacity increase of up to 50%
 Fuel savings of up to 50%
 Reduction of CO
2
emissions by up to 50%
 Reduction of NO
X
emissions
 Reduction of scaling losses (improving the material yield)
Compared with conventional oxyfuel, flameless oxyfuel provides even higher production
rates, excellent temperature uniformity and very low NO
X
emissions. Since its commercial
introduction in 2003, the leading supplier has made more than 30 installations of the
flameless oxyfuel technology, some using a low calorific fuel.
This chapter describes the state-of-the-art of oxygen technologies, including results from
installations in the steel industry, and discusses their future very interesting possibilities to
make the steel production more effective. Oxyfuel combustion has begun to make the steel
industry more energy efficiency, but more can be done and, moreover, those technologies
can be employed also in other branches of the industry, there as well making improvements
of 20-50%.
2. Oxyfuel combustion technologies
Oxyfuel combustion refers to the use of pure, that is industrial grade, oxygen instead of air
for combustion of fossil fuels. Oxyfuel technology offers a number of advantages over air-
fuel combustion. In air-fuel combustion the burner flame contains nitrogen from the
combustion air. A significant amount of the fuel energy is used to heat up this nitrogen. The
hot nitrogen leaves through the stack, creating energy losses. When avoiding the nitrogen
ballast, by the use of industrial grade oxygen, then not only is the combustion itself more
efficient but also the heat transfer. Oxyfuel combustion influences the combustion process in
a number of ways. The first obvious result is the increase in thermal efficiency due to the
reduced exhaust gas volume, a result that is fundamental and valid for all types of oxyfuel
burners. In combustion gases, heat radiation is mainly from CO
2
and H
2
O molecules. As
there is no, or very low, nitrogen content in an oxyfuel furnace atmosphere, the
concentration of highly radiating CO
2
and H
2
O will be very high, a fact which considerably
increases heat transfer by gas radiation. A striking feature of oxyfuel combustion is the very
high thermal efficiency even at high flue-gas temperatures and no preheating of fuel or
oxygen.

Fig. 1. An ingot for bearing steel production is lifted out of a soaking pit furnace at
Ascométal in France. The furnace is fired with flameless oxyfuel, heating the ingots
uniformly to over 1200°C.

In addition to using a burner for the combustion, which normally is operated at
stoichiometric conditions, two other technologies should be mentioned: lancing, and post-
combustion. Lancing refers to injecting oxygen, sometimes at very high velocities into
furnace free-space or a melt. It is done to intensify the air-fuel combustion, either to combust
for example carbon into CO, or achieve a complete combustion of a fuel into products like
CO
2
and H
2
O. Typically it is employed an Electric Arc Furnace (EAF) for scrap melting, but
it could also be like in the case of the REBOX
®
HLL technology to improve reheating.
Post-combustion does in most cases in this context refer to the reaction CO+½O
2
=CO
2
,
which is strongly exothermic; the released energy is typically used to improve melting.
Generally speaking, the prerequisites for a beneficial post-combustion are CO generation,
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 85
used to combust a fuel, all the heat transfer mechanisms (convection, conduction and
radiation) can be promoted at the same time. Air contains 21% oxygen and 79% ballast. In a
combustion process, this ballast, practically all nitrogen, has to be heated, without taking
part in the process. By using oxygen instead of air we get the beneficial oxyfuel combustion.
New demands and challenges from the industry have been met by a continuous
development work. As a result, in parallel to the conventional oxyfuel – for example widely
used to boost melting in electric arc furnaces – there are today established very interesting
technologies. Among those, the most important ones seem to be flameless combustion and
direct flame impingement. These new technologies not only fulfil the existing needs with
astonishing results, they also open up for completely new areas of application.
Flameless oxyfuel is today applied in drying and preheating of ladles and converters, for
heating in reheat furnaces and annealing lines, and for melting when avoiding oxidation. It
provides excellent temperature uniformity and reduced NO
X
emissions. Additionally, it can
be applied in, for example, preheating of air in the blast furnace hot stoves.
The use of direct flame impingement has so far been limited to boosting of strip annealing
and galvanizing lines, but its opportunities are almost uncountable. For example, there are
ideas about applying this technology to substantially shorten process routes by omitting
process steps, or using it in the iron-making step.
In reheating, today’s best air-fuel solutions need at least 1.3 GJ (360 kWh) for heating a tonne
of steel to the right temperature for rolling or forging; employing oxyfuel the comparable
figure is below 1 GJ, a saving of 25%. For continuous heating operations it is also possible to
economically operate the furnace at a higher temperature at the entry (loading) side of the
furnace. This will even further increase the possible throughput in any furnace unit. Oxyfuel
combustion allows all installation pipes and flow trains to be compact without any need for
recuperative or regenerative heat recovery solutions. Combustion air-blowers and related
low frequency noise problems are avoided.
Oxyfuel solutions deliver a unique combination of advantages in reheat and annealing.
Thanks to improved thermal efficiency (about 80% compared with 40-60% for air-fuel), the
heating rate and productivity are increased and less fuel is required to heat the product to
the desired temperature, at the same time saving on CO
2
and NO
X
emissions. In summary
the results include:
 Throughput capacity increase of up to 50%
 Fuel savings of up to 50%
 Reduction of CO
2
emissions by up to 50%
 Reduction of NO
X
emissions
 Reduction of scaling losses (improving the material yield)
Compared with conventional oxyfuel, flameless oxyfuel provides even higher production
rates, excellent temperature uniformity and very low NO
X
emissions. Since its commercial
introduction in 2003, the leading supplier has made more than 30 installations of the
flameless oxyfuel technology, some using a low calorific fuel.
This chapter describes the state-of-the-art of oxygen technologies, including results from
installations in the steel industry, and discusses their future very interesting possibilities to
make the steel production more effective. Oxyfuel combustion has begun to make the steel
industry more energy efficiency, but more can be done and, moreover, those technologies
can be employed also in other branches of the industry, there as well making improvements
of 20-50%.
2. Oxyfuel combustion technologies
Oxyfuel combustion refers to the use of pure, that is industrial grade, oxygen instead of air
for combustion of fossil fuels. Oxyfuel technology offers a number of advantages over air-
fuel combustion. In air-fuel combustion the burner flame contains nitrogen from the
combustion air. A significant amount of the fuel energy is used to heat up this nitrogen. The
hot nitrogen leaves through the stack, creating energy losses. When avoiding the nitrogen
ballast, by the use of industrial grade oxygen, then not only is the combustion itself more
efficient but also the heat transfer. Oxyfuel combustion influences the combustion process in
a number of ways. The first obvious result is the increase in thermal efficiency due to the
reduced exhaust gas volume, a result that is fundamental and valid for all types of oxyfuel
burners. In combustion gases, heat radiation is mainly from CO
2
and H
2
O molecules. As
there is no, or very low, nitrogen content in an oxyfuel furnace atmosphere, the
concentration of highly radiating CO
2
and H
2
O will be very high, a fact which considerably
increases heat transfer by gas radiation. A striking feature of oxyfuel combustion is the very
high thermal efficiency even at high flue-gas temperatures and no preheating of fuel or
oxygen.

Fig. 1. An ingot for bearing steel production is lifted out of a soaking pit furnace at
Ascométal in France. The furnace is fired with flameless oxyfuel, heating the ingots
uniformly to over 1200°C.

In addition to using a burner for the combustion, which normally is operated at
stoichiometric conditions, two other technologies should be mentioned: lancing, and post-
combustion. Lancing refers to injecting oxygen, sometimes at very high velocities into
furnace free-space or a melt. It is done to intensify the air-fuel combustion, either to combust
for example carbon into CO, or achieve a complete combustion of a fuel into products like
CO
2
and H
2
O. Typically it is employed an Electric Arc Furnace (EAF) for scrap melting, but
it could also be like in the case of the REBOX
®
HLL technology to improve reheating.
Post-combustion does in most cases in this context refer to the reaction CO+½O
2
=CO
2
,
which is strongly exothermic; the released energy is typically used to improve melting.
Generally speaking, the prerequisites for a beneficial post-combustion are CO generation,
Energy Effciency 86
oxygen available, and a high heat transfer. For example, charging coal with the scrap in an
EAF so that it dissolves into the hot heel and blowing oxygen into the hot heel at
simultaneous over-stoichiometric operation of the oxyfuel burners when there is scrap in the
furnace, provides such wanted conditions. Post-combustion at flat bath operation, on the
other hand, normally provides too low heat transfer efficiency.

2.1. Flameless oxyfuel combustion
Some very interesting technologies have emerged in parallel with conventional oxyfuel,
which is widely used to boost melting in electric arc furnaces. The most important ones are
flameless combustion and Direct Flame Impingement (DFI). These new technologies not
only fulfil existing needs with astonishing results, they also open up completely new areas
of application. The flameless combustion creates a huge practically invisible oxyfuel flame
whereas the DFI technology uses small, well-defined sharp flames.
Increasingly stricter legislation on emissions led to the development of flameless oxyfuel,
which was introduced for the first time in 2003 in continuous furnaces for strip annealing
and slabs reheating, both at the stainless steel producer Outokumpu. The expression
'flameless combustion' communicates the visual aspect of the combustion type, that is, the
flame is no longer seen or easily detected by the human eye. Another description might be
that combustion is 'extended' in time and space – it is spread out in large volumes, and this
is why it is sometimes referred to as 'volume combustion'. Such a flame has a uniform and
lower temperature, yet containing same amount of energy.
In flameless oxyfuel the mixture of fuel and oxidant reacts uniformly through flame volume,
with the rate controlled by partial pressures of reactants and their temperature. The
flameless oxyfuel burners effectively disperse the combustion gases throughout the furnace,
ensuring more effective and uniform heating of the material even with a limited number of
burners installed. The lower flame temperature is substantially reducing the NO
X
formation.
Low NO
X
emission is also important from a global warming perspective; NO
2
has a so-
called Global Warming Potential that is almost 300 times that of CO
2
.


Fig. 2. The principle way of creating flameless combustion; the flame is diluted by the hot
furnace gases. This reduces the flame temperature to avoid creation of thermal NO
X
and to
achieve a more homogenous heating of the steel.

Compared with conventional oxyfuel, flameless oxyfuel provides even higher production
rates, excellent temperature uniformity and very low NO
X
emissions. The first installations
of this innovative flameless oxyfuel technology were made by Linde. Since 2003 over 30
installations of this technology have been made at more than a dozen sites, some even using
a low calorific fuel. There seems to be an increasing need to combust low calorific fuels; for
fuels containing below 2 kWh/m
3
, use of oxygen is an absolute requirement for flame
temperature and stability. At integrated steel mills use of blast furnace top gas (<1
kWh/m
3
), alone or in combination with other external or internal fuels, is becoming
increasingly important. Flameless oxyfuel can be successfully employed here.
The first installations of flameless oxyfuel took place in reheating and annealing, but it was
quickly adopted also for preheating of ladles and converters. The next area to be exploited,
with substantial positive impact, is the blast furnace hot stoves.


Fig. 3. A comparison of the results from installations at Ovako’s Hofors Works, Sweden
using different combustion technologies. When conventional oxyfuel was installed the
heating time decreased by 30%, but with flameless oxyfuel it was possible to run a heating
time half of the original one with air-fuel. It should be noted that the power has not been
increased, but decreased.


Fig. 4. A flameless oxyfuel flame; the flame is diluted and almost transparent. The
combustion in this photograph is using a low calorific fuel.
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 87
oxygen available, and a high heat transfer. For example, charging coal with the scrap in an
EAF so that it dissolves into the hot heel and blowing oxygen into the hot heel at
simultaneous over-stoichiometric operation of the oxyfuel burners when there is scrap in the
furnace, provides such wanted conditions. Post-combustion at flat bath operation, on the
other hand, normally provides too low heat transfer efficiency.

2.1. Flameless oxyfuel combustion
Some very interesting technologies have emerged in parallel with conventional oxyfuel,
which is widely used to boost melting in electric arc furnaces. The most important ones are
flameless combustion and Direct Flame Impingement (DFI). These new technologies not
only fulfil existing needs with astonishing results, they also open up completely new areas
of application. The flameless combustion creates a huge practically invisible oxyfuel flame
whereas the DFI technology uses small, well-defined sharp flames.
Increasingly stricter legislation on emissions led to the development of flameless oxyfuel,
which was introduced for the first time in 2003 in continuous furnaces for strip annealing
and slabs reheating, both at the stainless steel producer Outokumpu. The expression
'flameless combustion' communicates the visual aspect of the combustion type, that is, the
flame is no longer seen or easily detected by the human eye. Another description might be
that combustion is 'extended' in time and space – it is spread out in large volumes, and this
is why it is sometimes referred to as 'volume combustion'. Such a flame has a uniform and
lower temperature, yet containing same amount of energy.
In flameless oxyfuel the mixture of fuel and oxidant reacts uniformly through flame volume,
with the rate controlled by partial pressures of reactants and their temperature. The
flameless oxyfuel burners effectively disperse the combustion gases throughout the furnace,
ensuring more effective and uniform heating of the material even with a limited number of
burners installed. The lower flame temperature is substantially reducing the NO
X
formation.
Low NO
X
emission is also important from a global warming perspective; NO
2
has a so-
called Global Warming Potential that is almost 300 times that of CO
2
.


Fig. 2. The principle way of creating flameless combustion; the flame is diluted by the hot
furnace gases. This reduces the flame temperature to avoid creation of thermal NO
X
and to
achieve a more homogenous heating of the steel.

Compared with conventional oxyfuel, flameless oxyfuel provides even higher production
rates, excellent temperature uniformity and very low NO
X
emissions. The first installations
of this innovative flameless oxyfuel technology were made by Linde. Since 2003 over 30
installations of this technology have been made at more than a dozen sites, some even using
a low calorific fuel. There seems to be an increasing need to combust low calorific fuels; for
fuels containing below 2 kWh/m
3
, use of oxygen is an absolute requirement for flame
temperature and stability. At integrated steel mills use of blast furnace top gas (<1
kWh/m
3
), alone or in combination with other external or internal fuels, is becoming
increasingly important. Flameless oxyfuel can be successfully employed here.
The first installations of flameless oxyfuel took place in reheating and annealing, but it was
quickly adopted also for preheating of ladles and converters. The next area to be exploited,
with substantial positive impact, is the blast furnace hot stoves.


Fig. 3. A comparison of the results from installations at Ovako’s Hofors Works, Sweden
using different combustion technologies. When conventional oxyfuel was installed the
heating time decreased by 30%, but with flameless oxyfuel it was possible to run a heating
time half of the original one with air-fuel. It should be noted that the power has not been
increased, but decreased.


Fig. 4. A flameless oxyfuel flame; the flame is diluted and almost transparent. The
combustion in this photograph is using a low calorific fuel.
Energy Effciency 88
2.2. Direct Flame Impingement
It is also possible to fire with oxyfuel flames directly onto a material. This is what we call
DFI, Direct Flame Impingement. DFI Oxyfuel is a fascinating compact high heat transfer
technology, which since 2002 provides enhanced operation in strip processing lines, for
example at galvanizing. It is patented by Linde. So far the use of DFI Oxyfuel has been to
boost strip annealing and hot dip metal coating lines. Use of DFI Oxyfuel reduces the
specific fuel consumption while delivering a powerful 30% capacity increase, or more.
Installations are found at Outokumpu’s Nyby Works in Sweden and ThyssenKrupp at
Finnentrop and Bruckhausen in Germany. The latest installation is in a continuous
annealing line at POSCO in Pohang, South Korea. Due to DFI’s effective pre-cleaning
properties there are also benefits relating to surface appearance and improved adhesion of
metal coatings.
The main benefits of DFI Oxyfuel are:
 Significantly higher heat transfer resulting in increased production capacity
 Lower fuel consumption
 Ability to use high power input in limited furnace volume
 Compact and powerful unit for easy retrofit
 Heating and cleaning in one operation
 Option to modify surface conditions
Tests have verified the higher level of local heat flux for the DFI Oxyfuel technology. In
general, the use of oxyfuel combustion substantially increases the thermal efficiency of a
furnace. This is primarily due to the fact that radiant heat transfer of furnace gases produced
by oxyfuel combustion is significantly more efficient than those of air-fuel. Due to the
absence of nitrogen in the combustion mixture, which does not need to be heated, the
volume of exhaust gas is also substantially reduced, thus lowering total heat loss through
the exhaust gas. Thanks to improved thermal efficiency, the heating rate and productivity
are increased and less fuel is required to heat the product to a given temperature, while at
the same time economizing on fuel and reducing CO
2
emissions.


Fig. 5. The principle of DFI with many small oxyfuel flames heating directly onto the
material.
The DFI unit has a thermal efficiency of around 80%. This reduces the specific fuel
consumption while delivering a powerful 30% capacity increase in an existing strip
processing line. In galvanizing, zinc adhesion and surface appearance are also improved
due to DFI’s effective pre-cleaning properties, leaving both strip and furnace rolls cleaner
than before.
It is important to note that applying a DFI Oxyfuel system for preheating a strip does not
create an oxidation problem; for example, experience with preheating up to 300°C shows no
problems. In metal coating lines, the thin oxide layer formed is reduced in the subsequent
reduction zone. It is also possible to influence the oxidation level to a certain extent by
adjusting the stoichiometry of the flames.

3. At the integrated iron and steel processes
To produce iron in solid or liquid form from iron ore a reductant is needed. The two most
suitable reductants, from a technical and economic perspective, are carbon and hydrogen.
The use of pure hydrogen is today frequently not realistic; there are a few exceptions but
these are linked to unusual localised conditions. In practice, the reductants used in today’s
iron-making are coal and natural gas; the coal being in the form of coke or pulverized coal.
Blast furnaces inevitably require coke to a certain extent. It should be noted that the lowest
operational limit of coke in a shaft furnace has been estimated at 150-200 kg/t. This is
determined by the requirements for the carburization of iron, direct reduction with carbon
and, in particular, shaft permeability and burden support.
To improve throughput and decrease of CO
2
emissions, so-called Full Oxygen Blast Furnace
processes are frequently discussed as a possible alternative. The idea of the Full Oxygen
Blast Furnace (FOBF) is not new; some researchers discussed it back in the 1930s and 1950s.
The modern ideas were presented in the 1980s. They are based on two main principles:
using pure oxygen as blast instead of air to create oxyfuel combustion of the injected
pulverized coal, and acquiring a pre-reduction degree of iron (for example, 90%). However,
FOBF processes are hampered by the so-called “hot bottom and cold top problem”. Since
Fink presented a proposal for an FOBF process in a patent in 1978, the idea of using
recirculated top gas for compensating the decreasing amount of shaft gas and adjusting the
very high flame temperature has been the basis for most other proposals.
Lately, FOBF processes – which have not yet been taken into full-scale operation – have
experienced a renaissance, seen by many as the best way to decrease CO
2
emissions. The
potential benefit of FOBF processes lies in the possibility of achieving a top gas with low
nitrogen content (with a calorific value of 2 kWh/Nm
3
, more than twice that of conventional
blast furnaces) from which the CO
2
can then be removed reasonably effectively. The top gas
is then recirculated back into the furnace as part of the fuel input. The captured CO
2
can be
disposed of so that it is not discharged into the atmosphere or, for example, used in oil and
natural gas fields. Using FOBF processes is a possible solution, but it is not yet a proven
alternative.
Large benefits can be achieved from improved utilization of gases from other facilities on
site, like coke oven gas from the coke-making and BOF gas from the steel-making converter
(Basic Oxygen Furnaces – BOF). As the energy content of those gases are rather low, from
half that of natural gas and downwards, combustion with oxygen is very beneficial or even
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 89
2.2. Direct Flame Impingement
It is also possible to fire with oxyfuel flames directly onto a material. This is what we call
DFI, Direct Flame Impingement. DFI Oxyfuel is a fascinating compact high heat transfer
technology, which since 2002 provides enhanced operation in strip processing lines, for
example at galvanizing. It is patented by Linde. So far the use of DFI Oxyfuel has been to
boost strip annealing and hot dip metal coating lines. Use of DFI Oxyfuel reduces the
specific fuel consumption while delivering a powerful 30% capacity increase, or more.
Installations are found at Outokumpu’s Nyby Works in Sweden and ThyssenKrupp at
Finnentrop and Bruckhausen in Germany. The latest installation is in a continuous
annealing line at POSCO in Pohang, South Korea. Due to DFI’s effective pre-cleaning
properties there are also benefits relating to surface appearance and improved adhesion of
metal coatings.
The main benefits of DFI Oxyfuel are:
 Significantly higher heat transfer resulting in increased production capacity
 Lower fuel consumption
 Ability to use high power input in limited furnace volume
 Compact and powerful unit for easy retrofit
 Heating and cleaning in one operation
 Option to modify surface conditions
Tests have verified the higher level of local heat flux for the DFI Oxyfuel technology. In
general, the use of oxyfuel combustion substantially increases the thermal efficiency of a
furnace. This is primarily due to the fact that radiant heat transfer of furnace gases produced
by oxyfuel combustion is significantly more efficient than those of air-fuel. Due to the
absence of nitrogen in the combustion mixture, which does not need to be heated, the
volume of exhaust gas is also substantially reduced, thus lowering total heat loss through
the exhaust gas. Thanks to improved thermal efficiency, the heating rate and productivity
are increased and less fuel is required to heat the product to a given temperature, while at
the same time economizing on fuel and reducing CO
2
emissions.


Fig. 5. The principle of DFI with many small oxyfuel flames heating directly onto the
material.
The DFI unit has a thermal efficiency of around 80%. This reduces the specific fuel
consumption while delivering a powerful 30% capacity increase in an existing strip
processing line. In galvanizing, zinc adhesion and surface appearance are also improved
due to DFI’s effective pre-cleaning properties, leaving both strip and furnace rolls cleaner
than before.
It is important to note that applying a DFI Oxyfuel system for preheating a strip does not
create an oxidation problem; for example, experience with preheating up to 300°C shows no
problems. In metal coating lines, the thin oxide layer formed is reduced in the subsequent
reduction zone. It is also possible to influence the oxidation level to a certain extent by
adjusting the stoichiometry of the flames.

3. At the integrated iron and steel processes
To produce iron in solid or liquid form from iron ore a reductant is needed. The two most
suitable reductants, from a technical and economic perspective, are carbon and hydrogen.
The use of pure hydrogen is today frequently not realistic; there are a few exceptions but
these are linked to unusual localised conditions. In practice, the reductants used in today’s
iron-making are coal and natural gas; the coal being in the form of coke or pulverized coal.
Blast furnaces inevitably require coke to a certain extent. It should be noted that the lowest
operational limit of coke in a shaft furnace has been estimated at 150-200 kg/t. This is
determined by the requirements for the carburization of iron, direct reduction with carbon
and, in particular, shaft permeability and burden support.
To improve throughput and decrease of CO
2
emissions, so-called Full Oxygen Blast Furnace
processes are frequently discussed as a possible alternative. The idea of the Full Oxygen
Blast Furnace (FOBF) is not new; some researchers discussed it back in the 1930s and 1950s.
The modern ideas were presented in the 1980s. They are based on two main principles:
using pure oxygen as blast instead of air to create oxyfuel combustion of the injected
pulverized coal, and acquiring a pre-reduction degree of iron (for example, 90%). However,
FOBF processes are hampered by the so-called “hot bottom and cold top problem”. Since
Fink presented a proposal for an FOBF process in a patent in 1978, the idea of using
recirculated top gas for compensating the decreasing amount of shaft gas and adjusting the
very high flame temperature has been the basis for most other proposals.
Lately, FOBF processes – which have not yet been taken into full-scale operation – have
experienced a renaissance, seen by many as the best way to decrease CO
2
emissions. The
potential benefit of FOBF processes lies in the possibility of achieving a top gas with low
nitrogen content (with a calorific value of 2 kWh/Nm
3
, more than twice that of conventional
blast furnaces) from which the CO
2
can then be removed reasonably effectively. The top gas
is then recirculated back into the furnace as part of the fuel input. The captured CO
2
can be
disposed of so that it is not discharged into the atmosphere or, for example, used in oil and
natural gas fields. Using FOBF processes is a possible solution, but it is not yet a proven
alternative.
Large benefits can be achieved from improved utilization of gases from other facilities on
site, like coke oven gas from the coke-making and BOF gas from the steel-making converter
(Basic Oxygen Furnaces – BOF). As the energy content of those gases are rather low, from
half that of natural gas and downwards, combustion with oxygen is very beneficial or even
Energy Effciency 90
necessary. This could take place when injecting into the blast furnace or when using them as
fuel in different types of heating operations. This also applies to use of blast furnace top gas.


Fig. 6. Blast furnace hot stoves, the large heat exchangers for heating the air-blast to over
1000°C prior to injection into the blast furnace tuyères.

An area where increased use of blast furnace top gas could be very beneficial is at the hot
stoves. Use of flameless oxyfuel in blast furnace hot stoves is now under way. An evaluation
of applying the technology in a large modern and efficient blast furnace, which produces a
low calorific top gas, includes the following key observations:
 25% of flue-gas can be recycled and this leads to a modified flue-gas composition
containing 60% CO
2
, accordingly increasingly suitable for Carbon Capture and
Sequestration.
 The heat transfer by radiation is increased by 15% relative to conventional practices
and this will manifest as improved stove efficiency.
 60% of the fuel gas enrichment can be eliminated.
 The total energy use at the hot stove is reduced by 5%.
 The temperature of the blast increases by about 15°C.

Use of flameless oxyfuel in blast furnace hot stoves could, thus, replace combustion of coke
oven gas or natural gas in this process with blast furnace top gas. This would typically cover
the cost for the oxygen, or even provide a minor cost saving. The coke savings arising from
an increase blast temperature will be substantial. In addition to the energy and
environmental benefits, the stoves campaign life will be increased.

4. At the electric arc furnace
Today’s electric arc furnace (EAF) for producing steel from scrap can be considered as a
very sophisticated piece of equipment. During the 20
th
century, the development of electric
steelmaking has been tremendous. In 1910, the electric furnaces, including both EAFs and
induction furnaces, produced 0.2% of the world steel production, today this figure is nearly
35%. The two main factors explaining this evolution are the increased scrap availability and
the development of ladle metallurgy, especially with the introduction of the ladle furnace
(ASEA-SKF in 1965), which made it possible to increase both the production rate and the
product quality. We should also bare in mind the favour of a much lower capital
requirement as compared with the integrated steel mills.
When operating a modern EAF, the energy turnover is about 650 kWh/t, but only 60% of
that energy is needed to heat and melt the scrap. A decrease of the energy turnover as such
can be a goal, but many EAF mills consider a decrease of the electricity consumption as a
more important way to cut costs and to enable an increased production rate. Decreased
electricity consumption also offers additional advantages, such as lowered costs for
electrodes and less disturbances on the grid. Moreover, electricity prices are at many places
increasing. The electricity consumption can be lowered either by decreasing the total energy
turnover or by replacing the electricity with energy in another form. It should be noted, that
the electrical transmission losses are a direct function of the electricity consumption,
representing >6% of the electricity supplied in an AC furnace and even more in a DC
furnace.
The increasing use of oxygen has been very important for the development of EAF
steelmaking. It begun with the (manual) oxygen lancing, in a first step used to replace the
iron ore added during the refining period, but via oxyfuel burners and post-combustion it
has developed into a number of more and more sophisticated applications. Today there are
EAFs with a specific oxygen use above 50 Nm
3
/t, more than the Basic Oxygen Furnace
(BOF) in integrated steel mills.
The average ratio between the electricity savings and the oxygen use, should be about 3.5
kWh/Nm
3
O
2
. When introducing oxygen into an EAF, oxyfuel burners and oxygen lancing
are employed in a first stage up to a total use of some 20 to 25 Nm
3
/t usually with savings in
electricity of about 5 kWh/Nm
3
O
2
or more and with a corresponding increase of the
production rate. When evaluating the overall reaction for oxygen lancing, (C+½O
2
=CO), one
should expect electricity savings to be maximum about 2 kWh/Nm
3
O
2
even taking into
account the higher contribution from dissolved carbon in steel scrap and adding energy
corresponding to a possible post-combustion value of 8% in the bath-slag system. However,
the much higher savings actually achieved, can be explained as follows. The overall reaction
takes place in two steps: (1) the injected oxygen immediately combines with iron to form
iron oxide, a strongly exothermic reaction, and (2) iron oxide in the slag is reduced by
carbon, an endothermic reaction. The first reaction releases almost four times more heat per
Nm
3
O
2
than the overall reaction and this heat will be absorbed by surrounding scrap and
significantly speed up the melt-down process.
Operating an EAF with under-pressure and especially with the slag door open during most
of the operation leads to a heavy in-leakage of air. The oxygen part of this air could of
course be of use inside the furnace, but the nitrogen (and argon) part is only to be
considered as ballast. The energy demand for heating-up the ballast nitrogen, due to in-
leakage of air, is 50-60 kWh/t. Even much higher figures, above 100 kWh/t, have been
found at several EAF shops. The solution to this is to keep the slag door shut during the
main part of the operation and run the furnace with a slight overpressure.
Since oxygen lancing was introduced, it has at most EAF shops been carried out by lancing
through the slag door. Even if this way of lancing allows moving the injection point in all
directions, the oxygen introduced will not be equally distributed throughout the bath. The
trend of the EAF becoming more and more air-tight and the dynamic impact of a shorter
meltdown time made it increasingly harder to use the conventional way of lancing oxygen
and coal through the slag door. Nowadays we find combined equipment including all the
functions: oxygen lancing, coal lancing, oxyfuel burners, and post-combustion. This
equipment can be considered as combined lance-burners often with a coherent jet function
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 91
necessary. This could take place when injecting into the blast furnace or when using them as
fuel in different types of heating operations. This also applies to use of blast furnace top gas.


Fig. 6. Blast furnace hot stoves, the large heat exchangers for heating the air-blast to over
1000°C prior to injection into the blast furnace tuyères.

An area where increased use of blast furnace top gas could be very beneficial is at the hot
stoves. Use of flameless oxyfuel in blast furnace hot stoves is now under way. An evaluation
of applying the technology in a large modern and efficient blast furnace, which produces a
low calorific top gas, includes the following key observations:
 25% of flue-gas can be recycled and this leads to a modified flue-gas composition
containing 60% CO
2
, accordingly increasingly suitable for Carbon Capture and
Sequestration.
 The heat transfer by radiation is increased by 15% relative to conventional practices
and this will manifest as improved stove efficiency.
 60% of the fuel gas enrichment can be eliminated.
 The total energy use at the hot stove is reduced by 5%.
 The temperature of the blast increases by about 15°C.

Use of flameless oxyfuel in blast furnace hot stoves could, thus, replace combustion of coke
oven gas or natural gas in this process with blast furnace top gas. This would typically cover
the cost for the oxygen, or even provide a minor cost saving. The coke savings arising from
an increase blast temperature will be substantial. In addition to the energy and
environmental benefits, the stoves campaign life will be increased.

4. At the electric arc furnace
Today’s electric arc furnace (EAF) for producing steel from scrap can be considered as a
very sophisticated piece of equipment. During the 20
th
century, the development of electric
steelmaking has been tremendous. In 1910, the electric furnaces, including both EAFs and
induction furnaces, produced 0.2% of the world steel production, today this figure is nearly
35%. The two main factors explaining this evolution are the increased scrap availability and
the development of ladle metallurgy, especially with the introduction of the ladle furnace
(ASEA-SKF in 1965), which made it possible to increase both the production rate and the
product quality. We should also bare in mind the favour of a much lower capital
requirement as compared with the integrated steel mills.
When operating a modern EAF, the energy turnover is about 650 kWh/t, but only 60% of
that energy is needed to heat and melt the scrap. A decrease of the energy turnover as such
can be a goal, but many EAF mills consider a decrease of the electricity consumption as a
more important way to cut costs and to enable an increased production rate. Decreased
electricity consumption also offers additional advantages, such as lowered costs for
electrodes and less disturbances on the grid. Moreover, electricity prices are at many places
increasing. The electricity consumption can be lowered either by decreasing the total energy
turnover or by replacing the electricity with energy in another form. It should be noted, that
the electrical transmission losses are a direct function of the electricity consumption,
representing >6% of the electricity supplied in an AC furnace and even more in a DC
furnace.
The increasing use of oxygen has been very important for the development of EAF
steelmaking. It begun with the (manual) oxygen lancing, in a first step used to replace the
iron ore added during the refining period, but via oxyfuel burners and post-combustion it
has developed into a number of more and more sophisticated applications. Today there are
EAFs with a specific oxygen use above 50 Nm
3
/t, more than the Basic Oxygen Furnace
(BOF) in integrated steel mills.
The average ratio between the electricity savings and the oxygen use, should be about 3.5
kWh/Nm
3
O
2
. When introducing oxygen into an EAF, oxyfuel burners and oxygen lancing
are employed in a first stage up to a total use of some 20 to 25 Nm
3
/t usually with savings in
electricity of about 5 kWh/Nm
3
O
2
or more and with a corresponding increase of the
production rate. When evaluating the overall reaction for oxygen lancing, (C+½O
2
=CO), one
should expect electricity savings to be maximum about 2 kWh/Nm
3
O
2
even taking into
account the higher contribution from dissolved carbon in steel scrap and adding energy
corresponding to a possible post-combustion value of 8% in the bath-slag system. However,
the much higher savings actually achieved, can be explained as follows. The overall reaction
takes place in two steps: (1) the injected oxygen immediately combines with iron to form
iron oxide, a strongly exothermic reaction, and (2) iron oxide in the slag is reduced by
carbon, an endothermic reaction. The first reaction releases almost four times more heat per
Nm
3
O
2
than the overall reaction and this heat will be absorbed by surrounding scrap and
significantly speed up the melt-down process.
Operating an EAF with under-pressure and especially with the slag door open during most
of the operation leads to a heavy in-leakage of air. The oxygen part of this air could of
course be of use inside the furnace, but the nitrogen (and argon) part is only to be
considered as ballast. The energy demand for heating-up the ballast nitrogen, due to in-
leakage of air, is 50-60 kWh/t. Even much higher figures, above 100 kWh/t, have been
found at several EAF shops. The solution to this is to keep the slag door shut during the
main part of the operation and run the furnace with a slight overpressure.
Since oxygen lancing was introduced, it has at most EAF shops been carried out by lancing
through the slag door. Even if this way of lancing allows moving the injection point in all
directions, the oxygen introduced will not be equally distributed throughout the bath. The
trend of the EAF becoming more and more air-tight and the dynamic impact of a shorter
meltdown time made it increasingly harder to use the conventional way of lancing oxygen
and coal through the slag door. Nowadays we find combined equipment including all the
functions: oxygen lancing, coal lancing, oxyfuel burners, and post-combustion. This
equipment can be considered as combined lance-burners often with a coherent jet function
Energy Effciency 92
enabling high-velocity injection, with a device for coal injection, where the burner also can
be run overstoichiometrically to provide post-combustion or there is a separate nozzle for
oxygen injection. To secure a good distribution of the heat supply throughout the furnace,
including also the rear end of an Eccentric Bottom Tapping (EBT) type furnace, and the
advantage of combining oxygen injection with oxyfuel burner operation, we end up with a
minimum of four wall-mounted injection devices (assuming an AC furnace) – one at each
cold spot between the electrodes and one in the EBT area.
The main factor limiting the energy supply from oxyfuel burners in an EAF is the heat
transfer efficiency, which decreases with increased scrap temperature - we here have to
compare with heat transfer from the electric arcs. However, as long as this heat transfer
efficiency enables a decreased average cost for the production, it is of course beneficial to
run the oxyfuel burners. Generally speaking, this normally means operation of the oxyfuel
burners during about half of the time needed for the melting of each bucket of scrap
charged, but the time is also a function of the production rate demand.
The CO/CO
2
ratio in equilibrium with liquid steel is high, even at low carbon contents. This
result in a CO-rich gas leaving the bath-slag system in the furnace providing a potential for
large energy recovery if this CO can be burnt with O
2
into CO
2
and the heat released be
transferred to the metal. To illustrate the potential of post-combustion, we can say that in an
EAF operation with a high coal injection, the energy released from the formation of CO is
about 25 kWh/t, but if this entire CO can be transferred into CO
2
the total amount of energy
released will be about 140 kWh/t. This should preferably be done with pure oxygen in order
to minimize losses to the flue-gases.
Electricity savings from post-combustion are in the range 3-5 kWh/Nm
3
O
2
, and can be
obtained with rather simple means such as oxygen injection at fixed flow rates through
existing oxyfuel burners during fixed periods of time, or by running the oxyfuel burners
overstoichiometrically. For reaching high values, oxygen flow control through on-line flue-
gas analysis and separate post-combustion lances can be used, making a heat recovery of 60-
75% reasonable.

5. At vessel preheating
The use of oxyfuel to preheat vessels such as torpedoes, ladles and converters has been
around for several decades. However, the number of installations is still surprisingly low
given its potential. Using oxyfuel instead of air-fuel would reduce the fuel consumption
drastically by approximately 50%, which would bring about a proportional decrease in CO
2

emissions. However, it would also have additional benefits such as a shorter heating time
and hotter vessels. These would, for example, lead to fewer ladles in circulation and the
possibility of reducing tapping temperatures. The latter directly saves energy in the furnace,
but it could also decrease the tap-to-tap time of the furnace. The time saving would lead to
additional energy savings as the specific (time dependent) heat losses from a furnace, would
then be lowered.
If an oxyfuel ladle preheating system is installed adjacent to the EAF, preferably just a few
metres away from the tapping position, very hot ladles can be used. Experience shows that
such a measure would allow 20 minutes decreased ladle cycle and a 15°C lowered EAF
tapping temperature, providing electricity savings at 5-6 kWh/t.
Let us look at a proven example of what this could lead to. The operating power with
oxyfuel for a 60t ladle is approximately 1.2 MW. The average annual level is 0.8 MW, which
at 7,500 h/y means 6 GWh/y. This is around half of what would be required with air-fuel;
thus the annual saving is 6 GWh. Assuming the fuel is natural gas, the resulting decrease in
CO
2
emissions would be 1,200 t/y, and this is only for one preheating station; normally
there are multiple at each site.


Fig. 7. Ladle preheating using flameless oxyfuel at Ovako’s Hofors Works, Sweden.

Conventional oxyfuel delivers a simple, compact and low weight installation as compared
to an air-fuel system with a recuperator or regenerative solution. However, in preheating of
vessels flameless oxyfuel brings additional strong advantages. Flameless oxyfuel is seen as
the best available technology for heating and not only allows for ultra low NO
X
emissions,
but brings extended refractory life through more uniform temperature distribution. The first
installation took place in 2003. Today more than 15 installations of flameless oxyfuel are in
operation, two recent cases are found at Outokumpu at Tornio, Finland and SKF at
Katrineholm, Sweden.
In 2008 flameless oxyfuel preheating was installed at Outokumpu’s 90 tonnes ferrochrome
converter. The 2.5 MW flameless oxyfuel system is used for drying and preheating of the
converter, and provides the Tornio Works with greater energy efficiency, lower fuel
consumption, and reduced emissions CO
2
and NO
X
.
At SKF a similar type of flameless oxyfuel technology was installed last year, but for
preheating ladles. And the size is here completely different; the ladles are for just 1 tonne of
steel. Six ladle preheating stands were equipped with OXYGON
®
flameless oxyfuel
preheating systems. This installation shows that a new energy saving and environmentally
friendly technology also can be viable in a smaller scale production.

6. At reheating
Prompted by rapidly rising fuel prices in the 1970s, the steel industry began to consider
methods to reduce fuel consumption in reheating and annealing. This laid the foundation
for the use of oxyfuel solutions in rolling mills and forge shops. In the mid 1980s, some of
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 93
enabling high-velocity injection, with a device for coal injection, where the burner also can
be run overstoichiometrically to provide post-combustion or there is a separate nozzle for
oxygen injection. To secure a good distribution of the heat supply throughout the furnace,
including also the rear end of an Eccentric Bottom Tapping (EBT) type furnace, and the
advantage of combining oxygen injection with oxyfuel burner operation, we end up with a
minimum of four wall-mounted injection devices (assuming an AC furnace) – one at each
cold spot between the electrodes and one in the EBT area.
The main factor limiting the energy supply from oxyfuel burners in an EAF is the heat
transfer efficiency, which decreases with increased scrap temperature - we here have to
compare with heat transfer from the electric arcs. However, as long as this heat transfer
efficiency enables a decreased average cost for the production, it is of course beneficial to
run the oxyfuel burners. Generally speaking, this normally means operation of the oxyfuel
burners during about half of the time needed for the melting of each bucket of scrap
charged, but the time is also a function of the production rate demand.
The CO/CO
2
ratio in equilibrium with liquid steel is high, even at low carbon contents. This
result in a CO-rich gas leaving the bath-slag system in the furnace providing a potential for
large energy recovery if this CO can be burnt with O
2
into CO
2
and the heat released be
transferred to the metal. To illustrate the potential of post-combustion, we can say that in an
EAF operation with a high coal injection, the energy released from the formation of CO is
about 25 kWh/t, but if this entire CO can be transferred into CO
2
the total amount of energy
released will be about 140 kWh/t. This should preferably be done with pure oxygen in order
to minimize losses to the flue-gases.
Electricity savings from post-combustion are in the range 3-5 kWh/Nm
3
O
2
, and can be
obtained with rather simple means such as oxygen injection at fixed flow rates through
existing oxyfuel burners during fixed periods of time, or by running the oxyfuel burners
overstoichiometrically. For reaching high values, oxygen flow control through on-line flue-
gas analysis and separate post-combustion lances can be used, making a heat recovery of 60-
75% reasonable.

5. At vessel preheating
The use of oxyfuel to preheat vessels such as torpedoes, ladles and converters has been
around for several decades. However, the number of installations is still surprisingly low
given its potential. Using oxyfuel instead of air-fuel would reduce the fuel consumption
drastically by approximately 50%, which would bring about a proportional decrease in CO
2

emissions. However, it would also have additional benefits such as a shorter heating time
and hotter vessels. These would, for example, lead to fewer ladles in circulation and the
possibility of reducing tapping temperatures. The latter directly saves energy in the furnace,
but it could also decrease the tap-to-tap time of the furnace. The time saving would lead to
additional energy savings as the specific (time dependent) heat losses from a furnace, would
then be lowered.
If an oxyfuel ladle preheating system is installed adjacent to the EAF, preferably just a few
metres away from the tapping position, very hot ladles can be used. Experience shows that
such a measure would allow 20 minutes decreased ladle cycle and a 15°C lowered EAF
tapping temperature, providing electricity savings at 5-6 kWh/t.
Let us look at a proven example of what this could lead to. The operating power with
oxyfuel for a 60t ladle is approximately 1.2 MW. The average annual level is 0.8 MW, which
at 7,500 h/y means 6 GWh/y. This is around half of what would be required with air-fuel;
thus the annual saving is 6 GWh. Assuming the fuel is natural gas, the resulting decrease in
CO
2
emissions would be 1,200 t/y, and this is only for one preheating station; normally
there are multiple at each site.


Fig. 7. Ladle preheating using flameless oxyfuel at Ovako’s Hofors Works, Sweden.

Conventional oxyfuel delivers a simple, compact and low weight installation as compared
to an air-fuel system with a recuperator or regenerative solution. However, in preheating of
vessels flameless oxyfuel brings additional strong advantages. Flameless oxyfuel is seen as
the best available technology for heating and not only allows for ultra low NO
X
emissions,
but brings extended refractory life through more uniform temperature distribution. The first
installation took place in 2003. Today more than 15 installations of flameless oxyfuel are in
operation, two recent cases are found at Outokumpu at Tornio, Finland and SKF at
Katrineholm, Sweden.
In 2008 flameless oxyfuel preheating was installed at Outokumpu’s 90 tonnes ferrochrome
converter. The 2.5 MW flameless oxyfuel system is used for drying and preheating of the
converter, and provides the Tornio Works with greater energy efficiency, lower fuel
consumption, and reduced emissions CO
2
and NO
X
.
At SKF a similar type of flameless oxyfuel technology was installed last year, but for
preheating ladles. And the size is here completely different; the ladles are for just 1 tonne of
steel. Six ladle preheating stands were equipped with OXYGON
®
flameless oxyfuel
preheating systems. This installation shows that a new energy saving and environmentally
friendly technology also can be viable in a smaller scale production.

6. At reheating
Prompted by rapidly rising fuel prices in the 1970s, the steel industry began to consider
methods to reduce fuel consumption in reheating and annealing. This laid the foundation
for the use of oxyfuel solutions in rolling mills and forge shops. In the mid 1980s, some of
Energy Effciency 94
these furnaces got equipped with oxygen-enrichment systems, which increased the oxygen
content of the combustion air to 23-24%. The results were encouraging: fuel consumption
was reduced and the output, in terms of tons per hour, increased.
Oxyfuel solutions deliver a unique combination of advantages in reheat and annealing.
Thanks to improved thermal efficiency (about 80% compared with 40-60% for air-fuel), the
heating rate and productivity are increased and less fuel is required to heat the product to
the desired temperature, at the same time saving on CO
2
and NO
X
emissions. In summary
the results include:
 Throughput capacity increase of up to 50%
 Fuel savings of up to 50%
 Reduction of CO
2
emissions by up to 50%
 Reduction of NO
X
emissions
 Reduction of scaling losses
In 1990, Linde converted the first steel reheating furnace in the world to operate with 100%
oxygen at Timken in the USA Since then, Linde has been pioneering the use of oxyfuel for
this application. Today there are 120 reheat furnaces and annealing lines using Linde’s
oxyfuel solutions. The best air-fuel solutions need at least 1.3 GJ for heating a tonne of steel
to the right temperature for rolling or forging. When using the REBOX oxyfuel solutions the
comparable figure is below 1 GJ, a saving of 25%. For continuous heating operations it is
also possible to economically operate the furnace at a higher temperature at the entry side of
the furnace. This will even further increase the possible throughput in any furnace unit.
Oxyfuel combustion allows all installation pipes and flow trains to be compact without any
need for recuperative or regenerative heat recovery solutions. Combustion air-blowers and
related low frequency noise problems are avoided.


Table 1. With oxyfuel it is possible to achieve an 80% thermal efficiency, as compared with
60% in the best air-fuel cases. Even if also adding the energy needed to produce the required
oxygen, we would reach 285 kWh/tonne, thus still close to 1 GJ, a saving of 20%.

During the last years flameless oxyfuel have been employed, for example in Brazil, China,
France, Sweden, and the USA. Here follows some examples from those installations.

Soaking pit furnaces at Ascométal
There are flameless oxyfuel installations at two sites belonging to the bearing steel producer
Ascométal in France, which is part of the Severstal Group. At Fos-sur-Mer, a turnkey
delivery in 2005-2007 converted nine soaking pit furnaces into all flameless oxyfuel. The
delivery included a combustion system with flameless burners, furnace upgrade, new flue-
gas system, flow train, and a control system. The furnace sizes are 80 to 120 tonne heating
capacity each. The results include 50% more heating capacity, 40% fuel savings, NO
X

emission reduced by 40%, and scale formation reduced with 3 tonne per 1,000 tonne heated.
In a second and similar project in France in 2007-2008, four soaking pit furnaces at the Les
Dunes plant were also converted into all flameless oxyfuel operation.


Fig. 8. Total average fuel consumption in the 13 soaking pit furnaces at Ascométal, Fos-sur-
Mer. 2001-2004 was all air-fuel combustion. The first conversion into oxyfuel took place in
2005. In 2007 nine out of 13 furnaces were operated with all oxyfuel. The average fuel
consumption per tonne for all furnaces was reduced by 100 kWh or 10 Nm
3
of natural gas.

15 installations at Outokumpu
At Outokumpu’s sites in Sweden there are a total of 15 installations. In 2003, a walking
beam furnace in Degerfors was rebuilt and refurbished in a Linde turnkey project with
performance guarantees. It entailed replacing the air-fuel system, including recuperator,
with flameless oxyfuel, and installation of essential control systems. The resultant 40-50%
increase in heating capacity provided increased loading of the rolling mill, reduction of over
25% in fuel consumption and NO
X
emissions below 70 mg/MJ.
At the Nyby plant, there are two catenary furnaces, originally installed in 1955 and 1960
respectively. The catenary furnace on the first annealing-pickling line, for hot or cold rolled
strips, was converted to all oxyfuel operation in 2003. Requirements for increased
production combined with stricter requirements for low NO
X
emissions led to this decision.
The furnace, 18 m long, was equipped with flameless oxyfuel burners. The total power
input, 16 MW, was not altered when converting from air-fuel to oxyfuel, but with oxyfuel
the heat transfer efficiency increased from 46% to 76%. The replacement of the air-fuel
system, combustion blowers and recuperators resulted in a 50% increase in heating capacity
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 95
these furnaces got equipped with oxygen-enrichment systems, which increased the oxygen
content of the combustion air to 23-24%. The results were encouraging: fuel consumption
was reduced and the output, in terms of tons per hour, increased.
Oxyfuel solutions deliver a unique combination of advantages in reheat and annealing.
Thanks to improved thermal efficiency (about 80% compared with 40-60% for air-fuel), the
heating rate and productivity are increased and less fuel is required to heat the product to
the desired temperature, at the same time saving on CO
2
and NO
X
emissions. In summary
the results include:
 Throughput capacity increase of up to 50%
 Fuel savings of up to 50%
 Reduction of CO
2
emissions by up to 50%
 Reduction of NO
X
emissions
 Reduction of scaling losses
In 1990, Linde converted the first steel reheating furnace in the world to operate with 100%
oxygen at Timken in the USA Since then, Linde has been pioneering the use of oxyfuel for
this application. Today there are 120 reheat furnaces and annealing lines using Linde’s
oxyfuel solutions. The best air-fuel solutions need at least 1.3 GJ for heating a tonne of steel
to the right temperature for rolling or forging. When using the REBOX oxyfuel solutions the
comparable figure is below 1 GJ, a saving of 25%. For continuous heating operations it is
also possible to economically operate the furnace at a higher temperature at the entry side of
the furnace. This will even further increase the possible throughput in any furnace unit.
Oxyfuel combustion allows all installation pipes and flow trains to be compact without any
need for recuperative or regenerative heat recovery solutions. Combustion air-blowers and
related low frequency noise problems are avoided.


Table 1. With oxyfuel it is possible to achieve an 80% thermal efficiency, as compared with
60% in the best air-fuel cases. Even if also adding the energy needed to produce the required
oxygen, we would reach 285 kWh/tonne, thus still close to 1 GJ, a saving of 20%.

During the last years flameless oxyfuel have been employed, for example in Brazil, China,
France, Sweden, and the USA. Here follows some examples from those installations.

Soaking pit furnaces at Ascométal
There are flameless oxyfuel installations at two sites belonging to the bearing steel producer
Ascométal in France, which is part of the Severstal Group. At Fos-sur-Mer, a turnkey
delivery in 2005-2007 converted nine soaking pit furnaces into all flameless oxyfuel. The
delivery included a combustion system with flameless burners, furnace upgrade, new flue-
gas system, flow train, and a control system. The furnace sizes are 80 to 120 tonne heating
capacity each. The results include 50% more heating capacity, 40% fuel savings, NO
X

emission reduced by 40%, and scale formation reduced with 3 tonne per 1,000 tonne heated.
In a second and similar project in France in 2007-2008, four soaking pit furnaces at the Les
Dunes plant were also converted into all flameless oxyfuel operation.


Fig. 8. Total average fuel consumption in the 13 soaking pit furnaces at Ascométal, Fos-sur-
Mer. 2001-2004 was all air-fuel combustion. The first conversion into oxyfuel took place in
2005. In 2007 nine out of 13 furnaces were operated with all oxyfuel. The average fuel
consumption per tonne for all furnaces was reduced by 100 kWh or 10 Nm
3
of natural gas.

15 installations at Outokumpu
At Outokumpu’s sites in Sweden there are a total of 15 installations. In 2003, a walking
beam furnace in Degerfors was rebuilt and refurbished in a Linde turnkey project with
performance guarantees. It entailed replacing the air-fuel system, including recuperator,
with flameless oxyfuel, and installation of essential control systems. The resultant 40-50%
increase in heating capacity provided increased loading of the rolling mill, reduction of over
25% in fuel consumption and NO
X
emissions below 70 mg/MJ.
At the Nyby plant, there are two catenary furnaces, originally installed in 1955 and 1960
respectively. The catenary furnace on the first annealing-pickling line, for hot or cold rolled
strips, was converted to all oxyfuel operation in 2003. Requirements for increased
production combined with stricter requirements for low NO
X
emissions led to this decision.
The furnace, 18 m long, was equipped with flameless oxyfuel burners. The total power
input, 16 MW, was not altered when converting from air-fuel to oxyfuel, but with oxyfuel
the heat transfer efficiency increased from 46% to 76%. The replacement of the air-fuel
system, combustion blowers and recuperators resulted in a 50% increase in heating capacity
Energy Effciency 96
without any increase in the length of the furnace, a 40% reduction in specific fuel
consumption and NO
X
levels below the guaranteed level of 70 mg/MJ.
At Avesta we find the world’s largest oxyfuel fired furnace, 40 MW. The old 24 m catenary
furnace had a 75 tph capacity, but the requirement was to double this whilst at same time
meeting strict requirements for emissions. The refurbishment included a 10 m extension, yet
production capacity was increased to 150 tph. The conversion involved the removal of air-
fuel burners and recuperators and the installation of all oxyfuel. The oxyfuel technology
used involved staged combustion. The conversion reduced fuel consumption by 40%, and
NO
X
levels are below 65 mg/MJ. This furnace is an example of another route to flameless;
having been converted from conventional oxyfuel to flameless oxyfuel last year and
resulting in an additional 50% reduction of the NO
X
levels.


Fig. 9. A heated slab is discharged from the walking beam furnace at Outokumpu’s
Degerfors Works. Here flameless oxyfuel has increased the heating capacity by 40-50%.

50% fuel savings at ArcelorMittal
There have been several successful installations in rotary hearth furnaces. One is found at
ArcelorMittal Shelby in Ohio, USA. In 2007, Linde delivered a turnkey conversion of a 15-
metre diameter rotary hearth furnace at this seamless tube producer. It included combustion
system with flameless burners, furnace upgrade, new flue-gas system, flow train, and a
control system. The former air-fuel fired furnace was converted in two steps, first using
oxygen-enrichment for a period of time and then implementation of the flameless oxyfuel
operation. Excellent results have been achieved, meeting all performance guarantees. These
included >25% more throughput, 50% fuel savings compared with oxygen-enrichment (60%
below the prior air-fuel performance), CO
2
emissions dropped accordingly, NO
X
emission
<70 mg/MJ corresponding to 92% less on an annual basis, and 50% reduced scale formation.
In January 2010, ArcelorMittal’s received the Association for Iron & Steel Technology’s
Energy Achievement Award for its efforts to reduce fuel consumption and emissions using
the REBOX oxyfuel technical solution at its Shelby mill.


Fig. 10. Outside view of the rotary hearth furnace at ArcelorMittal Shelby after the
conversion into all flameless oxyfuel operation. Please note that all bulky equipment and
piping relating to the previously used air-fuel system have been removed as it no longer is
of any use.

SSAB Walking Beam Furnace with REBOX HLL
At SSAB in Sweden REBOX HLL is used. The slabs are reheated in walking beam furnaces
with a capacity of 300 tph per furnace, from ambient temperature to 1,230°C. The air-fuel
combustion system uses a recuperation system to preheat air to 400°C. The fuel is oil, and
prior to the HLL installation the consumption was 440 kWh/tonne, or 1.58 GJ/tonne.
REBOX HLL creates a type of flameless oxyfuel without replacing the existing air-fuel
burners. By reducing the air flow and substituting high velocity oxygen injection into the
combustion, great benefits can be achieved. 75% of the oxygen needed for the combustion is
supplied with this technique. The flue-gas volume is less than 45% that of air-fuel. The
system start-up took just one day. The installation in only one zone has increased the
heating capacity from 300 to 320 tph.
The installation of HLL is rather easy because it does not include any replacement of
burners or installation of additional burners, which minimizes the installation down time.
The air-fuel system can at any time be brought back into operation as it was before. This
eliminates any potential risk relating to the implemention, and it enables operation to be
flexible and optimized in response to fluctuating fuel cost and production requirements.
Some important results from this installation are:
 No negative impact on the surface quality.
 A positive impact on the temperature uniformity of the slabs.
 The ideal heating curve suggested by the control system can be achieved more
easily.
 Less smoke emanating from the furnace, greatly improving the plant environment.
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 97
without any increase in the length of the furnace, a 40% reduction in specific fuel
consumption and NO
X
levels below the guaranteed level of 70 mg/MJ.
At Avesta we find the world’s largest oxyfuel fired furnace, 40 MW. The old 24 m catenary
furnace had a 75 tph capacity, but the requirement was to double this whilst at same time
meeting strict requirements for emissions. The refurbishment included a 10 m extension, yet
production capacity was increased to 150 tph. The conversion involved the removal of air-
fuel burners and recuperators and the installation of all oxyfuel. The oxyfuel technology
used involved staged combustion. The conversion reduced fuel consumption by 40%, and
NO
X
levels are below 65 mg/MJ. This furnace is an example of another route to flameless;
having been converted from conventional oxyfuel to flameless oxyfuel last year and
resulting in an additional 50% reduction of the NO
X
levels.


Fig. 9. A heated slab is discharged from the walking beam furnace at Outokumpu’s
Degerfors Works. Here flameless oxyfuel has increased the heating capacity by 40-50%.

50% fuel savings at ArcelorMittal
There have been several successful installations in rotary hearth furnaces. One is found at
ArcelorMittal Shelby in Ohio, USA. In 2007, Linde delivered a turnkey conversion of a 15-
metre diameter rotary hearth furnace at this seamless tube producer. It included combustion
system with flameless burners, furnace upgrade, new flue-gas system, flow train, and a
control system. The former air-fuel fired furnace was converted in two steps, first using
oxygen-enrichment for a period of time and then implementation of the flameless oxyfuel
operation. Excellent results have been achieved, meeting all performance guarantees. These
included >25% more throughput, 50% fuel savings compared with oxygen-enrichment (60%
below the prior air-fuel performance), CO
2
emissions dropped accordingly, NO
X
emission
<70 mg/MJ corresponding to 92% less on an annual basis, and 50% reduced scale formation.
In January 2010, ArcelorMittal’s received the Association for Iron & Steel Technology’s
Energy Achievement Award for its efforts to reduce fuel consumption and emissions using
the REBOX oxyfuel technical solution at its Shelby mill.


Fig. 10. Outside view of the rotary hearth furnace at ArcelorMittal Shelby after the
conversion into all flameless oxyfuel operation. Please note that all bulky equipment and
piping relating to the previously used air-fuel system have been removed as it no longer is
of any use.

SSAB Walking Beam Furnace with REBOX HLL
At SSAB in Sweden REBOX HLL is used. The slabs are reheated in walking beam furnaces
with a capacity of 300 tph per furnace, from ambient temperature to 1,230°C. The air-fuel
combustion system uses a recuperation system to preheat air to 400°C. The fuel is oil, and
prior to the HLL installation the consumption was 440 kWh/tonne, or 1.58 GJ/tonne.
REBOX HLL creates a type of flameless oxyfuel without replacing the existing air-fuel
burners. By reducing the air flow and substituting high velocity oxygen injection into the
combustion, great benefits can be achieved. 75% of the oxygen needed for the combustion is
supplied with this technique. The flue-gas volume is less than 45% that of air-fuel. The
system start-up took just one day. The installation in only one zone has increased the
heating capacity from 300 to 320 tph.
The installation of HLL is rather easy because it does not include any replacement of
burners or installation of additional burners, which minimizes the installation down time.
The air-fuel system can at any time be brought back into operation as it was before. This
eliminates any potential risk relating to the implemention, and it enables operation to be
flexible and optimized in response to fluctuating fuel cost and production requirements.
Some important results from this installation are:
 No negative impact on the surface quality.
 A positive impact on the temperature uniformity of the slabs.
 The ideal heating curve suggested by the control system can be achieved more
easily.
 Less smoke emanating from the furnace, greatly improving the plant environment.
Energy Effciency 98
Based on the results of current installation in one zone, SSAB has estimated that a full
implementation would provide the following:
 A reduction of NO
X
emission by 45%.
 Fuel consumption can be decreased by 25%, leading to the same reductions in SO
2

and CO
2
emissions.
 Production throughput can be increased by 15-20%.


Fig. 11. “Semi-flameless” oxyfuel combustion in a 300 tph walking beam furnace at SSAB,
Sweden.

Stainless wire annealing in China
At Dongbei Special Steel Group in China, a new state-of-the-art annealing furnace for
stainless steel wire has been taken into operation in 2010. It applies a combined technology
called REBOX DST (Direct Solution Treatment), the benefits compared with a conventional
solution are extremely huge, for example the treatment time is drastically reduced. The
flameless combustion here uses a low calorific fuel with an energy content of 1.75 kWh/Nm
3

(6.3 MJ/Nm
3
).

7. At strip processing
Flameless oxyfuel can be used for heating at strip processing, but the real difference here is
made by applying DFI Oxyfuel, a fascinating, compact, high-heat transfer technology, which
provides enhanced operation in strip processing lines such as galvanizing. DFI Oxyfuel has
been used to boost capacity of strip annealing and hot dip metal coating lines by 30% or
more, while reducing the specific fuel consumption. Systems are in operation at
Outokumpu’s Nyby Works in Sweden and ThyssenKrupp’s works at Finnentrop and
Bruckhausen in Germany. In mid 2010 a unit was installed in a continuous annealing line at
POSCO in Pohang, South Korea.
Since the beginning of the 1990s, Linde has pioneered the use of 100% oxyfuel applications
in reheat furnaces in close cooperation with customers such as Outokumpu. At
Outokumpu’s Nyby site in Sweden, the company wanted to increase the production
capacity of a stainless strip annealing line, but the furnace already contained an oxyfuel
combustion system and had extremely limited physical space available. In 2002, the first
compact DFI Oxyfuel unit was installed, making it possible to increase the production by
50% (from 23 to 35 tph) without extending the furnace length. This DFI Oxyfuel installation
consisted of a 2-metre long DFI unit at the entry side with four burner row units including a
total of 4 MW installed power distributed on 120 oxyfuel flames.
In 2007, the REBOX DFI system was installed at ThyssenKrupp Steel’s (TKS) galvanizing
and aluminizing line in Bruckhausen, Germany. Earlier, Linde had installed a DFI unit at
the TKS galvanizing line at Finnentrop, and increased production from 82 to 105 tph, or over
30%. The results at the Bruckhausen installation matched those in Finnentrop: increasing
capacity from 70 to 90 tph. Oxyfuel not only effectively heats – contributing to a reduction of
fuel consumption – but also cleans, thus eliminating the need for the pre-cleaning section. In
addition, the process made it possible for ThyssenKrupp to pre-oxidize steel strips in a
precise and controlled manner. Prior to the DFI installation, the Finnentrop plant had a 25 m
long pre-cleaning section with electrolytic cleaning and brushes.
At Finnentrop, to minimize line downtime, the design resulted in a 3-metre long DFI unit
equipped with four burner row units, with a total of 120 oxyfuel flames and 5 MW installed
power, with an option of two more row sets for an additional 2.5 MW. Three metres of the
existing recuperative entry section was removed to fit the DFI Oxyfuel unit. The number of
burner row units and burners employed depend on set preheating temperatures and the
actual strip width and tonnage. At 105 tph, DFI Oxyfuel results in an immediate steel strip
surface temperature increase of more than 200°C.
With the DFI unit the capacity of the Finnentrop line increased from 82 to 109 tph. The DFI
Oxyfuel unit also manages to burn off residue, particles, grease and oil from the strip rolling
process, providing a cleaner strip than the long electrolytic and brush strip pre-cleaning
section used to do. At a production level of 36,000 tonnes per month at Finnentrop, results
include an over 5% reduction in natural gas consumption, almost 20% less NO
X
emissions,
and a reduction of 1200 tonnes per year in CO
2
emissions.


Fig. 12. REBOX DFI installation in a galvanizing line at ThyssenKrupp Steel at Finnentrop,
Germany. The 3-metre long DFI unit was fitted into the previous (non-fired) dark-zone.

The oxidation is lower than normal at a specific strip temperature since the dwell time is
very limited; applying DFI Oxyfuel for preheating a strip up to 300°C does not create
oxidation problems. In metal coating lines, the thin oxide layer formed is reduced in the
subsequent reduction zone. It is also possible to influence the oxidation by adjusting the
stoichiometry of the flames, for example by changing the lambda value from 1.0 to 0.9.
The oxide layer thicknesses have been measured to be in the range of 50-100 nanometres,
even at high strip temperatures. A well performing reduction zone should be able to reduce
the scaling further. For high strength steel, a small formed oxide layer, for instance, 200 nm,
may be beneficial, since after reduction in the Radiant Tube Furnace section, pure iron will
form on the surface for improve zinc adhesion.
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 99
Based on the results of current installation in one zone, SSAB has estimated that a full
implementation would provide the following:
 A reduction of NO
X
emission by 45%.
 Fuel consumption can be decreased by 25%, leading to the same reductions in SO
2

and CO
2
emissions.
 Production throughput can be increased by 15-20%.


Fig. 11. “Semi-flameless” oxyfuel combustion in a 300 tph walking beam furnace at SSAB,
Sweden.

Stainless wire annealing in China
At Dongbei Special Steel Group in China, a new state-of-the-art annealing furnace for
stainless steel wire has been taken into operation in 2010. It applies a combined technology
called REBOX DST (Direct Solution Treatment), the benefits compared with a conventional
solution are extremely huge, for example the treatment time is drastically reduced. The
flameless combustion here uses a low calorific fuel with an energy content of 1.75 kWh/Nm
3

(6.3 MJ/Nm
3
).

7. At strip processing
Flameless oxyfuel can be used for heating at strip processing, but the real difference here is
made by applying DFI Oxyfuel, a fascinating, compact, high-heat transfer technology, which
provides enhanced operation in strip processing lines such as galvanizing. DFI Oxyfuel has
been used to boost capacity of strip annealing and hot dip metal coating lines by 30% or
more, while reducing the specific fuel consumption. Systems are in operation at
Outokumpu’s Nyby Works in Sweden and ThyssenKrupp’s works at Finnentrop and
Bruckhausen in Germany. In mid 2010 a unit was installed in a continuous annealing line at
POSCO in Pohang, South Korea.
Since the beginning of the 1990s, Linde has pioneered the use of 100% oxyfuel applications
in reheat furnaces in close cooperation with customers such as Outokumpu. At
Outokumpu’s Nyby site in Sweden, the company wanted to increase the production
capacity of a stainless strip annealing line, but the furnace already contained an oxyfuel
combustion system and had extremely limited physical space available. In 2002, the first
compact DFI Oxyfuel unit was installed, making it possible to increase the production by
50% (from 23 to 35 tph) without extending the furnace length. This DFI Oxyfuel installation
consisted of a 2-metre long DFI unit at the entry side with four burner row units including a
total of 4 MW installed power distributed on 120 oxyfuel flames.
In 2007, the REBOX DFI system was installed at ThyssenKrupp Steel’s (TKS) galvanizing
and aluminizing line in Bruckhausen, Germany. Earlier, Linde had installed a DFI unit at
the TKS galvanizing line at Finnentrop, and increased production from 82 to 105 tph, or over
30%. The results at the Bruckhausen installation matched those in Finnentrop: increasing
capacity from 70 to 90 tph. Oxyfuel not only effectively heats – contributing to a reduction of
fuel consumption – but also cleans, thus eliminating the need for the pre-cleaning section. In
addition, the process made it possible for ThyssenKrupp to pre-oxidize steel strips in a
precise and controlled manner. Prior to the DFI installation, the Finnentrop plant had a 25 m
long pre-cleaning section with electrolytic cleaning and brushes.
At Finnentrop, to minimize line downtime, the design resulted in a 3-metre long DFI unit
equipped with four burner row units, with a total of 120 oxyfuel flames and 5 MW installed
power, with an option of two more row sets for an additional 2.5 MW. Three metres of the
existing recuperative entry section was removed to fit the DFI Oxyfuel unit. The number of
burner row units and burners employed depend on set preheating temperatures and the
actual strip width and tonnage. At 105 tph, DFI Oxyfuel results in an immediate steel strip
surface temperature increase of more than 200°C.
With the DFI unit the capacity of the Finnentrop line increased from 82 to 109 tph. The DFI
Oxyfuel unit also manages to burn off residue, particles, grease and oil from the strip rolling
process, providing a cleaner strip than the long electrolytic and brush strip pre-cleaning
section used to do. At a production level of 36,000 tonnes per month at Finnentrop, results
include an over 5% reduction in natural gas consumption, almost 20% less NO
X
emissions,
and a reduction of 1200 tonnes per year in CO
2
emissions.


Fig. 12. REBOX DFI installation in a galvanizing line at ThyssenKrupp Steel at Finnentrop,
Germany. The 3-metre long DFI unit was fitted into the previous (non-fired) dark-zone.

The oxidation is lower than normal at a specific strip temperature since the dwell time is
very limited; applying DFI Oxyfuel for preheating a strip up to 300°C does not create
oxidation problems. In metal coating lines, the thin oxide layer formed is reduced in the
subsequent reduction zone. It is also possible to influence the oxidation by adjusting the
stoichiometry of the flames, for example by changing the lambda value from 1.0 to 0.9.
The oxide layer thicknesses have been measured to be in the range of 50-100 nanometres,
even at high strip temperatures. A well performing reduction zone should be able to reduce
the scaling further. For high strength steel, a small formed oxide layer, for instance, 200 nm,
may be beneficial, since after reduction in the Radiant Tube Furnace section, pure iron will
form on the surface for improve zinc adhesion.
Energy Effciency 100
Cleaning tests show that the carbon and iron fines contaminations can be drastically reduced
by use of DFI. With the DFI Oxyfuel technology the cleaning section can be shortened to a
spray cleaning section, one brush machine and a final rinsing section. The final cleaning
operation is transferred to the DFI Oxyfuel inside the thermal section. The elimination of one
brush machine and the electrolytic cleaning section brings considerable cost savings in
maintenance and operation due to energy savings and less wear parts. Furthermore, DFI gives
potential to reduce investment and operating costs in the furnace section since the furnace
length can be reduced; the preheating and one heating zone can be saved.
This year, 2010, REBOX DFI is for the first time employed in a continuous annealing line for
carbon steel, at POSCO’s large integrated steel mill at Pohang, South Korea. The DFI unit
provides a guaranteed level of preheating which will be capable of achieving approximately
15% higher capacity in the annealing furnace. The natural gas fired DFI unit consists of four
oxyfuel burner row units with a combined capacity of close to 6 MW.

8. Opportunities for decreasing CO
2
emissions
There is a strong political will to decrease CO
2
emission. The steel industry only accounts for
some 3% of worldwide CO
2
emissions, which totals roughly 30 billion tonnes per annum
relating to the human activity of burning of fossil fuels, but seems to be strongly affected by
this. To radically change existing processes and production routes to decrease the CO
2

emissions would be extremely expensive, even if it were possible.
However, there exist today a number of proven solutions and technologies which, if fully
implemented, could substantially decrease CO
2
emissions without seriously altering current
methods of operation and are therefore short-term viable solutions. If these solutions are
fully implemented, the combined impact on CO
2
emissions from the steel industry
worldwide is estimated to be a reduction of 100 million tonnes of CO
2
per annum within a
relatively short time span. Among these solutions, the most viable is oxyfuel combustion.


Fig. 13. A look through the furnace door of the rotary hearth furnace at ArcelorMittal
Shelby, USA; a flameless oxyfuel burner is firing straight towards the open door. Here the
conversion from air-fuel to flameless oxyfuel led to a 60% reduction of the CO
2
emission.

CO
2
emissions from the steel industry have two main sources: reduction processes, and
melting and heating processes. It is well known that reduction processes are the dominant
source. The two main routes for steel production account for quite different impacts on CO
2

emissions: integrated steel mills, including all upstream processes, average approximately 2
tonnes of CO
2
per tonne of hot rolled plate; for mini-mills, the corresponding figure is 0.5-0.6
tonnes. However, the contribution from heating processes is not negligible; each piece of
steel is on average heated twice on its journey through the production chain, and this is far
from the only heating process. Accordingly, by increasing the energy efficiency in the
heating processes, a large impact can be made on reducing the carbon footprint. An
additional advantage is the low flue-gas volumes with high concentration of CO
2
, which
enable directing it to capturing and potentially sequestration.
Use of a fuel with a low calorific value is of interest in this context. It could, for example, be
internally produced gas streams at a plant, like blast furnace top gas and BOF gas. In many
places, at least some of the latter gases are not used but put to flaring. What is frequently
hampering their greater use is the flame temperature required in heating applications.
However, using oxyfuel instead of air-fuel would in many cases make it possible to even
run solely a low calorific gas as fuel. Where these gases are being flared today, the resultant
impact on the site’s CO
2
emissions of using them in this way would be very positive and
would replace other energy sources. A practical example of an increased use of a low grade
fuel can be found in blast furnace hot stoves, where due to the oxygen-enrichment it leads to
improved fuel economy and reduced CO
2
emissions.
As the examples and solutions discussed in this chapter all use oxygen, it is appropriate to
comment on the CO
2
emissions relating to oxygen production. The production of 1 Nm
3
of
gaseous oxygen requires approximately 0.5 kWh of electricity. If this electricity is produced
by hydro or nuclear power plants, it “carries” no CO
2
. However, if produced using fossil
fuel it would correspond to 0.5 kg CO
2
per Nm
3
of oxygen. Thus, in the worst case scenario,
oxyfuel combustion contributes (from oxygen production) 0.1 kg CO
2
per kWh. Turning that
worst case scenario into practice, it is known that oxyfuel combustion (compared with air-
fuel) would reduce the fuel consumption by an average of 40%, and the combined effect on
CO
2
emissions would then be a reduction of approximately 35%.

9. Conclusions
The traditional use of oxyfuel in steel-making is in the electric arc furnace. Today
sophisticated wall-mounted equipment is used combining the functions of oxygen and coal
lancing, oxyfuel burner, and post-combustion. The level of oxygen use could reach above 50
Nm
3
/t, more than in the steel-making converter in integrated steel mills.
Mainly due to the strive to reduce CO
2
emissions the Full Oxygen Blast Furnace concept is
now being tested. Here oxygen is completely replacing the air-blast. However, in a short-
term perspective it seems advantageous to instead focus on the hot stoves, where low
calorific fuel can be used to an increased extent, a typical benefit from oxyfuel.
Oxyfuel provides an overall thermal efficiency in the heating of 80%, air-fuel reaches 40-
60%. With flameless oxyfuel, compared to air-fuel, the energy savings in a reheating furnace
are at least 25%, but many times 50% or even more. It is possible to operate a reheat furnace
with fuel consumption below 1 GJ per tonne. The corresponding reduction in CO
2
emissions
is also 25-50%. Savings in terms of NO
X
emissions are substantial. Flameless oxyfuel
combustion has major advantages over conventional oxyfuel and, even more, over any kind
Oxyfuel combustion in the steel industry: energy effciency and decrease of co2 emissions 101
Cleaning tests show that the carbon and iron fines contaminations can be drastically reduced
by use of DFI. With the DFI Oxyfuel technology the cleaning section can be shortened to a
spray cleaning section, one brush machine and a final rinsing section. The final cleaning
operation is transferred to the DFI Oxyfuel inside the thermal section. The elimination of one
brush machine and the electrolytic cleaning section brings considerable cost savings in
maintenance and operation due to energy savings and less wear parts. Furthermore, DFI gives
potential to reduce investment and operating costs in the furnace section since the furnace
length can be reduced; the preheating and one heating zone can be saved.
This year, 2010, REBOX DFI is for the first time employed in a continuous annealing line for
carbon steel, at POSCO’s large integrated steel mill at Pohang, South Korea. The DFI unit
provides a guaranteed level of preheating which will be capable of achieving approximately
15% higher capacity in the annealing furnace. The natural gas fired DFI unit consists of four
oxyfuel burner row units with a combined capacity of close to 6 MW.

8. Opportunities for decreasing CO
2
emissions
There is a strong political will to decrease CO
2
emission. The steel industry only accounts for
some 3% of worldwide CO
2
emissions, which totals roughly 30 billion tonnes per annum
relating to the human activity of burning of fossil fuels, but seems to be strongly affected by
this. To radically change existing processes and production routes to decrease the CO
2

emissions would be extremely expensive, even if it were possible.
However, there exist today a number of proven solutions and technologies which, if fully
implemented, could substantially decrease CO
2
emissions without seriously altering current
methods of operation and are therefore short-term viable solutions. If these solutions are
fully implemented, the combined impact on CO
2
emissions from the steel industry
worldwide is estimated to be a reduction of 100 million tonnes of CO
2
per annum within a
relatively short time span. Among these solutions, the most viable is oxyfuel combustion.


Fig. 13. A look through the furnace door of the rotary hearth furnace at ArcelorMittal
Shelby, USA; a flameless oxyfuel burner is firing straight towards the open door. Here the
conversion from air-fuel to flameless oxyfuel led to a 60% reduction of the CO
2
emission.

CO
2
emissions from the steel industry have two main sources: reduction processes, and
melting and heating processes. It is well known that reduction processes are the dominant
source. The two main routes for steel production account for quite different impacts on CO
2

emissions: integrated steel mills, including all upstream processes, average approximately 2
tonnes of CO
2
per tonne of hot rolled plate; for mini-mills, the corresponding figure is 0.5-0.6
tonnes. However, the contribution from heating processes is not negligible; each piece of
steel is on average heated twice on its journey through the production chain, and this is far
from the only heating process. Accordingly, by increasing the energy efficiency in the
heating processes, a large impact can be made on reducing the carbon footprint. An
additional advantage is the low flue-gas volumes with high concentration of CO
2
, which
enable directing it to capturing and potentially sequestration.
Use of a fuel with a low calorific value is of interest in this context. It could, for example, be
internally produced gas streams at a plant, like blast furnace top gas and BOF gas. In many
places, at least some of the latter gases are not used but put to flaring. What is frequently
hampering their greater use is the flame temperature required in heating applications.
However, using oxyfuel instead of air-fuel would in many cases make it possible to even
run solely a low calorific gas as fuel. Where these gases are being flared today, the resultant
impact on the site’s CO
2
emissions of using them in this way would be very positive and
would replace other energy sources. A practical example of an increased use of a low grade
fuel can be found in blast furnace hot stoves, where due to the oxygen-enrichment it leads to
improved fuel economy and reduced CO
2
emissions.
As the examples and solutions discussed in this chapter all use oxygen, it is appropriate to
comment on the CO
2
emissions relating to oxygen production. The production of 1 Nm
3
of
gaseous oxygen requires approximately 0.5 kWh of electricity. If this electricity is produced
by hydro or nuclear power plants, it “carries” no CO
2
. However, if produced using fossil
fuel it would correspond to 0.5 kg CO
2
per Nm
3
of oxygen. Thus, in the worst case scenario,
oxyfuel combustion contributes (from oxygen production) 0.1 kg CO
2
per kWh. Turning that
worst case scenario into practice, it is known that oxyfuel combustion (compared with air-
fuel) would reduce the fuel consumption by an average of 40%, and the combined effect on
CO
2
emissions would then be a reduction of approximately 35%.

9. Conclusions
The traditional use of oxyfuel in steel-making is in the electric arc furnace. Today
sophisticated wall-mounted equipment is used combining the functions of oxygen and coal
lancing, oxyfuel burner, and post-combustion. The level of oxygen use could reach above 50
Nm
3
/t, more than in the steel-making converter in integrated steel mills.
Mainly due to the strive to reduce CO
2
emissions the Full Oxygen Blast Furnace concept is
now being tested. Here oxygen is completely replacing the air-blast. However, in a short-
term perspective it seems advantageous to instead focus on the hot stoves, where low
calorific fuel can be used to an increased extent, a typical benefit from oxyfuel.
Oxyfuel provides an overall thermal efficiency in the heating of 80%, air-fuel reaches 40-
60%. With flameless oxyfuel, compared to air-fuel, the energy savings in a reheating furnace
are at least 25%, but many times 50% or even more. It is possible to operate a reheat furnace
with fuel consumption below 1 GJ per tonne. The corresponding reduction in CO
2
emissions
is also 25-50%. Savings in terms of NO
X
emissions are substantial. Flameless oxyfuel
combustion has major advantages over conventional oxyfuel and, even more, over any kind
Energy Effciency 102
of air-fuel combustion. The improved temperature uniformity is a very important benefit,
which also reduces the fuel consumption further.
With oxyfuel it is possible to increase the throughput rate by up to 50%. This can be used for
increased production, less number of furnaces in operation, increased flexibility, etc. It is
also of interest when ramping up production; two furnaces can cover the previous
production of 2.5-3 furnaces, meaning possibility to post start-up of the third furnace and,
additionally, resulting in decreased fuel consumption. Increased capacity can also be used to
prolong soaking times. Thanks to the reduced time at elevated temperatures, oxyfuel leads
to reduced scale losses, at many installations as high as 50%.
Using DFI Oxyfuel, where the flames heat directly onto the moving material, a very compact
solution has been established. Installations show the production throughput can be
increased by 30%, but it also provides other important benefits. This technology is
particularly suitable for strip processing.
The experiences from converting furnaces into all oxyfuel operation show energy savings
ranging from 20% to 70%, excluding savings in energy needed for bringing the fuel to the
site. The use flameless oxyfuel in ladle and converter preheating is extremely advantageous.
Now we also see that this innovative technology can be used at blast furnace hot stoves to
improve energy and production efficiencies and reduce environmental impact.
There exist today a number of solutions and technologies which could substantially
decrease CO
2
emissions without seriously altering current methods of operation and are
therefore short-term viable solutions. Additionally, they would lead to improved fuel
economics and reduced processing times. In heating and melting, oxyfuel combustion offers
clear advantages over state-of-the-art air-fuel combustion, for example regenerative
technology, in terms of energy use, maintenance costs and utilization of existing production
facilities. If all the reheating and annealing furnaces would employ oxyfuel combustion, the
CO
2
emissions from the world’s steel industry would be reduced by 100 million tonnes per
annum. Additionally, a small off-gas volume and a high concentration of CO
2
make it
increasingly suitable for Carbon Capture and Sequestration.
Using oxyfuel instead of air-fuel combustion for all kinds of melting and heating operations
opens up tremendous opportunities, as it leads to fuel savings, reduces the time required for
the process and reduces emissions. Numerous results from installations have proven this.
Low-energy buildings – scientifc trends and developments 103
Low-energy buildings – scientifc trends and developments
Dr. Patrik Rohdin, Dr. Wiktoria Glad and Dr. Jenny Palm
x

Low-energy buildings – scientific
trends and developments

Dr. Patrik Rohdin
1
, Dr. Wiktoria Glad
2
and Dr. Jenny Palm
2

1
Energy systems, Linköping University
2
Tema T, Linköping University
Sweden

1. Introduction
Over the past twenty years primary energy demand in the world has increased drastically,
while during the same time demand for electrical energy has increased even more. This, in
combination with the impact of global warming, is forcing policy-makers to formulate goals
to meet this threat. The EU Commission has recently stated that one of its highest priority
tasks is to address global warming, with special focus on reducing greenhouse gases. The
EU Commission states in the directive for energy efficiency in the built environment that the
building sector must decrease its use of energy to reduce CO
2
emissions. In addition a goal
for energy efficiency within the Union states that a 20% increase in energy efficiency shall be
met by 2020. The Swedish parliament has also set a national goal for space heating, which
states that by 2020 the use per floor area should be reduced by 20% and by 2050 this figure
should be 50% compared to use during 1995.
To be able to meet these goals, many different activities must strive towards the same goal.
One major part is the building and service sector, which accounts for about 35% of total
Swedish national energy use. A large part of that use is concentrated in cities, which
underlines the importance of working with such areas. The connection between CO
2

emissions and the use of energy is also an important motive for promoting a more efficient
use of energy and reducing the total energy demand. This means that there is a need to
choose the correct primary energy and energy conservation measures as well as to reduce
the total electrical usage in the built environment.
Furthermore, the consequences of global warming are introducing changing conditions to
be met by future buildings with increasing temperatures, and for Sweden increasing
precipitation as well. In IPCC (2007) the temperature increase is predicted to be 1-2°C with
an increase in precipitation by 20% for the 2020-2029 scenarios relative to 1980-1999. For the
long-term scenario until 2090-2099 the predictions are of the order of 4-5°C. Effects like this
should be included in the analysis of future energy systems and design criteria, since it will
reduce heat demand and increase the risk of overheating in buildings. Poor indoor
environmental conditions in buildings is an important factor which costs large amounts of
money in healthcare and administration, while a well-functioning indoor environment plays
6
Energy Effciency 104

an important part in a convenient and modern life. It is also important to include the
environmental impact from building materials.
One key component in achieving a more sustainable building sector is to introduce different
forms of energy-efficient or renewable buildings. In this chapter a review of the literature
published within the Web of Science databases on low-energy houses, passive houses, zero-
energy houses and passive solar houses is presented. The aim is to analyze trends in the
scientific literature concerning sustainable buildings and to discuss which issues have been
in focus and which have been neglected in earlier studies. This will create a basis for
discussing knowledge gaps and future research needs. Our scope is to focus on the
development of research on dwellings.

2. Field overview
The field of low-energy buildings is broad and complex. The first article included in this
review is from 1978. A total of 83 relevant hits were found within Web of Science for the
seven search words (1) low-energy buildings; (2) low-energy architecture; (3) low-energy
house; (4) passive house; (5) passive solar building; (6) passive solar house; and (7) zero
energy house. The number of unique hits for each search word is seen in Figure 4.
The trend of increasing interest in low-energy buildings can also be seen in the increase in
production of scientific papers within the scope of this review, see Figure 1. During the last
five years the number of publications has moved up from about one or two per year to
between eight and ten. This should also be seen in the light of an increase in the general
production of papers, but at the same time it shows that there is strong focus on low-energy
solutions for the built environment in the scientific community too.
10
7 7
10
6
0
3
0
3
1 1
4
0
4
0
4
1 1 1
2
1
0
1
3
2
1
2
1
0
2
0
1
0
2
4
6
8
10
12
1975 1980 1985 1990 1995 2000 2005 2010 2015
N
u
m
b
e
r

o
f

p
u
b
l
i
c
a
t
i
o
n
s

e
a
c
h

y
e
a
r

Fig. 1. Overview of the number of publications each year from the first article reported in a
journal in Web of Science in 1978 to today (2010). The red line indicates the five-year moving
average.


We have used the Web of Science database when we searched for relevant articles. This
database is dominated by journals with a technical focus, which may partly explain that
when examining the structure and focus of the reviewed articles in terms of the main
method used it turned out to be a highly technical field. But we also noticed that within
these journals examples of broader articles including policy issues, interdisciplinary studies
and economic studies have become increasingly common in the last few years. But the main
part of the field still remains technical in nature, with focus on building energy simulation
(BES), component studies of thermal walls and solar applications and measurements, see
Figure 3. There is also a strong tendency for the field to employ case studies and
experimental setups either in laboratory form or as in real constructions, see Figure 2.
Technical
45%
Case
19%
Component
9%
Policy
9%
Economic
6%
Other
6%
Interdisciplinary
3%
Environmental
3%

Fig. 2. A distinction between different types of article within the review. The categorization
is not unambiguous since several articles may be relevant for more than one of the
suggested groups.
CFD
4%
BES
40%
Measurement
33%
Interview
5%
Questionnaire
2%
Document
5%
Statistical
11%

Fig. 3. The relative difference in number of publications using different methods. This
distinction is however not unambiguous since several papers can be argued to have more
than one of the above suggested methods.
Low-energy buildings – scientifc trends and developments 105

an important part in a convenient and modern life. It is also important to include the
environmental impact from building materials.
One key component in achieving a more sustainable building sector is to introduce different
forms of energy-efficient or renewable buildings. In this chapter a review of the literature
published within the Web of Science databases on low-energy houses, passive houses, zero-
energy houses and passive solar houses is presented. The aim is to analyze trends in the
scientific literature concerning sustainable buildings and to discuss which issues have been
in focus and which have been neglected in earlier studies. This will create a basis for
discussing knowledge gaps and future research needs. Our scope is to focus on the
development of research on dwellings.

2. Field overview
The field of low-energy buildings is broad and complex. The first article included in this
review is from 1978. A total of 83 relevant hits were found within Web of Science for the
seven search words (1) low-energy buildings; (2) low-energy architecture; (3) low-energy
house; (4) passive house; (5) passive solar building; (6) passive solar house; and (7) zero
energy house. The number of unique hits for each search word is seen in Figure 4.
The trend of increasing interest in low-energy buildings can also be seen in the increase in
production of scientific papers within the scope of this review, see Figure 1. During the last
five years the number of publications has moved up from about one or two per year to
between eight and ten. This should also be seen in the light of an increase in the general
production of papers, but at the same time it shows that there is strong focus on low-energy
solutions for the built environment in the scientific community too.
10
7 7
10
6
0
3
0
3
1 1
4
0
4
0
4
1 1 1
2
1
0
1
3
2
1
2
1
0
2
0
1
0
2
4
6
8
10
12
1975 1980 1985 1990 1995 2000 2005 2010 2015
N
u
m
b
e
r

o
f

p
u
b
l
i
c
a
t
i
o
n
s

e
a
c
h

y
e
a
r

Fig. 1. Overview of the number of publications each year from the first article reported in a
journal in Web of Science in 1978 to today (2010). The red line indicates the five-year moving
average.


We have used the Web of Science database when we searched for relevant articles. This
database is dominated by journals with a technical focus, which may partly explain that
when examining the structure and focus of the reviewed articles in terms of the main
method used it turned out to be a highly technical field. But we also noticed that within
these journals examples of broader articles including policy issues, interdisciplinary studies
and economic studies have become increasingly common in the last few years. But the main
part of the field still remains technical in nature, with focus on building energy simulation
(BES), component studies of thermal walls and solar applications and measurements, see
Figure 3. There is also a strong tendency for the field to employ case studies and
experimental setups either in laboratory form or as in real constructions, see Figure 2.
Technical
45%
Case
19%
Component
9%
Policy
9%
Economic
6%
Other
6%
Interdisciplinary
3%
Environmental
3%

Fig. 2. A distinction between different types of article within the review. The categorization
is not unambiguous since several articles may be relevant for more than one of the
suggested groups.
CFD
4%
BES
40%
Measurement
33%
Interview
5%
Questionnaire
2%
Document
5%
Statistical
11%

Fig. 3. The relative difference in number of publications using different methods. This
distinction is however not unambiguous since several papers can be argued to have more
than one of the above suggested methods.
Energy Effciency 106

Low energy house
16%
Passive house
23%
Passive solar
building
12%
Passive solar
house
14%
Zero energy house
6%
Low energy
building
28%
Low energy
architecture
1%

Fig. 4. The relative magnitude of hits within the review on different search words.

3. Main methods cited in the reviewed papers
This section will present an overview of the method characterisation used in the review and
introduce the concepts of the different methods. The main methods identified in the articles
are: (1) Computational Fluid Dynamics; (2) Building Energy Simulation; (3) Measurements;
(4) Interviews; (5) Questionnaires; (6) Statistical; or (7) Environmental or Life Cycle-Focused
Studies.

3.1 Computational Fluid Dynamics (CFD)
Computational Fluid Dynamics (CFD) has been extensively used as a scientific tool in many
application and research situations since the 1950s. The use is widespread in many fields,
such as aerodynamics, hydraulics, combustion engineering, meteorology, electronic cooling
and biomedical engineering, and in predicting the external and internal environment of
buildings. In Versteeg and Malalasekera (1995) the authors give a rather broad definition of
CFD: “Computational fluid dynamics (CFD) is the analysis of systems involving fluid flow,
heat transfer or associated phenomena such as chemical reactions by means of computer
based simulation.”
The use of CFD to simulate ventilation and air movements in rooms is becoming more and
more common. One of the earliest publications where CFD was used to simulate air flow in
rooms was made by Nielsen in 1974. Due to the increase in computer resources, the use of
CFD as a scientific tool has increased and continues to increase as it is possible to solve more
complex and challenging problems. As the cost and time needed to perform real
experiments in many cases are high, CFD has become more and more extensively used. This
method is of course of special interest for cases where it is not possible to obtain
measurements, such as situations where the object has not yet been built. However, to
ensure the validity and reliability of CFD models, measurements are still very much needed.
An often-used approach is to compare results from numerical simulations with
measurements; if the results coincide, a numerical approach in predicting similar situations
may be used.

3.2 Building Energy Simulation (BES)
Building Energy Simulation (BES) is a frequently used tool to predict energy use in
buildings within the academic sphere as well as in the design process in the construction
industry. Similar to other types of simulations, BES is a numerical experiment using a
mathematical model. The aim is to predict or forecast a future or an otherwise presently
unknown situation. For energy simulation programs, issues such as predicting energy use,
either in a future building not yet built, or after a change in a system has been made in a
present building, are of interest. In Bergsten (2001) a comparison of different energy
simulation software is presented, and a classification of the software is made depending on
whether it is a general simulation program or has multi-zone capabilities and if it is static or
dynamic. The software compared, considered the most important energy simulation
software used in Sweden, Norway and Denmark, were Bsim 2000, BV2, EiB, IDA ICE,
Energikiosken, Enorm 2000, Huset, OPERA, Villaenergi, VIP+ and Värmeenergi (Bergsten,
2001). In Crawley et al. (2005) a more extensive review of the performance and capabilities
of building energy simulation programs is presented. The review includes BLAST, BSim,
DeST, DOE-2. IE, ECOTECT, Energy-10, Energy Express, Ener-Win, EnergyPlus, eQUEST,
ESP-r, IDA ICE, IES<VE>, HAP, HEED, PowerDomus, SUNREL, Tas, TRACE and TRNSYS.

3.3 Measurements
Studies of indoor climate and energy efficiency often include measurements of temperature,
moisture, air velocity, turbulence intensity, carbon dioxide, radon and other pollutants in
addition to power and energy. When measuring spatial distributions there is a problem with
creating a comprehensive view as it is time-consuming and is costly. Measuring the climate
in a room with arbitrary accuracy is virtually impossible because it would require too many
data points. It is also true that it will be time consuming and expensive to measure over long
periods of time. Measurements as evaluation instruments are of course invaluable, but the
very nature of the measurement in itself does not give any idea of the future, as it only says
something about the past. At this stage different types of models are needed in order to
make statements about the future. All measurements are also affected by different
measurement errors. These vary greatly depending on the type of equipment used and the
manner in which measurements are made.

3.4 Interviews
Interviewing is a common data collection method in social science qualitative research,
among with observations and document analyses. The aim of qualitative inquiries is to
explore the qualities of phenomena and provide data to gain deeper understanding (Lincoln
and Guba, 1985). Using interviews to acquire data is usually preceded by a process of letting
the problem at hand determine what type of inquiry is suitable and how the problem is best
explored. A structured interview could in some cases generate similar data as a
questionnaire, while a more open-ended, semi-structured interview requires more
attentiveness and flexibility from the interviewer but can provide detailed descriptions and
interpretations of phenomena in the world (Kvale and Brinkmann, 2009). While quantative
data concern more or less of a studied entity, qualitative data concern similarities or
dissimilarities. Analysis of interviews is descriptive, but the purpose is to reach beyond the
description of the questions in the interview. The analysis means that, through reflection,
Low-energy buildings – scientifc trends and developments 107

Low energy house
16%
Passive house
23%
Passive solar
building
12%
Passive solar
house
14%
Zero energy house
6%
Low energy
building
28%
Low energy
architecture
1%

Fig. 4. The relative magnitude of hits within the review on different search words.

3. Main methods cited in the reviewed papers
This section will present an overview of the method characterisation used in the review and
introduce the concepts of the different methods. The main methods identified in the articles
are: (1) Computational Fluid Dynamics; (2) Building Energy Simulation; (3) Measurements;
(4) Interviews; (5) Questionnaires; (6) Statistical; or (7) Environmental or Life Cycle-Focused
Studies.

3.1 Computational Fluid Dynamics (CFD)
Computational Fluid Dynamics (CFD) has been extensively used as a scientific tool in many
application and research situations since the 1950s. The use is widespread in many fields,
such as aerodynamics, hydraulics, combustion engineering, meteorology, electronic cooling
and biomedical engineering, and in predicting the external and internal environment of
buildings. In Versteeg and Malalasekera (1995) the authors give a rather broad definition of
CFD: “Computational fluid dynamics (CFD) is the analysis of systems involving fluid flow,
heat transfer or associated phenomena such as chemical reactions by means of computer
based simulation.”
The use of CFD to simulate ventilation and air movements in rooms is becoming more and
more common. One of the earliest publications where CFD was used to simulate air flow in
rooms was made by Nielsen in 1974. Due to the increase in computer resources, the use of
CFD as a scientific tool has increased and continues to increase as it is possible to solve more
complex and challenging problems. As the cost and time needed to perform real
experiments in many cases are high, CFD has become more and more extensively used. This
method is of course of special interest for cases where it is not possible to obtain
measurements, such as situations where the object has not yet been built. However, to
ensure the validity and reliability of CFD models, measurements are still very much needed.
An often-used approach is to compare results from numerical simulations with
measurements; if the results coincide, a numerical approach in predicting similar situations
may be used.

3.2 Building Energy Simulation (BES)
Building Energy Simulation (BES) is a frequently used tool to predict energy use in
buildings within the academic sphere as well as in the design process in the construction
industry. Similar to other types of simulations, BES is a numerical experiment using a
mathematical model. The aim is to predict or forecast a future or an otherwise presently
unknown situation. For energy simulation programs, issues such as predicting energy use,
either in a future building not yet built, or after a change in a system has been made in a
present building, are of interest. In Bergsten (2001) a comparison of different energy
simulation software is presented, and a classification of the software is made depending on
whether it is a general simulation program or has multi-zone capabilities and if it is static or
dynamic. The software compared, considered the most important energy simulation
software used in Sweden, Norway and Denmark, were Bsim 2000, BV2, EiB, IDA ICE,
Energikiosken, Enorm 2000, Huset, OPERA, Villaenergi, VIP+ and Värmeenergi (Bergsten,
2001). In Crawley et al. (2005) a more extensive review of the performance and capabilities
of building energy simulation programs is presented. The review includes BLAST, BSim,
DeST, DOE-2. IE, ECOTECT, Energy-10, Energy Express, Ener-Win, EnergyPlus, eQUEST,
ESP-r, IDA ICE, IES<VE>, HAP, HEED, PowerDomus, SUNREL, Tas, TRACE and TRNSYS.

3.3 Measurements
Studies of indoor climate and energy efficiency often include measurements of temperature,
moisture, air velocity, turbulence intensity, carbon dioxide, radon and other pollutants in
addition to power and energy. When measuring spatial distributions there is a problem with
creating a comprehensive view as it is time-consuming and is costly. Measuring the climate
in a room with arbitrary accuracy is virtually impossible because it would require too many
data points. It is also true that it will be time consuming and expensive to measure over long
periods of time. Measurements as evaluation instruments are of course invaluable, but the
very nature of the measurement in itself does not give any idea of the future, as it only says
something about the past. At this stage different types of models are needed in order to
make statements about the future. All measurements are also affected by different
measurement errors. These vary greatly depending on the type of equipment used and the
manner in which measurements are made.

3.4 Interviews
Interviewing is a common data collection method in social science qualitative research,
among with observations and document analyses. The aim of qualitative inquiries is to
explore the qualities of phenomena and provide data to gain deeper understanding (Lincoln
and Guba, 1985). Using interviews to acquire data is usually preceded by a process of letting
the problem at hand determine what type of inquiry is suitable and how the problem is best
explored. A structured interview could in some cases generate similar data as a
questionnaire, while a more open-ended, semi-structured interview requires more
attentiveness and flexibility from the interviewer but can provide detailed descriptions and
interpretations of phenomena in the world (Kvale and Brinkmann, 2009). While quantative
data concern more or less of a studied entity, qualitative data concern similarities or
dissimilarities. Analysis of interviews is descriptive, but the purpose is to reach beyond the
description of the questions in the interview. The analysis means that, through reflection,
Energy Effciency 108

the researcher abstracts from the description and searches for patterns and dysfunctional
ties in relation to earlier studies or theories (Kvale, 1996).

3.5 Questionnaires
Questionnaires are an important part of survey research since this is the most common data
collection method. Structured interviews might sometimes be similar to questionnaires with
standardized answers although they may be closed or open-ended. When the objective of
the research is specific, for example to determine which factors influence a certain
phenomenon, a statistical analysis of data from questionnaires might be used. Recently,
assumptions taken for granted in survey research have been questioned (Krosnick 1999).
Response rates, pretesting of questionnaires and questionnaire design are among the
contested aspects of this research. Research has shown, for example, that low response rates
might show more accurate results than studies with high response rates (Visser et al., 1996).

3.6 Document studies
Analyses of written texts are important research tools in, for example, content analyses.
Other types of analyses using texts as the primary source of data are discourse and
argumentation analyses. The various types might be used differently in different fields of
research. For example, linguists use discourse analysis to find syntax and choice of words in
relation to contexts and focus is on language as a power tool, while social scientists might
focus on discourse analysis as a tool in itself to reveal social practice (Bergström and Boréus,
2005). This is often referred to as critical discourse analysis (Fairclough, 2003). These texts
might be minutes, records, protocols, letters, articles, books, newspapers, journals found in
libraries, on the internet, in public or private archives and files, etc. Some texts are superior
to other data collection methods in terms of reliability since they are real-time
documentations of a phenomenon, compared to interviews which construct phenomena
during the interview. Also, large amounts of written texts enable researchers to do structural
analyses in relation to a research question (Lincoln and Guba, 1985)

3.7 Statistical
Statistics is a whole scientific field with several subfields, but in general, statistical methods
include ways to collect, process and draw conclusions from statistical figures to enable, for
example, sampling and analyse differences (Körner et al., 1984). Focus is on empirical data
and statistical methods are used in both natural and social sciences. Descriptive statistics
include mapping and providing information, while inferential statistics aim to present
causal explanations and factors that influence a certain phenomenon. The basis for
explorations is elements in a population (Moses, 1986). Populations and elements might be
individuals but can also refer to material or physical entities. A census is a study using
whole populations, but usually a sample of a population is selected in order to limit the
research due to resource constraints. Social and economic sciences use questionnaires or
observations to collect data, while the most common data collection method in natural
sciences is different types of observations. Data can be analysed using computer programs
such as SPSS.

3.8 Environmental or Life Cycle-Focused Studies
Different types of environmental or life cycle-focused studies form an eclectic category of
research. A definition of life cycle assessment is “compilation and evaluation of the inputs,
outputs and potential environmental impacts of a product system throughout its life cycle”
(Guinée, 2002). Industrial ecology covers a similar field of exploration (Frosch and
Gallopoulos, 1989). Some of these methods claim to have a “holistic” approach but in reality
it is impossible to include all environmental aspects “from cradle to grave” and methods
with this approach might be criticized because they promise too much. Also, a “system”
approach is common among the different methods and may receive the same critique (ISO ,
1997). The definition of system boundaries is a constant issue of debate. Another critical
point in these analyses is the validity of data (Cooper and Fava, 2006). Common concepts
included in energy-related research with these approaches are “embodied energy” (Odum,
2007) and exergy (Dinçer and Rosen, 2007). Social issues and phenomena that are hard to
put in figures are generally missing in these studies.

4. Articles by method
In this section the articles from the review will be presented under a characterization of the
method used in the paper. As previously stated, this is not an unambiguous characterization
but should be seen as a way of grouping the studies reviewed within this chapter. Below a
short presentation of the studies is made and a discussion of the main lessons learned will
follow.

4.1 Computational Fluid Dynamics
In this section only two studies are represented. The first study was made for conditions in
Japan. A constructed CFD code is made to study the transient effects of a thermal storage
wall. The authors suggest that this technique may be viable for low-energy houses with
hybrid systems (Onishi et al., 2001).
In Karlsson and Moshfegh (2006) the authors present an overview and results from a low-
energy building built in Lindås, Sweden. The authors use CFD to study the air velocity and
temperature pattern in one of the rooms in the building. In addition, building simulation is
used, in the form of ESP-r, to study dynamic effects. Furthermore the paper investigates the
importance of high-performance windows, and concludes that this is an important factor,
both to decrease energy use and for the thermal environment. The paper also discusses the
problem of overheating during summer.

4.2 Building Energy Simulation (BES)
Building energy simulation is a commonly used tool for predicting energy use and other
parameters for buildings and to conduct parametric studies in different forms. A total of 18
papers are cited in this section. In Ohanessian and Charters (1978) a thermal simulation of a
solar passive house with a Trombe-Michel wall is presented. The technique as well as
measurements and simulations are presented, and an energy savings potential of about 40%
is demonstrated for the passive solar house.

Low-energy buildings – scientifc trends and developments 109

the researcher abstracts from the description and searches for patterns and dysfunctional
ties in relation to earlier studies or theories (Kvale, 1996).

3.5 Questionnaires
Questionnaires are an important part of survey research since this is the most common data
collection method. Structured interviews might sometimes be similar to questionnaires with
standardized answers although they may be closed or open-ended. When the objective of
the research is specific, for example to determine which factors influence a certain
phenomenon, a statistical analysis of data from questionnaires might be used. Recently,
assumptions taken for granted in survey research have been questioned (Krosnick 1999).
Response rates, pretesting of questionnaires and questionnaire design are among the
contested aspects of this research. Research has shown, for example, that low response rates
might show more accurate results than studies with high response rates (Visser et al., 1996).

3.6 Document studies
Analyses of written texts are important research tools in, for example, content analyses.
Other types of analyses using texts as the primary source of data are discourse and
argumentation analyses. The various types might be used differently in different fields of
research. For example, linguists use discourse analysis to find syntax and choice of words in
relation to contexts and focus is on language as a power tool, while social scientists might
focus on discourse analysis as a tool in itself to reveal social practice (Bergström and Boréus,
2005). This is often referred to as critical discourse analysis (Fairclough, 2003). These texts
might be minutes, records, protocols, letters, articles, books, newspapers, journals found in
libraries, on the internet, in public or private archives and files, etc. Some texts are superior
to other data collection methods in terms of reliability since they are real-time
documentations of a phenomenon, compared to interviews which construct phenomena
during the interview. Also, large amounts of written texts enable researchers to do structural
analyses in relation to a research question (Lincoln and Guba, 1985)

3.7 Statistical
Statistics is a whole scientific field with several subfields, but in general, statistical methods
include ways to collect, process and draw conclusions from statistical figures to enable, for
example, sampling and analyse differences (Körner et al., 1984). Focus is on empirical data
and statistical methods are used in both natural and social sciences. Descriptive statistics
include mapping and providing information, while inferential statistics aim to present
causal explanations and factors that influence a certain phenomenon. The basis for
explorations is elements in a population (Moses, 1986). Populations and elements might be
individuals but can also refer to material or physical entities. A census is a study using
whole populations, but usually a sample of a population is selected in order to limit the
research due to resource constraints. Social and economic sciences use questionnaires or
observations to collect data, while the most common data collection method in natural
sciences is different types of observations. Data can be analysed using computer programs
such as SPSS.

3.8 Environmental or Life Cycle-Focused Studies
Different types of environmental or life cycle-focused studies form an eclectic category of
research. A definition of life cycle assessment is “compilation and evaluation of the inputs,
outputs and potential environmental impacts of a product system throughout its life cycle”
(Guinée, 2002). Industrial ecology covers a similar field of exploration (Frosch and
Gallopoulos, 1989). Some of these methods claim to have a “holistic” approach but in reality
it is impossible to include all environmental aspects “from cradle to grave” and methods
with this approach might be criticized because they promise too much. Also, a “system”
approach is common among the different methods and may receive the same critique (ISO ,
1997). The definition of system boundaries is a constant issue of debate. Another critical
point in these analyses is the validity of data (Cooper and Fava, 2006). Common concepts
included in energy-related research with these approaches are “embodied energy” (Odum,
2007) and exergy (Dinçer and Rosen, 2007). Social issues and phenomena that are hard to
put in figures are generally missing in these studies.

4. Articles by method
In this section the articles from the review will be presented under a characterization of the
method used in the paper. As previously stated, this is not an unambiguous characterization
but should be seen as a way of grouping the studies reviewed within this chapter. Below a
short presentation of the studies is made and a discussion of the main lessons learned will
follow.

4.1 Computational Fluid Dynamics
In this section only two studies are represented. The first study was made for conditions in
Japan. A constructed CFD code is made to study the transient effects of a thermal storage
wall. The authors suggest that this technique may be viable for low-energy houses with
hybrid systems (Onishi et al., 2001).
In Karlsson and Moshfegh (2006) the authors present an overview and results from a low-
energy building built in Lindås, Sweden. The authors use CFD to study the air velocity and
temperature pattern in one of the rooms in the building. In addition, building simulation is
used, in the form of ESP-r, to study dynamic effects. Furthermore the paper investigates the
importance of high-performance windows, and concludes that this is an important factor,
both to decrease energy use and for the thermal environment. The paper also discusses the
problem of overheating during summer.

4.2 Building Energy Simulation (BES)
Building energy simulation is a commonly used tool for predicting energy use and other
parameters for buildings and to conduct parametric studies in different forms. A total of 18
papers are cited in this section. In Ohanessian and Charters (1978) a thermal simulation of a
solar passive house with a Trombe-Michel wall is presented. The technique as well as
measurements and simulations are presented, and an energy savings potential of about 40%
is demonstrated for the passive solar house.

Energy Effciency 110

In Clarke et al. (1998) an integrated model of a low-energy building is presented. The case
study is a city centre building in Glasgow, where an optimum mix of low, passive and active
renewable energy technologies is sought. The main method used is the well-known
simulation software ESP-r. The paper describes results from simulation and outlines a
method by which “the replication potential of beneficial outcomes can be assessed”.
A methodology for computational support during all phases of a design process with a
focus on energy is presented in Shaviv (1996). The paper combines a procedural simulation
approach with a knowledge-based heuristic approach in one integrated system. The overall
aim is to provide architects and other actors in the design phase with a tool that can be used
throughout the entire process (Shaviv, 1996). Holm (1996) also focuses on architects and the
status of low-energy architecture in South Africa is presented. Chlela et al. (2009) discuss the
design phase and a new methodology for design is proposed. Instead of parametric studies
of design criteria using building energy simulation to optimize building envelope and
HVAC, a Design of Experiments approach is suggested. The methodology is tested on three
French cities with cold, moderate and hot climate. The models show “rather good results”
for the annual heating demand and final energy use for the building as a whole. However,
less accurate results were obtained for the cooling demand. The authors point out that
further improvements may be made on the model.
In Lomas (1996) an application study of thermal simulation programs for passive solar
house design from the U.K. is presented. This type of simulation program has been
extensively used within the passive solar program in the U.K, and is also used in: (1)
domestic and non-domestic building design studies; (2) in the assessment of innovative
material and design techniques; (3) development of design guidelines; and (4) the design
and interpretation of building monitoring studies. This makes it important that the
programs must be reliable when it comes to: (1) the changes in energy demand of the
building when making changes to the building such as changing window size; (2) the
energy savings as a result of a design change, this to be able to predict pay-off times or other
investment criteria; (3) the absolute energy demand of the house and the internal
temperatures, this to be able to compare with energy targets and for example the risk of
overheating, etc. The paper presents among other things (1) a methodology for structuring
inter-program variability studies; (2) an overview of the U.K. application study project; and
(3) a proposal of a Simulation Resolution (SR) concept. The authors argue that the SR could
be taken as an estimate for the best (smallest) value that can be achieved for U.K. domestic
buildings. The value provides a basis for estimating the significance of thermal predictions
by this type of transient simulations. In the study ESP-r and HTB2 and SERIES were used.
Kalogirou et al. (2000) use an artificial neural network to predict energy use in a passive
solar building. According to the authors, the model presented was able to predict energy use
with acceptable accuracy. The model also proved to be much faster than dynamic simulation
programs.
Solar heating is central in Badescu (2005) where active solar heating systems for passive
houses are investigated using simulation. The suggested systems were tested on a passive
house, and a suggested control scheme is outlined. In Badescu and Sicre (2003a) a
description of the case is made, which is based on measurements and input from the
Pirmasens passive house in Germany. Detailed input in the form of standardized data for a
typical German family is used. Badescu and Sicre (2003b) reports on a model for predicting
the thermal behaviour of this passive house. The topic of the paper is evaluation of

renewable energy in the context of passive houses. The renewable energy alternatives in
focus in the paper are (1) passive solar heating with large windows facing the south; (2)
active solar collectors for space heating and heating of domestic hot water; and (3) ground
heat exchanger to preheat the supply air. The authors argue that the computational effort of
transient simulation for this type of problem is valuable.
Feist et al. (2005) introduce and summarize the Passive House Standard and results from the
EU project “Cost Efficient Passive Houses as European Standards” (CEPHEUS). The aim,
according to the authors, is to “provide an acceptable and even improved indoor
environment in terms of indoor air quality (IAQ) and thermal comfort at minimum energy
demand and cost”. This is achieved by improving the thermal performance of the envelope
in such a way that the heating system can be simplified, thus keeping costs at a minimum.
One important factor of the high-performance envelope is that the temperatures on the
inside surface are close to the room air temperature and thus the radiation asymmetry is
small. This enables high thermal comfort by the use of supply air heating instead of
conventional radiator systems that usually compensate for both down draught and
radiation asymmetry. If the thermal properties of the wall are low enough it is also possible
to only use the IAQ-based supply air for heating, without exceeding 50°C which is a
possible temperature to supply air without complications. The thermal transmittance for the
wall is proposed to be <0.15 W/m
2
K for Central Europe; for air tightness a value of <0.6 h
-1

at 50 Pa; and for heat exchanger efficiency >75%. For a climate in central Europe a
requirement of less than 15 kWh/m
2
a is also set and a maximum power demand of 10
W/m
2
. A more detailed description of values for different building components is found in
Feist et al. (2005). The CEPHEUS project includes 221 housing units from five countries that
comply with the passive house standards. The aim of the projects is, according to the
authors, “to demonstrate the technical feasibility (in terms of achieving the target energy
performance indices) at low extra cost (target: compensation of extra investment cost by cost
savings in operation) for a variety of different buildings, constructions and designs
implemented by architects and developers in several European countries.”. Results in the
study show no correlation between types of heating system and mean indoor temperature.
Supply air heating was found suitable for passive houses. A comparison between the
passive houses with other newly built conventional buildings show a reduction in useful
energy by 56%, final energy 52% and primal energy by 56%. The thermal comfort is reported
to be good to very good for these buildings built in central Europe.
In Persson et al. (2006) the authors investigate the influence of window size on the energy
balance of low-energy buildings. The aim of the paper is to “investigate how decreasing the
window size facing south and increasing window size facing north” would affect energy
demand. A building energy simulation tool (DEROB-LTH) was used in the study. The
authors conclude that the size of the energy-efficient windows does not have any major
influence on heating demand during the cold season. However, the authors show that
window size is important during summer, as it will affect the solar gains during this season.
The main conclusion of the paper is that, according to the authors, it is possible to build
windows in a more traditional way even in low-energy buildings and thereby gain better
indoor lighting conditions. Al-Sallal (1998), a case study for a one-storey house in Fresno,
California, also focuses on windows. Here the effects of window size on passive cooling,
and passive heating in day lighting are investigated for hot, arid regions. Wall (2006)
investigates the first Swedish passive house project, twenty terraced passive houses in
Low-energy buildings – scientifc trends and developments 111

In Clarke et al. (1998) an integrated model of a low-energy building is presented. The case
study is a city centre building in Glasgow, where an optimum mix of low, passive and active
renewable energy technologies is sought. The main method used is the well-known
simulation software ESP-r. The paper describes results from simulation and outlines a
method by which “the replication potential of beneficial outcomes can be assessed”.
A methodology for computational support during all phases of a design process with a
focus on energy is presented in Shaviv (1996). The paper combines a procedural simulation
approach with a knowledge-based heuristic approach in one integrated system. The overall
aim is to provide architects and other actors in the design phase with a tool that can be used
throughout the entire process (Shaviv, 1996). Holm (1996) also focuses on architects and the
status of low-energy architecture in South Africa is presented. Chlela et al. (2009) discuss the
design phase and a new methodology for design is proposed. Instead of parametric studies
of design criteria using building energy simulation to optimize building envelope and
HVAC, a Design of Experiments approach is suggested. The methodology is tested on three
French cities with cold, moderate and hot climate. The models show “rather good results”
for the annual heating demand and final energy use for the building as a whole. However,
less accurate results were obtained for the cooling demand. The authors point out that
further improvements may be made on the model.
In Lomas (1996) an application study of thermal simulation programs for passive solar
house design from the U.K. is presented. This type of simulation program has been
extensively used within the passive solar program in the U.K, and is also used in: (1)
domestic and non-domestic building design studies; (2) in the assessment of innovative
material and design techniques; (3) development of design guidelines; and (4) the design
and interpretation of building monitoring studies. This makes it important that the
programs must be reliable when it comes to: (1) the changes in energy demand of the
building when making changes to the building such as changing window size; (2) the
energy savings as a result of a design change, this to be able to predict pay-off times or other
investment criteria; (3) the absolute energy demand of the house and the internal
temperatures, this to be able to compare with energy targets and for example the risk of
overheating, etc. The paper presents among other things (1) a methodology for structuring
inter-program variability studies; (2) an overview of the U.K. application study project; and
(3) a proposal of a Simulation Resolution (SR) concept. The authors argue that the SR could
be taken as an estimate for the best (smallest) value that can be achieved for U.K. domestic
buildings. The value provides a basis for estimating the significance of thermal predictions
by this type of transient simulations. In the study ESP-r and HTB2 and SERIES were used.
Kalogirou et al. (2000) use an artificial neural network to predict energy use in a passive
solar building. According to the authors, the model presented was able to predict energy use
with acceptable accuracy. The model also proved to be much faster than dynamic simulation
programs.
Solar heating is central in Badescu (2005) where active solar heating systems for passive
houses are investigated using simulation. The suggested systems were tested on a passive
house, and a suggested control scheme is outlined. In Badescu and Sicre (2003a) a
description of the case is made, which is based on measurements and input from the
Pirmasens passive house in Germany. Detailed input in the form of standardized data for a
typical German family is used. Badescu and Sicre (2003b) reports on a model for predicting
the thermal behaviour of this passive house. The topic of the paper is evaluation of

renewable energy in the context of passive houses. The renewable energy alternatives in
focus in the paper are (1) passive solar heating with large windows facing the south; (2)
active solar collectors for space heating and heating of domestic hot water; and (3) ground
heat exchanger to preheat the supply air. The authors argue that the computational effort of
transient simulation for this type of problem is valuable.
Feist et al. (2005) introduce and summarize the Passive House Standard and results from the
EU project “Cost Efficient Passive Houses as European Standards” (CEPHEUS). The aim,
according to the authors, is to “provide an acceptable and even improved indoor
environment in terms of indoor air quality (IAQ) and thermal comfort at minimum energy
demand and cost”. This is achieved by improving the thermal performance of the envelope
in such a way that the heating system can be simplified, thus keeping costs at a minimum.
One important factor of the high-performance envelope is that the temperatures on the
inside surface are close to the room air temperature and thus the radiation asymmetry is
small. This enables high thermal comfort by the use of supply air heating instead of
conventional radiator systems that usually compensate for both down draught and
radiation asymmetry. If the thermal properties of the wall are low enough it is also possible
to only use the IAQ-based supply air for heating, without exceeding 50°C which is a
possible temperature to supply air without complications. The thermal transmittance for the
wall is proposed to be <0.15 W/m
2
K for Central Europe; for air tightness a value of <0.6 h
-1

at 50 Pa; and for heat exchanger efficiency >75%. For a climate in central Europe a
requirement of less than 15 kWh/m
2
a is also set and a maximum power demand of 10
W/m
2
. A more detailed description of values for different building components is found in
Feist et al. (2005). The CEPHEUS project includes 221 housing units from five countries that
comply with the passive house standards. The aim of the projects is, according to the
authors, “to demonstrate the technical feasibility (in terms of achieving the target energy
performance indices) at low extra cost (target: compensation of extra investment cost by cost
savings in operation) for a variety of different buildings, constructions and designs
implemented by architects and developers in several European countries.”. Results in the
study show no correlation between types of heating system and mean indoor temperature.
Supply air heating was found suitable for passive houses. A comparison between the
passive houses with other newly built conventional buildings show a reduction in useful
energy by 56%, final energy 52% and primal energy by 56%. The thermal comfort is reported
to be good to very good for these buildings built in central Europe.
In Persson et al. (2006) the authors investigate the influence of window size on the energy
balance of low-energy buildings. The aim of the paper is to “investigate how decreasing the
window size facing south and increasing window size facing north” would affect energy
demand. A building energy simulation tool (DEROB-LTH) was used in the study. The
authors conclude that the size of the energy-efficient windows does not have any major
influence on heating demand during the cold season. However, the authors show that
window size is important during summer, as it will affect the solar gains during this season.
The main conclusion of the paper is that, according to the authors, it is possible to build
windows in a more traditional way even in low-energy buildings and thereby gain better
indoor lighting conditions. Al-Sallal (1998), a case study for a one-storey house in Fresno,
California, also focuses on windows. Here the effects of window size on passive cooling,
and passive heating in day lighting are investigated for hot, arid regions. Wall (2006)
investigates the first Swedish passive house project, twenty terraced passive houses in
Energy Effciency 112

Lindås, Gothenburg. The houses were constructed to meet the target peak power of 10
W/m
2
, and were to use supply air heating. The focus of the project was low transmission
losses (U=0.16 W/m
2
K) and low ventilation losses, which meant using a high-performance
heat exchanger (80%) and a high degree of insulation. Special focus was also applied to get
the buildings airtight. The average airtightness at 50 Pa was measured to 0.3 l/s.m
2
, and
should be compared to the common praxis in Sweden, 0.8 l/s.m
2
. In addition to these
passive measures a solar heating system was installed to provide about 40% of the heat
needed for hot tap water. The paper also include a parametric study of space heating and
power demand as a function of set point for heating and cooling, infiltration rate, etc. The
simulations are compared to monitored energy performance for the building. The author
concludes that the air tightness of the building envelope is essential to meet the targets of
10W/m
2
and the low heating demand of about 15 kWh/m
2
.a.
Material selection in passive solar buildings is addressed in Thomas et al. (2006). The paper
presents a combination of analytical, experimental and computational studies for selecting
affordable materials and designing buildings with the aim of high thermal comfort. The
models are validated using measurements in two housing complexes in Egypt.
In Karlsson et al. (2007) the authors make a comparison between three different energy
simulation codes, and use a low-energy building as a case. All three models use dynamic
models to calculate energy demand for heating and indoor temperature. The low-energy
case is a well-known and extensively measured low-energy building in Lindås, Sweden. A
parameter of interest in the paper was the small difference between the software’s in terms
of deviation of energy use. Thus, the paper shows that the relative importance in terms of
choice of software is small compared to the large difference in terms of deviation between
different households within the studied low-energy buildings. The deviation between
software’s is shown to be as low as 2%, but the deviation between different households
ranges from 6,000 kWh/year to 12,000 kWh/year with an average of 8,020 kWh/year.
Furthermore, occupant behaviour, heat exchanger efficiencies as well as air flow control are
shown to be important factors. The authors also stress the need for more detailed
information about activities and more input data from manufacturers.
In Wang et al. (2009) the authors present a case study of zero-energy house design in the
U.K. Zero-energy buildings are defined in the paper as “a building with a net consumption
of zero over a typical year.” This means that the energy use for heat and electricity is
reduced at the same time as this demand is met on an annual basis from renewable energy
supply. The renewable can either be building integrated or part of a community renewable
energy supply system. A combination of TRANSYS and EnergyPlus is used in the paper,
where EnergyPlus is used for building envelope design and TRANSYS for the installations
as well as the renewable energy system design. The conclusion of the study is that it is
theoretically possible to build zero-energy houses in the U.K. The study also suggests a
methodology for the design process where three steps are summarized: (1) analysis of the
local climate; (2) application of passive design methods; and (3) investigation of various
systems for supply and installations such as PV, wind turbines and solar hot water to
optimize the design.
Zhu et al. (2009a) present an energy and economic analysis of a zero-energy house and
compare this with a conventional house in Las Vegas. Two houses were built side by side,
one zero-energy house and one baseline house, and energy performance measurements
were made. The energy contribution from the different components in the building was

obtained using Energy10 and eQUEST3.6. The results from these two models gave similar
results. The study concludes that four components were clearly economically under given
constraints: high-performance windows, compact fluorescent lighting, well insulated roofs
and AC units with water-cooled condensers. If financial support was included PV tiles were
also considered to have good financial return. Thermal mass walls were found to be too
costly. Walls are in focus in Zhu (2009b) where a detailed energy-saving analysis of a high
thermal mass wall is presented. This is demonstrated in an actual construction project and
compared to a conventional wall. It is shown that for this wall construction the heating use
in the building was much lower, but the load slightly higher. According to the study this is
due to the effect that more heat is stored during the day than can be returned during the
night, increasing the cooling demand. The simulation software used is Energy10 and the
experimental part of the study was carried out in Las Vegas.
In Heim et al. (2010) the authors investigate isothermal storage of solar energy in building
construction with focus on passive houses. A storage system with phase change materials
that absorbs heat during the hot period and releases heat during the cold period is analyzed.
The material behaviour is studied using numerical techniques. These methods are then
implemented in a general building simulation tool, ESP-r. The paper investigates the effect
of a PCM wall and the influence on internal surface temperature. The case is compared with
a conventional wall without the PCM material. Both diurnal as well as seasonal latent heat
storage is studied. The authors conclude that isothermal heat storage may improve thermal
conditions on internal surfaces, but emphasize that the effect of the latent heat storage will
depend on its structure, phase change temperature range and total latent heat of the phase
change.

4.3 Measurements
Measurements as a method for investigating low-energy buildings is another commonly
used method for the papers included in this review. A total of 14 papers are connected to
this section. In Dallaire (1980) the concept of zero-energy houses is introduced as a bold low-
cost breakthrough that may revolutionize housing. The benefits of super-insulated houses
are described in a U.S. and Canadian context, and performance and cost of different
components are also described. The background to the development of the concept was the
increasing prices of oil in the U.S. The paper includes a series of empirical studies, and
among other things the importance of keeping infiltration to a minimum was emphasized.
In Starr et al. (1980) a passive solar house research project is presented. The project
demonstrates significant savings in energy. The studied house use 82% less energy than the
average California house at the time. The importance of thermal mass, window location and
direction as well as insulation is shown. The study contains monitoring for over one year,
including both winter and summer conditions.
Nieminen (1994) presents results from the Finnish demonstration houses within IEA Task 13
“Advanced solar low-energy houses”. This demonstration project shows an estimated
energy use of about 20kWh/m
2
, which was below 10% of the average value for small houses
in Finland at the time. The main focus of the project was to reduce space heating demand.
Filippin et al. (1998) present the first two years of experiences from a passive solar house in
Argentina. The paper presents measurements and the authors conclude that simulation
during the design phase had significant advantages, and the internal gains in the form of
equipment use patterns have an important influence on the performance.
Low-energy buildings – scientifc trends and developments 113

Lindås, Gothenburg. The houses were constructed to meet the target peak power of 10
W/m
2
, and were to use supply air heating. The focus of the project was low transmission
losses (U=0.16 W/m
2
K) and low ventilation losses, which meant using a high-performance
heat exchanger (80%) and a high degree of insulation. Special focus was also applied to get
the buildings airtight. The average airtightness at 50 Pa was measured to 0.3 l/s.m
2
, and
should be compared to the common praxis in Sweden, 0.8 l/s.m
2
. In addition to these
passive measures a solar heating system was installed to provide about 40% of the heat
needed for hot tap water. The paper also include a parametric study of space heating and
power demand as a function of set point for heating and cooling, infiltration rate, etc. The
simulations are compared to monitored energy performance for the building. The author
concludes that the air tightness of the building envelope is essential to meet the targets of
10W/m
2
and the low heating demand of about 15 kWh/m
2
.a.
Material selection in passive solar buildings is addressed in Thomas et al. (2006). The paper
presents a combination of analytical, experimental and computational studies for selecting
affordable materials and designing buildings with the aim of high thermal comfort. The
models are validated using measurements in two housing complexes in Egypt.
In Karlsson et al. (2007) the authors make a comparison between three different energy
simulation codes, and use a low-energy building as a case. All three models use dynamic
models to calculate energy demand for heating and indoor temperature. The low-energy
case is a well-known and extensively measured low-energy building in Lindås, Sweden. A
parameter of interest in the paper was the small difference between the software’s in terms
of deviation of energy use. Thus, the paper shows that the relative importance in terms of
choice of software is small compared to the large difference in terms of deviation between
different households within the studied low-energy buildings. The deviation between
software’s is shown to be as low as 2%, but the deviation between different households
ranges from 6,000 kWh/year to 12,000 kWh/year with an average of 8,020 kWh/year.
Furthermore, occupant behaviour, heat exchanger efficiencies as well as air flow control are
shown to be important factors. The authors also stress the need for more detailed
information about activities and more input data from manufacturers.
In Wang et al. (2009) the authors present a case study of zero-energy house design in the
U.K. Zero-energy buildings are defined in the paper as “a building with a net consumption
of zero over a typical year.” This means that the energy use for heat and electricity is
reduced at the same time as this demand is met on an annual basis from renewable energy
supply. The renewable can either be building integrated or part of a community renewable
energy supply system. A combination of TRANSYS and EnergyPlus is used in the paper,
where EnergyPlus is used for building envelope design and TRANSYS for the installations
as well as the renewable energy system design. The conclusion of the study is that it is
theoretically possible to build zero-energy houses in the U.K. The study also suggests a
methodology for the design process where three steps are summarized: (1) analysis of the
local climate; (2) application of passive design methods; and (3) investigation of various
systems for supply and installations such as PV, wind turbines and solar hot water to
optimize the design.
Zhu et al. (2009a) present an energy and economic analysis of a zero-energy house and
compare this with a conventional house in Las Vegas. Two houses were built side by side,
one zero-energy house and one baseline house, and energy performance measurements
were made. The energy contribution from the different components in the building was

obtained using Energy10 and eQUEST3.6. The results from these two models gave similar
results. The study concludes that four components were clearly economically under given
constraints: high-performance windows, compact fluorescent lighting, well insulated roofs
and AC units with water-cooled condensers. If financial support was included PV tiles were
also considered to have good financial return. Thermal mass walls were found to be too
costly. Walls are in focus in Zhu (2009b) where a detailed energy-saving analysis of a high
thermal mass wall is presented. This is demonstrated in an actual construction project and
compared to a conventional wall. It is shown that for this wall construction the heating use
in the building was much lower, but the load slightly higher. According to the study this is
due to the effect that more heat is stored during the day than can be returned during the
night, increasing the cooling demand. The simulation software used is Energy10 and the
experimental part of the study was carried out in Las Vegas.
In Heim et al. (2010) the authors investigate isothermal storage of solar energy in building
construction with focus on passive houses. A storage system with phase change materials
that absorbs heat during the hot period and releases heat during the cold period is analyzed.
The material behaviour is studied using numerical techniques. These methods are then
implemented in a general building simulation tool, ESP-r. The paper investigates the effect
of a PCM wall and the influence on internal surface temperature. The case is compared with
a conventional wall without the PCM material. Both diurnal as well as seasonal latent heat
storage is studied. The authors conclude that isothermal heat storage may improve thermal
conditions on internal surfaces, but emphasize that the effect of the latent heat storage will
depend on its structure, phase change temperature range and total latent heat of the phase
change.

4.3 Measurements
Measurements as a method for investigating low-energy buildings is another commonly
used method for the papers included in this review. A total of 14 papers are connected to
this section. In Dallaire (1980) the concept of zero-energy houses is introduced as a bold low-
cost breakthrough that may revolutionize housing. The benefits of super-insulated houses
are described in a U.S. and Canadian context, and performance and cost of different
components are also described. The background to the development of the concept was the
increasing prices of oil in the U.S. The paper includes a series of empirical studies, and
among other things the importance of keeping infiltration to a minimum was emphasized.
In Starr et al. (1980) a passive solar house research project is presented. The project
demonstrates significant savings in energy. The studied house use 82% less energy than the
average California house at the time. The importance of thermal mass, window location and
direction as well as insulation is shown. The study contains monitoring for over one year,
including both winter and summer conditions.
Nieminen (1994) presents results from the Finnish demonstration houses within IEA Task 13
“Advanced solar low-energy houses”. This demonstration project shows an estimated
energy use of about 20kWh/m
2
, which was below 10% of the average value for small houses
in Finland at the time. The main focus of the project was to reduce space heating demand.
Filippin et al. (1998) present the first two years of experiences from a passive solar house in
Argentina. The paper presents measurements and the authors conclude that simulation
during the design phase had significant advantages, and the internal gains in the form of
equipment use patterns have an important influence on the performance.
Energy Effciency 114

Schnieders and Hermelink (2006) present the results from the CEPHEUS project, where the
material includes measurements and occupant satisfaction for passive houses. The Cost
Efficient Passive Houses as European Standard (CEPHEUS) project here includes over 100
dwellings that have been studied. All the projects within the program exhibit
extraordinarily low energy use according to the authors, as they can save 80% of space
heating compared to ordinary buildings and the total primary energy use was down 50%
when compared. It is also concluded that this is achievable with high performance in terms
of thermal comfort both in summer and winter for the houses in the project.
In Liu and Henze (2006b) an experimental analysis of simulated learning control for active
and passive building thermal storage is reported. In Liu and Henze (2006a) the theoretical
foundation is presented and in Liu and Henze (2006b) the results and analysis are found.
The work was conducted at the Energy Resource Center Station in Iowa.
The next article discusses how to address the effects of climate change and thermal comfort
while at the same time meeting the design challenges of the twenty-first century (Holmes et
al., 2007). The authors demonstrate the effect of a changing climate with increasing
temperatures using predictions and outline a series of principles in terms of load
management, cooling and heating using alternative systems. The paper also shows some of
the effects of solar shading as a way of controlling internal gains, as well as the effect of
night cooling and other ventilation applications. They also state that the study shows that
high-mass buildings are able to provide a higher quality in terms of “internal environment”.
Tommerup (2007) presents the results from measurements and discusses how to develop
typical single-family houses to meet new energy requirements without compromising on
either economy or construction. Tommerup has studied energy-efficient houses built
according to the new energy performance requirements in Denmark. The purpose of the
project was to demonstrate that it is possible to produce energy-efficient single-family
houses that meet existing standards without compromising on economy or construction.
Tommerup presents the houses within the project as well as the energy efficiency measures
that were applied. A full presentation of energy use, thermal comfort and airtightness is also
included. The energy used by these buildings is about 50% to 75% of the typical energy use
in Danish buildings in general.
Makaka et al. (2008) present a case study where the building was monitored for a period
covering all the South African seasons. The performance of the building was seen to depend
mainly on how the occupants used the house. The type of house presented was shown to
represent a lower rate of temperature and humidity variations. The thermal behaviour and
ventilation efficiency of a low-cost passive solar energy efficient house is investigated. The
low-cost houses in South Africa are categorized by poor craftsmanship in terms of energy-
efficient design and passive solar features, thermal climate and ventilation efficiency. If this
type of design is used, large savings are possible.
Chandel and Aggarwal (2008) have done an evaluation of the performance of a passive solar
building. The building is located in the Western Himalayas. The heat losses were shown to
be reduced by approximately 35% with passive solar measures. In Wojdyga (2009) the
author investigates the heat demand in a low-energy building in Poland. Results from a
five-year study of the energy consumption in a single-storey terraced low-energy house are
presented.
Maier et al. (2009) combines methods when presenting a comparison of physical
performance of a ventilation system in residential low-energy buildings. To analyze the

influence of ventilation systems on comfort, the authors used a combination of
measurements and a questionnaire. The measurement part of the project included 22
residential houses in Germany, chosen and equipped with four different types of ventilation
systems: (1) natural ventilation; (2) air heating system; (3) mechanical ventilation with
supply and exhaust with heat exchanging; and (4) mechanical ventilation with single
ventilators. The monitor’s parameters were CO
2
, relative humidity, air temperature,
electricity, gas and heat. The use of the window openings, use of ventilation and number of
residents present were also presented. The mechanical ventilation performed better in terms
of CO
2
concentration than the naturally ventilated cases.
Feist and Schnieders (2009) explain the concept of the passive house technique and issues
such as design methods, components in a passive house such as thermal bridges, windows,
junctions of roof and wall, etc., the importance of internal gains and ventilation and air
tightness issues. Practical experience is also summarized. In a similar way Nicoletti (1998)
discusses low-energy design from an architectural point of view. Form as a tool for energy
control is discussed, and several examples and objects such as the University building in
Udine, Casa Moncada and a headquater of a bank in Rome. A special discussion for tall
buildings is included.
Kalz et al. (2010) have also done a long-term study. The authors present what they define as
a holistic approach when evaluating heating and cooling using building signatures. The
study includes a comprehensive analysis of eleven low-energy buildings in terms of energy
use and thermal comfort. The long-term study is presented using detailed time series for
between two and five years depending on object. In the paper a methodology is described
for evaluating heating and cooling concepts, not only by focusing on thermal comfort but
also by including the useful energy consumption and energy efficiency, generation,
distribution and deliveries.

4.4 Interviews
Two papers are related to this section, one interdisciplinary study of the low-energy houses
in Lindås, Sweden and one article investigating the attitude of large construction companies.
Isaksson and Karlsson (2006) present an interdisciplinary study of the indoor climate in the
low-energy buildings in Lindås in Sweden. The paper presents results from an investigation
of the 20 terraced houses in Gothenburg in Sweden. Qualitative interviews with occupants
are combined with physical measurements of the thermal environment. The results show
that when occupants are present and appliances are used, the temperature can be managed
within acceptable limits even during cold days. One main outcome of the study is the
importance of information given to households about the functionality of the heating
system. In addition to this the authors state that temperature control could be improved.
The paper also gives a wider view of how activities within these houses change compared to
normal houses where power is less of a problem. Special focus should also be placed on
gable houses in terms of thermal comfort.

In Hamza and Greenwood (2009) the impact of the new energy conservation regulations
and its impact on low-energy buildings is studied. Data collection was made by semi-
structured interviews with a sample representing large construction companies,
architectural practitioners and building performance consultants. The authors express that
“overall, it appears likely that the legislation is already having a profound effect on the
Low-energy buildings – scientifc trends and developments 115

Schnieders and Hermelink (2006) present the results from the CEPHEUS project, where the
material includes measurements and occupant satisfaction for passive houses. The Cost
Efficient Passive Houses as European Standard (CEPHEUS) project here includes over 100
dwellings that have been studied. All the projects within the program exhibit
extraordinarily low energy use according to the authors, as they can save 80% of space
heating compared to ordinary buildings and the total primary energy use was down 50%
when compared. It is also concluded that this is achievable with high performance in terms
of thermal comfort both in summer and winter for the houses in the project.
In Liu and Henze (2006b) an experimental analysis of simulated learning control for active
and passive building thermal storage is reported. In Liu and Henze (2006a) the theoretical
foundation is presented and in Liu and Henze (2006b) the results and analysis are found.
The work was conducted at the Energy Resource Center Station in Iowa.
The next article discusses how to address the effects of climate change and thermal comfort
while at the same time meeting the design challenges of the twenty-first century (Holmes et
al., 2007). The authors demonstrate the effect of a changing climate with increasing
temperatures using predictions and outline a series of principles in terms of load
management, cooling and heating using alternative systems. The paper also shows some of
the effects of solar shading as a way of controlling internal gains, as well as the effect of
night cooling and other ventilation applications. They also state that the study shows that
high-mass buildings are able to provide a higher quality in terms of “internal environment”.
Tommerup (2007) presents the results from measurements and discusses how to develop
typical single-family houses to meet new energy requirements without compromising on
either economy or construction. Tommerup has studied energy-efficient houses built
according to the new energy performance requirements in Denmark. The purpose of the
project was to demonstrate that it is possible to produce energy-efficient single-family
houses that meet existing standards without compromising on economy or construction.
Tommerup presents the houses within the project as well as the energy efficiency measures
that were applied. A full presentation of energy use, thermal comfort and airtightness is also
included. The energy used by these buildings is about 50% to 75% of the typical energy use
in Danish buildings in general.
Makaka et al. (2008) present a case study where the building was monitored for a period
covering all the South African seasons. The performance of the building was seen to depend
mainly on how the occupants used the house. The type of house presented was shown to
represent a lower rate of temperature and humidity variations. The thermal behaviour and
ventilation efficiency of a low-cost passive solar energy efficient house is investigated. The
low-cost houses in South Africa are categorized by poor craftsmanship in terms of energy-
efficient design and passive solar features, thermal climate and ventilation efficiency. If this
type of design is used, large savings are possible.
Chandel and Aggarwal (2008) have done an evaluation of the performance of a passive solar
building. The building is located in the Western Himalayas. The heat losses were shown to
be reduced by approximately 35% with passive solar measures. In Wojdyga (2009) the
author investigates the heat demand in a low-energy building in Poland. Results from a
five-year study of the energy consumption in a single-storey terraced low-energy house are
presented.
Maier et al. (2009) combines methods when presenting a comparison of physical
performance of a ventilation system in residential low-energy buildings. To analyze the

influence of ventilation systems on comfort, the authors used a combination of
measurements and a questionnaire. The measurement part of the project included 22
residential houses in Germany, chosen and equipped with four different types of ventilation
systems: (1) natural ventilation; (2) air heating system; (3) mechanical ventilation with
supply and exhaust with heat exchanging; and (4) mechanical ventilation with single
ventilators. The monitor’s parameters were CO
2
, relative humidity, air temperature,
electricity, gas and heat. The use of the window openings, use of ventilation and number of
residents present were also presented. The mechanical ventilation performed better in terms
of CO
2
concentration than the naturally ventilated cases.
Feist and Schnieders (2009) explain the concept of the passive house technique and issues
such as design methods, components in a passive house such as thermal bridges, windows,
junctions of roof and wall, etc., the importance of internal gains and ventilation and air
tightness issues. Practical experience is also summarized. In a similar way Nicoletti (1998)
discusses low-energy design from an architectural point of view. Form as a tool for energy
control is discussed, and several examples and objects such as the University building in
Udine, Casa Moncada and a headquater of a bank in Rome. A special discussion for tall
buildings is included.
Kalz et al. (2010) have also done a long-term study. The authors present what they define as
a holistic approach when evaluating heating and cooling using building signatures. The
study includes a comprehensive analysis of eleven low-energy buildings in terms of energy
use and thermal comfort. The long-term study is presented using detailed time series for
between two and five years depending on object. In the paper a methodology is described
for evaluating heating and cooling concepts, not only by focusing on thermal comfort but
also by including the useful energy consumption and energy efficiency, generation,
distribution and deliveries.

4.4 Interviews
Two papers are related to this section, one interdisciplinary study of the low-energy houses
in Lindås, Sweden and one article investigating the attitude of large construction companies.
Isaksson and Karlsson (2006) present an interdisciplinary study of the indoor climate in the
low-energy buildings in Lindås in Sweden. The paper presents results from an investigation
of the 20 terraced houses in Gothenburg in Sweden. Qualitative interviews with occupants
are combined with physical measurements of the thermal environment. The results show
that when occupants are present and appliances are used, the temperature can be managed
within acceptable limits even during cold days. One main outcome of the study is the
importance of information given to households about the functionality of the heating
system. In addition to this the authors state that temperature control could be improved.
The paper also gives a wider view of how activities within these houses change compared to
normal houses where power is less of a problem. Special focus should also be placed on
gable houses in terms of thermal comfort.

In Hamza and Greenwood (2009) the impact of the new energy conservation regulations
and its impact on low-energy buildings is studied. Data collection was made by semi-
structured interviews with a sample representing large construction companies,
architectural practitioners and building performance consultants. The authors express that
“overall, it appears likely that the legislation is already having a profound effect on the
Energy Effciency 116

contractual and procurement arrangements of U.K. construction projects.” In Hamza and
Greenwood (2009) a number of interesting impacts of the new legislation is seen on: (1)
tendering practice and documentation; (2) procurement practice; (3) post-tender engineering
and “value engineering”; and (4) collaborative working.

4.5 Questionnaires
Only one article fell under this category. In Thomsen et al. (2005) twelve demonstration
projects within IEI Task 13 “advanced solar low-energy buildings” are presented. The paper
includes a brief presentation of the houses. The study is a follow-up study three years after
Task 13 ended, and is made in the form of a questionnaire sent to the former participants
within the task. The paper states that the measured energy use was in general higher than
expected due mainly to unforeseen technical problems but that energy savings of 60% were
achieved compared to a typical building. The question of overheating in summer was
specifically addressed, and it was shown that with proper planning and design this can be
avoided. However, within the project this was a problem in two cases, one in Norway and
one in Denmark. The paper summarizes a series of lessons to be learned: (1) Special
consideration should be given to heat losses in partitions between apartments in highly
insulated buildings; (2) obtaining the needed air tightness of a house requires careful
planning and control of seals and barriers; (3) ventilation should be designed carefully with
regards to sound and draught; (4) overheating can be prevented in moderate climates by
means of thermal mass, solar shading and ventilation, if they are designed properly; and (5)
heat losses from ducts and pipes are important and should be minimized.

4.6 Environmental or life cycle focus studies
In this final section nine papers are reviewed. In Chwieduk (1999) a study of thermal
modernisation and refurbishment of existing buildings are presented in a Polish context.
The paper outlines a view that a transition to low-energy buildings in Poland is a natural
progression. It also includes some remarks and recommendations for Polish low-energy
buildings.
Tombazis and Preuss (2001) discuss design of solar buildings in an urban context. The study
emphasizes the building’s access to natural resources while taking into account the negative
influence that may prevail around the site. The associated constraints, according to the
authors, are challenging but very interesting and rewarding from an architectural point of
view. The paper exemplifies different design options for three different cases. Even though
the buildings are different they share some features, such as: (1) well insulated; (2) shallow
plan, so daylight is able to penetrate and also to achieve well functioning natural ventilation;
and (3) hybrid ventilation systems are used and some form of intelligent control is included.
Zimmermann et al. (2005) present a benchmark of sustainable construction. The paper
addresses the policy field and has the aim of being a contribution to developing a standard
in the field. The paper also shows that buildings designed to the passive house standard
may comply with the requirements for sustainable construction if the electricity generation
is based on environmentally friendly generation. However, for other parts where a high
degree of fossil fuel is used the authors find it much harder to meet the requirements.
In Rabah (2005) a design strategy for energy-efficient passive solar buildings in Cyprus is
presented. The methodology includes: (1) initial pre-design considerations; (2) initial climate

analysis; (3) determination of passive solar design strategies; (4) analysis of the control zone
(comfort, etc.). The aim of the paper to provide general information at pre-design phases to
be able to more effectively implement passive solar energy. Krishan et al. (1996) also focus
on the design phase and discuss climate responsible design for two cases, one high-altitude
“cold-dry” case and one “hot and dry” case. The paper includes design principles for this
“indigenous architecture of two Indian deserts”.
In Thormark (2006) the effect of choice of material on total energy use and recycling
potential is reported. The author addresses both the need for reducing energy use as well as
the maximization of the recycling potential. Since the embodied energy for a low-energy
house accounts for a large part of the total energy use during the life span of the building, it
is important to consider the choice of material. The article presents the impact of material
choice on the passive houses built in Lindås, Sweden.
In Sartori and Hestnes (2007) energy use in conventional building is compared to low-
energy buildings using a review approach. The review includes 60 cases from nine
countries, and showed that by far the largest part of energy use is related to the operating
phase. The results presented in the paper show that the solar houses proved to be more
energy efficient than the houses within the studies that used “green” materials.
Furthermore, it was shown that solar houses decreased life-cycle energy use by half
compared to a conventional building. A passive house proved to be more efficient than the
solar houses in the studies.
In Aste et al. (2010) the low-energy residential settlement in Borgo Solare, Italy is presented.
The project is not just an experimental operation; instead Borgo Solare is a real urban
district. However, the project may be considered to be an advanced and innovative
residential area designed on sustainable architectural grounds. The paper presents a techno-
economical analysis of the project. The analysis shows that the higher initial embodied
energy in a low-energy building may be paid back well within the life span of the building.
In the economical analysis the authors argue that the higher initial costs may be effective in
the long term.
In Verbeeck and Hens (2010) a life-cycle inventory of buildings is presented. The paper
presents results of a contribution analysis of the life-cycle inventory of four typical
buildings. The location of the objects is Belgium. The paper shows the small importance of
the embodied energy when comparing energy use during the buildings' entire usage phase.
This is also shown to be even more so for energy efficiency measures, when comparing
embodied energy of the measures with the reduction in use. Only extreme low-energy
buildings may have a higher embodied energy than the energy use during the phase when it
is used; for a normal building this represents about one third of the total energy use during
the life cycle. The total savings, however, are still shown to be large for low-energy
dwellings.

5. Concluding discussion
The attention and research in this field is characterized by a strong increase in the number of
articles during the last five years, not least within the scope of low-energy buildings and solar
buildings. However, it is also important to note that most of the development in the field is
taking place outside the scientific community, in construction companies, national programs and
housing companies. The research field, as presented here, is a clearly technical field with a strong
Low-energy buildings – scientifc trends and developments 117

contractual and procurement arrangements of U.K. construction projects.” In Hamza and
Greenwood (2009) a number of interesting impacts of the new legislation is seen on: (1)
tendering practice and documentation; (2) procurement practice; (3) post-tender engineering
and “value engineering”; and (4) collaborative working.

4.5 Questionnaires
Only one article fell under this category. In Thomsen et al. (2005) twelve demonstration
projects within IEI Task 13 “advanced solar low-energy buildings” are presented. The paper
includes a brief presentation of the houses. The study is a follow-up study three years after
Task 13 ended, and is made in the form of a questionnaire sent to the former participants
within the task. The paper states that the measured energy use was in general higher than
expected due mainly to unforeseen technical problems but that energy savings of 60% were
achieved compared to a typical building. The question of overheating in summer was
specifically addressed, and it was shown that with proper planning and design this can be
avoided. However, within the project this was a problem in two cases, one in Norway and
one in Denmark. The paper summarizes a series of lessons to be learned: (1) Special
consideration should be given to heat losses in partitions between apartments in highly
insulated buildings; (2) obtaining the needed air tightness of a house requires careful
planning and control of seals and barriers; (3) ventilation should be designed carefully with
regards to sound and draught; (4) overheating can be prevented in moderate climates by
means of thermal mass, solar shading and ventilation, if they are designed properly; and (5)
heat losses from ducts and pipes are important and should be minimized.

4.6 Environmental or life cycle focus studies
In this final section nine papers are reviewed. In Chwieduk (1999) a study of thermal
modernisation and refurbishment of existing buildings are presented in a Polish context.
The paper outlines a view that a transition to low-energy buildings in Poland is a natural
progression. It also includes some remarks and recommendations for Polish low-energy
buildings.
Tombazis and Preuss (2001) discuss design of solar buildings in an urban context. The study
emphasizes the building’s access to natural resources while taking into account the negative
influence that may prevail around the site. The associated constraints, according to the
authors, are challenging but very interesting and rewarding from an architectural point of
view. The paper exemplifies different design options for three different cases. Even though
the buildings are different they share some features, such as: (1) well insulated; (2) shallow
plan, so daylight is able to penetrate and also to achieve well functioning natural ventilation;
and (3) hybrid ventilation systems are used and some form of intelligent control is included.
Zimmermann et al. (2005) present a benchmark of sustainable construction. The paper
addresses the policy field and has the aim of being a contribution to developing a standard
in the field. The paper also shows that buildings designed to the passive house standard
may comply with the requirements for sustainable construction if the electricity generation
is based on environmentally friendly generation. However, for other parts where a high
degree of fossil fuel is used the authors find it much harder to meet the requirements.
In Rabah (2005) a design strategy for energy-efficient passive solar buildings in Cyprus is
presented. The methodology includes: (1) initial pre-design considerations; (2) initial climate

analysis; (3) determination of passive solar design strategies; (4) analysis of the control zone
(comfort, etc.). The aim of the paper to provide general information at pre-design phases to
be able to more effectively implement passive solar energy. Krishan et al. (1996) also focus
on the design phase and discuss climate responsible design for two cases, one high-altitude
“cold-dry” case and one “hot and dry” case. The paper includes design principles for this
“indigenous architecture of two Indian deserts”.
In Thormark (2006) the effect of choice of material on total energy use and recycling
potential is reported. The author addresses both the need for reducing energy use as well as
the maximization of the recycling potential. Since the embodied energy for a low-energy
house accounts for a large part of the total energy use during the life span of the building, it
is important to consider the choice of material. The article presents the impact of material
choice on the passive houses built in Lindås, Sweden.
In Sartori and Hestnes (2007) energy use in conventional building is compared to low-
energy buildings using a review approach. The review includes 60 cases from nine
countries, and showed that by far the largest part of energy use is related to the operating
phase. The results presented in the paper show that the solar houses proved to be more
energy efficient than the houses within the studies that used “green” materials.
Furthermore, it was shown that solar houses decreased life-cycle energy use by half
compared to a conventional building. A passive house proved to be more efficient than the
solar houses in the studies.
In Aste et al. (2010) the low-energy residential settlement in Borgo Solare, Italy is presented.
The project is not just an experimental operation; instead Borgo Solare is a real urban
district. However, the project may be considered to be an advanced and innovative
residential area designed on sustainable architectural grounds. The paper presents a techno-
economical analysis of the project. The analysis shows that the higher initial embodied
energy in a low-energy building may be paid back well within the life span of the building.
In the economical analysis the authors argue that the higher initial costs may be effective in
the long term.
In Verbeeck and Hens (2010) a life-cycle inventory of buildings is presented. The paper
presents results of a contribution analysis of the life-cycle inventory of four typical
buildings. The location of the objects is Belgium. The paper shows the small importance of
the embodied energy when comparing energy use during the buildings' entire usage phase.
This is also shown to be even more so for energy efficiency measures, when comparing
embodied energy of the measures with the reduction in use. Only extreme low-energy
buildings may have a higher embodied energy than the energy use during the phase when it
is used; for a normal building this represents about one third of the total energy use during
the life cycle. The total savings, however, are still shown to be large for low-energy
dwellings.

5. Concluding discussion
The attention and research in this field is characterized by a strong increase in the number of
articles during the last five years, not least within the scope of low-energy buildings and solar
buildings. However, it is also important to note that most of the development in the field is
taking place outside the scientific community, in construction companies, national programs and
housing companies. The research field, as presented here, is a clearly technical field with a strong
Energy Effciency 118

focus on the technologies and development of techniques for improving energy performance of
buildings. The number of studies with focus on the end users and how they interpret and interact
with this new technology is scarce, but there are examples, such as Isaksson and Karlsson (2006),
Schnieders and Hermelink (2006) and Feist et al. (2005), to name a few important contributions.
This is a field which is important, especially when several authors, among them Karlsson et al.
(2007), stress the importance of the activities and internal gains for low-energy buildings. The
relative importance of this factor is so much greater since the losses from the building are so
much smaller. It is therefore even more important to understand and be able to predict activities
when designing this type of building.
The general trend of publication can also be said to have started to shift if looking at the process
from 1978 to the present day. In the late 1970s and early 1980s the focus in the presented articles
was in general on single technology investigations or building oriented with energy use and cost
as key focus. The main driver was to remove oil-fired burners or to minimize their use, as an
effect of the oil crisis. A shift can be seen in this main focus, as a large part of the articles
presented in the late 1990s and after 2000 have a more environmental focus with greenhouse
emissions as a key focus. This is in line with the general trend. However, what may be of interest
is the increasing number of policy articles that argue for low-energy buildings when looking at
long-term scenarios for sustainable buildings or regions. The number of publications where
sustainable city parts like Borgo Solare in Italy are reported is also increasing. So the general
trend may be argued to be moving from single technology and individual case studies of
buildings to a more regional and large-scale production of energy-efficient city parts. In Lindås a
in comparison small-scale production of 20 terraced passive houses was constructed in Sweden.
These houses are investigated from multiple perspectives and some of the material is reported in
the references here, such as Karlsson and Moshfegh (2006) and Wall (2006) for a technical
description of the implementation and Isaksson and Karlsson (2005) for an interdisciplinary
study of the buildings and also Thormark (2006) for a study of embodied energy and life-cycle
analysis of these buildings.
For Sweden the buildings in Lindås are important as they mark the starting point for building
passive houses in Sweden. Due to that they also represent a starting point in terms of learning to
build this type of building with low infiltration rates and a high level of insulation, etc. which
requires different approaches from the construction industry in terms of process. One interesting
factor is that the German standard for passive houses sees e.g. Fiest and Schnieders (2009) or Fiest
et al. (2005), has been adapted to Swedish conditions by a national forum for energy-efficient
buildings funded by the Swedish energy agency (FEBY). This trend is similar for several other
European countries. One point of interest is how the national standards use the requirements
within the German standard in cases such as for Sweden, where the climate is different. For
Sweden the certification process has the same requirements on maximum power and energy.
These are based on electric heating, using the indoor air quality designed airflow for the building
(10 W/m2) and a maximum energy use of 15 kWh/m
2
. This is of course harder to achieve in a
Nordic climate than for a central European climate. This issue is also connected to the issue of
thermal comfort in passive houses in Nordic regions, where relatively few studies are reported.
However, the issue of thermal comfort in general is something that is becoming more common to
investigate, see Feist (2009) for example, but further studies are still very much needed especially
for cold climates.
Along with users’ interaction and interpretation of low-energy buildings, user satisfaction was
expected to be a main focus of articles in this compilation. However, only one (Isaksson and

Karlsson 2005) explicitly tried to explore this. There are no internationally standardized methods
to evaluate user satisfaction, but a closed-end questionnaire on indoor climate in dwellings has
been developed in, for example, Sweden (Andersson et al., 1988). However, research results
from low-energy buildings in Web of Science using this questionnaire are lacking. Also in
Germany, questionnaires have been used in research about user satisfaction, but in office
buildings (Pfafferott et al., 2007; Wagner et al., 2007). The method has been developed by
University of California’s Center for Environmental Design Research, Berkeley and according to
the authors it addresses “all relevant aspects of occupant satisfaction with indoor environments”
(Wagner et al., p. 764). Results show how user satisfaction corresponds to control abilities for
users which are supported by results in Pfafferott et al. (2007). Actual temperature and
temperature sensations had less effect on user satisfaction in this study. The perceived flexibility
of low-energy buildings is something that future research could address. Post-occupancy
evaluations of office buildings might offer methodological inspiration.
Research focusing on the construction sector, clients, design teams and the organization of
construction processes are in this compilation mainly found in the U.K. (cf. Hamza and
Greenwood, 2009; Hamza and Horne, 2007). Although the articles analyse phenomena specific to
the U.K. (new energy conservation regulations and low-energy architecture in higher education),
some general conclusions can be made. When designing low-energy buildings, more relational
thinking is needed because of the increased complexity in the design phase (Hamza and Horne,
2007). Students in architecture might not have sufficient training in this higher level of
approaching tasks, which includes critical thinking. Modules are being developed, however, to
incorporate and facilitate relational thinking. A new regulation on energy conservation in the
U.K. has also proved to support collaborations between design and construction teams, which is
considered most welcome (Hamza and Greenwood, 2009). As noted in Hamza and Greenwood
(2009), it is important not only to study user satisfaction post occupancy, but also the experiences
of design and construction teams, in order to improve present regulations and practice in
construction processes. Groups that should be addressed are practitioners, educators and policy-
makers and publications in Web of Science journals might not be the most effective way to
disseminate this feedback.

6. References
Al-Sallal, KA. (1998). Sizing windows to achieve passive cooling, passive heating, and
daylighting in hot arid regions, In: Renewable energy, 14 (1-4): MAY-AUG 365-371
Andersson, K.; Fagerlund, I.; Bodin, L.; Ydreborg, B. (1988). Questionnaire as an instrument when
evaluating indoor climate. In: Healthy Buildings´88 Stockholm 1988, Vol 1:139-146
Aste, N.; Adhikari, RS.; Buzzetti, M. (2010). Beyond the EPBD: The low energy residential
settlement Borgo Solare, In: Applied Energy, 87 (2): FEB 629-642
Babbie, E. (1990) Survey research methods (2nd ed.),: Wadsworth. 0-534-12672-3 Belmont CA
Badescu, V. (2005). Simulation analysis for the active solar heating system of a passive house, In:
Applied Thermal Engineering, 25 (17-18): DEC 2754-2763
Badescu, V.; Sicre, B. (2003). In: Renewable energy for passive house heating II. Model, In: Energy
and Buildings, 35 (11): DEC 1085-1096
Badescu, V.; Sicre, B. (2003). Renewable energy for passive house heating Part I. Building
description, In: Energy and Buildings, 35 (11): DEC 1077-1084
Low-energy buildings – scientifc trends and developments 119

focus on the technologies and development of techniques for improving energy performance of
buildings. The number of studies with focus on the end users and how they interpret and interact
with this new technology is scarce, but there are examples, such as Isaksson and Karlsson (2006),
Schnieders and Hermelink (2006) and Feist et al. (2005), to name a few important contributions.
This is a field which is important, especially when several authors, among them Karlsson et al.
(2007), stress the importance of the activities and internal gains for low-energy buildings. The
relative importance of this factor is so much greater since the losses from the building are so
much smaller. It is therefore even more important to understand and be able to predict activities
when designing this type of building.
The general trend of publication can also be said to have started to shift if looking at the process
from 1978 to the present day. In the late 1970s and early 1980s the focus in the presented articles
was in general on single technology investigations or building oriented with energy use and cost
as key focus. The main driver was to remove oil-fired burners or to minimize their use, as an
effect of the oil crisis. A shift can be seen in this main focus, as a large part of the articles
presented in the late 1990s and after 2000 have a more environmental focus with greenhouse
emissions as a key focus. This is in line with the general trend. However, what may be of interest
is the increasing number of policy articles that argue for low-energy buildings when looking at
long-term scenarios for sustainable buildings or regions. The number of publications where
sustainable city parts like Borgo Solare in Italy are reported is also increasing. So the general
trend may be argued to be moving from single technology and individual case studies of
buildings to a more regional and large-scale production of energy-efficient city parts. In Lindås a
in comparison small-scale production of 20 terraced passive houses was constructed in Sweden.
These houses are investigated from multiple perspectives and some of the material is reported in
the references here, such as Karlsson and Moshfegh (2006) and Wall (2006) for a technical
description of the implementation and Isaksson and Karlsson (2005) for an interdisciplinary
study of the buildings and also Thormark (2006) for a study of embodied energy and life-cycle
analysis of these buildings.
For Sweden the buildings in Lindås are important as they mark the starting point for building
passive houses in Sweden. Due to that they also represent a starting point in terms of learning to
build this type of building with low infiltration rates and a high level of insulation, etc. which
requires different approaches from the construction industry in terms of process. One interesting
factor is that the German standard for passive houses sees e.g. Fiest and Schnieders (2009) or Fiest
et al. (2005), has been adapted to Swedish conditions by a national forum for energy-efficient
buildings funded by the Swedish energy agency (FEBY). This trend is similar for several other
European countries. One point of interest is how the national standards use the requirements
within the German standard in cases such as for Sweden, where the climate is different. For
Sweden the certification process has the same requirements on maximum power and energy.
These are based on electric heating, using the indoor air quality designed airflow for the building
(10 W/m2) and a maximum energy use of 15 kWh/m
2
. This is of course harder to achieve in a
Nordic climate than for a central European climate. This issue is also connected to the issue of
thermal comfort in passive houses in Nordic regions, where relatively few studies are reported.
However, the issue of thermal comfort in general is something that is becoming more common to
investigate, see Feist (2009) for example, but further studies are still very much needed especially
for cold climates.
Along with users’ interaction and interpretation of low-energy buildings, user satisfaction was
expected to be a main focus of articles in this compilation. However, only one (Isaksson and

Karlsson 2005) explicitly tried to explore this. There are no internationally standardized methods
to evaluate user satisfaction, but a closed-end questionnaire on indoor climate in dwellings has
been developed in, for example, Sweden (Andersson et al., 1988). However, research results
from low-energy buildings in Web of Science using this questionnaire are lacking. Also in
Germany, questionnaires have been used in research about user satisfaction, but in office
buildings (Pfafferott et al., 2007; Wagner et al., 2007). The method has been developed by
University of California’s Center for Environmental Design Research, Berkeley and according to
the authors it addresses “all relevant aspects of occupant satisfaction with indoor environments”
(Wagner et al., p. 764). Results show how user satisfaction corresponds to control abilities for
users which are supported by results in Pfafferott et al. (2007). Actual temperature and
temperature sensations had less effect on user satisfaction in this study. The perceived flexibility
of low-energy buildings is something that future research could address. Post-occupancy
evaluations of office buildings might offer methodological inspiration.
Research focusing on the construction sector, clients, design teams and the organization of
construction processes are in this compilation mainly found in the U.K. (cf. Hamza and
Greenwood, 2009; Hamza and Horne, 2007). Although the articles analyse phenomena specific to
the U.K. (new energy conservation regulations and low-energy architecture in higher education),
some general conclusions can be made. When designing low-energy buildings, more relational
thinking is needed because of the increased complexity in the design phase (Hamza and Horne,
2007). Students in architecture might not have sufficient training in this higher level of
approaching tasks, which includes critical thinking. Modules are being developed, however, to
incorporate and facilitate relational thinking. A new regulation on energy conservation in the
U.K. has also proved to support collaborations between design and construction teams, which is
considered most welcome (Hamza and Greenwood, 2009). As noted in Hamza and Greenwood
(2009), it is important not only to study user satisfaction post occupancy, but also the experiences
of design and construction teams, in order to improve present regulations and practice in
construction processes. Groups that should be addressed are practitioners, educators and policy-
makers and publications in Web of Science journals might not be the most effective way to
disseminate this feedback.

6. References
Al-Sallal, KA. (1998). Sizing windows to achieve passive cooling, passive heating, and
daylighting in hot arid regions, In: Renewable energy, 14 (1-4): MAY-AUG 365-371
Andersson, K.; Fagerlund, I.; Bodin, L.; Ydreborg, B. (1988). Questionnaire as an instrument when
evaluating indoor climate. In: Healthy Buildings´88 Stockholm 1988, Vol 1:139-146
Aste, N.; Adhikari, RS.; Buzzetti, M. (2010). Beyond the EPBD: The low energy residential
settlement Borgo Solare, In: Applied Energy, 87 (2): FEB 629-642
Babbie, E. (1990) Survey research methods (2nd ed.),: Wadsworth. 0-534-12672-3 Belmont CA
Badescu, V. (2005). Simulation analysis for the active solar heating system of a passive house, In:
Applied Thermal Engineering, 25 (17-18): DEC 2754-2763
Badescu, V.; Sicre, B. (2003). In: Renewable energy for passive house heating II. Model, In: Energy
and Buildings, 35 (11): DEC 1085-1096
Badescu, V.; Sicre, B. (2003). Renewable energy for passive house heating Part I. Building
description, In: Energy and Buildings, 35 (11): DEC 1077-1084
Energy Effciency 120

Bergsten, B. (2001) Energiberäkningsprogram för byggnader – en jämförelse utifrån funktions-
och användaraspekter, Effektivrapport.
Bergström, G.; Boréus, K. (2005). Textens mening och makt: metodbok i samhällsvetenskaplig
text- och diskursanalys. (2., [omarb.] uppl.) Studentlitteratur. ISBN: 91-44-04274-4, Lund
Chandel, S.; Aggarwal, R. (2008) Performance evaluation of a passive solar building in Western
Himalayas, In: Renewable energy, 33 (10): OCT 2166-2173
Chlela, F.; Husaunndee, A .; Inard, C.; RiedeFer, P. (2009) A new methodology for the design of
low energy buildings, In: Energy and Buildings, 41 (9): SEP 982-990
Chwieduk, D. (1999). Prospects for low energy buildings in Poland, In: Renewable energy, 16 (1-4):
JAN-APR 1196-1199
Clarke, J.; Grant, A.; Johnstone, C.; Macdonald, I. (1998). Integrated modelling of low energy
buildings, Renewable energy, 15 (1-4): SEP-DEC 151-156
Cooper, J.S.; Fava, J. (2006). Life Cycle Assessment Practitioner Survey: Summary of Results, In:
Journal of Industrial Ecology, 10(4) 12 -14
Crawley, D. Hand, J. Kummert, M. Griffith, B. (2005) Contrasting the capabilities of building
energy performance simulation programs. In proceedings of international IBPSA conference
8, Montreal, Canada, 231-238.
Dallaire, G. (1980). Zero-energy house: bold, low-cost breakthrough that may revolutionize
housing, In: Civil Engineering –ASCE, 52, 47-59
Dinçer, I.; Rosen, M.A. (2007). Exergy: energy, environment and sustainable development. (1. ed.),
Elsevier, ISBN: 978-0-08-044529-8, Oxford
Fairclough, N. (2003). Analysing discourse: textual analysis for social research. Routledge, ISBN:
0-415-25893-6, New York
Feist, W.; Schnieders, J. (2009). Energy efficiency - a key to sustainable housing, In: European
Physical Journal – Special Topics, 176: SEP 141-153
Feist, W.; Schnieders, J.; Dorer, V.; Haas, A. (2005) Re-inventing air heating: Convenient and
comfortable within the frame of the Passive House concept, In: Energy and Buildings, 37
(11): NOV 1186-1203
Filippin, C.; Beascochea, A.; Esteves, A.; De Rosa, C.; Cortegoso, L.; Estelrich, D. (1998) A passive
solar building for ecological research in Argentina: The first two years experience, In:
Solar Energy, 63 (2): AUG 105-115
Frosch, RA.; Nicholas EG. (1989). Strategies for Manufacturing, In: Scientific American 261(3):
144-152
Guinée, JB. (red.) (2002). Handbook on life cycle assessment: operational guide to the ISO
standards, Kluwer, ISBN: 1-4020-0228-9, Dordrecht
Hamza, N.; Greenwood, D. (2009) Energy conservation regulations: Impacts on design and
procurement of low energy buildings, In: Building and Environment, , 44 (5): MAY 929-
936
Hamza, N.; Horne, M. (2007). Educating the designer: An operational model for visualizing low-
energy architecture, In: Building and Environment, 42 (11): NOV 3841-3847
Heim, D. (2010) Isothermal storage of solar energy in building construction, In: Renewable energy,
35 (4): APR 788-796
Holm, D. (1996). The status of low energy architecture in South Africa, In: Renewable energy, 8 (1-
4): MAY-AUG 301-304
Holmes, M.; Hacker, J. (2007). Climate change, thermal comfort and energy: Meeting the design
challenges of the 21st century, In: Energy and Buildings, 39 (7): 802-814

International Organization for Standardization (ISO) (1997). Environmental management - Life cycle
assessment - Principles and framework. Geneva: ISO
IPCC (ed. Terry Baker) (2007). Climate Change 2007: Synthesis Report,
http://www.ipcc.ch/ipccreports/ar4-syr.htm (April 21 2009).
Isaksson, C.; Karlsson, F. (2006). Indoor climate in low-energy houses - an interdisciplinary
investigation, In: Building and Environment, 41 (12): DEC 1678-1690
Kalogirou, SA.; Bojic, M. (2000). Artificial neural networks for the prediction of the energy
consumption of a passive solar building, In: Energy, 25 (5): MAY 479-491
Kalz, D.; Pfafferott, J.; Herkel, S. (2010). Building signatures: A holistic approach to the
evaluation of heating and cooling concepts, In: Building and Environment, 45 (3): MAR
632-646
Karlsson, F.; Rohdin, P.; Persson, M-L. (2007). Measured and predicted energy demand of a low
energy building: Important aspects when using building energy simulation, In: Building
Services Engineering Research and Technology, 28 (3): 223-235
Karlsson, J.; Moshfegh, B. (2006) Energy demand and indoor climate in a low energy building-
changed control strategies and boundary conditions, In: Energy and Buildings, 38 (4):
APR 315-326
Krishan, A.; Jain, K.; Tewari, P. (1996). Indigenous architecture of two Indian deserts and modern
climatic responsive solutions, In: Renewable energy, 8 (1-4): MAY-AUG 272-277
Krosnick, JA. (1999). Survey Research, In: Annual Review of Psychology, 50: 537-567
Kvale, S (1996). Interviews An Introduction to Qualitative Research Interviewing, Sage Publications,
Thousand Oaks.
Kvale, S.; Brinkmann, S. (2009). InterViews: learning the craft of qualitative research interviewing. (2nd
ed.) Sage Publications, ISBN: 978-0-7619-2542-2, Los Angeles
Körner, S.; Ek, L.; Berg, S. (1984). Deskriptiv statistik. (2. ed.) Studentlitt., ISBN: 91-44-15392-9,
Lund.
Lincoln, Y.S.; Guba, EG. (1985). Naturalistic inquiry. Beverly Hills, Calif.: Sage.
Liu, S.; Henze, G. (2006a). Experimental analysis of simulated reinforcement learning control for
active and passive building thermal storage inventory Part 1. Theoretical foundation, In:
Energy and Buildings, 38 (2): FEB 142-147
Liu, S.; Henze, G. (2006b). Experimental analysis of simulated reinforcement learning control for
active and passive building thermal storage inventory Part 2: Results and analysis, In:
Energy and Buildings, 38 (2): FEB 148-161
Lomas, K. (1996). The UK Applicability Study: An evaluation of thermal simulation programs for
passive solar house design, In: Building and Environment, 31 (3): MAY 197-206
Maier, T.; Krzaczek, M.; Tejchman, J. (2009). Comparison of physical performances of the
ventilation systems in low-energy residential houses, In: Energy and Buildings, 41 (3):
MAR 337-353
Makaka, G Meyer, E McPherson, M, Thermal behaviour and ventilation efficiency of a low-cost
passive solar energy efficient house, In: Renewable energy, 33 (9): 1959-1973 SEP 2008
Moses, LE. (1986). Think and explain with statistics. Addison-Wesley ISBN 0-201-15619-9, Reading,
Mass.
Nicoletti, M. (1998). Architectural expression and low energy design, In: Renewable energy, 15 (1-
4): SEP-DEC 32-41
Nieminen, J. (1994). Low-energy residential housing, In: Energy and Buildings, 21 (3): 187-197
Low-energy buildings – scientifc trends and developments 121

Bergsten, B. (2001) Energiberäkningsprogram för byggnader – en jämförelse utifrån funktions-
och användaraspekter, Effektivrapport.
Bergström, G.; Boréus, K. (2005). Textens mening och makt: metodbok i samhällsvetenskaplig
text- och diskursanalys. (2., [omarb.] uppl.) Studentlitteratur. ISBN: 91-44-04274-4, Lund
Chandel, S.; Aggarwal, R. (2008) Performance evaluation of a passive solar building in Western
Himalayas, In: Renewable energy, 33 (10): OCT 2166-2173
Chlela, F.; Husaunndee, A .; Inard, C.; RiedeFer, P. (2009) A new methodology for the design of
low energy buildings, In: Energy and Buildings, 41 (9): SEP 982-990
Chwieduk, D. (1999). Prospects for low energy buildings in Poland, In: Renewable energy, 16 (1-4):
JAN-APR 1196-1199
Clarke, J.; Grant, A.; Johnstone, C.; Macdonald, I. (1998). Integrated modelling of low energy
buildings, Renewable energy, 15 (1-4): SEP-DEC 151-156
Cooper, J.S.; Fava, J. (2006). Life Cycle Assessment Practitioner Survey: Summary of Results, In:
Journal of Industrial Ecology, 10(4) 12 -14
Crawley, D. Hand, J. Kummert, M. Griffith, B. (2005) Contrasting the capabilities of building
energy performance simulation programs. In proceedings of international IBPSA conference
8, Montreal, Canada, 231-238.
Dallaire, G. (1980). Zero-energy house: bold, low-cost breakthrough that may revolutionize
housing, In: Civil Engineering –ASCE, 52, 47-59
Dinçer, I.; Rosen, M.A. (2007). Exergy: energy, environment and sustainable development. (1. ed.),
Elsevier, ISBN: 978-0-08-044529-8, Oxford
Fairclough, N. (2003). Analysing discourse: textual analysis for social research. Routledge, ISBN:
0-415-25893-6, New York
Feist, W.; Schnieders, J. (2009). Energy efficiency - a key to sustainable housing, In: European
Physical Journal – Special Topics, 176: SEP 141-153
Feist, W.; Schnieders, J.; Dorer, V.; Haas, A. (2005) Re-inventing air heating: Convenient and
comfortable within the frame of the Passive House concept, In: Energy and Buildings, 37
(11): NOV 1186-1203
Filippin, C.; Beascochea, A.; Esteves, A.; De Rosa, C.; Cortegoso, L.; Estelrich, D. (1998) A passive
solar building for ecological research in Argentina: The first two years experience, In:
Solar Energy, 63 (2): AUG 105-115
Frosch, RA.; Nicholas EG. (1989). Strategies for Manufacturing, In: Scientific American 261(3):
144-152
Guinée, JB. (red.) (2002). Handbook on life cycle assessment: operational guide to the ISO
standards, Kluwer, ISBN: 1-4020-0228-9, Dordrecht
Hamza, N.; Greenwood, D. (2009) Energy conservation regulations: Impacts on design and
procurement of low energy buildings, In: Building and Environment, , 44 (5): MAY 929-
936
Hamza, N.; Horne, M. (2007). Educating the designer: An operational model for visualizing low-
energy architecture, In: Building and Environment, 42 (11): NOV 3841-3847
Heim, D. (2010) Isothermal storage of solar energy in building construction, In: Renewable energy,
35 (4): APR 788-796
Holm, D. (1996). The status of low energy architecture in South Africa, In: Renewable energy, 8 (1-
4): MAY-AUG 301-304
Holmes, M.; Hacker, J. (2007). Climate change, thermal comfort and energy: Meeting the design
challenges of the 21st century, In: Energy and Buildings, 39 (7): 802-814

International Organization for Standardization (ISO) (1997). Environmental management - Life cycle
assessment - Principles and framework. Geneva: ISO
IPCC (ed. Terry Baker) (2007). Climate Change 2007: Synthesis Report,
http://www.ipcc.ch/ipccreports/ar4-syr.htm (April 21 2009).
Isaksson, C.; Karlsson, F. (2006). Indoor climate in low-energy houses - an interdisciplinary
investigation, In: Building and Environment, 41 (12): DEC 1678-1690
Kalogirou, SA.; Bojic, M. (2000). Artificial neural networks for the prediction of the energy
consumption of a passive solar building, In: Energy, 25 (5): MAY 479-491
Kalz, D.; Pfafferott, J.; Herkel, S. (2010). Building signatures: A holistic approach to the
evaluation of heating and cooling concepts, In: Building and Environment, 45 (3): MAR
632-646
Karlsson, F.; Rohdin, P.; Persson, M-L. (2007). Measured and predicted energy demand of a low
energy building: Important aspects when using building energy simulation, In: Building
Services Engineering Research and Technology, 28 (3): 223-235
Karlsson, J.; Moshfegh, B. (2006) Energy demand and indoor climate in a low energy building-
changed control strategies and boundary conditions, In: Energy and Buildings, 38 (4):
APR 315-326
Krishan, A.; Jain, K.; Tewari, P. (1996). Indigenous architecture of two Indian deserts and modern
climatic responsive solutions, In: Renewable energy, 8 (1-4): MAY-AUG 272-277
Krosnick, JA. (1999). Survey Research, In: Annual Review of Psychology, 50: 537-567
Kvale, S (1996). Interviews An Introduction to Qualitative Research Interviewing, Sage Publications,
Thousand Oaks.
Kvale, S.; Brinkmann, S. (2009). InterViews: learning the craft of qualitative research interviewing. (2nd
ed.) Sage Publications, ISBN: 978-0-7619-2542-2, Los Angeles
Körner, S.; Ek, L.; Berg, S. (1984). Deskriptiv statistik. (2. ed.) Studentlitt., ISBN: 91-44-15392-9,
Lund.
Lincoln, Y.S.; Guba, EG. (1985). Naturalistic inquiry. Beverly Hills, Calif.: Sage.
Liu, S.; Henze, G. (2006a). Experimental analysis of simulated reinforcement learning control for
active and passive building thermal storage inventory Part 1. Theoretical foundation, In:
Energy and Buildings, 38 (2): FEB 142-147
Liu, S.; Henze, G. (2006b). Experimental analysis of simulated reinforcement learning control for
active and passive building thermal storage inventory Part 2: Results and analysis, In:
Energy and Buildings, 38 (2): FEB 148-161
Lomas, K. (1996). The UK Applicability Study: An evaluation of thermal simulation programs for
passive solar house design, In: Building and Environment, 31 (3): MAY 197-206
Maier, T.; Krzaczek, M.; Tejchman, J. (2009). Comparison of physical performances of the
ventilation systems in low-energy residential houses, In: Energy and Buildings, 41 (3):
MAR 337-353
Makaka, G Meyer, E McPherson, M, Thermal behaviour and ventilation efficiency of a low-cost
passive solar energy efficient house, In: Renewable energy, 33 (9): 1959-1973 SEP 2008
Moses, LE. (1986). Think and explain with statistics. Addison-Wesley ISBN 0-201-15619-9, Reading,
Mass.
Nicoletti, M. (1998). Architectural expression and low energy design, In: Renewable energy, 15 (1-
4): SEP-DEC 32-41
Nieminen, J. (1994). Low-energy residential housing, In: Energy and Buildings, 21 (3): 187-197
Energy Effciency 122

Odum, H.T. (2007). Environment, power and society for the twenty-first century: the hierarchy of energy.
(New ed.) University Press. ISBN: 978-0-231-12886-5 New York: Columbia
Onishi, J.; Soeda, H.; Mizuno, P. (2001). Numerical study on a low energy architecture based
upon distributed heat storage system, In: Renewable energy, 22 (1-3): JAN-MAR 61-66
Osgood, CE.; Suci, GJ.; Tannenbaum, P. (1957). The Measurement of Meaning, University of Illinois
Press, Urbana, IL
Persson, M-L.; Roos A.; Wall, M. (2006). Influence of window size on the energy balance of low
energy houses, In: Energy and Buildings, 38 (3): MAR 181-188
Pfafferott, JÜ.; Herkel, S.;. Kalz, DE.; Zeuschner, A. (2007) Comparison of low-energy office
buildings in summer using different thermal comfort criteria, In: Energy and Buildings,
39 pp. 750-757
Rabah, K. (2005). Development of energy-efficient passive solar building design in Nicosia
Cyprus, In: Renewable energy, 30 (6): MAY 937-956
Sartori, I.; Hestnes, A. (2007) Energy use in the life cycle of conventional and low-energy
buildings: A review article, In: Energy and Buildings, 39 (3): MAR 249-257
Schnieders, E.; Hermelink, A. (2006). CEPHEUS results: measurements and occupants'
satisfaction provide evidence for Passive Houses being an option for sustainable
building, In: Energy Policy, 34 (2): JAN 151-171
Shaviv, E.; Yezioro, A.; Capeluto, IG.; Peleg, UJ.; Kalay, YE. (1996). Simulations and knowledge-
based computer-aided architectural design (CAAD) systems for passive and low energy
architecture, In: Energy and Buildings, 23 (3): MAR 257-269
Starr, G.; Neubauer, L.; Melzer, B. (1980). Temperature Constrol by Passive Solar House Design
in Calefonia, In: Transactions of the ASAE, , 23 (2): 449-456
Thomas, J.; Algohary, S.; Hammad, F.; Soboyejo, W. (2006). Materials selection for thermal
comfort in passive solar buildings, In: Journal of Materials Science, 41 (21): NOV 6897-6907
Thomsen, K.; Schultz, J.; Poel, B. (2005). Measured performance of 12 demonstration projects -
IEA Task 13 "advanced solar low energy buildings", In: Energy and Buildings, 37(2):
FEB 111-119
Thormark, C. (2006). The effect of material choice on the total energy need and recycling potential
of a building, In: Building and Environment, 41 (8): AUG 1019-1026
Tombazis, A.; Preuss, S. (2001). Design of passive solar buildings in urban areas, In: Solar Energy,
70 (3): 311-318
Tommerup, H.; Rose, J.; Svendsen, S. (2007). Energy-efficient houses built according to the
energy performance requirements introduced in Denmark in 2006, In: Energy and
Buildings, 39 (10): OCT 1123-1130
Verbeeck, G.; Hens, H. (2010). Life cycle inventory of buildings: A contribution analysis, In:
Building and Environment, 45 (4): APR 964-967
Versteeg, H. K. Malalasekera, W. (1995) An introduction to computational fluid dynamics: the finite
volume method. Harlow: Longman Scientific & Technical.
Visser, PS.; Krosnick, JA.; Marquette, J.; Curtin M. (1996). Mail surveys for election forecasting?
An evaluation of the Columbus Dispatch poll. In: Public Opin. Q. 60:181.227
Wagner, A.; Gossauer, E.; Moosmann, C.;. Gropp, Th.; Leonhart, R. (2007) Thermal comfort and
workplace occupant satisfaction—Results of field studies in German low energy office
buildings, In: Energy and Buildings , 39, pp. 758-769
Wall, M. (2006). Energy-efficient terrace houses in Sweden - Simulations and measurements, In:
Energy and Buildings, 38 (6): JUN 627-634

Wang, L.; Gwilliam, J.; Jones, P. (2009) Case study of zero energy house design in UK, In: Energy
and Buildings, 41 (11): NOV 1215-1222
Wojdyga, K. (2009). An investigation into the heat consumption in a low-energy building, In:
Renewable energy, 34 (12): DEC 2935-2939
Zhu, L.; Hurt, R.; Correa, D.; Boehm, R. (2009a). Comprehensive energy and economic analyses
on a zero energy house versus a conventional house, In: Energy, 34 (9): SEP 1043-1053
Zhu, L.; Hurt, R.; Correia, D.; Boehm, R. (2009b). Detailed energy saving performance analyses
on thermal mass walls demonstrated in a zero energy house, In: Energy and Buildings, 41
(3): MAR 303-310
Zimmermann, M.; Althaus, H.; Haas, A. (2005). Benchmarks for sustainable construction - A
contribution to develop a standard, In: Energy and Buildings, 37 (11): NOV 1147-1157
Low-energy buildings – scientifc trends and developments 123

Odum, H.T. (2007). Environment, power and society for the twenty-first century: the hierarchy of energy.
(New ed.) University Press. ISBN: 978-0-231-12886-5 New York: Columbia
Onishi, J.; Soeda, H.; Mizuno, P. (2001). Numerical study on a low energy architecture based
upon distributed heat storage system, In: Renewable energy, 22 (1-3): JAN-MAR 61-66
Osgood, CE.; Suci, GJ.; Tannenbaum, P. (1957). The Measurement of Meaning, University of Illinois
Press, Urbana, IL
Persson, M-L.; Roos A.; Wall, M. (2006). Influence of window size on the energy balance of low
energy houses, In: Energy and Buildings, 38 (3): MAR 181-188
Pfafferott, JÜ.; Herkel, S.;. Kalz, DE.; Zeuschner, A. (2007) Comparison of low-energy office
buildings in summer using different thermal comfort criteria, In: Energy and Buildings,
39 pp. 750-757
Rabah, K. (2005). Development of energy-efficient passive solar building design in Nicosia
Cyprus, In: Renewable energy, 30 (6): MAY 937-956
Sartori, I.; Hestnes, A. (2007) Energy use in the life cycle of conventional and low-energy
buildings: A review article, In: Energy and Buildings, 39 (3): MAR 249-257
Schnieders, E.; Hermelink, A. (2006). CEPHEUS results: measurements and occupants'
satisfaction provide evidence for Passive Houses being an option for sustainable
building, In: Energy Policy, 34 (2): JAN 151-171
Shaviv, E.; Yezioro, A.; Capeluto, IG.; Peleg, UJ.; Kalay, YE. (1996). Simulations and knowledge-
based computer-aided architectural design (CAAD) systems for passive and low energy
architecture, In: Energy and Buildings, 23 (3): MAR 257-269
Starr, G.; Neubauer, L.; Melzer, B. (1980). Temperature Constrol by Passive Solar House Design
in Calefonia, In: Transactions of the ASAE, , 23 (2): 449-456
Thomas, J.; Algohary, S.; Hammad, F.; Soboyejo, W. (2006). Materials selection for thermal
comfort in passive solar buildings, In: Journal of Materials Science, 41 (21): NOV 6897-6907
Thomsen, K.; Schultz, J.; Poel, B. (2005). Measured performance of 12 demonstration projects -
IEA Task 13 "advanced solar low energy buildings", In: Energy and Buildings, 37(2):
FEB 111-119
Thormark, C. (2006). The effect of material choice on the total energy need and recycling potential
of a building, In: Building and Environment, 41 (8): AUG 1019-1026
Tombazis, A.; Preuss, S. (2001). Design of passive solar buildings in urban areas, In: Solar Energy,
70 (3): 311-318
Tommerup, H.; Rose, J.; Svendsen, S. (2007). Energy-efficient houses built according to the
energy performance requirements introduced in Denmark in 2006, In: Energy and
Buildings, 39 (10): OCT 1123-1130
Verbeeck, G.; Hens, H. (2010). Life cycle inventory of buildings: A contribution analysis, In:
Building and Environment, 45 (4): APR 964-967
Versteeg, H. K. Malalasekera, W. (1995) An introduction to computational fluid dynamics: the finite
volume method. Harlow: Longman Scientific & Technical.
Visser, PS.; Krosnick, JA.; Marquette, J.; Curtin M. (1996). Mail surveys for election forecasting?
An evaluation of the Columbus Dispatch poll. In: Public Opin. Q. 60:181.227
Wagner, A.; Gossauer, E.; Moosmann, C.;. Gropp, Th.; Leonhart, R. (2007) Thermal comfort and
workplace occupant satisfaction—Results of field studies in German low energy office
buildings, In: Energy and Buildings , 39, pp. 758-769
Wall, M. (2006). Energy-efficient terrace houses in Sweden - Simulations and measurements, In:
Energy and Buildings, 38 (6): JUN 627-634

Wang, L.; Gwilliam, J.; Jones, P. (2009) Case study of zero energy house design in UK, In: Energy
and Buildings, 41 (11): NOV 1215-1222
Wojdyga, K. (2009). An investigation into the heat consumption in a low-energy building, In:
Renewable energy, 34 (12): DEC 2935-2939
Zhu, L.; Hurt, R.; Correa, D.; Boehm, R. (2009a). Comprehensive energy and economic analyses
on a zero energy house versus a conventional house, In: Energy, 34 (9): SEP 1043-1053
Zhu, L.; Hurt, R.; Correia, D.; Boehm, R. (2009b). Detailed energy saving performance analyses
on thermal mass walls demonstrated in a zero energy house, In: Energy and Buildings, 41
(3): MAR 303-310
Zimmermann, M.; Althaus, H.; Haas, A. (2005). Benchmarks for sustainable construction - A
contribution to develop a standard, In: Energy and Buildings, 37 (11): NOV 1147-1157
Energy Effciency 124
Energy transformed: building capacity in the engineering profession in australia 125
Energy transformed: building capacity in the engineering profession in
australia
Cheryl Desha and Karlson ‘Charlie’ Hargroves
x

Energy transformed: building capacity in the
engineering profession in Australia

Cheryl Desha and Karlson ‘Charlie’ Hargroves
Griffith University, Curtin University
Australia

1. Introduction
Global pressures of burgeoning population growth and consumption are threatening efforts
to reduce negative environmental pressures associated with development such as
atmospheric, land and water pollution. For example, the world’s population is now growing
at over 70 million per year or 1 billion per decade (Brown, 2007), increasing from 3.5 billion
in 1970, to 5 billion in 1990, to 7 billion by 2010 (United Nations, 2002). In 1990 only 13
percent of the global population lived in cities, while in 2007 more than half did.

More than
60 percent of the global population lives within 100 kilometers of the coastline (World
Resources Institute, 2005) and nearly all of the population growth hereon is forecast to
happen in developing countries (Postel, 1999). Future levels of stress on the global
environment are therefore likely to increase if current trends are used for forecasting, which
is particularly challenging as scientists are already observing significant signs of
degradation and failure in environmental systems. For example, the Intergovernmental
Panel on Climate Change Fourth Assessment Report (IPCC, 2007) provided an unequivocal
link between climate change and current human activities, in particular: the burning of fossil
fuels; deforestation and land clearing; the use of synthetic greenhouse gases; and
decomposition of wastes from landfill. The UK Stern Review concluded that within our
lifetime there is between a 77 to 99 percent chance (depending on the climate model used) of
the global average temperature rising by more than 2 degrees Celsius (Stern, 2006), with a
likely greenhouse gas concentration in the atmosphere of 550 parts per million (ppm) or
more by around 2100.
Hence, the way in which the human race deals with energy over the next 30 years will
determine the quality of life for generations to come. This includes increasing the level of
non-fossil fuel energy generation, improving the efficiency with which energy is supplied
(i.e. reducing supply losses, for example from the power station to the end user), and
increasing the efficiency of the end user (i.e. reducing demand, for example improving the
efficiency of machinery and appliances to perform, for example heating, cooling, and
washing). However, the reality is that there are significant skills shortages – particularly
within the engineering profession – to address these issues. As highlighted in a United
Nations Environment Program report on working in a low-carbon world, ‘… companies in the
fledgling green economy are struggling to find workers with the skills needed to perform the work
that needs to be done. Indeed, there are signs that shortages of skilled labor could put the brakes on
7
Energy Effciency 126

green expansion…There is thus a need to put appropriate education and training arrangements in
place’ (United Nations Environment Programme, 2008).
In Australia for example, considering energy efficiency, according to a national study,
‘Given the wide range of technical issues associated with energy efficiency, gaps in the skill
sets of specialists such as engineers or trades people could prevent the uptake of these
options across a range of sectors’ (Garnaut, 2008). Modelling by the CSIRO has shown that 3
million Australians (of a population of just over 20 million), will need training or re-training
in energy efficiency, green building technologies, sustainable energy and more sustainable
agricultural systems to enable Australia to achieve the IPCC’s recommended targets for
greenhouse gas reductions (Hatfield-Dodds et al., 2008). Furthermore, surveys are
highlighting that the state of knowledge, understanding and implementation of even basic
environmental and energy management systems in the business sector is poor. For instance,
a 2008 survey of 300 Australian business CEOs regarding operating in a carbon-constrained
economy found that two-thirds (67 percent) of businesses were concerned or unsure about
compliance obligations, and only a handful of businesses (less than 3 percent) had
implemented a strategic response to climate change (Price Waterhouse Coopers, 2008). A
2007 survey of the Australian mining and metals sector also highlighted an alarmingly slow
adoption of energy demand management practices, with nearly half (43 percent) of
companies still not having implemented an official energy policy. In the same context, only
10 percent of companies responding to a 2007 national Australian Industry Group survey on
climate change practices felt informed enough to manage the risks associated with climate-
related impacts (Australian Industry Group, 2007). Australia’s peak engineering
professional body, the Institution of Engineers Australia, has also acknowledged that, ‘The
need to make changes in the way energy is used and supplied throughout the world
represents the greatest challenge to engineers in moving toward sustainability’ (Institution
of Engineers Australia, undated).
Within this context, this chapter overviews the need to transform engineering education, to
deliver graduates and capacity build professionals who can address such energy supply and
demand challenges, highlighting the complexity of transforming such education systems.
We begin by discussing the problem of a time lag dilemma facing education worldwide,
whereby the timeframe for capacity building the profession is converging with the
requirement for global action. We also briefly discuss related risks and benefits facing
organizations in light of this dilemma. The chapter then focuses on requirements for
capacity building the engineering profession, drawing on two research initiatives that we
have previously led: a 2007 Australian survey of the state of energy efficiency education in
engineering education, and a 2009 investigation into increasing the extent of energy
efficiency content in curriculum. Finally, we discuss a peer reviewed, online and freely
accessible resource that has been developed from this increased understanding, to assist
with capacity building, focusing on sustainable energy solutions for climate change
mitigation.

2. A time lag facing engineering education
With considerable advances over the last century, the effectiveness of the education system
to deliver skilled professionals would appear to be self-evident. Yet signals now clearly
suggest that the focus of much of higher education requires a significant update, with an

emerging time related imperative facing the engineering community. Despite an absence of
discussion in the literature, anecdotal evidence from discussions with engineering educators
suggests that a typical (or ‘standard’) process of curriculum renewal in the higher education
sector may take 3-4 accreditation cycles (of approximately 5-year intervals) for engineering
departments to fully integrate a substantial new set of knowledge and skills within all year
levels of a degree as required; i.e. between 15-20 years.
Given that the average pathway to graduate from an engineering and built environment
program is approximately 3-5 years, from enrolment to graduation, followed by 3-5 years of
on-the-job graduate development, if HEIs take the typical approach over a 15-20 year period
to fully renew such bachelor programs, this has the potential to result in a time lag of
around 21-29 years – 2-3 decades – before students graduating from fully integrated
programs will be in decision-making positions. Clearly this is well beyond the timeframes
needed to address immediate climate change issues. For postgraduate students the time lag
may be shorter as students may already be practising in their field and studies span just 1-2
years. However the time lag may still be in the order of 5-10 years depending on the
curriculum renewal process, which still potentially results in a lack of capacity in the
professional sector over the next decade to address urgent climate change and sustainable
development issues.
Along with understanding that current education systems are yet to be prepared to rapidly
develop knowledge and skills related to reducing environmental pressures, it is important
to understand that it is logistically impossible for the education system to change
‘overnight’, as programs need to balance the current student demands and expectations
with industry expectations for graduate attributes. Figure 1 highlights how the transition
might occur, with a period of rapid curriculum renewal followed by continual program
improvement which follows a regular improvement cycle of research, curriculum
development, trial, evaluation and review.

Fig. 1. An illustrative scenario for integrating energy efficiency knowledge and skills to
match industry requirements over time
2000 2020 2040 2010 2030
20
40
60
80
100
Year
Curriculum for “Old Industry”
Curriculum for “New Industry”
Period of
rapid
curriculum
renewal
I
n
t
e
g
r
a
t
i
o
n

o
f

E
n
e
r
g
y

E
f
f
i
c
i
e
n
c
y

C
o
n
t
e
n
t

(
%
)

Energy transformed: building capacity in the engineering profession in australia 127

green expansion…There is thus a need to put appropriate education and training arrangements in
place’ (United Nations Environment Programme, 2008).
In Australia for example, considering energy efficiency, according to a national study,
‘Given the wide range of technical issues associated with energy efficiency, gaps in the skill
sets of specialists such as engineers or trades people could prevent the uptake of these
options across a range of sectors’ (Garnaut, 2008). Modelling by the CSIRO has shown that 3
million Australians (of a population of just over 20 million), will need training or re-training
in energy efficiency, green building technologies, sustainable energy and more sustainable
agricultural systems to enable Australia to achieve the IPCC’s recommended targets for
greenhouse gas reductions (Hatfield-Dodds et al., 2008). Furthermore, surveys are
highlighting that the state of knowledge, understanding and implementation of even basic
environmental and energy management systems in the business sector is poor. For instance,
a 2008 survey of 300 Australian business CEOs regarding operating in a carbon-constrained
economy found that two-thirds (67 percent) of businesses were concerned or unsure about
compliance obligations, and only a handful of businesses (less than 3 percent) had
implemented a strategic response to climate change (Price Waterhouse Coopers, 2008). A
2007 survey of the Australian mining and metals sector also highlighted an alarmingly slow
adoption of energy demand management practices, with nearly half (43 percent) of
companies still not having implemented an official energy policy. In the same context, only
10 percent of companies responding to a 2007 national Australian Industry Group survey on
climate change practices felt informed enough to manage the risks associated with climate-
related impacts (Australian Industry Group, 2007). Australia’s peak engineering
professional body, the Institution of Engineers Australia, has also acknowledged that, ‘The
need to make changes in the way energy is used and supplied throughout the world
represents the greatest challenge to engineers in moving toward sustainability’ (Institution
of Engineers Australia, undated).
Within this context, this chapter overviews the need to transform engineering education, to
deliver graduates and capacity build professionals who can address such energy supply and
demand challenges, highlighting the complexity of transforming such education systems.
We begin by discussing the problem of a time lag dilemma facing education worldwide,
whereby the timeframe for capacity building the profession is converging with the
requirement for global action. We also briefly discuss related risks and benefits facing
organizations in light of this dilemma. The chapter then focuses on requirements for
capacity building the engineering profession, drawing on two research initiatives that we
have previously led: a 2007 Australian survey of the state of energy efficiency education in
engineering education, and a 2009 investigation into increasing the extent of energy
efficiency content in curriculum. Finally, we discuss a peer reviewed, online and freely
accessible resource that has been developed from this increased understanding, to assist
with capacity building, focusing on sustainable energy solutions for climate change
mitigation.

2. A time lag facing engineering education
With considerable advances over the last century, the effectiveness of the education system
to deliver skilled professionals would appear to be self-evident. Yet signals now clearly
suggest that the focus of much of higher education requires a significant update, with an

emerging time related imperative facing the engineering community. Despite an absence of
discussion in the literature, anecdotal evidence from discussions with engineering educators
suggests that a typical (or ‘standard’) process of curriculum renewal in the higher education
sector may take 3-4 accreditation cycles (of approximately 5-year intervals) for engineering
departments to fully integrate a substantial new set of knowledge and skills within all year
levels of a degree as required; i.e. between 15-20 years.
Given that the average pathway to graduate from an engineering and built environment
program is approximately 3-5 years, from enrolment to graduation, followed by 3-5 years of
on-the-job graduate development, if HEIs take the typical approach over a 15-20 year period
to fully renew such bachelor programs, this has the potential to result in a time lag of
around 21-29 years – 2-3 decades – before students graduating from fully integrated
programs will be in decision-making positions. Clearly this is well beyond the timeframes
needed to address immediate climate change issues. For postgraduate students the time lag
may be shorter as students may already be practising in their field and studies span just 1-2
years. However the time lag may still be in the order of 5-10 years depending on the
curriculum renewal process, which still potentially results in a lack of capacity in the
professional sector over the next decade to address urgent climate change and sustainable
development issues.
Along with understanding that current education systems are yet to be prepared to rapidly
develop knowledge and skills related to reducing environmental pressures, it is important
to understand that it is logistically impossible for the education system to change
‘overnight’, as programs need to balance the current student demands and expectations
with industry expectations for graduate attributes. Figure 1 highlights how the transition
might occur, with a period of rapid curriculum renewal followed by continual program
improvement which follows a regular improvement cycle of research, curriculum
development, trial, evaluation and review.

Fig. 1. An illustrative scenario for integrating energy efficiency knowledge and skills to
match industry requirements over time
2000 2020 2040 2010 2030
20
40
60
80
100
Year
Curriculum for “Old Industry”
Curriculum for “New Industry”
Period of
rapid
curriculum
renewal
I
n
t
e
g
r
a
t
i
o
n

o
f

E
n
e
r
g
y

E
f
f
i
c
i
e
n
c
y

C
o
n
t
e
n
t

(
%
)

Energy Effciency 128

A key consideration in timing the transition, is the shift in focus from ‘old industry’ to ‘new
industry’ curriculum, matching changing educational needs with the pace of emerging
demand for such graduate attributes by employers. As part of the transition towards more
sustainable infrastructure and societies, ‘old industry’ plant and equipment will require
service and maintenance by professionals with ‘old industry’ knowledge and skills.
However as with any major adjustment such as the information technology revolution, there
needs to be a staged approach, where the balance of ‘old’ and ‘new’ needs to be carefully
managed in relation to the emerging needs of society and employer demands. As the large
amount of embedded infrastructure (for example buildings, power stations, electricity grids
etc) needs to be managed, maintained and transitioned, this requires ‘old industry’
education. Hence the process to integrate ‘new industry’ knowledge and skills needs to be
appropriately staged, as if it is too quick, this could be problematic as graduates may not
have the skills that the employment market needs at the time that they graduate.
Hence, the timeframe for updating undergraduate engineering curriculum using standard
methods may be too long to ensure that engineering professionals will be equipped with
knowledge and skills that can address such immediate 21
st
Century challenges while still
being able to maintain current systems. The extent of the time lag will depend on how
quickly the new knowledge and skills are embedded into engineering curriculum, to the
point where a student can begin studies in first year, and fully develop the new set of
desired knowledge and skills (or ‘graduate attributes’) by the time they graduate.
This observed time lag dilemma facing engineering education has significant implications
for society if the need for curriculum renewal is not addressed. Furthermore, there are
implications for university engineering departments as they make decisions about the scale
and pace of curriculum renewal as regulations and the market continue to change.
Engineering departments may also be exposed to potential risks with regard to both student
demand for the programs, and tightening accreditation requirements. However,
departments need to be wary of keeping pace with graduate demand (i.e. not stepping too
far in front) to ensure that their graduates remain employable and in demand throughout
the process.
Drawing on the literature, Figure 2 presents an illustrative representation of the relationship
between a department’s commitment to engineering education for sustainable development
and potential risk and reward implications. Risks include for example falling student
numbers, increasing accreditation difficulties, poaching of key staff. Rewards include for
example attracting the best students and staff, staying ahead of accreditation requirements,
attracting research funding, securing key academic appointments and industry funding.
For the last 20 years, there has been relatively low risks and benefits from seeking to
accelerate curriculum renewal in this area, evidenced by the relative lack of action on the
whole in the sector apart from a small number of outstanding cases (Desha et al., 2009).
However, recent market, regulatory and institutional shifts around environmental and
sustainable development related issues, together with the significant shift in public opinion
on these matters, and the increasing competition among higher education institutions, have
caused the level of both the risks and the benefits to increase dramatically over the coming
decades.



Fig. 2. A stylistic representation of risk and reward scenarios for curriculum renewal in the
higher education sector
Source: (Desha & Hargroves, 2009a)

This situation presents significant cause for universities and engineering departments to
rethink their strategies related to curriculum reform in order to minimize the risks and
capture the rewards. In short, over the coming years, departments who do not transition
their programs with topic areas such as energy efficiency are likely to find it increasingly
difficult to operate. Furthermore, their traditional roles as providers of education for
engineers may be challenged by private training providers who explore niche business
opportunities in capacity building in these topic areas, along with engineering firms and
government departments developing in-house capacity building programs that assume a
base-line graduate capacity.

3. The state of energy efficiency education
In the face of such a time lag dilemma, the literature suggests that engineering educators
need to undertake rapid curriculum renewal to update what is taught, within 2-3
accreditation cycles in undergraduate programs. Furthermore, rapid curriculum renewal in
postgraduate engineering education also needs to occur; equipping practitioners and
decision-makers with knowledge and skills surround energy efficiency. With this in mind,
we now consider the state of engineering education for energy efficiency, for which a full
account is provided by Desha et al. (2007). We also identify challenges and opportunities for
energy efficiency education within universities, for which a full literature review is available
online (Desha, Hargroves & Reeve, 2009) and a summary is provided by Desha and
Hargroves (2009b).

Energy transformed: building capacity in the engineering profession in australia 129

A key consideration in timing the transition, is the shift in focus from ‘old industry’ to ‘new
industry’ curriculum, matching changing educational needs with the pace of emerging
demand for such graduate attributes by employers. As part of the transition towards more
sustainable infrastructure and societies, ‘old industry’ plant and equipment will require
service and maintenance by professionals with ‘old industry’ knowledge and skills.
However as with any major adjustment such as the information technology revolution, there
needs to be a staged approach, where the balance of ‘old’ and ‘new’ needs to be carefully
managed in relation to the emerging needs of society and employer demands. As the large
amount of embedded infrastructure (for example buildings, power stations, electricity grids
etc) needs to be managed, maintained and transitioned, this requires ‘old industry’
education. Hence the process to integrate ‘new industry’ knowledge and skills needs to be
appropriately staged, as if it is too quick, this could be problematic as graduates may not
have the skills that the employment market needs at the time that they graduate.
Hence, the timeframe for updating undergraduate engineering curriculum using standard
methods may be too long to ensure that engineering professionals will be equipped with
knowledge and skills that can address such immediate 21
st
Century challenges while still
being able to maintain current systems. The extent of the time lag will depend on how
quickly the new knowledge and skills are embedded into engineering curriculum, to the
point where a student can begin studies in first year, and fully develop the new set of
desired knowledge and skills (or ‘graduate attributes’) by the time they graduate.
This observed time lag dilemma facing engineering education has significant implications
for society if the need for curriculum renewal is not addressed. Furthermore, there are
implications for university engineering departments as they make decisions about the scale
and pace of curriculum renewal as regulations and the market continue to change.
Engineering departments may also be exposed to potential risks with regard to both student
demand for the programs, and tightening accreditation requirements. However,
departments need to be wary of keeping pace with graduate demand (i.e. not stepping too
far in front) to ensure that their graduates remain employable and in demand throughout
the process.
Drawing on the literature, Figure 2 presents an illustrative representation of the relationship
between a department’s commitment to engineering education for sustainable development
and potential risk and reward implications. Risks include for example falling student
numbers, increasing accreditation difficulties, poaching of key staff. Rewards include for
example attracting the best students and staff, staying ahead of accreditation requirements,
attracting research funding, securing key academic appointments and industry funding.
For the last 20 years, there has been relatively low risks and benefits from seeking to
accelerate curriculum renewal in this area, evidenced by the relative lack of action on the
whole in the sector apart from a small number of outstanding cases (Desha et al., 2009).
However, recent market, regulatory and institutional shifts around environmental and
sustainable development related issues, together with the significant shift in public opinion
on these matters, and the increasing competition among higher education institutions, have
caused the level of both the risks and the benefits to increase dramatically over the coming
decades.



Fig. 2. A stylistic representation of risk and reward scenarios for curriculum renewal in the
higher education sector
Source: (Desha & Hargroves, 2009a)

This situation presents significant cause for universities and engineering departments to
rethink their strategies related to curriculum reform in order to minimize the risks and
capture the rewards. In short, over the coming years, departments who do not transition
their programs with topic areas such as energy efficiency are likely to find it increasingly
difficult to operate. Furthermore, their traditional roles as providers of education for
engineers may be challenged by private training providers who explore niche business
opportunities in capacity building in these topic areas, along with engineering firms and
government departments developing in-house capacity building programs that assume a
base-line graduate capacity.

3. The state of energy efficiency education
In the face of such a time lag dilemma, the literature suggests that engineering educators
need to undertake rapid curriculum renewal to update what is taught, within 2-3
accreditation cycles in undergraduate programs. Furthermore, rapid curriculum renewal in
postgraduate engineering education also needs to occur; equipping practitioners and
decision-makers with knowledge and skills surround energy efficiency. With this in mind,
we now consider the state of engineering education for energy efficiency, for which a full
account is provided by Desha et al. (2007). We also identify challenges and opportunities for
energy efficiency education within universities, for which a full literature review is available
online (Desha, Hargroves & Reeve, 2009) and a summary is provided by Desha and
Hargroves (2009b).

Energy Effciency 130

3.1 Understanding the state of engineering education for energy efficiency
The sub-topic of energy efficiency is a prime example for a new area of practice that needs to
be rapidly integrated into engineering courses, while also addressing a knowledge gap in a
highly topical content area. However, there is an absence of literature documenting the state
of affairs, to provide a robust platform on which to act. Hence, in 2007 the National
Framework for Energy Efficiency (NFEE) funded researchers from Griffith University (The
Natural Edge Project, TNEP) to undertake the first survey of energy efficiency education
across all Australian universities teaching engineering education, which asked, ‘What is the
state of education for energy efficiency in Australian engineering education?’ (Desha et al., 2007).
The subsequent research project used a paper-based questionnaire which was issued in hard
copy and electronic format to the heads of department of all 32 Australian universities
providing engineering undergraduate and/or post-graduate programs. It included an
invitation to every Dean for completion by every lecturer teaching energy related material
within engineering education. The project also included a student questionnaire, which was
provided to all lecturers who received the lecturer questionnaire, to distribute and collect in
one or more of their classes where energy related material were taught. The results of the
two questionnaires were cross-checked for additional context and validity of interpretation
through semi-structured telephone interviews with a subset of Australian academics who
were experienced in engineering education for energy efficiency.
With excellent participation by 27 of the 32 universities teaching higher education
(comprising 62 lecturers and 261 students), the survey identified that even though energy
efficiency education was highly variable and ad hoc, there were a range of preferred options
for improvement (Desha et al., 2007; Desha & Hargroves, 2009b). In summary, for more than
half of the surveyed courses (55 percent), lecturers reported that their course could include
more (in-depth) energy efficiency content, while most respondents (74 percent) thought that
the increase in content should be in the specific area of applying energy efficiency theory and
knowledge. More than half (52 percent) thought their course could include more
information about energy efficiency opportunities. The survey also showed a clear
preference for resources to be available through open access, online learning modules
(90 percent) as opposed to restricted access sources (6 percent) or intensive short courses
undertaken in person (13 percent) or remotely (10 percent).
While there was clearly a desire to integrate energy efficiency content, the 2007 Australian
survey indicated a substantial shortfall in the inclusion of energy efficiency theory,
knowledge, application and assessment in engineering education on the whole. Even
mainstream contextual topics such as ‘carbon dioxide and other greenhouse gas emissions
from energy generation’ and ‘the link between greenhouse gas emissions and global
temperature change’ were only covered in detail by up to a third of surveyed courses, and
mentioned by less than half. Moreover, student survey results indicated only a low to
moderate appreciation of how energy efficiency might be directly related to their future
careers. Lecturers and students agreed that there was little if any coverage of topics such as
‘product stewardship and responsibility’, ‘decoupling energy utility profits from kilowatt-
hours sold’ or ‘incremental efficiency versus whole system design’. The survey results
indicated that this disconnect – between lecturers recognizing an absence of content, and a
lack of action in integrating the content – was likely to be due to the presence of a variety of
barriers to implementation. For example, nearly two thirds (58 percent) considered the
potential for course content overload to be an issue, while more than half (52 percent)

considered having insufficient time to prepare new materials as a challenge to such
curriculum renewal.
This survey contributes to a growing global understanding of the current state of education
in this sustainability topic. There is clearly an urgent need to embed energy efficiency
knowledge and skills into engineering curriculum, beyond once-off courses, special interest
topics in later years, or highly specialized masters programs. These survey findings are also
immediately relevant for senior management in engineering departments, Australian
professional organizations, and government departments considering future programs and
funding allocations, as they provide an indication of the preferred options for increasing
energy efficiency education.

3.2 Societal drivers promoting and impeding education for sustainable development
Reports such as the Higher Education Funding Council for England’s 2006 report on the
‘Barriers and Challenges to Education for Sustainable Development’ (Levett-Therivel, 2006)
suggest that although actual progress in curriculum renewal has been slow for engineering
education, there is increasing pressure for curriculum renewal towards engineering
education for sustainable development from a range of actors. This includes pressure from
the ‘top down’ (for example from accrediting institutions, professional organizations,
advisory boards, education institutions and government) and from the ‘bottom up’ (for
example from faculty members and students themselves). Table 1 provides a brief
explanation of the drivers that are promoting such education, synthesizing the literature.

Driver Factors promoting engineering education for sustainable development
Market/
Business
– Shifting requirements by potential employers - increasing requirements
for engineers to demonstrate sustainable development capacity.
– Increasing cost of resources and associated taxes/markets – increasing
demand for capacity to reduce water and energy consumption.
– Shifting investment preferences – increasing attraction to engineers
who can reduce energy demand and environmental liabilities.
– Introduction of ‘sustainability’ rankings – increasing pressure to
improve rankings in indexes (e.g. Dow Jones Sustainability Index).
– Market leadership opportunities – increasing pressure to achieve/
maintain leadership position and capture early mover advantages.
– Increasing student demand and market potential - students seeking
sustainable development content within their institutions of study.
Information/
Technology
– Increased scientific understanding – accumulating scientific knowledge
regarding environmental issues, creating pressure for performance
improvement in all sectors.
– New technologies – increasing calls for incorporating a range of new
technologies into designs (e.g. renewable energy options).
– New examples of leadership – emerging examples of leading efforts
across sectors will drive competitors.
– Increasing faculty interest in related research and teaching innovation –
increasing incentives offered by governments and organizations.
– Increasing focus in declarations and conference action plans - creating
benchmarks for new kinds of engineering professionals.
Energy transformed: building capacity in the engineering profession in australia 131

3.1 Understanding the state of engineering education for energy efficiency
The sub-topic of energy efficiency is a prime example for a new area of practice that needs to
be rapidly integrated into engineering courses, while also addressing a knowledge gap in a
highly topical content area. However, there is an absence of literature documenting the state
of affairs, to provide a robust platform on which to act. Hence, in 2007 the National
Framework for Energy Efficiency (NFEE) funded researchers from Griffith University (The
Natural Edge Project, TNEP) to undertake the first survey of energy efficiency education
across all Australian universities teaching engineering education, which asked, ‘What is the
state of education for energy efficiency in Australian engineering education?’ (Desha et al., 2007).
The subsequent research project used a paper-based questionnaire which was issued in hard
copy and electronic format to the heads of department of all 32 Australian universities
providing engineering undergraduate and/or post-graduate programs. It included an
invitation to every Dean for completion by every lecturer teaching energy related material
within engineering education. The project also included a student questionnaire, which was
provided to all lecturers who received the lecturer questionnaire, to distribute and collect in
one or more of their classes where energy related material were taught. The results of the
two questionnaires were cross-checked for additional context and validity of interpretation
through semi-structured telephone interviews with a subset of Australian academics who
were experienced in engineering education for energy efficiency.
With excellent participation by 27 of the 32 universities teaching higher education
(comprising 62 lecturers and 261 students), the survey identified that even though energy
efficiency education was highly variable and ad hoc, there were a range of preferred options
for improvement (Desha et al., 2007; Desha & Hargroves, 2009b). In summary, for more than
half of the surveyed courses (55 percent), lecturers reported that their course could include
more (in-depth) energy efficiency content, while most respondents (74 percent) thought that
the increase in content should be in the specific area of applying energy efficiency theory and
knowledge. More than half (52 percent) thought their course could include more
information about energy efficiency opportunities. The survey also showed a clear
preference for resources to be available through open access, online learning modules
(90 percent) as opposed to restricted access sources (6 percent) or intensive short courses
undertaken in person (13 percent) or remotely (10 percent).
While there was clearly a desire to integrate energy efficiency content, the 2007 Australian
survey indicated a substantial shortfall in the inclusion of energy efficiency theory,
knowledge, application and assessment in engineering education on the whole. Even
mainstream contextual topics such as ‘carbon dioxide and other greenhouse gas emissions
from energy generation’ and ‘the link between greenhouse gas emissions and global
temperature change’ were only covered in detail by up to a third of surveyed courses, and
mentioned by less than half. Moreover, student survey results indicated only a low to
moderate appreciation of how energy efficiency might be directly related to their future
careers. Lecturers and students agreed that there was little if any coverage of topics such as
‘product stewardship and responsibility’, ‘decoupling energy utility profits from kilowatt-
hours sold’ or ‘incremental efficiency versus whole system design’. The survey results
indicated that this disconnect – between lecturers recognizing an absence of content, and a
lack of action in integrating the content – was likely to be due to the presence of a variety of
barriers to implementation. For example, nearly two thirds (58 percent) considered the
potential for course content overload to be an issue, while more than half (52 percent)

considered having insufficient time to prepare new materials as a challenge to such
curriculum renewal.
This survey contributes to a growing global understanding of the current state of education
in this sustainability topic. There is clearly an urgent need to embed energy efficiency
knowledge and skills into engineering curriculum, beyond once-off courses, special interest
topics in later years, or highly specialized masters programs. These survey findings are also
immediately relevant for senior management in engineering departments, Australian
professional organizations, and government departments considering future programs and
funding allocations, as they provide an indication of the preferred options for increasing
energy efficiency education.

3.2 Societal drivers promoting and impeding education for sustainable development
Reports such as the Higher Education Funding Council for England’s 2006 report on the
‘Barriers and Challenges to Education for Sustainable Development’ (Levett-Therivel, 2006)
suggest that although actual progress in curriculum renewal has been slow for engineering
education, there is increasing pressure for curriculum renewal towards engineering
education for sustainable development from a range of actors. This includes pressure from
the ‘top down’ (for example from accrediting institutions, professional organizations,
advisory boards, education institutions and government) and from the ‘bottom up’ (for
example from faculty members and students themselves). Table 1 provides a brief
explanation of the drivers that are promoting such education, synthesizing the literature.

Driver Factors promoting engineering education for sustainable development
Market/
Business
– Shifting requirements by potential employers - increasing requirements
for engineers to demonstrate sustainable development capacity.
– Increasing cost of resources and associated taxes/markets – increasing
demand for capacity to reduce water and energy consumption.
– Shifting investment preferences – increasing attraction to engineers
who can reduce energy demand and environmental liabilities.
– Introduction of ‘sustainability’ rankings – increasing pressure to
improve rankings in indexes (e.g. Dow Jones Sustainability Index).
– Market leadership opportunities – increasing pressure to achieve/
maintain leadership position and capture early mover advantages.
– Increasing student demand and market potential - students seeking
sustainable development content within their institutions of study.
Information/
Technology
– Increased scientific understanding – accumulating scientific knowledge
regarding environmental issues, creating pressure for performance
improvement in all sectors.
– New technologies – increasing calls for incorporating a range of new
technologies into designs (e.g. renewable energy options).
– New examples of leadership – emerging examples of leading efforts
across sectors will drive competitors.
– Increasing faculty interest in related research and teaching innovation –
increasing incentives offered by governments and organizations.
– Increasing focus in declarations and conference action plans - creating
benchmarks for new kinds of engineering professionals.
Energy Effciency 132

Driver Factors promoting engineering education for sustainable development
Institutional/
Civil Society
– Shifting accreditation requirements for graduate engineers - formalising
sustainability knowledge and skill requirements.
– Mandatory disclosure and reporting – increasing disclosure and
reporting requirements (e.g. greenhouse gas emissions).
– Increasing professional advocacy - with leaders stating the pivotal role
of engineering in addressing 21st Century challenges.
– Shifting requirements for practising engineers by professional
organizations - where mission statements, code of ethics statements
and codes of practice are being updated.
– Increasing commitment and action by highly regarded university peers -
increasingly vocal commitments and alliances.
Table 1. Key factors promoting engineering education for sustainable development

A number of key barriers are also evident in the literature, which appear to be limiting
efforts by engineering educators to undertake significant and rapid engineering curriculum
renewal, as summarized in Table 2.

Barrier Factors limiting engineering education for sustainable development
Market/
Business
– Persistent ‘old economy’ industry practices, wherein employers
continue to employ graduates to undertake unsustainable practices.
– Uncertainty around future requirements to change – where varying
government messages create considerable uncertainty around
impending requirements to change.
– Perceived threat to employability and position, from taking action ahead
of market or sector wide requirements to do so.
– Short-termism in the Higher Education Institution (HEI) sector, where
short-term pressures demand increasing staff to student ratios, and
increasing student intake, rather than program innovation.
– A shortage of engineering graduates, resulting in a ‘take what you can
get’ scenario, to then up-skill internally.
Information/
Technology
– Growing disconnect between engineering and science, where engineering
professionals may not be ‘in-step’ in understanding the complexity
and interdisciplinary nature of 21st Century challenges.
– Lack of convenient access to emerging and rigorously reviewed information,
where academics may have difficulty getting information and those
who have good access may be overwhelmed.
– Lack of access to information in foreign languages, which may impede
the integration of emerging technologies and innovations.
Institutional/
Civil Society
– Lack of strong requirements for change, where there is a lack of certainty
about current and future legislative requirements and support.
– Lack of academic staff competencies in EESD, with a relatively low rate
of professional development among educators.
Table 2. Key drivers limiting engineering education for sustainable development


Hence, there exist a number of significant societal drivers promoting curriculum renewal
within engineering education, which are being tempered by a number of barriers that are
limiting the progress. These barriers and others have been strong enough to-date, to prevent
a transition towards engineering education for sustainable development in the majority of
universities around the world. Many engineering departments are doing little more than
including one or two ‘sustainability’ courses within existing programs, leaving isolated
individuals or small teams within departments to undertake ad hoc curriculum renewal
efforts. In reality, most current engineering degrees are still focused on what could broadly
be classified as ‘fossil fuel based old industry’, involving linear ‘heat, beat and treat’
processes that don’t tend to consider rethinking waste, minimizing inputs, maximizing
productivity, capturing synergies or other externalities as part of the process (Benyus, 1997).

3.3 Curriculum drivers promoting and impeding energy efficiency education
Given these observations regarding societal drivers promoting and limiting engineering
education for sustainable development, in 2009 the NFEE funded an investigation into
identify challenges and opportunities for timely curriculum renewal in energy efficiency
education, at the level of the lecturer (Desha & Hargroves, 2009b). Specifically, the project
focused on developing and releasing a strategic document to assist the curriculum renewal
process for energy efficiency education, drawing upon a behavior change methodology
developed by McKenzie-Mohr and Smith (2007). The findings were intended for use by
engineering departments, accreditation agencies, professional bodies and government, to
identify opportunities for moving forward, and then to strategically plan the transition. The
project also provided a significant opportunity to explore options to support lecturers,
program co-ordinators and staff to strategically approach, in an informed way, the challenge
of increasing the levels of education for energy efficiency as a proxy for other sustainable
development topics.
Through a comprehensive literature review followed by a national survey of engineering
educators, the researchers short-listed 10 favored options amongst HEIs to integrate
emerging energy efficiency content within current engineering programs, as shown below
(in order of priority):
1. Including a case study on energy efficiency.
2. Including a guest lecturer to teach a sub-topic.
3. Offering supervised research topics on energy efficiency themes.
4. Offering energy efficiency as a topic in a problem-based learning course.
5. Including assessment that aligns with the energy efficiency theme within the course
(e.g. exam questions and assignments)
6. Including tutorials that align with the energy efficiency theme in the course (e.g.
presentations/ discussions/ problem solving)
7. Overhauling the course to embed energy efficiency
8. Including one workshop on energy efficiency in the course (i.e. experiments)
9. Including a field trip related to energy efficiency
10. Developing a new course on energy efficiency
Table 2 provides a summary of the identified common barriers to one or more of the
shortlisted options, highlighting that putting in place mechanisms to address a particular
barrier can have multiple flow-on benefits for addressing other barriers. For example, for
key staff who are tasked with integrating new content, setting up an annual allocation of
Energy transformed: building capacity in the engineering profession in australia 133

Driver Factors promoting engineering education for sustainable development
Institutional/
Civil Society
– Shifting accreditation requirements for graduate engineers - formalising
sustainability knowledge and skill requirements.
– Mandatory disclosure and reporting – increasing disclosure and
reporting requirements (e.g. greenhouse gas emissions).
– Increasing professional advocacy - with leaders stating the pivotal role
of engineering in addressing 21st Century challenges.
– Shifting requirements for practising engineers by professional
organizations - where mission statements, code of ethics statements
and codes of practice are being updated.
– Increasing commitment and action by highly regarded university peers -
increasingly vocal commitments and alliances.
Table 1. Key factors promoting engineering education for sustainable development

A number of key barriers are also evident in the literature, which appear to be limiting
efforts by engineering educators to undertake significant and rapid engineering curriculum
renewal, as summarized in Table 2.

Barrier Factors limiting engineering education for sustainable development
Market/
Business
– Persistent ‘old economy’ industry practices, wherein employers
continue to employ graduates to undertake unsustainable practices.
– Uncertainty around future requirements to change – where varying
government messages create considerable uncertainty around
impending requirements to change.
– Perceived threat to employability and position, from taking action ahead
of market or sector wide requirements to do so.
– Short-termism in the Higher Education Institution (HEI) sector, where
short-term pressures demand increasing staff to student ratios, and
increasing student intake, rather than program innovation.
– A shortage of engineering graduates, resulting in a ‘take what you can
get’ scenario, to then up-skill internally.
Information/
Technology
– Growing disconnect between engineering and science, where engineering
professionals may not be ‘in-step’ in understanding the complexity
and interdisciplinary nature of 21st Century challenges.
– Lack of convenient access to emerging and rigorously reviewed information,
where academics may have difficulty getting information and those
who have good access may be overwhelmed.
– Lack of access to information in foreign languages, which may impede
the integration of emerging technologies and innovations.
Institutional/
Civil Society
– Lack of strong requirements for change, where there is a lack of certainty
about current and future legislative requirements and support.
– Lack of academic staff competencies in EESD, with a relatively low rate
of professional development among educators.
Table 2. Key drivers limiting engineering education for sustainable development


Hence, there exist a number of significant societal drivers promoting curriculum renewal
within engineering education, which are being tempered by a number of barriers that are
limiting the progress. These barriers and others have been strong enough to-date, to prevent
a transition towards engineering education for sustainable development in the majority of
universities around the world. Many engineering departments are doing little more than
including one or two ‘sustainability’ courses within existing programs, leaving isolated
individuals or small teams within departments to undertake ad hoc curriculum renewal
efforts. In reality, most current engineering degrees are still focused on what could broadly
be classified as ‘fossil fuel based old industry’, involving linear ‘heat, beat and treat’
processes that don’t tend to consider rethinking waste, minimizing inputs, maximizing
productivity, capturing synergies or other externalities as part of the process (Benyus, 1997).

3.3 Curriculum drivers promoting and impeding energy efficiency education
Given these observations regarding societal drivers promoting and limiting engineering
education for sustainable development, in 2009 the NFEE funded an investigation into
identify challenges and opportunities for timely curriculum renewal in energy efficiency
education, at the level of the lecturer (Desha & Hargroves, 2009b). Specifically, the project
focused on developing and releasing a strategic document to assist the curriculum renewal
process for energy efficiency education, drawing upon a behavior change methodology
developed by McKenzie-Mohr and Smith (2007). The findings were intended for use by
engineering departments, accreditation agencies, professional bodies and government, to
identify opportunities for moving forward, and then to strategically plan the transition. The
project also provided a significant opportunity to explore options to support lecturers,
program co-ordinators and staff to strategically approach, in an informed way, the challenge
of increasing the levels of education for energy efficiency as a proxy for other sustainable
development topics.
Through a comprehensive literature review followed by a national survey of engineering
educators, the researchers short-listed 10 favored options amongst HEIs to integrate
emerging energy efficiency content within current engineering programs, as shown below
(in order of priority):
1. Including a case study on energy efficiency.
2. Including a guest lecturer to teach a sub-topic.
3. Offering supervised research topics on energy efficiency themes.
4. Offering energy efficiency as a topic in a problem-based learning course.
5. Including assessment that aligns with the energy efficiency theme within the course
(e.g. exam questions and assignments)
6. Including tutorials that align with the energy efficiency theme in the course (e.g.
presentations/ discussions/ problem solving)
7. Overhauling the course to embed energy efficiency
8. Including one workshop on energy efficiency in the course (i.e. experiments)
9. Including a field trip related to energy efficiency
10. Developing a new course on energy efficiency
Table 2 provides a summary of the identified common barriers to one or more of the
shortlisted options, highlighting that putting in place mechanisms to address a particular
barrier can have multiple flow-on benefits for addressing other barriers. For example, for
key staff who are tasked with integrating new content, setting up an annual allocation of
Energy Effciency 134

teaching buy-out funds, or having an avenue for temporarily altering staff teaching-
research-service workload allocation to engage in rapid curriculum renewal, would help to
address the barrier of insufficient time for preparation, which affects 7 of the 10 options.
Similarly, an annual small-grants program available for educators to pilot rapid curriculum
renewal initiatives would help to address the barrier of prohibitive cost. A ‘tiered’ approach
could be applied, where the first three options, including the use of case studies, guest
lecturers and supervised research, may immediately be targeted, with other options then
implemented among various programs in the following budget cycles.

Key Issues
for Implementation
Shortlisted Options for Curriculum Renewal
1
.
C
a
s
e

S
t
u
d
y

2
.

G
u
e
s
t

L
e
c
t
u
r
e
r

3
.

S
u
p
e
r
v
i
s
e
d

R
e
s
e
a
r
c
h

4
.

P
B
L

T
o
p
i
c

5
.

I
n
c
l
u
d
e

A
s
s
e
s
s
m
e
n
t

6
.

T
u
t
o
r
i
a
l
s

7
.

C
o
u
r
s
e

O
v
e
r
h
a
u
l


8
.

W
o
r
k
s
h
o
p

9
.

F
i
e
l
d

T
r
i
p

1
0
.

N
e
w

C
o
u
r
s
e

Common Barriers
Lack of available data/ information        
Lack of time for preparation       
An overcrowded curriculum      
Prohibitive cost        
Lack of knowledge        
Lack of value attached   
Lack of industry contacts   
Resistance to top-down directive  
Students’ prior learning habits  
Lecturer apathy  
Administrative coordination   
Common Benefits
Improved marketability     
Cross-functionality of content   
Additional research opportunities  
Networking opportunities for students   
Networking opportunities for lecturers   
Experience in incorporating emerging
concepts into curriculum
 

Addressing the time-lag for graduates  
Improved pedagogy - problem based
learning
  



Improved pedagogy – generic skills    
Lecturer professional development  
Table 1. Identified key barriers and benefits to timely curriculum renewal in energy
efficiency education Source: (Desha et al., 2009b)


4. Enabling capacity building for energy efficiency
With such considerations in mind, higher education institutions can strategically allocate
budget and human resourcing to integrate new content – in this case energy efficiency
knowledge and skills – into existing education and training programs. However, the
successful transition of engineering education to incorporate such new material is reliant on
a number of factors as discussed in the following paragraphs.

4.1 Institutional leadership and support
According to a study by an American campus sustainability assessment project, higher
education institutions which are leading in embedding sustainable development knowledge
and skills within the curriculum share a number of characteristics: “First, these ‘sustainability
leaders’ have adopted serious strategies for systematically addressing the sustainability of the
institution. They have policies stating their commitment to sustainability goals, and they have
specific plans in place that explain how they intend to achieve them. Second, these institutions have
provided the resources needed to implement their sustainability plans. They hire staff, form
committees, allocate budgets, and show clear administrative support for sustainability initiatives.
Third, these sustainability leaders know where they have been, where they are, and where they are
headed in terms of sustainability. They measure and track their progress toward sustainability, and
regularly meet and update goals and targets” (The Campus Sustainability Assessment Project,
undated).
A 2008 report to the Australian Teaching and Learning Council on addressing the supply
and quality of engineering graduates for the new century observed four supporting actions
that were common in institutions facilitating significant change, namely: 1) vision; 2)
leadership; 3) stakeholder engagement; and 4) resources (King, 2008). Hence, where a period
of rapid curriculum renewal is required, it needs to be supported with appropriate
resources for the relevant staff members, and undertaken in a realistic timeframe. Staff
members need to be encouraged to consider their own strengths and professional
development opportunities in contributing to decisions about how their courses embed
sustainability knowledge and skills. Existing and proactive efforts by staff in curriculum
renewal (i.e. the ‘leaders’ or ‘champions’ to date) should be acknowledged, supported and
rewarded. A strong collaborative foundation across sub-communities (for example across
different disciplines, or different campuses) is also an important mechanism to successfully
address surprises or issues as they arise during the curriculum renewal process.
University support could include the provision of funding, marketing and flexibility in rules
regarding developing new courses and modifying existing courses. A number of these
suggestions involve investing funds, which can be a challenge. However, institutional
benefits are clear and in the short term opportunities could be creatively explored for
example through industry course sponsorship, the appointment of funded ‘sustainability
chairs’ and professional development bursaries.

4.2 Strategic planning and implementation
For the various curriculum renewal options to be successful, an overarching strategic plan is
needed, which maps out timeframes, responsibilities and resource requirements. In the
NFEE investigation, a number of key components were identified that might be considered
Energy transformed: building capacity in the engineering profession in australia 135

teaching buy-out funds, or having an avenue for temporarily altering staff teaching-
research-service workload allocation to engage in rapid curriculum renewal, would help to
address the barrier of insufficient time for preparation, which affects 7 of the 10 options.
Similarly, an annual small-grants program available for educators to pilot rapid curriculum
renewal initiatives would help to address the barrier of prohibitive cost. A ‘tiered’ approach
could be applied, where the first three options, including the use of case studies, guest
lecturers and supervised research, may immediately be targeted, with other options then
implemented among various programs in the following budget cycles.

Key Issues
for Implementation
Shortlisted Options for Curriculum Renewal
1
.
C
a
s
e

S
t
u
d
y

2
.

G
u
e
s
t

L
e
c
t
u
r
e
r

3
.

S
u
p
e
r
v
i
s
e
d

R
e
s
e
a
r
c
h

4
.

P
B
L

T
o
p
i
c

5
.

I
n
c
l
u
d
e

A
s
s
e
s
s
m
e
n
t

6
.

T
u
t
o
r
i
a
l
s

7
.

C
o
u
r
s
e

O
v
e
r
h
a
u
l


8
.

W
o
r
k
s
h
o
p

9
.

F
i
e
l
d

T
r
i
p

1
0
.

N
e
w

C
o
u
r
s
e

Common Barriers
Lack of available data/ information        
Lack of time for preparation       
An overcrowded curriculum      
Prohibitive cost        
Lack of knowledge        
Lack of value attached   
Lack of industry contacts   
Resistance to top-down directive  
Students’ prior learning habits  
Lecturer apathy  
Administrative coordination   
Common Benefits
Improved marketability     
Cross-functionality of content   
Additional research opportunities  
Networking opportunities for students   
Networking opportunities for lecturers   
Experience in incorporating emerging
concepts into curriculum
 

Addressing the time-lag for graduates  
Improved pedagogy - problem based
learning
  



Improved pedagogy – generic skills    
Lecturer professional development  
Table 1. Identified key barriers and benefits to timely curriculum renewal in energy
efficiency education Source: (Desha et al., 2009b)


4. Enabling capacity building for energy efficiency
With such considerations in mind, higher education institutions can strategically allocate
budget and human resourcing to integrate new content – in this case energy efficiency
knowledge and skills – into existing education and training programs. However, the
successful transition of engineering education to incorporate such new material is reliant on
a number of factors as discussed in the following paragraphs.

4.1 Institutional leadership and support
According to a study by an American campus sustainability assessment project, higher
education institutions which are leading in embedding sustainable development knowledge
and skills within the curriculum share a number of characteristics: “First, these ‘sustainability
leaders’ have adopted serious strategies for systematically addressing the sustainability of the
institution. They have policies stating their commitment to sustainability goals, and they have
specific plans in place that explain how they intend to achieve them. Second, these institutions have
provided the resources needed to implement their sustainability plans. They hire staff, form
committees, allocate budgets, and show clear administrative support for sustainability initiatives.
Third, these sustainability leaders know where they have been, where they are, and where they are
headed in terms of sustainability. They measure and track their progress toward sustainability, and
regularly meet and update goals and targets” (The Campus Sustainability Assessment Project,
undated).
A 2008 report to the Australian Teaching and Learning Council on addressing the supply
and quality of engineering graduates for the new century observed four supporting actions
that were common in institutions facilitating significant change, namely: 1) vision; 2)
leadership; 3) stakeholder engagement; and 4) resources (King, 2008). Hence, where a period
of rapid curriculum renewal is required, it needs to be supported with appropriate
resources for the relevant staff members, and undertaken in a realistic timeframe. Staff
members need to be encouraged to consider their own strengths and professional
development opportunities in contributing to decisions about how their courses embed
sustainability knowledge and skills. Existing and proactive efforts by staff in curriculum
renewal (i.e. the ‘leaders’ or ‘champions’ to date) should be acknowledged, supported and
rewarded. A strong collaborative foundation across sub-communities (for example across
different disciplines, or different campuses) is also an important mechanism to successfully
address surprises or issues as they arise during the curriculum renewal process.
University support could include the provision of funding, marketing and flexibility in rules
regarding developing new courses and modifying existing courses. A number of these
suggestions involve investing funds, which can be a challenge. However, institutional
benefits are clear and in the short term opportunities could be creatively explored for
example through industry course sponsorship, the appointment of funded ‘sustainability
chairs’ and professional development bursaries.

4.2 Strategic planning and implementation
For the various curriculum renewal options to be successful, an overarching strategic plan is
needed, which maps out timeframes, responsibilities and resource requirements. In the
NFEE investigation, a number of key components were identified that might be considered
Energy Effciency 136

in a strategic plan to rapidly develop graduates who can fill critical energy efficiency
knowledge and skills gaps (Desha et al., 2009b):
– Planning from the outset, the best approach for the department given the opportunities
and risks with niche degrees versus embedding content throughout programs and
offering short courses.
– Building a strong collaborative foundation across campus sub-communities to
successfully address surprises or issues as they arise.
– Accessing the growing online library of academically rigorous open-access teaching
and learning resources to accelerate course development and renewal;
– Undertaking bridging and outreach opportunities across industry and government,
undergraduate and postgraduate programs, and high schools and the community, to
recruit students to the renewed programs;
– Making use of national and international collaboration with other academic institutions
and non-profit organizations, to jointly deliver courses on energy efficiency topics.
– Integrating such capacity building into campus operations as a two-way collaboration
between academics and students.

4.3 Catalysts for accelerating curriculum renewal
To address the existing time lag dilemma evident within engineering education, it is
important to set clear timeframes for capacity building processes. Three catalysts that can set
such timeframes are briefly discussed here:
– Program accreditation: Within regulated disciplines such as engineering, accreditation is
a strong driver of change, setting a review period of 3-5 years for universities to
continually reflect on and demonstrate how they have addressed existing and emerging
accreditation requirements in their programs, in order for their programs to remain
endorsed by the accrediting institution. However accreditation is quite a weak driver
for engineering education for sustainable development in reality, due to the lack of
clear direction on how much or within what timeframe to embed sustainability into
engineering curriculum. Furthermore, accreditation agencies and their academic
representatives on accreditation committees and boards do not necessarily have
adequate understanding of future needs and expectations for curriculum, resulting in a
lack of ability to change accreditation requirements. This situation was highlighted
more than a decade ago by the Australian Higher Education Council in their report on
Professional Education and Credentialism (Higher Education Council, 1996), which
outlined difficulties facing universities and professional bodies when defining
pathways for professional education.
– Employment: Both government and industry are significant potential catalysts in their
role as current and future employers of undergraduate and postgraduate students,
setting clear expectations about changing future employment and training needs. For
example, both government and industry could assist professional organizations and the
universities themselves (for example through advisory boards) to identify current and
future industry demands for graduates with specific knowledge and skill capabilities,
and in the demands of undergraduate and postgraduate students themselves.
Government and industry could require employees who are undertaking professional
development, to include a certain number of hours each year learning about
sustainability related technology and innovations.

– Regulation and policy: Government can play a role in catalyzing rapid curriculum
renewal through providing both penalties and incentives. This could be for example
through regulation, requiring industry to accelerate efforts such as energy efficiency
assessments. Government could also play a role in influencing professional
accreditation requirements to provide the necessary ‘calls for action’ in priority
knowledge and skills areas, to review and revise the coverage and extent of
accreditation requirements. Government could change the criteria and selection for
research funding, and link a portion of federal funding for higher education institutions
to institutional learning and teaching performance with regard to integrating energy
efficiency knowledge and skills into curricula.
An example of a government catalyst role can be seen in the example of the Australian
federal government’s ‘Energy Efficiency Opportunities’ program, launched in July 2006,
which required more than 220 businesses (representing around 45 percent of national
energy demand) that use more than 0.5 PJ (approximately 139,000 MWh) of energy per year,
to undertake an energy efficiency assessment and report publically on opportunities with a
payback period of up to 4 years (DRET, undated). Further to this, Victoria was the first state
to require all EPA license holders using more than 0.1 PJ (27,800 MWh) to implement
opportunities with a payback period of up to 3 years, through its ‘Industry Greenhouse
Program’ (Victorian Environmental Protection Agency, undated). As a result of
implementing these programs, both state and federal government has identified a
significant skills shortage in the area of undertaking energy efficiency assessments.
Subsequently the federal government initiated a ‘Long Term Training Strategy for the
Development of Energy Efficiency Assessment Skills’, beginning in 2009 with an extensive
survey process across the energy intensive industries, energy service providers, and
universities (Council of Australian Governments, 2009). In 2007, the CSIRO (Commonwealth
Scientific and Industrial Research Organization) through its ‘Energy Transformed Flagship’
engaged researchers from The Natural Edge Project to provide capacity building notes for
professionals and students looking to up-skill in energy efficiency opportunities, aimed at
both undergraduate education and professional development, as discussed below.

5. Capacity building resources
In 2007, the CSIRO funded the development of three education and training modules (30
lectures) in line with its goal for its ‘energy transformed’ program, ‘to facilitate the
development and implementation of stationary and transport technologies so as to halve greenhouse
gas emissions, double the efficiency of the nation’s new energy generation, supply and end use, and to
position Australia for a future hydrogen economy’. It was intended that these modules would
provide a base capacity-building training program that would prepare
engineers/technicians/facilities managers/architects etc. to address the issues of
greenhouse gas emissions and work towards creating sustainable energy solutions
throughout the course of their professional life. Within this context the modules would
provide an introduction to energy efficiency and low emissions technologies.
The resultant Energy Transformed education package (Smith et al., 2007) contains over 600
pages of peer-reviewed content that is freely available online, covering a wide range of
issues related to energy for use in undergraduate education, providing industry, business
and households with the knowledge they need to realize at least 30 percent energy
Energy transformed: building capacity in the engineering profession in australia 137

in a strategic plan to rapidly develop graduates who can fill critical energy efficiency
knowledge and skills gaps (Desha et al., 2009b):
– Planning from the outset, the best approach for the department given the opportunities
and risks with niche degrees versus embedding content throughout programs and
offering short courses.
– Building a strong collaborative foundation across campus sub-communities to
successfully address surprises or issues as they arise.
– Accessing the growing online library of academically rigorous open-access teaching
and learning resources to accelerate course development and renewal;
– Undertaking bridging and outreach opportunities across industry and government,
undergraduate and postgraduate programs, and high schools and the community, to
recruit students to the renewed programs;
– Making use of national and international collaboration with other academic institutions
and non-profit organizations, to jointly deliver courses on energy efficiency topics.
– Integrating such capacity building into campus operations as a two-way collaboration
between academics and students.

4.3 Catalysts for accelerating curriculum renewal
To address the existing time lag dilemma evident within engineering education, it is
important to set clear timeframes for capacity building processes. Three catalysts that can set
such timeframes are briefly discussed here:
– Program accreditation: Within regulated disciplines such as engineering, accreditation is
a strong driver of change, setting a review period of 3-5 years for universities to
continually reflect on and demonstrate how they have addressed existing and emerging
accreditation requirements in their programs, in order for their programs to remain
endorsed by the accrediting institution. However accreditation is quite a weak driver
for engineering education for sustainable development in reality, due to the lack of
clear direction on how much or within what timeframe to embed sustainability into
engineering curriculum. Furthermore, accreditation agencies and their academic
representatives on accreditation committees and boards do not necessarily have
adequate understanding of future needs and expectations for curriculum, resulting in a
lack of ability to change accreditation requirements. This situation was highlighted
more than a decade ago by the Australian Higher Education Council in their report on
Professional Education and Credentialism (Higher Education Council, 1996), which
outlined difficulties facing universities and professional bodies when defining
pathways for professional education.
– Employment: Both government and industry are significant potential catalysts in their
role as current and future employers of undergraduate and postgraduate students,
setting clear expectations about changing future employment and training needs. For
example, both government and industry could assist professional organizations and the
universities themselves (for example through advisory boards) to identify current and
future industry demands for graduates with specific knowledge and skill capabilities,
and in the demands of undergraduate and postgraduate students themselves.
Government and industry could require employees who are undertaking professional
development, to include a certain number of hours each year learning about
sustainability related technology and innovations.

– Regulation and policy: Government can play a role in catalyzing rapid curriculum
renewal through providing both penalties and incentives. This could be for example
through regulation, requiring industry to accelerate efforts such as energy efficiency
assessments. Government could also play a role in influencing professional
accreditation requirements to provide the necessary ‘calls for action’ in priority
knowledge and skills areas, to review and revise the coverage and extent of
accreditation requirements. Government could change the criteria and selection for
research funding, and link a portion of federal funding for higher education institutions
to institutional learning and teaching performance with regard to integrating energy
efficiency knowledge and skills into curricula.
An example of a government catalyst role can be seen in the example of the Australian
federal government’s ‘Energy Efficiency Opportunities’ program, launched in July 2006,
which required more than 220 businesses (representing around 45 percent of national
energy demand) that use more than 0.5 PJ (approximately 139,000 MWh) of energy per year,
to undertake an energy efficiency assessment and report publically on opportunities with a
payback period of up to 4 years (DRET, undated). Further to this, Victoria was the first state
to require all EPA license holders using more than 0.1 PJ (27,800 MWh) to implement
opportunities with a payback period of up to 3 years, through its ‘Industry Greenhouse
Program’ (Victorian Environmental Protection Agency, undated). As a result of
implementing these programs, both state and federal government has identified a
significant skills shortage in the area of undertaking energy efficiency assessments.
Subsequently the federal government initiated a ‘Long Term Training Strategy for the
Development of Energy Efficiency Assessment Skills’, beginning in 2009 with an extensive
survey process across the energy intensive industries, energy service providers, and
universities (Council of Australian Governments, 2009). In 2007, the CSIRO (Commonwealth
Scientific and Industrial Research Organization) through its ‘Energy Transformed Flagship’
engaged researchers from The Natural Edge Project to provide capacity building notes for
professionals and students looking to up-skill in energy efficiency opportunities, aimed at
both undergraduate education and professional development, as discussed below.

5. Capacity building resources
In 2007, the CSIRO funded the development of three education and training modules (30
lectures) in line with its goal for its ‘energy transformed’ program, ‘to facilitate the
development and implementation of stationary and transport technologies so as to halve greenhouse
gas emissions, double the efficiency of the nation’s new energy generation, supply and end use, and to
position Australia for a future hydrogen economy’. It was intended that these modules would
provide a base capacity-building training program that would prepare
engineers/technicians/facilities managers/architects etc. to address the issues of
greenhouse gas emissions and work towards creating sustainable energy solutions
throughout the course of their professional life. Within this context the modules would
provide an introduction to energy efficiency and low emissions technologies.
The resultant Energy Transformed education package (Smith et al., 2007) contains over 600
pages of peer-reviewed content that is freely available online, covering a wide range of
issues related to energy for use in undergraduate education, providing industry, business
and households with the knowledge they need to realize at least 30 percent energy
Energy Effciency 138

efficiency savings as rapidly as possible. The text also provides an updated overview of the
latest advances in low carbon technologies, renewable energy and sustainable transport.
The contents of the Energy Transformed program are separated into three ‘modules’:
– Module A: Understanding, identifying and implementing energy efficiency
opportunities for industrial/commercial users – by technology.
– Module B: Understanding, identifying and implementing energy efficiency
opportunities for industrial/commercial users – by sector.
– Module C: Integrated approaches to energy efficiency and low emissions electricity,
transport and distributed energy.
The component chapters and lessons in the package are summarized in Table 4.

Module Content
A
Chapter 1: Climate Change Mitigation in Australia's Energy Sector
1.1: Achieving a 60 percent reduction in greenhouse gas emissions by 2050
1.2: Carbon down, profits up – multiple benefits for Australia
1.3: Integrated approaches to energy efficiency & low carbon technologies
1.4: A whole systems approach to energy efficiency in new & existing systems
Chapter 2: Energy Efficiency Opportunities for Commercial Users
2.1: The importance & benefits of a front-loaded design process
2.2: Opportunities for energy efficiency in commercial buildings
2.3: Opportunities for improving the efficiency of HVAC systems
Chapter 3: Energy Efficiency Opportunities for Industrial Users
3.1: Opportunities for improving the efficiency of motor systems
3.2: Opportunities for improving the efficiency of boiler and steam Distribution systems
3.3: Energy efficiency improvements available through co-generation
B
Chapter 4: Responding to Increasing Demand for Electricity
4.1: What factors are causing rising peak and base load electricity demand in Australia?
4.2: Demand management approaches to reduce rising ‘peak load’ electricity demand
4.3: Demand management approaches to reduce rising ‘base load’ electricity
4.4: Making energy efficiency opportunities a win-win for customers and the utility:
decoupling energy utility profits from electricity sales
Chapter 5: Energy Efficiency Opportunities in Large Energy Using Industry Sectors
5.1: Opportunities for energy efficiency in the aluminum, steel and cement sectors
5.2: Opportunities for energy efficiency in manufacturing industries
5.3: Opportunities for energy efficiency in the it industry and services sector
Chapter 6: Energy Efficiency Opportunities in Light Industry and Commercial Sectors
6.1: Opportunities for energy efficiency in the tourism and hospitality sectors
6.2: Opportunities for energy efficiency in the food processing and retail sector
6.3: Opportunities for energy efficiency in the fast food industry
C
Chapter 7: Integrated Approaches to Energy Efficiency and Low Emissions Electricity
7.1: Opportunities & technologies to produce low emission electricity from fossil fuels
7.2: Can renewable energy supply peak electricity demand?
7.3: Can renewable energy supply base electricity demand?
7.4: Hidden benefits of distributed generation to supply base electricity demand
Chapter 8: Integrated Approaches to Energy Efficiency and Transport
8.1: Designing a sustainable transport future
8.2: Integrated approaches to energy efficiency & alternative transport fuels – passenger
8.3: Integrated approaches to energy efficiency and alternative transport fuels - trucking
Chapter 9: Integrated Approaches to Energy Efficiency and Distributed Energy

Module Content
9.1: Residential building energy efficiency and renewable energy opportunities:
towards a climate-neutral home
9.2: Commercial building energy efficiency and renewable energy opportunities:
towards climate-neutral commercial buildings
9.3: Beyond energy efficiency and distributed energy: options to offset emissions
Table 4. Energy Transformed: Sustainable Energy Solutions for Climate Change Mitigation
Source: (Smith et al., 2007)

In summary, these modules bring together the knowledge of how countries, specifically
Australia, can achieve at least 60 percent cuts to greenhouse gas emissions by 2050, in line
with the activities of the CSIRO Energy Transformed Flagship research program which is
focused on research that will assist Australia to achieve this target. The materials provide
industry, governments, business and households with knowledge to realize at least 30
percent energy efficiency savings in the short term while providing a strong basis for further
improvement. It also includes an overview of advances in low carbon technologies,
renewable energy and sustainable transport. While the package has an Australian focus, it
outlines sustainable energy strategies and provides links to numerous online reports which
can assist climate change mitigation efforts globally. It seeks to inform other initiatives that
are encouraging the reduction of greenhouse gas emissions, for example through behavior
change, sustainable consumption, and changes to economic incentives and policy.
The online format of this education and training program has been designed using the
results of the 2007 and 2009 NFEE funded research, including the following considerations:
– Extensive peer review was sought during the writing process (see acknowledgements)
to assist with creating awareness of the materials.
– The length of content (i.e. number of pages) for each lecture is intended to make it easy
for rapid uptake, using a 20-30 page highly structured and straightforward format that
suits most learning environments.
– Each lecture provides links to numerous online reports that outline sustainable energy
strategies and which assist climate change mitigation efforts in Australia and globally.
– The presentation of the content has been designed for flexibility, to cater for a range of
learning processes, from self-paced modular learning through to PowerPoint
presentations, tutorial discussions and problem-based learning.
Each ‘lecture’ begins with an ‘Educational Aim’ which provides an overview of the module.
This is followed by a section called, ‘Essential Reading’, wherein key references used in the
module that is readily accessible (i.e. with regard to language and layout) are listed and
hyperlinked where practical. The lecture then proceeds with around ten ‘Learning Points’
that are around 3-4 sentences each, which step through the core knowledge. Each learning
point has been worded so that key words can be easily extracted for PowerPoint slides or
handouts. Following the learning points, the lecture includes several pages of ‘Background
Information’, which provide both contextual information and deeper insights into the
knowledge area. This is also intended to provide a straightforward and short briefing to
lecturers/ trainers who may not have prior knowledge of the specific content. Following the
learning points, a list of ‘Optional Reading’ is provided as an additional resource for
assignments or further research, and a set of ‘Key Words for Searching Online’ are listed to
assist with beginning an internet exploration.
Energy transformed: building capacity in the engineering profession in australia 139

efficiency savings as rapidly as possible. The text also provides an updated overview of the
latest advances in low carbon technologies, renewable energy and sustainable transport.
The contents of the Energy Transformed program are separated into three ‘modules’:
– Module A: Understanding, identifying and implementing energy efficiency
opportunities for industrial/commercial users – by technology.
– Module B: Understanding, identifying and implementing energy efficiency
opportunities for industrial/commercial users – by sector.
– Module C: Integrated approaches to energy efficiency and low emissions electricity,
transport and distributed energy.
The component chapters and lessons in the package are summarized in Table 4.

Module Content
A
Chapter 1: Climate Change Mitigation in Australia's Energy Sector
1.1: Achieving a 60 percent reduction in greenhouse gas emissions by 2050
1.2: Carbon down, profits up – multiple benefits for Australia
1.3: Integrated approaches to energy efficiency & low carbon technologies
1.4: A whole systems approach to energy efficiency in new & existing systems
Chapter 2: Energy Efficiency Opportunities for Commercial Users
2.1: The importance & benefits of a front-loaded design process
2.2: Opportunities for energy efficiency in commercial buildings
2.3: Opportunities for improving the efficiency of HVAC systems
Chapter 3: Energy Efficiency Opportunities for Industrial Users
3.1: Opportunities for improving the efficiency of motor systems
3.2: Opportunities for improving the efficiency of boiler and steam Distribution systems
3.3: Energy efficiency improvements available through co-generation
B
Chapter 4: Responding to Increasing Demand for Electricity
4.1: What factors are causing rising peak and base load electricity demand in Australia?
4.2: Demand management approaches to reduce rising ‘peak load’ electricity demand
4.3: Demand management approaches to reduce rising ‘base load’ electricity
4.4: Making energy efficiency opportunities a win-win for customers and the utility:
decoupling energy utility profits from electricity sales
Chapter 5: Energy Efficiency Opportunities in Large Energy Using Industry Sectors
5.1: Opportunities for energy efficiency in the aluminum, steel and cement sectors
5.2: Opportunities for energy efficiency in manufacturing industries
5.3: Opportunities for energy efficiency in the it industry and services sector
Chapter 6: Energy Efficiency Opportunities in Light Industry and Commercial Sectors
6.1: Opportunities for energy efficiency in the tourism and hospitality sectors
6.2: Opportunities for energy efficiency in the food processing and retail sector
6.3: Opportunities for energy efficiency in the fast food industry
C
Chapter 7: Integrated Approaches to Energy Efficiency and Low Emissions Electricity
7.1: Opportunities & technologies to produce low emission electricity from fossil fuels
7.2: Can renewable energy supply peak electricity demand?
7.3: Can renewable energy supply base electricity demand?
7.4: Hidden benefits of distributed generation to supply base electricity demand
Chapter 8: Integrated Approaches to Energy Efficiency and Transport
8.1: Designing a sustainable transport future
8.2: Integrated approaches to energy efficiency & alternative transport fuels – passenger
8.3: Integrated approaches to energy efficiency and alternative transport fuels - trucking
Chapter 9: Integrated Approaches to Energy Efficiency and Distributed Energy

Module Content
9.1: Residential building energy efficiency and renewable energy opportunities:
towards a climate-neutral home
9.2: Commercial building energy efficiency and renewable energy opportunities:
towards climate-neutral commercial buildings
9.3: Beyond energy efficiency and distributed energy: options to offset emissions
Table 4. Energy Transformed: Sustainable Energy Solutions for Climate Change Mitigation
Source: (Smith et al., 2007)

In summary, these modules bring together the knowledge of how countries, specifically
Australia, can achieve at least 60 percent cuts to greenhouse gas emissions by 2050, in line
with the activities of the CSIRO Energy Transformed Flagship research program which is
focused on research that will assist Australia to achieve this target. The materials provide
industry, governments, business and households with knowledge to realize at least 30
percent energy efficiency savings in the short term while providing a strong basis for further
improvement. It also includes an overview of advances in low carbon technologies,
renewable energy and sustainable transport. While the package has an Australian focus, it
outlines sustainable energy strategies and provides links to numerous online reports which
can assist climate change mitigation efforts globally. It seeks to inform other initiatives that
are encouraging the reduction of greenhouse gas emissions, for example through behavior
change, sustainable consumption, and changes to economic incentives and policy.
The online format of this education and training program has been designed using the
results of the 2007 and 2009 NFEE funded research, including the following considerations:
– Extensive peer review was sought during the writing process (see acknowledgements)
to assist with creating awareness of the materials.
– The length of content (i.e. number of pages) for each lecture is intended to make it easy
for rapid uptake, using a 20-30 page highly structured and straightforward format that
suits most learning environments.
– Each lecture provides links to numerous online reports that outline sustainable energy
strategies and which assist climate change mitigation efforts in Australia and globally.
– The presentation of the content has been designed for flexibility, to cater for a range of
learning processes, from self-paced modular learning through to PowerPoint
presentations, tutorial discussions and problem-based learning.
Each ‘lecture’ begins with an ‘Educational Aim’ which provides an overview of the module.
This is followed by a section called, ‘Essential Reading’, wherein key references used in the
module that is readily accessible (i.e. with regard to language and layout) are listed and
hyperlinked where practical. The lecture then proceeds with around ten ‘Learning Points’
that are around 3-4 sentences each, which step through the core knowledge. Each learning
point has been worded so that key words can be easily extracted for PowerPoint slides or
handouts. Following the learning points, the lecture includes several pages of ‘Background
Information’, which provide both contextual information and deeper insights into the
knowledge area. This is also intended to provide a straightforward and short briefing to
lecturers/ trainers who may not have prior knowledge of the specific content. Following the
learning points, a list of ‘Optional Reading’ is provided as an additional resource for
assignments or further research, and a set of ‘Key Words for Searching Online’ are listed to
assist with beginning an internet exploration.
Energy Effciency 140

6. Conclusion
This chapter has discussed the need for urgent capacity building in the engineering
profession in the area of energy efficiency, focusing on higher education institutions. We
have considered the complexity of the issue within the higher education sector, where the
problem is two-fold: energy efficiency knowledge and skills are not yet being taught; and
the process for curriculum renewal is generally slow and ad hoc. Moreover, there are a
number of organisational and curriculum influences that are working to both promote and
impede capacity building in energy efficiency, requiring a strategic and systematic approach
to ensure that the engineering profession is being up-skilled as quickly as possible. This
includes leadership and support at an institutional level, strategic planning and
implementation of curriculum renewal initiatives, and within clear timeframes. Three
examples of energy efficiency capacity building initiatives in Australia have been
highlighted; namely two research initiatives undertaken through the National Framework
for Energy Efficiency (NFEE), and modular content developed through the CSIRO’s energy
transformed program.
In conclusion, the research undertaken to date provides a clear understanding of the state of
engineering education for energy efficiency in Australia. Furthermore, the Energy
Transformed education package funded by the CSIRO provides a significant tool for
engineering educators to access, to provide immediate and robust capacity building, from
undergraduate through to postgraduate education.

7. Acknowledgements
The 2007 survey, 2009 investigation and the Energy Transformed education package were
undertaken by The Natural Edge Project using funds provided by CSIRO and the National
Framework for Energy Efficiency. Non-staff related on-costs and administrative support
was also provided by the Centre for Environment and Systems Research and the Urban
Research Program at Griffith University, and the Fenner School of Environment and Society
and Engineering Department at the Australian National University.
Principal reviewers for the three research initiatives included: Adjunct Professor Alan Pears
– RMIT, Geoff Andrews – Director, Genesis Now Pty Ltd, Dr Mike Dennis – ANU,
Engineering Department, Victoria Hart – Basset Engineering Consultants, Molly Olsen and
Phillip Toyne - EcoFutures Pty Ltd, Glenn Platt – CSIRO, Energy Transformed Flagship, and
Francis Barram – Bond University. The following persons provided peer review for specific
lectures in the Energy Transformed program; Dr Barry Newell – Australian national
University, Dr Chris Dunstan - Clean Energy Council, D van den Dool - Manager, Jamieson
Foley Traffic & Transport Pty Ltd, Daniel Veryard - Sustainable Transport Expert, Dr David
Lindley – Academic Principal, ACS Education, Frank Hubbard – International Hotels
Group, Gavin Gilchrist – Director, BigSwitch Projects, Ian Dunlop - President, Australian
Association for the Study of Peak Oil, Dr James McGregor – CSIRO, Energy Transformed
Flagship, Jill Grant – Department of Industry Training and Resources, Commonwealth
Government, Leonardo Ribon – RMIT Global Sustainability, Professor Mark Diesendorf –
University of New South Wales, Melinda Watt - CRC for Sustainable Tourism, Dr Paul
Compston - ANU AutoCRC, Dr Dominique Hes - University of Melbourne, Penny Prasad -
Project Officer, UNEP Working Group for Cleaner Production, University of Queensland,

Rob Gell – President, Greening Australia, Dr Tom Worthington - Director of the Professional
Development Board, Australian Computer Society.

8. References
Australian Industry Group (2007). Environmental Sustainability and Industry, Road to a
Sustainable Future, AIG, Australia.
Australian Federal Department of Resources, Energy and Tourism (DRET) (undated)
‘Energy Efficiency Opportunities’, http://www.ret.gov.au/energy/efficiency/
eeo/pages/default.aspx, accessed 12 May 2010.
Benyus, J. (1997). Biomimicry: Innovation Inspired by Nature, HarperCollins Publishers Inc.,
New York.
Brown, L. (2007). Plan B 3.0: Mobilising to Save Civilisation, W.W. Norton & Company, New
York, 117 and 208.
Council of Australian Governments (2009). National Strategy on Energy Efficiency, July 2009,
Commonwealth of Australia.
Desha, C. & Hargroves, K. (2009a). Re-engineering Higher Education for Energy Efficiency
Solutions. ECOS, CSIRO, Vol. 151, (Oct-Nov 2009) 16-17, 0311-4546.
Desha, C. & Hargroves, K. (2009b). Surveying the State of Higher Education in Energy
Efficiency in Australian Engineering Curriculum. Journal of Cleaner Production,
Elsevier, Vol. 18, Issue. 7, 652-658, 0959-6526.
Desha, C. ; Hargroves, K. & Smith, M. (2009). Addressing the time lag dilemma in
curriculum renewal towards engineering education for sustainable development,
International Journal of Sustainability in Higher Education, Vol. 10, No. 2, 184-199,
Elsevier.
Desha, C. ; Hargroves, K. & Stephens, R. (2007). Energy Transformed: Australian University
Survey Summary of Questionnaire Results. The Natural Edge Project (TNEP),
Australia.
Desha, C. ; Hargroves, K. & Reeve, A. (2009). An Investigation into the Options for Increasing
the Extent of Energy Efficiency Knowledge and Skills in Engineering Education, Report to
the National Framework for Energy Efficiency, The Natural Edge Project (TNEP),
Australia.
Energy Futures Forum (2006). The Heat Is On: The Future of Energy in Australia, CSIRO
Garnaut, R. (2008) The Garnaut Climate Change Review, Report to the Australian Federal
Government, Canberra.
Hatfield-Dodds, S. ; Turner ,G. ; Schandl, H. & Doss, T. (2008). Growing the green collar
economy: skills and labour challenges in reducing our greenhouse emissions and national
environmental footprint, Report to the Dusseldorp Skills Forum, CSIRO Sustainable
Ecosystems, Canberra.
Higher Education Council (1996). Professional Education and Credentialism, National Board of
Employment, Education and Training (NBEET), Canberra, p4.
IPCC (2007) Climate Change 2007: Impacts, Adaptation & Vulnerability, Contribution of Working
Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change, Cambridge University Press, Cambridge.
Energy transformed: building capacity in the engineering profession in australia 141

6. Conclusion
This chapter has discussed the need for urgent capacity building in the engineering
profession in the area of energy efficiency, focusing on higher education institutions. We
have considered the complexity of the issue within the higher education sector, where the
problem is two-fold: energy efficiency knowledge and skills are not yet being taught; and
the process for curriculum renewal is generally slow and ad hoc. Moreover, there are a
number of organisational and curriculum influences that are working to both promote and
impede capacity building in energy efficiency, requiring a strategic and systematic approach
to ensure that the engineering profession is being up-skilled as quickly as possible. This
includes leadership and support at an institutional level, strategic planning and
implementation of curriculum renewal initiatives, and within clear timeframes. Three
examples of energy efficiency capacity building initiatives in Australia have been
highlighted; namely two research initiatives undertaken through the National Framework
for Energy Efficiency (NFEE), and modular content developed through the CSIRO’s energy
transformed program.
In conclusion, the research undertaken to date provides a clear understanding of the state of
engineering education for energy efficiency in Australia. Furthermore, the Energy
Transformed education package funded by the CSIRO provides a significant tool for
engineering educators to access, to provide immediate and robust capacity building, from
undergraduate through to postgraduate education.

7. Acknowledgements
The 2007 survey, 2009 investigation and the Energy Transformed education package were
undertaken by The Natural Edge Project using funds provided by CSIRO and the National
Framework for Energy Efficiency. Non-staff related on-costs and administrative support
was also provided by the Centre for Environment and Systems Research and the Urban
Research Program at Griffith University, and the Fenner School of Environment and Society
and Engineering Department at the Australian National University.
Principal reviewers for the three research initiatives included: Adjunct Professor Alan Pears
– RMIT, Geoff Andrews – Director, Genesis Now Pty Ltd, Dr Mike Dennis – ANU,
Engineering Department, Victoria Hart – Basset Engineering Consultants, Molly Olsen and
Phillip Toyne - EcoFutures Pty Ltd, Glenn Platt – CSIRO, Energy Transformed Flagship, and
Francis Barram – Bond University. The following persons provided peer review for specific
lectures in the Energy Transformed program; Dr Barry Newell – Australian national
University, Dr Chris Dunstan - Clean Energy Council, D van den Dool - Manager, Jamieson
Foley Traffic & Transport Pty Ltd, Daniel Veryard - Sustainable Transport Expert, Dr David
Lindley – Academic Principal, ACS Education, Frank Hubbard – International Hotels
Group, Gavin Gilchrist – Director, BigSwitch Projects, Ian Dunlop - President, Australian
Association for the Study of Peak Oil, Dr James McGregor – CSIRO, Energy Transformed
Flagship, Jill Grant – Department of Industry Training and Resources, Commonwealth
Government, Leonardo Ribon – RMIT Global Sustainability, Professor Mark Diesendorf –
University of New South Wales, Melinda Watt - CRC for Sustainable Tourism, Dr Paul
Compston - ANU AutoCRC, Dr Dominique Hes - University of Melbourne, Penny Prasad -
Project Officer, UNEP Working Group for Cleaner Production, University of Queensland,

Rob Gell – President, Greening Australia, Dr Tom Worthington - Director of the Professional
Development Board, Australian Computer Society.

8. References
Australian Industry Group (2007). Environmental Sustainability and Industry, Road to a
Sustainable Future, AIG, Australia.
Australian Federal Department of Resources, Energy and Tourism (DRET) (undated)
‘Energy Efficiency Opportunities’, http://www.ret.gov.au/energy/efficiency/
eeo/pages/default.aspx, accessed 12 May 2010.
Benyus, J. (1997). Biomimicry: Innovation Inspired by Nature, HarperCollins Publishers Inc.,
New York.
Brown, L. (2007). Plan B 3.0: Mobilising to Save Civilisation, W.W. Norton & Company, New
York, 117 and 208.
Council of Australian Governments (2009). National Strategy on Energy Efficiency, July 2009,
Commonwealth of Australia.
Desha, C. & Hargroves, K. (2009a). Re-engineering Higher Education for Energy Efficiency
Solutions. ECOS, CSIRO, Vol. 151, (Oct-Nov 2009) 16-17, 0311-4546.
Desha, C. & Hargroves, K. (2009b). Surveying the State of Higher Education in Energy
Efficiency in Australian Engineering Curriculum. Journal of Cleaner Production,
Elsevier, Vol. 18, Issue. 7, 652-658, 0959-6526.
Desha, C. ; Hargroves, K. & Smith, M. (2009). Addressing the time lag dilemma in
curriculum renewal towards engineering education for sustainable development,
International Journal of Sustainability in Higher Education, Vol. 10, No. 2, 184-199,
Elsevier.
Desha, C. ; Hargroves, K. & Stephens, R. (2007). Energy Transformed: Australian University
Survey Summary of Questionnaire Results. The Natural Edge Project (TNEP),
Australia.
Desha, C. ; Hargroves, K. & Reeve, A. (2009). An Investigation into the Options for Increasing
the Extent of Energy Efficiency Knowledge and Skills in Engineering Education, Report to
the National Framework for Energy Efficiency, The Natural Edge Project (TNEP),
Australia.
Energy Futures Forum (2006). The Heat Is On: The Future of Energy in Australia, CSIRO
Garnaut, R. (2008) The Garnaut Climate Change Review, Report to the Australian Federal
Government, Canberra.
Hatfield-Dodds, S. ; Turner ,G. ; Schandl, H. & Doss, T. (2008). Growing the green collar
economy: skills and labour challenges in reducing our greenhouse emissions and national
environmental footprint, Report to the Dusseldorp Skills Forum, CSIRO Sustainable
Ecosystems, Canberra.
Higher Education Council (1996). Professional Education and Credentialism, National Board of
Employment, Education and Training (NBEET), Canberra, p4.
IPCC (2007) Climate Change 2007: Impacts, Adaptation & Vulnerability, Contribution of Working
Group II to the Fourth Assessment Report of the Intergovernmental Panel on Climate
Change, Cambridge University Press, Cambridge.
Energy Effciency 142

King, R. (2008). Addressing the Supply and Quality of Engineering Graduates for the New Century,
The Carrick Institute for Learning and Teaching in Higher Education Ltd, Sydney,
p30.
Levett-Therivel (2006). Sustainability carrots and sticks: the benefits and risks of sustainable
development for HEIs, Report to the Higher Education Funding Council of England
(HEFCE).
McKenzie-Mohr, D. & Smith, W. (2007). Fostering Sustainable Behaviour: An Introduction to
Community-Based Social Marketing, 3rd Ed, New Society Press, 0-86571-406-1,
Gabriola Island, BC.
Postel, S. (1999). Pillar of Sand, W.W. Norton & Company, New York, 13-21.
Price Waterhouse Coopers (2008). Carbon countdown: a survey of executive opinion on climate
change in the countdown to a carbon economy, PWC.
Proudfoot Consulting (2007). Meeting the Corporate Energy Challenge: Are Companies Walking
the Talk on Energy Efficiency?, Proudfoot Consulting.
Smith, M. ; Hargroves, K. ; Stasinopoulos, P. ; Stephens, R. ; Desha, C. & Hargroves, S.
(2007). Energy Transformed: Sustainable Energy Solutions for Climate Change Mitigation,
The Natural Edge Project (TNEP), Australia.
Stern, N. (2006) The Stern Review: The Economics of Climate Change, Cambridge University
Press, Cambridge.
The Campus Sustainability Assessment Project (undated) ‘Introduction: The Sustainability
Imperative and Higher Education: The Challenge of Sustainability’,
http://csap.envs.wmich.edu/pages/intro_imperative.html, accessed 13 May 2010.
The Institution of Engineers Australia (undated) ‘Energy Efficiency: The Importance of
Energy Efficiency in Moving toward Sustainability’,
http://www.engineersaustralia.org.au, accessed 12 May 2010.
United Nations (2002) Report of the World Summit on Sustainable Development, United Nations,
Johannesburg, South Africa, 26 August – 4 September 2002.
United Nations Environment Programme (2008) Green Jobs: Towards Decent Work in a
Sustainable, Low-Carbon World, United Nations.
US Geological Survey (2007), Mineral Commodity Summaries 2007, cited in Brown, L. (2007)
Plan B 3.0: Mobilising to Save Civilisation, W.W. Norton and Company, New York,
43.
Victorian Environmental Protection Agency (undated) ‘Environment and Resource
Efficiency Plans – EREP’, http://www.epa.vic.gov.au/bus/erep/, accessed 12 May
2010.
World Resources Institute (2005) World Resources 2005: The Wealth of the Poor-Managing
Ecosystems to Fight Poverty, WRI, in collaboration with the UNDP, UNEP and the
World Bank, Washington, DC.
The energy effciency of onboard hydrogen storage 143
The energy effciency of onboard hydrogen storage
Jens Oluf Jensen, Qingfeng Li and Niels J. Bjerrum
x

The energy efficiency of
onboard hydrogen storage

Jens Oluf Jensen, Qingfeng Li and Niels J. Bjerrum
Technical University of Denmark
Denmark

1. Introduction
Hydrogen is often suggested as a versatile energy carrier in future energy systems.
Hydrogen can be extracted from water by electrical energy through electrolysis and later
when the energy is needed hydrogen can recombine with oxygen from the air and release
the same amount of energy. The end product is water and the cycle is closed. Hydrogen as
an energy carrier is typically associated with renewable energy technologies because it
provides a way to store energy. The need for energy storage is tremendous if wind turbines,
wave energy devices or photovoltaics are to be implemented on a large scale. This is because
of the fluctuating nature of the electricity production by these means and moreover, because
the energy might be meant for application in the transport sector. Batteries store electrical
energy efficiently, but they are not economic for large scale storage and for transportation
they are only practical in smaller vehicles with a limited driving range and certainly not in
trucks, ships or airplanes. The alternative is to store energy as hydrogen and hydrogen is an
ideal fuel in many aspects. It is easily combusted in an engine or converted back to
electricity in a fuel cell. It is not poisonous and the raw material for its production (water) is
practically unlimited.
Hydrogen is often said to have the highest energy content per unit mass, but since it is a low
density gas at ambient conditions it needs a storage tank that adds so much to the weight
and volume that the whole system ends up being both heavier and bulkier than a gasoline
tank with the same energy content. Therefore, hydrogen storage is a key issue, and in
particular, onboard hydrogen storage in vehicles. As a matter of fact, the production of
hydrogen from renewable sources only makes sense if hydrogen is stored for later use or for
use elsewhere. Otherwise, one might as well use the extracted electricity directly (one
exception could be the use of bio fuels in a fuel cell through a stage where hydrogen is
liberated by reforming for immediate use, but this is not really within the idea of hydrogen
as an energy carrier). Two recent monographs each provide a detailed introduction to all the
aspects of hydrogen energy with several chapters dealing with storage techniques (Leon,
2008), (Züttel et al., 2008).
Many different techniques have been developed to solve this fundamental problem, and any
one of them has its own energy balance to consider. Storage of hydrogen can be quite energy
consuming and so can the subsequent liberation of hydrogen. In some cases both processes
are energy intensive. The literature on hydrogen storage often focuses on the storage
8
Energy Effciency 144

density, and the question of round trip energy efficiency of the storage process may then be
forgotten. In small systems, such energy losses might, although significant, be of less
importance, but for vehicular applications, they cannot be neglected. After all, improved
efficiency is one of the arguments when future fuel cell vehicles are compared with
conventional ones. This work will review the most common hydrogen storage techniques
with the focus on energy efficiency for charging and discharging the system, i.e. the round
trip efficiency. It is an elaborated version of a previous study (Jensen et al., 2007).

2. Overview of storage techniques
Hydrogen is a volatile gas at ambient conditions, and the storage challenge is to fight the
kinetic energy of the hydrogen molecules. Basically there are three ways to go. (1) The gas
can be confined at high pressure by external physical forces. (2) The energy of the molecules
can be withdrawn by cooling and ultimately the gas condenses into a liquid. (3) The
molecules can be bound to a surface or inside a solid material. This way hydrogen is more
or less immobilized and like in the case of liquid hydrogen, most of its kinetic energy is
removed. The three fundamental storage techniques are visualised in the corners of the
triangle in figure 1. Between the corners combined techniques that utilize more than one of
the principles are plotted.
Compression
Cooling Binding
Pressurized H
2
Liquid H
2
Cryo-
sorbed
H
2
Ambient temp. sorbed H
2
Metal hydrides
Syn. Fuels
+ chem. hydr.
Compression
Cooling Binding
Pressurized H
2
Liquid H
2
Cryo-
sorbed
H
2
Ambient temp. sorbed H
2
Metal hydrides
Syn. Fuels
+ chem. hydr.

Fig. 1. The different storage techniques arranged qualitatively after degree of cooling,
binding and pressurization.

Compressed hydrogen is kept in a dense state by external physical forces only. This is what
happens in a pressure vessel. It takes mechanical energy to compress the gas, but the release
is free of charge. Liquid hydrogen is kept together by weak chemical forces (van der Waals)

at very low temperature but at ambient pressure. Heat must be supplied to release
hydrogen through boiling, but due to the low boiling point of 20 K, the heat can in principle
be taken from the surroundings or any waste heat. Liquefaction of hydrogen by
pressurization alone is not possible since the critical point is as low as 33 K (and 13 bar).
Hydrogen can bind to matter in many ways. It can be via adsorption on a large surface with
some affinity for hydrogen molecules. In order to obtain a reasonable storage capacity this is
always done in combination with either cooling (to reduce the energy of the hydrogen
molecules), pressurization or both. The binding forces are the weak van der Waals forces
like in liquid hydrogen, but the interaction is stronger due to the substrate. Release is
comparable to a combination of compressed and liquid hydrogen. Absorption of hydrogen
takes place in specialized solid materials into which hydrogen can diffuse and bind by
metallic, ionic or covalent bonds. These forces are much stronger than the van der Waals
forces and consequently, it takes more energy to release hydrogen afterwards. Examples are
interstitial metal hydrides and complex hydrides. Finally it is possible to store hydrogen by
making synthetic fuels like hydrocarbons, alcohols and ammonia. In this case the bonds are
mostly covalent and require a significant amount of energy for hydrogen release. Moreover,
in many cases, addition of water is needed too like for steam reforming. Synthetic fuels
cannot be recharged onboard. Instead they are manufactured through chemical synthesis in
a plant.
Another way to arrange the storage techniques is shown in figure 2, where they are ordered
in a line ranging from pure physical storage to a gradually more chemical technique. A
tendency that goes with this is that the more chemical the technique, the less easily available
is the hydrogen. This less easy availability of hydrogen is seen as higher energy demands for
hydrogen release and/or higher release temperatures.

Compressed
(200-700 bar)
Liquid
(20 K)
Adsorbed
(Surfaces)
Absorbed
(Metal hydrides)
Chemical
compounds
Physical Chemical
Easy access Chemical extraction
Compressed
(200-700 bar)
Liquid
(20 K)
Adsorbed
(Surfaces)
Absorbed
(Metal hydrides)
Chemical
compounds
Physical Chemical
Easy access Chemical extraction

Fig. 2. The sequence of hydrogen storage techniques from physical to increasingly chemical.

3. The approach
The different storage techniques are in the following treated in the same order as in figure 2 from
left to right. Although hydrogen storage does in principle not depend on the application,
onboard storage, e.g. on a vehicle, is assumed since here we have the most demanding situation
that may justify sophisticated and possibly expensive storage techniques. The aim of the study is
first of all to compare the minimum energies required for storing hydrogen and releasing
hydrogen. When energy is needed for the release, typically heat, it can in some cases be supplied
by otherwise wasted heat from an engine or a fuel cell, but it depends on the temperature of that
heat whether it is possible. Alternatively, the heat for release can be supplied by part of the
hydrogen via a burner. In the latter case the available hydrogen for the main purpose (e.g.
The energy effciency of onboard hydrogen storage 145

density, and the question of round trip energy efficiency of the storage process may then be
forgotten. In small systems, such energy losses might, although significant, be of less
importance, but for vehicular applications, they cannot be neglected. After all, improved
efficiency is one of the arguments when future fuel cell vehicles are compared with
conventional ones. This work will review the most common hydrogen storage techniques
with the focus on energy efficiency for charging and discharging the system, i.e. the round
trip efficiency. It is an elaborated version of a previous study (Jensen et al., 2007).

2. Overview of storage techniques
Hydrogen is a volatile gas at ambient conditions, and the storage challenge is to fight the
kinetic energy of the hydrogen molecules. Basically there are three ways to go. (1) The gas
can be confined at high pressure by external physical forces. (2) The energy of the molecules
can be withdrawn by cooling and ultimately the gas condenses into a liquid. (3) The
molecules can be bound to a surface or inside a solid material. This way hydrogen is more
or less immobilized and like in the case of liquid hydrogen, most of its kinetic energy is
removed. The three fundamental storage techniques are visualised in the corners of the
triangle in figure 1. Between the corners combined techniques that utilize more than one of
the principles are plotted.
Compression
Cooling Binding
Pressurized H
2
Liquid H
2
Cryo-
sorbed
H
2
Ambient temp. sorbed H
2
Metal hydrides
Syn. Fuels
+ chem. hydr.
Compression
Cooling Binding
Pressurized H
2
Liquid H
2
Cryo-
sorbed
H
2
Ambient temp. sorbed H
2
Metal hydrides
Syn. Fuels
+ chem. hydr.

Fig. 1. The different storage techniques arranged qualitatively after degree of cooling,
binding and pressurization.

Compressed hydrogen is kept in a dense state by external physical forces only. This is what
happens in a pressure vessel. It takes mechanical energy to compress the gas, but the release
is free of charge. Liquid hydrogen is kept together by weak chemical forces (van der Waals)

at very low temperature but at ambient pressure. Heat must be supplied to release
hydrogen through boiling, but due to the low boiling point of 20 K, the heat can in principle
be taken from the surroundings or any waste heat. Liquefaction of hydrogen by
pressurization alone is not possible since the critical point is as low as 33 K (and 13 bar).
Hydrogen can bind to matter in many ways. It can be via adsorption on a large surface with
some affinity for hydrogen molecules. In order to obtain a reasonable storage capacity this is
always done in combination with either cooling (to reduce the energy of the hydrogen
molecules), pressurization or both. The binding forces are the weak van der Waals forces
like in liquid hydrogen, but the interaction is stronger due to the substrate. Release is
comparable to a combination of compressed and liquid hydrogen. Absorption of hydrogen
takes place in specialized solid materials into which hydrogen can diffuse and bind by
metallic, ionic or covalent bonds. These forces are much stronger than the van der Waals
forces and consequently, it takes more energy to release hydrogen afterwards. Examples are
interstitial metal hydrides and complex hydrides. Finally it is possible to store hydrogen by
making synthetic fuels like hydrocarbons, alcohols and ammonia. In this case the bonds are
mostly covalent and require a significant amount of energy for hydrogen release. Moreover,
in many cases, addition of water is needed too like for steam reforming. Synthetic fuels
cannot be recharged onboard. Instead they are manufactured through chemical synthesis in
a plant.
Another way to arrange the storage techniques is shown in figure 2, where they are ordered
in a line ranging from pure physical storage to a gradually more chemical technique. A
tendency that goes with this is that the more chemical the technique, the less easily available
is the hydrogen. This less easy availability of hydrogen is seen as higher energy demands for
hydrogen release and/or higher release temperatures.

Compressed
(200-700 bar)
Liquid
(20 K)
Adsorbed
(Surfaces)
Absorbed
(Metal hydrides)
Chemical
compounds
Physical Chemical
Easy access Chemical extraction
Compressed
(200-700 bar)
Liquid
(20 K)
Adsorbed
(Surfaces)
Absorbed
(Metal hydrides)
Chemical
compounds
Physical Chemical
Easy access Chemical extraction

Fig. 2. The sequence of hydrogen storage techniques from physical to increasingly chemical.

3. The approach
The different storage techniques are in the following treated in the same order as in figure 2 from
left to right. Although hydrogen storage does in principle not depend on the application,
onboard storage, e.g. on a vehicle, is assumed since here we have the most demanding situation
that may justify sophisticated and possibly expensive storage techniques. The aim of the study is
first of all to compare the minimum energies required for storing hydrogen and releasing
hydrogen. When energy is needed for the release, typically heat, it can in some cases be supplied
by otherwise wasted heat from an engine or a fuel cell, but it depends on the temperature of that
heat whether it is possible. Alternatively, the heat for release can be supplied by part of the
hydrogen via a burner. In the latter case the available hydrogen for the main purpose (e.g.
Energy Effciency 146

propulsion) will be reduced comparatively and the effective storage capacity is thus lower than
predicted from the amount of hydrogen stored.
A true comparison would involve a detailed analysis of whole systems. Such analyses are
truly relevant but also complicated with numerous assumptions on which the outcome will
strongly depend. Instead, transparency is aimed at with the hope that the conclusions are
less questionable, although they do not tell the whole story. Throughout, the lower heating
value (LHV) of the fuel is used instead of the higher heating value (HHV). This is because in
several of the systems, heat for hydrogen liberation must be supplied at temperatures above
100ºC likely by combustion of hydrogen. It is also assumed that hydrogen or a hydrogen
mixture is released at no less than ambient pressure. The LHV used is 242.8 kJ/mol H
2
.

4. Compressed hydrogen
Despite many attempts to develop advanced techniques for compact, practical and safe
hydrogen storage, pressurization is still the dominating technique. This is a fact for onboard
hydrogen as well as for hydrogen storage in general. The standard pressure for steel
cylinders is 200 bar, but high pressure fibre composite tanks rated for up to 7-800 bar have
been developed. The gravimetric storage capacity ranges from 1-2 wt.% for 200 bar steel
tanks to 5-10 wt.% for high pressure fibre tanks. Fibre tanks are more expensive than steel
tanks.

4.1 Energy for storage
The theoretical minimum work needed for gas compression can be calculated based on
integration of the infinitesimal pressure-volume work, dw

Vdp dw 
(1)

where V is the tank volume and p the pressure. Assuming ideal gas behaviour integration of
(1) from p
0
to p
1
results in the expression of the work, W, of ideal isothermal compression










0
1
0
ln
p
p
V p W

(2)

where p
0
and p
1
are initial and final pressures. At hydrogen pressures over 100 bar,
deviations from ideality become significant in this connection, and the dimensionless
compression factor, Z, shall compensate for the non-ideality. The real gas equation is then

pV ZnRT  (3)

Z depends on both pressure and temperature and is tabulated elsewhere (Perry et al., 1984).
At 300 K and pressures up to 1000 bar, the compression factor is modelled well as









 

p
p
k Z
z 300 ,
1

(4)

where k
z,300
= 0.000631. Integration including (3) and (4) gives

(
(
¸
(


¸

|
|
.
|


\
|
+ ÷ =
0
1
0 1 300 , 0 0
ln ) (
p
p
p p k V p W
z

(5)

However, the compression is never isothermal, as heat is formed during the process. If the
compression is very slow, most heat will dissipate to the surroundings, but in practical high
pressure systems, a significant amount of heat is formed. The other extreme is adiabatic
compression in which all heat produced is kept in the gas by ideal insulation. The work of
adiabatic compression is

(
(
(
¸
(



¸

÷
|
|
.
|


\
|
÷
=
÷
1
1
1
0
1
0
¸
¸
¸
¸
p
p
V p W

(6)

where γ is the ratio of specific heats (C
p
/C
v
). γ = 1.41 for hydrogen. The work of adiabatic
compression to a fixed final density is much larger than the work of isothermal compression
because the heat accumulated creates a higher pressure for the compressor to work against.
Both isothermal and adiabatic compression is plotted in figure 3 as a function of the final
pressure. Isothermal compression is the absolute minimum theoretically possible, and in
reality, due to the discussed heat effect compression is performed in multiple stages with
inter-cooling of the gas. Consequently, the work of compression lies somewhere between
the two curves. The efficiency of a compressor system varies a lot and the curve in figure 3 is
assuming a satisfactory compressor technology (Bossel et al., 2003).
0
10
20
30
40
50
60
0 100 200 300 400 500 600 700 800 900 1000
Final pressure (bar)
C
o
m
p
.

w
o
r
k

(
k
J
/
m
o
l
)
0
5
10
15
20
%

o
f

L
H
V
Adiabatic
Ideal isothermic
Real isothermic
Practical multistage

Fig. 3. The energy required to compress hydrogen from 1 bar to the final pressure specified
on the primary axis. Re-plotted from (Jensen et al., 2007)

4.2 Energy for release
One strong advantage of compressed hydrogen is that it is easily available at a pressure high
enough for fast transport through tubes. Even though the pressure vessel will cool during
The energy effciency of onboard hydrogen storage 147

propulsion) will be reduced comparatively and the effective storage capacity is thus lower than
predicted from the amount of hydrogen stored.
A true comparison would involve a detailed analysis of whole systems. Such analyses are
truly relevant but also complicated with numerous assumptions on which the outcome will
strongly depend. Instead, transparency is aimed at with the hope that the conclusions are
less questionable, although they do not tell the whole story. Throughout, the lower heating
value (LHV) of the fuel is used instead of the higher heating value (HHV). This is because in
several of the systems, heat for hydrogen liberation must be supplied at temperatures above
100ºC likely by combustion of hydrogen. It is also assumed that hydrogen or a hydrogen
mixture is released at no less than ambient pressure. The LHV used is 242.8 kJ/mol H
2
.

4. Compressed hydrogen
Despite many attempts to develop advanced techniques for compact, practical and safe
hydrogen storage, pressurization is still the dominating technique. This is a fact for onboard
hydrogen as well as for hydrogen storage in general. The standard pressure for steel
cylinders is 200 bar, but high pressure fibre composite tanks rated for up to 7-800 bar have
been developed. The gravimetric storage capacity ranges from 1-2 wt.% for 200 bar steel
tanks to 5-10 wt.% for high pressure fibre tanks. Fibre tanks are more expensive than steel
tanks.

4.1 Energy for storage
The theoretical minimum work needed for gas compression can be calculated based on
integration of the infinitesimal pressure-volume work, dw

Vdp dw 
(1)

where V is the tank volume and p the pressure. Assuming ideal gas behaviour integration of
(1) from p
0
to p
1
results in the expression of the work, W, of ideal isothermal compression










0
1
0
ln
p
p
V p W

(2)

where p
0
and p
1
are initial and final pressures. At hydrogen pressures over 100 bar,
deviations from ideality become significant in this connection, and the dimensionless
compression factor, Z, shall compensate for the non-ideality. The real gas equation is then

pV ZnRT  (3)

Z depends on both pressure and temperature and is tabulated elsewhere (Perry et al., 1984).
At 300 K and pressures up to 1000 bar, the compression factor is modelled well as









 

p
p
k Z
z 300 ,
1

(4)

where k
z,300
= 0.000631. Integration including (3) and (4) gives

(
(
¸
(


¸

|
|
.
|


\
|
+ ÷ =
0
1
0 1 300 , 0 0
ln ) (
p
p
p p k V p W
z

(5)

However, the compression is never isothermal, as heat is formed during the process. If the
compression is very slow, most heat will dissipate to the surroundings, but in practical high
pressure systems, a significant amount of heat is formed. The other extreme is adiabatic
compression in which all heat produced is kept in the gas by ideal insulation. The work of
adiabatic compression is

(
(
(
¸
(



¸

÷
|
|
.
|


\
|
÷
=
÷
1
1
1
0
1
0
¸
¸
¸
¸
p
p
V p W

(6)

where γ is the ratio of specific heats (C
p
/C
v
). γ = 1.41 for hydrogen. The work of adiabatic
compression to a fixed final density is much larger than the work of isothermal compression
because the heat accumulated creates a higher pressure for the compressor to work against.
Both isothermal and adiabatic compression is plotted in figure 3 as a function of the final
pressure. Isothermal compression is the absolute minimum theoretically possible, and in
reality, due to the discussed heat effect compression is performed in multiple stages with
inter-cooling of the gas. Consequently, the work of compression lies somewhere between
the two curves. The efficiency of a compressor system varies a lot and the curve in figure 3 is
assuming a satisfactory compressor technology (Bossel et al., 2003).
0
10
20
30
40
50
60
0 100 200 300 400 500 600 700 800 900 1000
Final pressure (bar)
C
o
m
p
.

w
o
r
k

(
k
J
/
m
o
l
)
0
5
10
15
20
%

o
f

L
H
V
Adiabatic
Ideal isothermic
Real isothermic
Practical multistage

Fig. 3. The energy required to compress hydrogen from 1 bar to the final pressure specified
on the primary axis. Re-plotted from (Jensen et al., 2007)

4.2 Energy for release
One strong advantage of compressed hydrogen is that it is easily available at a pressure high
enough for fast transport through tubes. Even though the pressure vessel will cool during
Energy Effciency 148

release, the pressure will in most cases still be way above ambient pressure. Therefore, no
energy is needed for the release. In principle, part of the compression energy can even be
reclaimed via an expander, but as it adds to complexity and cost it can be argued whether or
not it is feasible.

4.3 Discussion
The work of compression in real systems is estimated by Bossel et al (Bossel et al., 2003) and
Weindorf et al (Weindorf et al.,2003). According to these studies, compression to 800 bar is
possible using 18 % (Bossel) or 13 % (Weindorf) of LHV. The estimated curve for a real
system added to figure 3 is between these values. Compression to a final pressure of 800 bar
then costs 15.5 % of LHV.
One way to minimize the work of compression is to produce hydrogen by high-pressure
electrolysis. The extra voltage (corresponding to extra energy) for reducing hydrogen at
high pressure is close to the theoretical value because the reaction kinetics is very fast.
It is evident from equation (2) that the minimum ideal work of compression of one mole
hydrogen from 100 to 1000 bar is the same as from 1 to 10 bar. This means that there is a
significant benefit even if the electrolyser is operated at just 10-50 bar. Industrial
electrolysers working at 32 bar are commercially available today from Statoil (former Norsk
Hydro) and IHT.
There may be additional sources for spending energy during filling of the tank. Filling
stations often store the gas to higher pressure than used onboard in order to facilitate fast
transfer. This “over compression” will naturally lead to some losses. Moreover, cooling may
be applied during the transfer to the vehicle and that costs energy too. These effects are not
considered in the above calculations.

5. Liquefied hydrogen
Liquid hydrogen has the advantages that it is quite dense and that fuelling is fast and in
principle as easy as for gasoline. The main drawbacks are that liquefaction is very energy
intensive and that hydrogen continuously evaporates due to influx of heat. The latter can be
reduced to a few percent per day or less by advanced thermal insulation, but it will always
have to be dealt with. Liquid hydrogen tanks are high cost items and at present liquid
hydrogen are only available in selected countries.

5.1 Energy for storage
Some gasses like propane and butane can be liquefied at room temperature by compression to
moderate pressures. Unfortunately this is not the case for hydrogen (as well as for oxygen and
nitrogen). The reason is that the critical point is situated at a temperature lower than ambient
temperature. The critical point of hydrogen is at 33 K and 13 bar. At this point in the phase
diagram the gas-liquid equilibrium line ends and above the critical temperature the substance
will never liquefy, but will act as a compressed gas even at extreme pressure (eventually
hydrogen will form a solid, but at room temperature it requires several thousand bar). This
means that for the liquefaction of hydrogen cooling at minimum to 33 K is mandatory.
A simple theoretical pathway for liquid hydrogen is to cool it from room temperature (298
K) to the boiling point at 20 K and then condense it. The average heat capacity in the interval

is 28.48 J/mol K, and the heat of vaporization at 20 K is 892 J/mol H
2
(Values by Air
Liquide). Based on this, the minimum energy required is 8.81 kJ/mol H
2
. To this value 1.06
kJ/mol H
2
should be added for ortho-para conversion of the hydrogen (see below) and the
total theoretical enthalpy change is then 9.87 kJ/mol H
2
or 4 % of LHV. In reality the process
is quite complicated and much more energy intensive.
Pressurized nitrogen can be liquefied by Joule-Thomson expansion through a valve. The
case of hydrogen is more complicated since the Joule-Thomson coefficient [μ
JT
= (∂T/∂p)
H
]
for hydrogen is negative at room temperature. This means that the gas would heat up
instead of cooling. The final complication is that molecular hydrogen exists in two forms
called ortho and para-hydrogen depending on the nuclear spin being parallel (ortho-H
2
) or
anti-parallel (para-H
2
). At room temperature hydrogen contains 75% ortho and 25% para-H
2

and at 20 K the stable form is para-H
2
. Hydrogen converts slowly between the two forms
and this process is exothermic in the direction ortho-para (-1.06 kJ/mol H
2
at 20 K). Over
time the equilibrium composition will be reached, but this takes a time orders of magnitude
longer than the liquefaction process. Consequently, heat will be produced in the liquid
hydrogen if this is not dealt with in the liquefaction process. The ortho-para conversion can
be accelerated by a catalyst, but the heat produced adds to the amount of heat that has to be
removed in the liquefaction process.
Practical hydrogen liquefaction plants provide the initial cooling of moderately compressed
hydrogen by a conventional cooling system. Further cooling is done by liquid nitrogen
followed by a mechanical expander and finally a Joule-Thomson valve. At low temperature
the Joule-Thomson coefficient becomes positive and hydrogen cools and liquefies. Along
this pathway ortho-para conversion catalysts are inserted at selected locations.
Bossel et al. (Bossel et al., 2003) refer to a detailed analysis concluding that the theoretical
energy demand is 14.2 MJ/kg H
2
. This equals 28.4 kJ/mol H
2
or 11.7 % of LHV.
In addition, hydrogen needs to be purified to high purity prior to liquefaction because all
gaseous impurities will solidify at the low temperature and possibly block the Joule-
Thomson valve.

5.2 Energy for release
Because the temperature is very low compared to the surroundings, all heat for hydrogen
evaporation should be available. Nevertheless, in practical systems, a built-in electrical
heater is often used because heat transfer fluids freeze if passed through heat exchange
tubes in the tank. In this study it is assumed that heat from the surroundings is used.
However, if electrical heating is used the latent heat of 892 J/mol H
2
should be sufficient to
liberate hydrogen gas. Further heating can take place in the external tubing.

5.3 Discussion
The practical energy demand for liquefaction is significantly larger and depends on the size
of the plant. Today, the energy demand in a modern plant is on the order of 40-45 % of LHV,
but according to Bossel (Bossel et al., 2003), 25 % and to Weindorf (Weindorf et al., 2003),
21% of LHV should be possible in very large liquefaction plants.

The energy effciency of onboard hydrogen storage 149

release, the pressure will in most cases still be way above ambient pressure. Therefore, no
energy is needed for the release. In principle, part of the compression energy can even be
reclaimed via an expander, but as it adds to complexity and cost it can be argued whether or
not it is feasible.

4.3 Discussion
The work of compression in real systems is estimated by Bossel et al (Bossel et al., 2003) and
Weindorf et al (Weindorf et al.,2003). According to these studies, compression to 800 bar is
possible using 18 % (Bossel) or 13 % (Weindorf) of LHV. The estimated curve for a real
system added to figure 3 is between these values. Compression to a final pressure of 800 bar
then costs 15.5 % of LHV.
One way to minimize the work of compression is to produce hydrogen by high-pressure
electrolysis. The extra voltage (corresponding to extra energy) for reducing hydrogen at
high pressure is close to the theoretical value because the reaction kinetics is very fast.
It is evident from equation (2) that the minimum ideal work of compression of one mole
hydrogen from 100 to 1000 bar is the same as from 1 to 10 bar. This means that there is a
significant benefit even if the electrolyser is operated at just 10-50 bar. Industrial
electrolysers working at 32 bar are commercially available today from Statoil (former Norsk
Hydro) and IHT.
There may be additional sources for spending energy during filling of the tank. Filling
stations often store the gas to higher pressure than used onboard in order to facilitate fast
transfer. This “over compression” will naturally lead to some losses. Moreover, cooling may
be applied during the transfer to the vehicle and that costs energy too. These effects are not
considered in the above calculations.

5. Liquefied hydrogen
Liquid hydrogen has the advantages that it is quite dense and that fuelling is fast and in
principle as easy as for gasoline. The main drawbacks are that liquefaction is very energy
intensive and that hydrogen continuously evaporates due to influx of heat. The latter can be
reduced to a few percent per day or less by advanced thermal insulation, but it will always
have to be dealt with. Liquid hydrogen tanks are high cost items and at present liquid
hydrogen are only available in selected countries.

5.1 Energy for storage
Some gasses like propane and butane can be liquefied at room temperature by compression to
moderate pressures. Unfortunately this is not the case for hydrogen (as well as for oxygen and
nitrogen). The reason is that the critical point is situated at a temperature lower than ambient
temperature. The critical point of hydrogen is at 33 K and 13 bar. At this point in the phase
diagram the gas-liquid equilibrium line ends and above the critical temperature the substance
will never liquefy, but will act as a compressed gas even at extreme pressure (eventually
hydrogen will form a solid, but at room temperature it requires several thousand bar). This
means that for the liquefaction of hydrogen cooling at minimum to 33 K is mandatory.
A simple theoretical pathway for liquid hydrogen is to cool it from room temperature (298
K) to the boiling point at 20 K and then condense it. The average heat capacity in the interval

is 28.48 J/mol K, and the heat of vaporization at 20 K is 892 J/mol H
2
(Values by Air
Liquide). Based on this, the minimum energy required is 8.81 kJ/mol H
2
. To this value 1.06
kJ/mol H
2
should be added for ortho-para conversion of the hydrogen (see below) and the
total theoretical enthalpy change is then 9.87 kJ/mol H
2
or 4 % of LHV. In reality the process
is quite complicated and much more energy intensive.
Pressurized nitrogen can be liquefied by Joule-Thomson expansion through a valve. The
case of hydrogen is more complicated since the Joule-Thomson coefficient [μ
JT
= (∂T/∂p)
H
]
for hydrogen is negative at room temperature. This means that the gas would heat up
instead of cooling. The final complication is that molecular hydrogen exists in two forms
called ortho and para-hydrogen depending on the nuclear spin being parallel (ortho-H
2
) or
anti-parallel (para-H
2
). At room temperature hydrogen contains 75% ortho and 25% para-H
2

and at 20 K the stable form is para-H
2
. Hydrogen converts slowly between the two forms
and this process is exothermic in the direction ortho-para (-1.06 kJ/mol H
2
at 20 K). Over
time the equilibrium composition will be reached, but this takes a time orders of magnitude
longer than the liquefaction process. Consequently, heat will be produced in the liquid
hydrogen if this is not dealt with in the liquefaction process. The ortho-para conversion can
be accelerated by a catalyst, but the heat produced adds to the amount of heat that has to be
removed in the liquefaction process.
Practical hydrogen liquefaction plants provide the initial cooling of moderately compressed
hydrogen by a conventional cooling system. Further cooling is done by liquid nitrogen
followed by a mechanical expander and finally a Joule-Thomson valve. At low temperature
the Joule-Thomson coefficient becomes positive and hydrogen cools and liquefies. Along
this pathway ortho-para conversion catalysts are inserted at selected locations.
Bossel et al. (Bossel et al., 2003) refer to a detailed analysis concluding that the theoretical
energy demand is 14.2 MJ/kg H
2
. This equals 28.4 kJ/mol H
2
or 11.7 % of LHV.
In addition, hydrogen needs to be purified to high purity prior to liquefaction because all
gaseous impurities will solidify at the low temperature and possibly block the Joule-
Thomson valve.

5.2 Energy for release
Because the temperature is very low compared to the surroundings, all heat for hydrogen
evaporation should be available. Nevertheless, in practical systems, a built-in electrical
heater is often used because heat transfer fluids freeze if passed through heat exchange
tubes in the tank. In this study it is assumed that heat from the surroundings is used.
However, if electrical heating is used the latent heat of 892 J/mol H
2
should be sufficient to
liberate hydrogen gas. Further heating can take place in the external tubing.

5.3 Discussion
The practical energy demand for liquefaction is significantly larger and depends on the size
of the plant. Today, the energy demand in a modern plant is on the order of 40-45 % of LHV,
but according to Bossel (Bossel et al., 2003), 25 % and to Weindorf (Weindorf et al., 2003),
21% of LHV should be possible in very large liquefaction plants.

Energy Effciency 150

6. Adsorbed hydrogen
Like any gas, hydrogen can absorb on surfaces. The molecules are held by the weak van der
Waals forces which are much smaller than those of real chemical bonds. Equilibrium is
established between adsorbed and free molecules and the surface coverage increases with
the gas pressure and with decreasing temperature. Materials like active carbon, carbon
nano-tubes, zeolites and metal-organic frameworks have been studied for sorption
capacities. The binding energy is 1-10 kJ/mol H
2
. It is the general experience that at room
temperature and pressures in the range 50-100 bar only up to 1 wt.% hydrogen storage is
possible. At liquid nitrogen temperature (77 K) 4-6 wt.% has been reported by different
groups. These values refer to the high surface area material only and do not take into
account a pressure tank or the insulation in case of cryogenic sorption.
The heat for adsorption is limited compared to the hydrides (treated below). This is an
advantage when filling the tank since only 1-10 kJ/mol H
2
is released (0.4-4 % of LHV).
However, in the cryogenic case this heat must be removed at 77 K and that requires more
energy than for cooling at room temperature or at elevated temperature.
A detailed calculation of the energies for storage and release is very complex. The sorption
energy and the storage pressure can both vary by an order of magnitude and if liquid
nitrogen is applied it requires an additional energy contribution that too depends on the
conditions. When, on top of that, hydrogen storage by adsorption has not yet shown
advantages over the other techniques discussed here, the calculation was not attempted.

7. Reversible metal hydrides
The term “reversible hydride” refers to both interstitial and complex hydrides as long as
they can be charged as well as discharged by direct solid/gas reactions (or liquid/gas).
“Reversible” should not be understood in a thermodynamic sense in this context, it only
means “capable of reversing”.
Hydrogen stored in interstitial metal hydrides is bound into interstitial positions in a host
metal alloy in a more or less metallic way. This bond is stronger than the van der Waals
forces mentioned before and a significant amount of heat is required to release hydrogen.
In the complex hydrides or other real chemical systems chemical bonds ranging from ionic
to covalent are formed between hydrogen and the carrier atoms. The hydrogen release
reactions in these cases typically require a significant energy input and also elevated
temperatures to overcome the activation energy.

7.1 Interstitial hydrides
Interstitial hydrides are the most studied metal hydride systems for hydrogen storage. Examples
are plentiful such as LaNi
5
H
6
, TiFeH
~2
, and LaNi
5
-based alloys for nickel metal hydride batteries.
They are considered very safe and easy to operate, and their main drawback apart from the price
in some cases is the fact that the hydrogen storage capacity (with a few exceptions) is below 2 wt.
%. One convenient characteristic is that the alloys can be tailored to a moderate equilibrium
pressure of a few bars at ambient temperature. The heat of desorption is then around 30 kJ/mol
H
2
or 12 % of LHV. During charging, this heat is liberated. In small canisters, the heat can be
exchanged with the surroundings, but in larger systems like in a vehicle, active cooling by water

is necessary. The energy balance of such a cooling system depends highly on the charging rate
aimed at. Consequently, only the sorption energy is considered.
When hydrogen is liberated, the hydride cools and the plateau pressure must still be above
ambient pressure to avoid subsequent compression of the released hydrogen. This implies that
the plateau pressure will be correspondingly higher when the hydride is heating up during
charging and the charging pressure must match that. A 20-50 bar charging pressure can be
suggested. Based on the discussion above, compression to 20 bar is set to 4-5 % of LHV (or 3 %
with isothermal compression).
The amount of heat for desorption is the same as for absorption. It can be taken from the excess
heat of the fuel cell or combustion engine provided that the temperature is high enough. The
interstitial hydride can be designed for that.

7.2 Other reversible hydrides
Other reversible hydrides obey the same thermodynamic laws but possibly with other
pressure-temperature characteristics. They are not as easily tailored, and the reaction
enthalpy is generally more or less fixed.
Examples of other reversible metal hydrides are MgH
2
, Mg
2
NiH
4
and NaAlH
4
. The two
magnesium-based hosts are both characterized by one flat plateau, while NaAlH
4
desorbs
hydrogen in two steps with different stabilities (Bogdanovic, 2000). The first step is:

NaAlH
4
↔ 1/3Na
3
AlH
6
+ 2/3Al + H
2
(7)
and the second step is

Na
3
AlH
6
↔3NaH + Al + 3/2H
2


(8)

The key desorption properties of the mentioned hydrides are listed in table 1. The column
“Temperature for 1 bar” is based on thermodynamics. For kinetic reasons, NaAlH
4
needs
temperatures of around 150ºC even when Ti-doped. This means that the charging hydrogen
pressure must be on the order of 100 bar which, assuming isothermal compression, takes 4.8
% of LHV or practically 7-8 % of LHV in reality. The other systems can be charged at low
pressures like the interstitial hydrides.

Hydride Rev. capacity Heat of desorption Temperature
for 1 bar
Interstitial MH 1-2 wt.% ~ 30 kJ/mol H2 (~12.4 % of LHV) Near room
temperature
MgH2 7.6 wt.% 74.5 kJ/mol H2 (30.8 % of LHV) 300ºC
Mg2NiH4 3.6 wt.% 64.5 kJ/mol H2 (26.7 % of LHV) 255ºC
NaAlH4
(one step)
3.7 wt.% 37 kJ/mol H2 (15.3 % of LHV) 35ºC
Na3AlH6 1.9 wt.% 47 kJ/mol H2 (19.4 % of LHV) 110ºC
NaAlH4
(two steps)
5.6 wt.% 40 kJ/mol H2 (16.5 % of LHV) 110ºC
Table 1. Selected metal hydrides and their hydrogen storage properties.
The energy effciency of onboard hydrogen storage 151

6. Adsorbed hydrogen
Like any gas, hydrogen can absorb on surfaces. The molecules are held by the weak van der
Waals forces which are much smaller than those of real chemical bonds. Equilibrium is
established between adsorbed and free molecules and the surface coverage increases with
the gas pressure and with decreasing temperature. Materials like active carbon, carbon
nano-tubes, zeolites and metal-organic frameworks have been studied for sorption
capacities. The binding energy is 1-10 kJ/mol H
2
. It is the general experience that at room
temperature and pressures in the range 50-100 bar only up to 1 wt.% hydrogen storage is
possible. At liquid nitrogen temperature (77 K) 4-6 wt.% has been reported by different
groups. These values refer to the high surface area material only and do not take into
account a pressure tank or the insulation in case of cryogenic sorption.
The heat for adsorption is limited compared to the hydrides (treated below). This is an
advantage when filling the tank since only 1-10 kJ/mol H
2
is released (0.4-4 % of LHV).
However, in the cryogenic case this heat must be removed at 77 K and that requires more
energy than for cooling at room temperature or at elevated temperature.
A detailed calculation of the energies for storage and release is very complex. The sorption
energy and the storage pressure can both vary by an order of magnitude and if liquid
nitrogen is applied it requires an additional energy contribution that too depends on the
conditions. When, on top of that, hydrogen storage by adsorption has not yet shown
advantages over the other techniques discussed here, the calculation was not attempted.

7. Reversible metal hydrides
The term “reversible hydride” refers to both interstitial and complex hydrides as long as
they can be charged as well as discharged by direct solid/gas reactions (or liquid/gas).
“Reversible” should not be understood in a thermodynamic sense in this context, it only
means “capable of reversing”.
Hydrogen stored in interstitial metal hydrides is bound into interstitial positions in a host
metal alloy in a more or less metallic way. This bond is stronger than the van der Waals
forces mentioned before and a significant amount of heat is required to release hydrogen.
In the complex hydrides or other real chemical systems chemical bonds ranging from ionic
to covalent are formed between hydrogen and the carrier atoms. The hydrogen release
reactions in these cases typically require a significant energy input and also elevated
temperatures to overcome the activation energy.

7.1 Interstitial hydrides
Interstitial hydrides are the most studied metal hydride systems for hydrogen storage. Examples
are plentiful such as LaNi
5
H
6
, TiFeH
~2
, and LaNi
5
-based alloys for nickel metal hydride batteries.
They are considered very safe and easy to operate, and their main drawback apart from the price
in some cases is the fact that the hydrogen storage capacity (with a few exceptions) is below 2 wt.
%. One convenient characteristic is that the alloys can be tailored to a moderate equilibrium
pressure of a few bars at ambient temperature. The heat of desorption is then around 30 kJ/mol
H
2
or 12 % of LHV. During charging, this heat is liberated. In small canisters, the heat can be
exchanged with the surroundings, but in larger systems like in a vehicle, active cooling by water

is necessary. The energy balance of such a cooling system depends highly on the charging rate
aimed at. Consequently, only the sorption energy is considered.
When hydrogen is liberated, the hydride cools and the plateau pressure must still be above
ambient pressure to avoid subsequent compression of the released hydrogen. This implies that
the plateau pressure will be correspondingly higher when the hydride is heating up during
charging and the charging pressure must match that. A 20-50 bar charging pressure can be
suggested. Based on the discussion above, compression to 20 bar is set to 4-5 % of LHV (or 3 %
with isothermal compression).
The amount of heat for desorption is the same as for absorption. It can be taken from the excess
heat of the fuel cell or combustion engine provided that the temperature is high enough. The
interstitial hydride can be designed for that.

7.2 Other reversible hydrides
Other reversible hydrides obey the same thermodynamic laws but possibly with other
pressure-temperature characteristics. They are not as easily tailored, and the reaction
enthalpy is generally more or less fixed.
Examples of other reversible metal hydrides are MgH
2
, Mg
2
NiH
4
and NaAlH
4
. The two
magnesium-based hosts are both characterized by one flat plateau, while NaAlH
4
desorbs
hydrogen in two steps with different stabilities (Bogdanovic, 2000). The first step is:

NaAlH
4
↔ 1/3Na
3
AlH
6
+ 2/3Al + H
2
(7)
and the second step is

Na
3
AlH
6
↔3NaH + Al + 3/2H
2


(8)

The key desorption properties of the mentioned hydrides are listed in table 1. The column
“Temperature for 1 bar” is based on thermodynamics. For kinetic reasons, NaAlH
4
needs
temperatures of around 150ºC even when Ti-doped. This means that the charging hydrogen
pressure must be on the order of 100 bar which, assuming isothermal compression, takes 4.8
% of LHV or practically 7-8 % of LHV in reality. The other systems can be charged at low
pressures like the interstitial hydrides.

Hydride Rev. capacity Heat of desorption Temperature
for 1 bar
Interstitial MH 1-2 wt.% ~ 30 kJ/mol H2 (~12.4 % of LHV) Near room
temperature
MgH2 7.6 wt.% 74.5 kJ/mol H2 (30.8 % of LHV) 300ºC
Mg2NiH4 3.6 wt.% 64.5 kJ/mol H2 (26.7 % of LHV) 255ºC
NaAlH4
(one step)
3.7 wt.% 37 kJ/mol H2 (15.3 % of LHV) 35ºC
Na3AlH6 1.9 wt.% 47 kJ/mol H2 (19.4 % of LHV) 110ºC
NaAlH4
(two steps)
5.6 wt.% 40 kJ/mol H2 (16.5 % of LHV) 110ºC
Table 1. Selected metal hydrides and their hydrogen storage properties.
Energy Effciency 152

7.3 Energy for storage
The energy for storage is basically the energy for pressurizing hydrogen to the charging
pressure, i.e. 10-50 bar for interstitial metal hydrides and 100 bar for NaAlH
4
as explained
above. Energy for active cooling is not estimated, but will be relevant, especially in case of
fast charging.

7.4 Energy for release
The energy for hydrogen release is the heat of desorption as listed in Table 1.

7.5 Discussion
The significant amount of heat liberated during charging will practically need to be
removed actively in a vehicle size tank for an acceptable filling rate. There is a separate
energy balance for this, but since it can be done in many different ways and at different rates
it has been omitted in this context.

8. Irreversible hydrides
8.1 NaBH
4

NaBH
4
does not easily liberate hydrogen like the hydrides discussed so far, but it reacts with
water over a catalyst. NaBH
4
is stored in an alkaline aqueous solution in which it is stable.
When passed over a catalyst the following reaction takes place

NaBH
4
+ 2H
2
O →NaBO
2
+ 4H
2
+ Heat (9)

The reaction is exothermic with the enthalpy -212 kJ/mol NaBH
4
or -53 kJ/mol H
2
(22% of
LHV). The hydrogen storage capacity is 21.2 wt.% disregarding the water, but the practical
capacity is much lower due to the water. Besides the role as a reactive solvent the water also
acts as a heat sink for the heat liberated during the process. The system is commercialized by
Millennium Cell
®
, and several demo cars have been fitted with such a system. The concept
can also be used directly in alkaline fuel cells with the catalyst being the anode catalyst (Li et
al., 2003).

8.2 Energy for storage
Being irreversible, NaBH
4
must be regenerated through other chemical pathways. As a
minimum, the 212 kJ/mol must be supplied during that process, but the real number is
significantly larger and depends on how regeneration is done.

8.3 Energy for release
No energy is needed for hydrogen liberation apart for pumping the liquid or for active heat
management.

8.4 Discussion
The round trip energy efficiency of this hydrogen storage system will most likely exclude it
from any application in which energy efficiency matters.

9. Methanol and ammonia
In this group, the hydrogen evolution reactions are characterized by equlibria with both
reactants and products in the gas phase. There is no such thing as a desorption temperature
at which the hydrogen pressure is 1 bar. The minimum hydrogen release temperature is
therefore chosen as the temperature at which kinetics are reasonably fast and the
equilibrium is strongly in favour of hydrogen formation.
The liberated hydrogen is in these cases mixed with either carbon dioxide or nitrogen. This
fact affects the way a fuel cell is fuelled. As the fuel part of the mixture is consumed, the
inert gas fraction increases, and this dilution effect can lead to local starvation of the
electrode and poor performance (and lead to electrode degradation). To overcome this
problem, fuel is fed in excess of at least 20 % (This problem can to some extent apply to any
fuel cell operating below the boiling point of water because of water vapour accumulation
followed by condensation. However, it can be solved by eventual purging without large
losses). The over-stoichiometry is labelled λ. λ = 1 means strictly stoichiometric and λ = 1.2
means 20% excess. The 20 % excess fuel is normally combusted in a burner, and the
resulting heat can then be used for fuel processing.

9.1 Methanol (CH
3
OH)
In order to extract hydrogen from methanol it can be steam reformed according to

CH
3
OH + H
2
O → CO
2
+ 3H
2
(10)

The hydrogen storage capacity is 18.8 wt.% disregarding the mass of the water. The process
is fast at 230-250ºC with a suitable catalyst, and the equilibrium is strongly in favour of
hydrogen. The enthalpy of reaction at 250ºC is +58.7 kJ/mol CH
3
OH or +19.6 kJ/mol H
2
or
8.6 % of LHV of methanol. Prior to reforming, methanol and water must be evaporated and
this takes another +75.8 kJ/mol H
2
or 11.1 % of LHV of methanol. The total minimum
requirement is then 19.7 % of LHV of methanol.
However, the LHV of the produced hydrogen (725.4 kJ/mol 3H
2
) is slightly higher than that
of the methanol (685.5 kJ/mol CH
3
OH). Taking this upgrading of 39.9 kJ into consideration,
the expense for reforming is only 58.7 + 75.8 - 39.9 = 94.6 kJ/mol CH
3
OH, which is only 13.8
% of LHV of CH
3
OH. Moreover, the energy for evaporation can be taken from the waste
heat of a fuel cell provided it is operated above 100ºC. In that case, only 2.7 % of LHV is
needed for fuel processing. This should be easily obtained from the excess stoichiometry
assuming λ =1.2.

9.2 Ammonia (NH3)
Ammonia is sometimes considered as an attractive onboard hydrogen carrier because of its
high hydrogen content of 16.6 wt.%, the absence of carbon, and the easy storage. At room
temperature, its vapour pressure is less than 10 bar and consequently it can be stored as a
liquid at moderate pressure. The major drawbacks are its chemical properties and its
stability. It is corrosive and poisonous. As a base, it reacts with acids and it is therefore
considered a poison to PEM fuel cells because it reacts with the perfluorosulphonic acid
membrane even at levels of 10 ppm (Halseid et al., 2006). Solid oxide fuel cells are able to
run on ammonia.
The energy effciency of onboard hydrogen storage 153

7.3 Energy for storage
The energy for storage is basically the energy for pressurizing hydrogen to the charging
pressure, i.e. 10-50 bar for interstitial metal hydrides and 100 bar for NaAlH
4
as explained
above. Energy for active cooling is not estimated, but will be relevant, especially in case of
fast charging.

7.4 Energy for release
The energy for hydrogen release is the heat of desorption as listed in Table 1.

7.5 Discussion
The significant amount of heat liberated during charging will practically need to be
removed actively in a vehicle size tank for an acceptable filling rate. There is a separate
energy balance for this, but since it can be done in many different ways and at different rates
it has been omitted in this context.

8. Irreversible hydrides
8.1 NaBH
4

NaBH
4
does not easily liberate hydrogen like the hydrides discussed so far, but it reacts with
water over a catalyst. NaBH
4
is stored in an alkaline aqueous solution in which it is stable.
When passed over a catalyst the following reaction takes place

NaBH
4
+ 2H
2
O →NaBO
2
+ 4H
2
+ Heat (9)

The reaction is exothermic with the enthalpy -212 kJ/mol NaBH
4
or -53 kJ/mol H
2
(22% of
LHV). The hydrogen storage capacity is 21.2 wt.% disregarding the water, but the practical
capacity is much lower due to the water. Besides the role as a reactive solvent the water also
acts as a heat sink for the heat liberated during the process. The system is commercialized by
Millennium Cell
®
, and several demo cars have been fitted with such a system. The concept
can also be used directly in alkaline fuel cells with the catalyst being the anode catalyst (Li et
al., 2003).

8.2 Energy for storage
Being irreversible, NaBH
4
must be regenerated through other chemical pathways. As a
minimum, the 212 kJ/mol must be supplied during that process, but the real number is
significantly larger and depends on how regeneration is done.

8.3 Energy for release
No energy is needed for hydrogen liberation apart for pumping the liquid or for active heat
management.

8.4 Discussion
The round trip energy efficiency of this hydrogen storage system will most likely exclude it
from any application in which energy efficiency matters.

9. Methanol and ammonia
In this group, the hydrogen evolution reactions are characterized by equlibria with both
reactants and products in the gas phase. There is no such thing as a desorption temperature
at which the hydrogen pressure is 1 bar. The minimum hydrogen release temperature is
therefore chosen as the temperature at which kinetics are reasonably fast and the
equilibrium is strongly in favour of hydrogen formation.
The liberated hydrogen is in these cases mixed with either carbon dioxide or nitrogen. This
fact affects the way a fuel cell is fuelled. As the fuel part of the mixture is consumed, the
inert gas fraction increases, and this dilution effect can lead to local starvation of the
electrode and poor performance (and lead to electrode degradation). To overcome this
problem, fuel is fed in excess of at least 20 % (This problem can to some extent apply to any
fuel cell operating below the boiling point of water because of water vapour accumulation
followed by condensation. However, it can be solved by eventual purging without large
losses). The over-stoichiometry is labelled λ. λ = 1 means strictly stoichiometric and λ = 1.2
means 20% excess. The 20 % excess fuel is normally combusted in a burner, and the
resulting heat can then be used for fuel processing.

9.1 Methanol (CH
3
OH)
In order to extract hydrogen from methanol it can be steam reformed according to

CH
3
OH + H
2
O → CO
2
+ 3H
2
(10)

The hydrogen storage capacity is 18.8 wt.% disregarding the mass of the water. The process
is fast at 230-250ºC with a suitable catalyst, and the equilibrium is strongly in favour of
hydrogen. The enthalpy of reaction at 250ºC is +58.7 kJ/mol CH
3
OH or +19.6 kJ/mol H
2
or
8.6 % of LHV of methanol. Prior to reforming, methanol and water must be evaporated and
this takes another +75.8 kJ/mol H
2
or 11.1 % of LHV of methanol. The total minimum
requirement is then 19.7 % of LHV of methanol.
However, the LHV of the produced hydrogen (725.4 kJ/mol 3H
2
) is slightly higher than that
of the methanol (685.5 kJ/mol CH
3
OH). Taking this upgrading of 39.9 kJ into consideration,
the expense for reforming is only 58.7 + 75.8 - 39.9 = 94.6 kJ/mol CH
3
OH, which is only 13.8
% of LHV of CH
3
OH. Moreover, the energy for evaporation can be taken from the waste
heat of a fuel cell provided it is operated above 100ºC. In that case, only 2.7 % of LHV is
needed for fuel processing. This should be easily obtained from the excess stoichiometry
assuming λ =1.2.

9.2 Ammonia (NH3)
Ammonia is sometimes considered as an attractive onboard hydrogen carrier because of its
high hydrogen content of 16.6 wt.%, the absence of carbon, and the easy storage. At room
temperature, its vapour pressure is less than 10 bar and consequently it can be stored as a
liquid at moderate pressure. The major drawbacks are its chemical properties and its
stability. It is corrosive and poisonous. As a base, it reacts with acids and it is therefore
considered a poison to PEM fuel cells because it reacts with the perfluorosulphonic acid
membrane even at levels of 10 ppm (Halseid et al., 2006). Solid oxide fuel cells are able to
run on ammonia.
Energy Effciency 154

The process of ammonia splitting is endothermic:

2NH
3
→ N
2
+ 3H
2

+30.6 kJ/mol H
2
or 12.7 % of LHV
(11)

and thus high temperatures and low pressure favours hydrogen formation.
If the pressure is set to 1 bar and only a few pct. ammonia are accepted for a subsequent
clean-up process, then the temperature must be at least 300-400ºC, and the reaction heat
must be supplied at that temperature (practically the temperature can not be lower for
kinetic reasons either).
Ammonia synthesis is exothermic, and in principle, the heat produced can be utilized.
Today however, ammonia is manufactured from natural gas and nitrogen from the air, and
the plants are consuming energy. A minimum energy required for synthesis is not
estimated. Normally, it is manufactured from natural gas instead of primary hydrogen.
One approach addressing the safety issue is to store ammonia as a complex with a salt, e.g.,
MgCl
2
. This idea was recently presented as hydrogen storage in tablet form (Christensen et
al., 2005). Dry MgCl
2
can reversibly take up 6 molecules of ammonia, and the vapour
pressure of ammonia becomes many orders of magnitude smaller. Moreover, in contrast to
the hydrides, the complex can be stored in air with only a slow liberation of ammonia. The
complex contains 9.1 wt.% hydrogen. Liberation of ammonia is endothermic and the
enthalpy for the process is +43 kJ/mol H
2
(Christensen et al., 2005) or 17.8% of LHV. When
the enthalpy for ammonia splitting is added, the overall minimum energy is +75 kJ/mol H
2

or 30.5 % of LHV. Liberation of all NH
3
requires a temperature of 350ºC, although 2/3 of the
ammonia is liberated at 200ºC. The system is reversible with respect to ammonia storage,
but not with respect to hydrogen storage. Ammonia can be charged onboard, but hydrogen
cannot, and ammonia must be synthesized in a plant.

10. Heat available from fuel cells
Fuel cell systems operate with different efficiencies, but in most cases, at least 50 % of the
fuel energy is liberated as heat due to different losses, mainly in the fuel cell. This
corresponds to 120.9 kJ/mol H
2
(LHV) and is plentiful for any of the storage systems
discussed here, even with large transfer losses. The determining factor is the working
temperature of the fuel cell (or the exhaust temperature of the combustion engine) because
the heat must be delivered at the working temperature of the storage device. Today the PEM
fuel cell is almost exclusively considered for vehicles, due to, among other things, the low
working temperature of around 80ºC. This advantage becomes a disadvantage when heat is
needed at higher temperatures, but today there are PEM fuel cells with working
temperatures up to 200ºC (Li et al., 2009).

11. Conclusion /Comparison
In Table 2, the maximum hydrogen storage efficiencies of the different storage systems are
compared. For each system, the onboard hydrogen content is arbitrarily set to 100%.


The column “Energy available onboard” shall be understood as follows: The first percentage
listed is the amount available when only heat at 25°C can be used for hydrogen liberation.
The value in brackets (when relevant) is the percentage available if a heat source at higher
temperature is available. If heat is required, but not available at the working temperature,
fuel must be burned to provide it, thus limiting the practical storage capacity.
The column “Energy used for storage process” lists energy consumption for storing the
hydrogen (e.g. for compression or charging a hydride). This is in principle an off board
energy consumption that does not affect the amount of hydrogen stored. The two sub-
columns contain minimum theoretical values and estimated practical values.
The column “Round trip efficiency” gives the overall efficiency for storing and liberating
hydrogen.
In all cases energy consumptions due to thermal losses, pumping and the like are omitted.

System Energy available
onboard
(% of stored H2)
Energy used for storage
process
(% of LHV of stored H2)
Round trip efficiency
(based on the assumptions made )
At 25ºC

Theoretical
minimum
Practical Theoretical
minimum
Practical
Compressed
200 bar
100% 5.4% 10.0% 94.9% 91%
Compressed
800 bar
100% 7.3% 15.5% 93.2% 87%
Liquid
hydrogen
~100% 11.8% 40% 89.4% 71%
Interstitial
hydride
88%
(100% @ ~50ºC)
3% 5% 85.4%
(97% at 50ºC)
84%
(95% at 50ºC)
MgH2 69.2%
(100% @ 300ºC)
0% 5% 69.2%
(100% @ 300°C)
66%
(95% @ 300°C)
Mg2NiH4 73.3%
(100% @ 255ºC)
0% 5% 73.3%
(100% @ 255°C)
70%
(95% @ 255°C)
NaAlH4
(3H available)
83.5%
(100% @ 150ºC)
4.8% 8% 79.7%
(95% @ 150°C)
77%
(93% @ 150°C)
NaBH4 +
water
100% 22% Very much 82% Very low
Methanol 81.4%
(92.5% @100ºC)
(100% @ 250ºC)
- - - -
Ammonia 87.3%
(100% @ 350ºC)
- - - -
Mg(NH3)6Cl2 69.5%
(100% @ 350ºC)
- - - -
Table 2. A summary of the efficiencies for storage and release of hydrogen. Note that heat
transfer losses and pumping might reduce the practical energy efficiency further.

12. Acknowledgements
The authors wish to thank Nordic Energy Research Programme and Danish Energy
Authority for the financial support of the present work.
The energy effciency of onboard hydrogen storage 155

The process of ammonia splitting is endothermic:

2NH
3
→ N
2
+ 3H
2

+30.6 kJ/mol H
2
or 12.7 % of LHV
(11)

and thus high temperatures and low pressure favours hydrogen formation.
If the pressure is set to 1 bar and only a few pct. ammonia are accepted for a subsequent
clean-up process, then the temperature must be at least 300-400ºC, and the reaction heat
must be supplied at that temperature (practically the temperature can not be lower for
kinetic reasons either).
Ammonia synthesis is exothermic, and in principle, the heat produced can be utilized.
Today however, ammonia is manufactured from natural gas and nitrogen from the air, and
the plants are consuming energy. A minimum energy required for synthesis is not
estimated. Normally, it is manufactured from natural gas instead of primary hydrogen.
One approach addressing the safety issue is to store ammonia as a complex with a salt, e.g.,
MgCl
2
. This idea was recently presented as hydrogen storage in tablet form (Christensen et
al., 2005). Dry MgCl
2
can reversibly take up 6 molecules of ammonia, and the vapour
pressure of ammonia becomes many orders of magnitude smaller. Moreover, in contrast to
the hydrides, the complex can be stored in air with only a slow liberation of ammonia. The
complex contains 9.1 wt.% hydrogen. Liberation of ammonia is endothermic and the
enthalpy for the process is +43 kJ/mol H
2
(Christensen et al., 2005) or 17.8% of LHV. When
the enthalpy for ammonia splitting is added, the overall minimum energy is +75 kJ/mol H
2

or 30.5 % of LHV. Liberation of all NH
3
requires a temperature of 350ºC, although 2/3 of the
ammonia is liberated at 200ºC. The system is reversible with respect to ammonia storage,
but not with respect to hydrogen storage. Ammonia can be charged onboard, but hydrogen
cannot, and ammonia must be synthesized in a plant.

10. Heat available from fuel cells
Fuel cell systems operate with different efficiencies, but in most cases, at least 50 % of the
fuel energy is liberated as heat due to different losses, mainly in the fuel cell. This
corresponds to 120.9 kJ/mol H
2
(LHV) and is plentiful for any of the storage systems
discussed here, even with large transfer losses. The determining factor is the working
temperature of the fuel cell (or the exhaust temperature of the combustion engine) because
the heat must be delivered at the working temperature of the storage device. Today the PEM
fuel cell is almost exclusively considered for vehicles, due to, among other things, the low
working temperature of around 80ºC. This advantage becomes a disadvantage when heat is
needed at higher temperatures, but today there are PEM fuel cells with working
temperatures up to 200ºC (Li et al., 2009).

11. Conclusion /Comparison
In Table 2, the maximum hydrogen storage efficiencies of the different storage systems are
compared. For each system, the onboard hydrogen content is arbitrarily set to 100%.


The column “Energy available onboard” shall be understood as follows: The first percentage
listed is the amount available when only heat at 25°C can be used for hydrogen liberation.
The value in brackets (when relevant) is the percentage available if a heat source at higher
temperature is available. If heat is required, but not available at the working temperature,
fuel must be burned to provide it, thus limiting the practical storage capacity.
The column “Energy used for storage process” lists energy consumption for storing the
hydrogen (e.g. for compression or charging a hydride). This is in principle an off board
energy consumption that does not affect the amount of hydrogen stored. The two sub-
columns contain minimum theoretical values and estimated practical values.
The column “Round trip efficiency” gives the overall efficiency for storing and liberating
hydrogen.
In all cases energy consumptions due to thermal losses, pumping and the like are omitted.

System Energy available
onboard
(% of stored H2)
Energy used for storage
process
(% of LHV of stored H2)
Round trip efficiency
(based on the assumptions made )
At 25ºC

Theoretical
minimum
Practical Theoretical
minimum
Practical
Compressed
200 bar
100% 5.4% 10.0% 94.9% 91%
Compressed
800 bar
100% 7.3% 15.5% 93.2% 87%
Liquid
hydrogen
~100% 11.8% 40% 89.4% 71%
Interstitial
hydride
88%
(100% @ ~50ºC)
3% 5% 85.4%
(97% at 50ºC)
84%
(95% at 50ºC)
MgH2 69.2%
(100% @ 300ºC)
0% 5% 69.2%
(100% @ 300°C)
66%
(95% @ 300°C)
Mg2NiH4 73.3%
(100% @ 255ºC)
0% 5% 73.3%
(100% @ 255°C)
70%
(95% @ 255°C)
NaAlH4
(3H available)
83.5%
(100% @ 150ºC)
4.8% 8% 79.7%
(95% @ 150°C)
77%
(93% @ 150°C)
NaBH4 +
water
100% 22% Very much 82% Very low
Methanol 81.4%
(92.5% @100ºC)
(100% @ 250ºC)
- - - -
Ammonia 87.3%
(100% @ 350ºC)
- - - -
Mg(NH3)6Cl2 69.5%
(100% @ 350ºC)
- - - -
Table 2. A summary of the efficiencies for storage and release of hydrogen. Note that heat
transfer losses and pumping might reduce the practical energy efficiency further.

12. Acknowledgements
The authors wish to thank Nordic Energy Research Programme and Danish Energy
Authority for the financial support of the present work.
Energy Effciency 156

13. References
Bogdanovic, B.; Brand, R. A.; Marjanovic, A.; Schwickardi, M. & Tolle, J (2000). Metal-doped
sodium aluminium hydrides as potential new hydrogen storage materials. J. Alloys
Comp. 302, 36-58.
Bossel, U.; Eliasson, B. & Taylor, G. (2003). The Future of the Hydrogen Economy: Bright or
Bleak? Version of 15. April 2003 updated for distribution at the 2003 Fuel Cell
Seminar 3 – 7 November 2003. www.efcf.com/reports.
Christensen, C. H.; Sørensen, R. Z.; Johannessen, T. ; Quaade, U. J.; Honkala, K.; Elmoe, T.
D.; Køhler, R. and Jens K. Nørskov (2005). Metal ammine complexes for hydrogen
storage. J. Mater. Chem. 15, 4106–4108
Halseid, R.; Vie, P. J. S. & Tunold, R. (2006). Effect of ammonia on the performance of
polymer electrolyte membrane fuel cells. J. Power Sources 154, 343–350.
Jensen, J. O.; Vestbø, A. P.; Li, Q. & Bjerrum, N. J. (2007). The Energy Efficiency of Onboard
Hydrogen Storage. J. Alloys Comp. 446–447 723–728
Leon, A. (2008). Hydrogen technology. Mobile and portable applications. Springer, ISBN:
978-3-540-79027-3, Berlin, Heidelberg.
Li, Q; Jensen, J. O.; Savinell, R. F. & Bjerrum, N. J. (2009). Acid-doped polybenzimidazole
(PBI) membranes for high temperature proton exchange membrane fuel cells. Prog.
Polymer Sci. 34, 449-477.
Li, Z. P.; Liu, B. H.; Arai, K.; Morigazaki, N. & Suda, S. (2003). Protide compounds in
hydrogen storage systems. J. Alloys Comp. 356-357 469-474.
Perry, R. H. & Green, D. W. (1984). Perry’s Chemical Engineers’ Handbook, 6. ed. Mc.Graw-
Hill.
Weindorf, W.; Bünger, U. & Schindler, J. (2003). Comments on the paper by Baldur Eliasson
and Ulf Bossel “The Future of the Hydrogen Economy: Bright or Bleak?”
L-B-Systemtechnik GmbH. On www.hyweb.de 15-07-2003.
Züttel, A.; Borgschulte, A. & Schlapbach, L. (2008). Hydrogen as an energy carrier. Wiley,
ISBN: 978-3-527-30817-0, Weinheim.
Energy effciency of Fuel Processor – PEM Fuel Cell systems 157
Energy effciency of Fuel Processor – PEM Fuel Cell systems
Lucia Salemme, Laura Menna and Marino Simeone
x

Energy efficiency of Fuel
Processor – PEM Fuel Cell systems

Lucia Salemme, Laura Menna and Marino Simeone
University of Naples “Federico II”, Department of Chemical Engineering
Italy

1. Introduction
As the world moved into the 21
st
century, a rapid development in industrial and
transportation sectors and improvements in living standards have been observed, leading to
a strong growth in the energy demand and in global emissions (Song, 2002). In this context,
fuel cell technology has been receiving an increasing attention, thanks to its lower emissions
and potentially higher energy efficiency if compared with internal combustion engines. A
fuel cell is defined as an electrochemical device in which the chemical energy stored in a fuel
is converted directly into electricity. Among all fuel cells, low temperature Proton Exchange
Membrane Fuel Cells (PEMFC) are promising devices for decentralized energy production,
both in stationary and automotive field, thanks to high compactness, low weight (high
power-to-weight ratio), high modularity and efficiency, fast start-up and response to load
changes.
The ideal fuel for PEMFC is hydrogen with low carbon monoxide content to avoid
poisoning of the fuel cell; in this way, PEMFC can achieve efficiency up to 60%, far higher if
compared to 20-35% efficiency of an internal combustion engine.
Hydrogen, though, is not a primary source. It is substantially an energy carrier, that can be
stored, transported and used as gaseous fuel, but, it needs to be produced from other fuels.
Today most of the hydrogen produced is obtained by hydrocarbons in large industrial
plants through the well-known Steam Reforming and Autothermal Reforming processes.
However, hydrogen distribution from industrial production plants to small-scale users
meets some limitations related to difficulties in hydrogen storage and transport. For its
chemical and physical properties, indeed, the development of an hydrogen infrastructure
seems to be not feasible in short term, while more reasonable seems to be the concept of
decentralized hydrogen production; in this way, an hydrogen source, such as methane, is
distributed through pipelines to the small-scale plant, placed nearby users, and the in situ
produced hydrogen is fed directly to the energy production system, avoiding hydrogen
storage and transportation. In this sense, a compact fuel processor, capable of generating a
hydrogen rich stream from an easily transportable fuel, is a potential root to accelerate
PEMFC deployment in the near future.
A typical fuel processor is constituted by a reforming unit coupled with a CO clean-up
section, introduced to guarantee hydrogen production with a CO content compatible with
9
Energy Effciency 158

PEMFC specifics. In particular, two different kinds of fuel processor are most frequently
described in the scientific literature; a conventional one, in which the reforming unit is
followed by two water gas shift (WGS) reactors and a preferential CO oxidation (PROX)
reactor (Ersoz et al, 2006), and an innovative one, in which the reforming unit is coupled
with highly selective hydrogen membranes to produce pure hydrogen, allowing to operate
the PEMFC without a purge stream, generally named as anode off-gas (Lattner et al, 2004).
The global energy efficiency of these systems strictly depends on fuel processor
configuration and on operating conditions; therefore, a comprehensive simulative analysis
of fuel processors coupled with a PEMFC can contribute to identify the conditions that
maximize system performance.
The following paragraphs provide a detailed description of conventional and membrane-
based fuel processors. In particular, section 2.1 describes the conventional fuel processors,
with details on the reforming technologies and on the typical CO clean-up techniques, while
section 2.2 describes innovative fuel processor and membrane technology. Section 2.3
reviews the state of art of the analysis of fuel processor – PEMFC system. Section 3 and 4
report the methodology employed to simulate system performance and the results obtained,
respectively. Finally, section 5 draws the main conclusions on the energy efficiency analysis
presented.

2. Fuel Processor - PEMFC systems
2.1 Conventional Fuel Processors
Fig. 1 shows the scheme of a conventional fuel processor for hydrogen production from
methane, which consists of a desulfurization unit (Des), a syngas production section and a
CO clean-up section.


Fig. 1. Conventional Fuel Processor

The desulfurization section is required to lower the sulfur content of the fuel to 0.2 ppm,
both for environmental and catalysts restrictions; it generally consists of an
hydrodesulphurization reactor, where hydrogen added to the fuel reacts with the sulfur
compounds to form H
2
S, followed by an adsorption bed to remove H
2
S.
The desulfurization process is a quite mature technology and its optimization is essentially
related to the catalytic system and it will not be analyzed further. A comprehensive
treatment of this unit can be found in Lampert et al, 2004.
The syngas production section is generally used to convert the fuel into syngas, a mixture of
H
2
and CO. Two main syngas production technologies are generally employed: Steam
Des SR/ATR HTS LTS PrOx
Burner
Fuel
Air
Q
SYNGAS PRODUCTION CO CLEAN-UP

Reforming and Autothermal Reforming. The thermodynamic analysis of reforming
processes is widely discussed in the literature (Seo et al, 2002), as well as the optimization of
catalyst formulation and operating conditions that maximize process performance (Xu et al,
2006, Semelsberger et al, 2004).
The Steam Reforming process is realized by feeding methane and steam to a catalytic
reactor, where the following reactions take place:
1) CH
4
+ H
2
O = CO + 3H
2
ΔH
o
R
= 49 Kcal/mol CH
4
2) CO + H
2
O = CO
2
+ H
2
ΔH
o
R
= -9.8 Kcal/mol CO

3) CH
4
= C + 2H
2
ΔH
o
R
= 18 Kcal/ mol CH
4

The operative parameters that influence this process are: pressure (P), temperature (T
SR
) and
steam to methane ratio (H
2
O/CH
4
) in the feed.
By observing reactions 1, 2 and 3, the reader will be easily convinced that the process occurs
with an increment of number of moles; therefore it is favored by low pressures.
The process is globally endothermic and it is favored by high temperatures. The heat
required for the reaction is supplied by an external burner fed with additional fuel and air.
Usually, reactor temperature does not exceed 800°C ca. due to catalyst and construction
materials constraints.
The value of H
2
O/CH
4
employed is usually higher than 2 (stoichiometric value), to reduce
coke formation and lower than 4, to limit operative cost and reactor size.
Due to its high selectivity and to the high concentration of hydrogen in the product stream,
steam reforming is the most common process to produce hydrogen from hydrocarbons.
However, when looked at from a “decentralized hydrogen production” perspective, it
shows some disadvantages essentially because of reduced compactness and slow response
to load changes. Both aspects should be attributed to the endothermicity of the reaction and
to the high residence times required.
Auto thermal Reforming is obtained by adding air to the inlet SR mixture; in this way, the
heat for the endothermic reforming reactions is supplied by the oxidation of part of the
methane inside the reactor itself.
The amount of air must be such that the energy generated by the oxidation reactions
balances the energy requirement of the reforming reaction, maintaining reactor temperature
to typical SR values (600-800°C).
The internal heat generation offers advantages in terms of reactor size and start up times;
however, the addition of air to the feed lowers hydrogen concentration in the reformate
stream due to the presence of large amounts of nitrogen, fed to the reactor as air.
Either through Steam reforming or Autothermal reforming, the outlet of the reactor has
potential of further hydrogen production. Indeed, being reaction 2 exothermic, it is limited
by the high temperatures typical of the reforming reactor. For this reason, another reaction
step usually follows the main reforming reactor and reduces CO content to less than 10
ppm. This CO clean-up section is constituted by two water gas shift (WGS) reactors and a
preferential CO oxidation (PROX) reactor.
The WGS process is a well-known technology, where the following reaction takes place:
4) CO + H
2
O = CO
2
+ H
2
ΔH°
R
= -9.8 Kcal/mol CO
WGS is realized in two stages with inter-cooling; the first stage is generally operated at 350-
420°C and is referred to as “high temperature stage” (HTS), whereas the second stage is
operated at 200-220°C and is referred to as “low temperature stage” (LTS). The outlet CO
concentration from LTS is 0.2 - 0.5% ca. and a further CO conversion stage must be present
Energy effciency of Fuel Processor – PEM Fuel Cell systems 159

PEMFC specifics. In particular, two different kinds of fuel processor are most frequently
described in the scientific literature; a conventional one, in which the reforming unit is
followed by two water gas shift (WGS) reactors and a preferential CO oxidation (PROX)
reactor (Ersoz et al, 2006), and an innovative one, in which the reforming unit is coupled
with highly selective hydrogen membranes to produce pure hydrogen, allowing to operate
the PEMFC without a purge stream, generally named as anode off-gas (Lattner et al, 2004).
The global energy efficiency of these systems strictly depends on fuel processor
configuration and on operating conditions; therefore, a comprehensive simulative analysis
of fuel processors coupled with a PEMFC can contribute to identify the conditions that
maximize system performance.
The following paragraphs provide a detailed description of conventional and membrane-
based fuel processors. In particular, section 2.1 describes the conventional fuel processors,
with details on the reforming technologies and on the typical CO clean-up techniques, while
section 2.2 describes innovative fuel processor and membrane technology. Section 2.3
reviews the state of art of the analysis of fuel processor – PEMFC system. Section 3 and 4
report the methodology employed to simulate system performance and the results obtained,
respectively. Finally, section 5 draws the main conclusions on the energy efficiency analysis
presented.

2. Fuel Processor - PEMFC systems
2.1 Conventional Fuel Processors
Fig. 1 shows the scheme of a conventional fuel processor for hydrogen production from
methane, which consists of a desulfurization unit (Des), a syngas production section and a
CO clean-up section.


Fig. 1. Conventional Fuel Processor

The desulfurization section is required to lower the sulfur content of the fuel to 0.2 ppm,
both for environmental and catalysts restrictions; it generally consists of an
hydrodesulphurization reactor, where hydrogen added to the fuel reacts with the sulfur
compounds to form H
2
S, followed by an adsorption bed to remove H
2
S.
The desulfurization process is a quite mature technology and its optimization is essentially
related to the catalytic system and it will not be analyzed further. A comprehensive
treatment of this unit can be found in Lampert et al, 2004.
The syngas production section is generally used to convert the fuel into syngas, a mixture of
H
2
and CO. Two main syngas production technologies are generally employed: Steam
Des SR/ATR HTS LTS PrOx
Burner
Fuel
Air
Q
SYNGAS PRODUCTION CO CLEAN-UP

Reforming and Autothermal Reforming. The thermodynamic analysis of reforming
processes is widely discussed in the literature (Seo et al, 2002), as well as the optimization of
catalyst formulation and operating conditions that maximize process performance (Xu et al,
2006, Semelsberger et al, 2004).
The Steam Reforming process is realized by feeding methane and steam to a catalytic
reactor, where the following reactions take place:
1) CH
4
+ H
2
O = CO + 3H
2
ΔH
o
R
= 49 Kcal/mol CH
4
2) CO + H
2
O = CO
2
+ H
2
ΔH
o
R
= -9.8 Kcal/mol CO

3) CH
4
= C + 2H
2
ΔH
o
R
= 18 Kcal/ mol CH
4

The operative parameters that influence this process are: pressure (P), temperature (T
SR
) and
steam to methane ratio (H
2
O/CH
4
) in the feed.
By observing reactions 1, 2 and 3, the reader will be easily convinced that the process occurs
with an increment of number of moles; therefore it is favored by low pressures.
The process is globally endothermic and it is favored by high temperatures. The heat
required for the reaction is supplied by an external burner fed with additional fuel and air.
Usually, reactor temperature does not exceed 800°C ca. due to catalyst and construction
materials constraints.
The value of H
2
O/CH
4
employed is usually higher than 2 (stoichiometric value), to reduce
coke formation and lower than 4, to limit operative cost and reactor size.
Due to its high selectivity and to the high concentration of hydrogen in the product stream,
steam reforming is the most common process to produce hydrogen from hydrocarbons.
However, when looked at from a “decentralized hydrogen production” perspective, it
shows some disadvantages essentially because of reduced compactness and slow response
to load changes. Both aspects should be attributed to the endothermicity of the reaction and
to the high residence times required.
Auto thermal Reforming is obtained by adding air to the inlet SR mixture; in this way, the
heat for the endothermic reforming reactions is supplied by the oxidation of part of the
methane inside the reactor itself.
The amount of air must be such that the energy generated by the oxidation reactions
balances the energy requirement of the reforming reaction, maintaining reactor temperature
to typical SR values (600-800°C).
The internal heat generation offers advantages in terms of reactor size and start up times;
however, the addition of air to the feed lowers hydrogen concentration in the reformate
stream due to the presence of large amounts of nitrogen, fed to the reactor as air.
Either through Steam reforming or Autothermal reforming, the outlet of the reactor has
potential of further hydrogen production. Indeed, being reaction 2 exothermic, it is limited
by the high temperatures typical of the reforming reactor. For this reason, another reaction
step usually follows the main reforming reactor and reduces CO content to less than 10
ppm. This CO clean-up section is constituted by two water gas shift (WGS) reactors and a
preferential CO oxidation (PROX) reactor.
The WGS process is a well-known technology, where the following reaction takes place:
4) CO + H
2
O = CO
2
+ H
2
ΔH°
R
= -9.8 Kcal/mol CO
WGS is realized in two stages with inter-cooling; the first stage is generally operated at 350-
420°C and is referred to as “high temperature stage” (HTS), whereas the second stage is
operated at 200-220°C and is referred to as “low temperature stage” (LTS). The outlet CO
concentration from LTS is 0.2 - 0.5% ca. and a further CO conversion stage must be present
Energy Effciency 160

before the mixture can be fed to a PEMFC. In conventional fuel processors, CO is reduced to
less than 50 ppm in a preferential CO Oxidation (PrOx) stage. The reactor is generally
adiabatic and catalyst as well as operating conditions must be carefully chosen, in order to
promote CO conversion whilst keeping hydrogen oxidation limited. This CO purification
technology is mature and well defined, although it has disadvantage in terms of
compactness and catalyst deactivation.
The stream leaving the fuel processor is generally named as reformate and contains the
hydrogen produced, as well as CO
2
, H
2
O, unreacted CH
4
and N
2
. This stream is ready to be
sent to a PEMFC.

2.2 Innovative Fuel Processors
Innovative Fuel Processors are characterized by the employment of a membrane reactor, in
which a high selective hydrogen separation membrane is coupled with a catalytic reactor to
produce pure hydrogen.
A typical membrane reactor is constituted by two co-axial tubes, with the internal one being
the hydrogen separation membrane; generally, the reaction happens in the annulus and the
permeate hydrogen flows in the inner tube.
The stream leaving the reaction is named retentate and the stream permeated through the
membrane is named permeate.
Membrane reactor is illustrated in Fig. 2 for the following generic reaction:
A + B = C + H
2

The membrane continuously removes the H
2
produced in the reaction zone, thus shifting
the chemical equilibrium towards the products; this allows obtaining higher conversions of
reactants to hydrogen with respect to a conventional reactor, working in the same operating
conditions.
A typical membrane used to separate hydrogen from a gas mixture is a Palladium or a
Palladium alloy membrane (Shu et al., 1991); this kind of membrane is able to separate
hydrogen with selectivity close to 100%. Hydrogen permeation through Palladium
membranes happens according to a solution/diffusion mechanism and the hydrogen flux
through the membrane, J
H2
is described by the following law:

 
P H2, R H2,
H2
H2
P P
δ
A
J  


(1)

where 
H2
is the permeability coefficient [mol/(m
2
s Pa
0.5
)], A is the membrane surface area
[m
2
], δ is the membrane thickness [m] and P
H2,R
and P
H2,P
are hydrogen partial pressures
[kPa] on the retentate side and on the permeate side of the membrane, respectively. Eq. 1 is
known as Sievert’s law and it is valid if the bulk phase diffusion of atomic hydrogen is the
rate limiting step in the hydrogen permeation process.
To increase the separation driving force, usually the retentate is kept at higher pressure than
the permeate. In common applications, permeate pressure is atmospheric and retentate
pressure is in the range 10-15 atm (compatibly with mechanical constraints).
A possible way to further increase the separation driving force is to reduce hydrogen partial
pressure in the permeate (P
H2,P
) by diluting the permeate stream with sweep gas (usually
superheated steam).

Sievert’s law shows that an increase of the hydrogen flux is achieved with reducing
membrane thickness. Palladium membranes should not be far thinner than 80-100 μm due
to mechanical stability of the layer and to the presence of defects and pinholes that reduce
hydrogen selectivity. To overcome this problem, current technologies foresee a thin layer
(20-50 μm) of Pd deposited on a porous ceramic or metal substrate.
Another important issue of Pd membranes (pure or supported) is thermal resistance.
Temperature should not be less than 200°C, to prevent hydrogen embrittlement and not
higher than 600°C ca. to prevent material damage.


Fig. 2. Membrane Reactor

Innovative fuel processors can be realized by combining the membrane either with the
reforming unit, generating the fuel processor reported in Fig. 3 (FP.1), or with a water gas
shift unit, generating the fuel processor reported in Fig. 3 (FP.2).


Fig. 3. Innovative Fuel Processors

FP.1 consists of a desulfurization unit followed by a membrane reforming reactor, with a
burner. This solution guarantees the highest compactness in terms of number of units, since
it allows to totally suppress the CO clean-up section; indeed, when the membrane is
integrated in the reforming reactor, the permeate stream is pure hydrogen, that can be
directly fed to a PEMFC.
A, B, C, H
2
RETENTATE
H
2
PERMEATE
A, B
H
H
H
H
A + B = C + H
2
REACTION SIDE MEMBRANE SIDE
Des
MEMBRANE
SR/ATR REACTOR
Burner
Fuel
Air
Q
R
e
t
e
n
t
a
t
e
H
2
FP.1
Des
MEMBRANE
WGS REACTOR
Burner
Air
Fuel
Q
R
e
t
e
n
t
a
t
e
H
2
SR/ATR
FP.2
Energy effciency of Fuel Processor – PEM Fuel Cell systems 161

before the mixture can be fed to a PEMFC. In conventional fuel processors, CO is reduced to
less than 50 ppm in a preferential CO Oxidation (PrOx) stage. The reactor is generally
adiabatic and catalyst as well as operating conditions must be carefully chosen, in order to
promote CO conversion whilst keeping hydrogen oxidation limited. This CO purification
technology is mature and well defined, although it has disadvantage in terms of
compactness and catalyst deactivation.
The stream leaving the fuel processor is generally named as reformate and contains the
hydrogen produced, as well as CO
2
, H
2
O, unreacted CH
4
and N
2
. This stream is ready to be
sent to a PEMFC.

2.2 Innovative Fuel Processors
Innovative Fuel Processors are characterized by the employment of a membrane reactor, in
which a high selective hydrogen separation membrane is coupled with a catalytic reactor to
produce pure hydrogen.
A typical membrane reactor is constituted by two co-axial tubes, with the internal one being
the hydrogen separation membrane; generally, the reaction happens in the annulus and the
permeate hydrogen flows in the inner tube.
The stream leaving the reaction is named retentate and the stream permeated through the
membrane is named permeate.
Membrane reactor is illustrated in Fig. 2 for the following generic reaction:
A + B = C + H
2

The membrane continuously removes the H
2
produced in the reaction zone, thus shifting
the chemical equilibrium towards the products; this allows obtaining higher conversions of
reactants to hydrogen with respect to a conventional reactor, working in the same operating
conditions.
A typical membrane used to separate hydrogen from a gas mixture is a Palladium or a
Palladium alloy membrane (Shu et al., 1991); this kind of membrane is able to separate
hydrogen with selectivity close to 100%. Hydrogen permeation through Palladium
membranes happens according to a solution/diffusion mechanism and the hydrogen flux
through the membrane, J
H2
is described by the following law:

 
P H2, R H2,
H2
H2
P P
δ
A
J  


(1)

where 
H2
is the permeability coefficient [mol/(m
2
s Pa
0.5
)], A is the membrane surface area
[m
2
], δ is the membrane thickness [m] and P
H2,R
and P
H2,P
are hydrogen partial pressures
[kPa] on the retentate side and on the permeate side of the membrane, respectively. Eq. 1 is
known as Sievert’s law and it is valid if the bulk phase diffusion of atomic hydrogen is the
rate limiting step in the hydrogen permeation process.
To increase the separation driving force, usually the retentate is kept at higher pressure than
the permeate. In common applications, permeate pressure is atmospheric and retentate
pressure is in the range 10-15 atm (compatibly with mechanical constraints).
A possible way to further increase the separation driving force is to reduce hydrogen partial
pressure in the permeate (P
H2,P
) by diluting the permeate stream with sweep gas (usually
superheated steam).

Sievert’s law shows that an increase of the hydrogen flux is achieved with reducing
membrane thickness. Palladium membranes should not be far thinner than 80-100 μm due
to mechanical stability of the layer and to the presence of defects and pinholes that reduce
hydrogen selectivity. To overcome this problem, current technologies foresee a thin layer
(20-50 μm) of Pd deposited on a porous ceramic or metal substrate.
Another important issue of Pd membranes (pure or supported) is thermal resistance.
Temperature should not be less than 200°C, to prevent hydrogen embrittlement and not
higher than 600°C ca. to prevent material damage.


Fig. 2. Membrane Reactor

Innovative fuel processors can be realized by combining the membrane either with the
reforming unit, generating the fuel processor reported in Fig. 3 (FP.1), or with a water gas
shift unit, generating the fuel processor reported in Fig. 3 (FP.2).


Fig. 3. Innovative Fuel Processors

FP.1 consists of a desulfurization unit followed by a membrane reforming reactor, with a
burner. This solution guarantees the highest compactness in terms of number of units, since
it allows to totally suppress the CO clean-up section; indeed, when the membrane is
integrated in the reforming reactor, the permeate stream is pure hydrogen, that can be
directly fed to a PEMFC.
A, B, C, H
2
RETENTATE
H
2
PERMEATE
A, B
H
H
H
H
A + B = C + H
2
REACTION SIDE MEMBRANE SIDE
Des
MEMBRANE
SR/ATR REACTOR
Burner
Fuel
Air
Q
R
e
t
e
n
t
a
t
e
H
2
FP.1
Des
MEMBRANE
WGS REACTOR
Burner
Air
Fuel
Q
R
e
t
e
n
t
a
t
e
H
2
SR/ATR
FP.2
Energy Effciency 162

However, this solution limits the choice of the operating temperature of the process that
must be compatible with the constraints imposed by the presence of a membrane.
FP.2 consists of a desulfurization unit followed by a reforming reactor and a membrane
water gas shift reactor. In this case, the membrane is placed in the low temperature zone of
the fuel processor, operating at thermal levels compatible with its stability. This solution,
although less compact than the previous one, allows to operate the syngas production
section at higher temperature.

2.3 PEMFC
A fuel cell is an electrochemical device that converts the chemical energy of a fuel directly
into electrical energy. Intermediate conversions of the fuel to thermal and mechanical
energy are not required. All fuel cells consist of two electrodes (anode and cathode) and an
electrolyte.
Proton exchange membrane fuel cells, also known as polymer electrolyte membrane fuel
cells (PEMFC), are a type of fuel cell in which the electrolyte is a polymeric membrane and
the electrodes are made of platinum.
In a PEMFC unit, hydrogen is supplied at one side of the membrane where it is split into
hydrogen protons and electrons, at anode electrode:
H
2
 2H
+
+ 2e
-
The protons permeate through the polymeric membrane to the reach the cathode electrode,
where oxygen is supplied and the following reactions takes place.
O
2
+ 4H
+
+ 4e
-
 2H
2
O
Electrons circulate in an external electric circuit under a potential difference.
The electric potential generated in a single unit is about 0.9V. To achieve a higher voltage,
several membrane units need to be connected in series, forming a fuel cell stack. The
electrical power output of the fuel cell is about 60% of its energy generation, the remaining
energy is released as heat.
Generally, oxygen is fed to the cathode as an air stream; in practical systems, an excess of
oxygen is fed to the cathode to avoid extremely low concentration at the exit. Frequently, a
50% or higher excess with respect to the stoichiometric oxygen is fed to the cathode.
For the anode, instead, it is not typically the stoichiometric ratio, but rather the amount of
hydrogen converted to the fuel cell as a percentage of the feed that is specified. This amount
is named as the hydrogen utilization factor U
f
; when pure hydrogen is fed to the PEMFC,
this factor can be assumed equal to unity.
For PEMFC systems running on reformate produced in a conventional fuel processor, this
factor can be assumed equal to 0.8. This implies that not all gas fed to the anode is converted
and unconverted hydrogen and the rest of the reformate is purged off as a stream named as
Anode Off-Gas (AOG). This stream presents a heating value due to the presence of
hydrogen and methane; therefore, it can be used in the burner of the conventional fuel
processor to eventually supply heat to the process.

2.4 System Analysis of Fuel Processor - PEMFC systems
Optimization of energy efficiency of a fuel processor PEMFC system is a central issue in
actual research studies. Since the efficiency of the PEMFC can be assumed as a constant

equal to 60%, the efficiency of the entire system depends on fuel processor efficiency and on
the integration between the fuel processor and the PEMFC.
The optimization of system efficiency is achieved by exploring the effect of the operating
parameters considering, at the same time, the heat recovery between the various streams
and units present in the system and the necessary driving force for heat exchange.
The optimization of conventional hydrocarbon-based fuel processors has been tackled by
several authors who have identified the most favorable operating conditions to maximize
the reforming efficiency. As a general outcome, SR-based fuel processors provide the
highest hydrogen concentration in the product stream, whereas the highest reforming
efficiency is reached with ATR-based fuel processors, due to the energy loss represented by
the latent heat of vaporization of the water that escapes with the combustion products in the
SR system (Ahmed et al, 2001).
However, as the system grows in complexity, due to the presence of the fuel cell,
optimization of the global energy efficiency must also take into account the recovery of the
energy contained in the spent gas released at the cell anode (anode off-gas). Ersoz et al.
(2006) performed an analysis of global energy efficiency on a fuel processor – PEMFC
system, considering methane as the fuel and steam reforming, partial oxidation and auto
thermal reforming as alternative processes to produce hydrogen. Their main conclusion is
that the highest global energy efficiency is reached when SR is used, essentially due to the
higher recovery of anode off-gas heating value.
As far as membrane-based fuel processor is concerned, only few contributions which
address the behavior of the entire system are available, that include not only the membrane-
based fuel processor, but also the fuel cell, the auxiliary power units and the heat
exchangers (Pearlman et al, Lattner et al, Manzolini et al, Campanari et al, Lyubovsky et al).
Most of these studies refer to liquid fuels and only few contributions are available when
methane is employed.
In particular, Campanari et al. (2008) analyzed an integrated membrane SR reactor coupled
with a PEMFC, showing that a higher global energy efficiency can be achieved, with respect
to conventional fuel processors, if a membrane reactor is employed.
Lyubovsky et al. (2006) analyzed a methane ATR-based fuel processor – PEMFC system,
with a membrane unit placed downstream the WGS unit and operating at high pressure,
concluding that high global energy efficiency can be obtained if a turbine is introduced in
the system to generate additional power from the expansion of the hot gases produced by
the combustion of the membrane retentate stream.
In order to have a complete vision of the effect of system configuration and of operating
parameters on the efficiency of fuel processor – PEMFC systems, a comprehensive analysis
of different configurations will be presented and compared in terms of energy efficiency; in
particular, methane will be considered as fuel and SR and ATR as reforming processes; the
focus of the discussion will be about the following fuel processor (FP) configurations, each
coupled with a PEMFC:
FP.A) SR reactor, followed by two WGS reactors and a PROX reactor.
FP.B) ATR reactor, followed by two WGS reactors and a PROX reactor.
FP.C) Integrated membrane-SR reactor.
FP.D) Integrated membrane-ATR reactor.
FP.E) SR reactor followed by a membrane WGS reactor.
FP.F) ATR reactor followed by a membrane WGS reactor.
Energy effciency of Fuel Processor – PEM Fuel Cell systems 163

However, this solution limits the choice of the operating temperature of the process that
must be compatible with the constraints imposed by the presence of a membrane.
FP.2 consists of a desulfurization unit followed by a reforming reactor and a membrane
water gas shift reactor. In this case, the membrane is placed in the low temperature zone of
the fuel processor, operating at thermal levels compatible with its stability. This solution,
although less compact than the previous one, allows to operate the syngas production
section at higher temperature.

2.3 PEMFC
A fuel cell is an electrochemical device that converts the chemical energy of a fuel directly
into electrical energy. Intermediate conversions of the fuel to thermal and mechanical
energy are not required. All fuel cells consist of two electrodes (anode and cathode) and an
electrolyte.
Proton exchange membrane fuel cells, also known as polymer electrolyte membrane fuel
cells (PEMFC), are a type of fuel cell in which the electrolyte is a polymeric membrane and
the electrodes are made of platinum.
In a PEMFC unit, hydrogen is supplied at one side of the membrane where it is split into
hydrogen protons and electrons, at anode electrode:
H
2
 2H
+
+ 2e
-
The protons permeate through the polymeric membrane to the reach the cathode electrode,
where oxygen is supplied and the following reactions takes place.
O
2
+ 4H
+
+ 4e
-
 2H
2
O
Electrons circulate in an external electric circuit under a potential difference.
The electric potential generated in a single unit is about 0.9V. To achieve a higher voltage,
several membrane units need to be connected in series, forming a fuel cell stack. The
electrical power output of the fuel cell is about 60% of its energy generation, the remaining
energy is released as heat.
Generally, oxygen is fed to the cathode as an air stream; in practical systems, an excess of
oxygen is fed to the cathode to avoid extremely low concentration at the exit. Frequently, a
50% or higher excess with respect to the stoichiometric oxygen is fed to the cathode.
For the anode, instead, it is not typically the stoichiometric ratio, but rather the amount of
hydrogen converted to the fuel cell as a percentage of the feed that is specified. This amount
is named as the hydrogen utilization factor U
f
; when pure hydrogen is fed to the PEMFC,
this factor can be assumed equal to unity.
For PEMFC systems running on reformate produced in a conventional fuel processor, this
factor can be assumed equal to 0.8. This implies that not all gas fed to the anode is converted
and unconverted hydrogen and the rest of the reformate is purged off as a stream named as
Anode Off-Gas (AOG). This stream presents a heating value due to the presence of
hydrogen and methane; therefore, it can be used in the burner of the conventional fuel
processor to eventually supply heat to the process.

2.4 System Analysis of Fuel Processor - PEMFC systems
Optimization of energy efficiency of a fuel processor PEMFC system is a central issue in
actual research studies. Since the efficiency of the PEMFC can be assumed as a constant

equal to 60%, the efficiency of the entire system depends on fuel processor efficiency and on
the integration between the fuel processor and the PEMFC.
The optimization of system efficiency is achieved by exploring the effect of the operating
parameters considering, at the same time, the heat recovery between the various streams
and units present in the system and the necessary driving force for heat exchange.
The optimization of conventional hydrocarbon-based fuel processors has been tackled by
several authors who have identified the most favorable operating conditions to maximize
the reforming efficiency. As a general outcome, SR-based fuel processors provide the
highest hydrogen concentration in the product stream, whereas the highest reforming
efficiency is reached with ATR-based fuel processors, due to the energy loss represented by
the latent heat of vaporization of the water that escapes with the combustion products in the
SR system (Ahmed et al, 2001).
However, as the system grows in complexity, due to the presence of the fuel cell,
optimization of the global energy efficiency must also take into account the recovery of the
energy contained in the spent gas released at the cell anode (anode off-gas). Ersoz et al.
(2006) performed an analysis of global energy efficiency on a fuel processor – PEMFC
system, considering methane as the fuel and steam reforming, partial oxidation and auto
thermal reforming as alternative processes to produce hydrogen. Their main conclusion is
that the highest global energy efficiency is reached when SR is used, essentially due to the
higher recovery of anode off-gas heating value.
As far as membrane-based fuel processor is concerned, only few contributions which
address the behavior of the entire system are available, that include not only the membrane-
based fuel processor, but also the fuel cell, the auxiliary power units and the heat
exchangers (Pearlman et al, Lattner et al, Manzolini et al, Campanari et al, Lyubovsky et al).
Most of these studies refer to liquid fuels and only few contributions are available when
methane is employed.
In particular, Campanari et al. (2008) analyzed an integrated membrane SR reactor coupled
with a PEMFC, showing that a higher global energy efficiency can be achieved, with respect
to conventional fuel processors, if a membrane reactor is employed.
Lyubovsky et al. (2006) analyzed a methane ATR-based fuel processor – PEMFC system,
with a membrane unit placed downstream the WGS unit and operating at high pressure,
concluding that high global energy efficiency can be obtained if a turbine is introduced in
the system to generate additional power from the expansion of the hot gases produced by
the combustion of the membrane retentate stream.
In order to have a complete vision of the effect of system configuration and of operating
parameters on the efficiency of fuel processor – PEMFC systems, a comprehensive analysis
of different configurations will be presented and compared in terms of energy efficiency; in
particular, methane will be considered as fuel and SR and ATR as reforming processes; the
focus of the discussion will be about the following fuel processor (FP) configurations, each
coupled with a PEMFC:
FP.A) SR reactor, followed by two WGS reactors and a PROX reactor.
FP.B) ATR reactor, followed by two WGS reactors and a PROX reactor.
FP.C) Integrated membrane-SR reactor.
FP.D) Integrated membrane-ATR reactor.
FP.E) SR reactor followed by a membrane WGS reactor.
FP.F) ATR reactor followed by a membrane WGS reactor.
Energy Effciency 164

Each system configuration is investigated by varying operating parameters, such as steam to
methane and oxygen to methane inlet ratios, reforming temperature, as well as pressure; the
effect of the addition of steam as sweep gas on the permeate side of the membrane reactors
will be also presented and discussed.

3. Methodology
The simulations were performed in stationary conditions, by using the commercial package
Aspen Plus®. The selected property method was Peng-Robinson and the component list
was restricted to CH
4
, O
2
, N
2
, H
2
O, CO, H
2
and CO
2
.
Methane was considered as fuel, fed at 25°C and 1 atm, with a constant flow rate of 1
kmol/h. Feed to the system was completed with a liquid water stream (25°C and 1 atm)
both in SR and ATR-based FPs; an air stream (25°C and 1 atm) is also present in the ATR-
based FPs.
The configurations simulated (flow sheets) are presented in the following sections, where
the assumptions and the model libraries used to simulate the process are presented. Section
3.1 is dedicated to conventional fuel processors, whereas membrane-based fuel processors
are described in section 3.2. The quantities employed to calculate energy efficiency are
defined in section 3.4.

3.1 Conventional fuel processor – PEMFC systems
Fig. 4 reports the flow sheet of a conventional SR-based fuel processor coupled with a
PEMFC (FP.A). The fuel processor consists of a reforming and a CO clean-up section.


Fig. 4. Flowsheet of fuel processor FP.A and FP.B coupled with a PEMFC

The reforming section is an isothermal reactor (SR), modeled by using the model library
RGIBBS.
The CO clean-up section consists of a high (HTS) and low (LTS) temperature water gas shift
reactor followed by a PROX reactor. HTS and LTS were modeled by using model library
RGIBBS; the reactors were considered as adiabatic and methane was considered as an inert
in order to eliminate the undesired methanation reaction, kinetically suppressed on a real
catalytic system.
CH4
H2O
AI R
ATR
H-WGS H-ATR
R
FP.A:  R=SR
FP.B:  R=ATR
FUEL
H2O
AIR
CH4
H2
AOG
AIRFC
OUT-FC
AIRPROX
AIRAOG
CH4-B
EXHAUST
HTS LTS
ANODE
PROX
CATHODE
H-B
BURNER
H-LTS H-PROX
H-PEMFC
H-EX
H‐HTS
H

The inlet temperature to the HTS reactor was fixed at 350°C, while the inlet temperature to
the LTS one at 200°C. The PROX reactor was modeled as an adiabatic stoichiometric reactor,
RSTOIC; this kind of reactor models a stoichiometric reactor with specified reaction extent
or conversion; in the case of PROX, two reactions were considered: oxidation of CO to CO
2

with complete conversion of CO and oxidation of H
2
to H
2
O; the air fed to the PROX reactor
(AIR-PROX) was calculated in order to achieve a 50% oxygen excess with respect to the
stoichiometric amount required to convert all the CO to CO
2
. The RSTOIC specifics were
completed with the assignment of total conversion of CO and O
2
. The inlet temperature to
the PROX reactor was fixed at 90°C.
The PEM fuel cell section is simulated as the sequence of the anode, modeled as an ideal
separator, SEP, and the cathode, modeled as an isothermal stoichiometric reactor, RSTOIC.
The presence of the SEP unit allows to model a purge gas (anode off-gas, AOG) required for
mass balance reasons, whenever the hydrogen stream sent to the PEM fuel cell is not 100%
pure. In agreement with the literature, the hydrogen split fraction in the stream H2 at the
outlet of the SEP was fixed at 0.75 (Francesconi et al, 2007), whereas the split fractions of all
the other components were taken as 0. The RSTOIC unit models the hydrogen oxidation
reaction occurring in the fuel cell. The reactor specifics were completed by considering an
operating temperature of 80°C and pressure of 1 atm; the inlet air at the cathode (AIR-FC),
fed at 25°C and 1 atm, guarantees a 50% excess of oxygen in the RSTOIC reactor. In
agreement with Ratnamala et al (2005), these conditions were considered as sufficient to
assign total hydrogen conversion. The anode off-gas is sent to a burner, modeled as an
adiabatic RSTOIC, working at atmospheric pressure with 50% excess air (AIR-B); the
complete combustion of all fuels contained (i.e. hydrogen, methane, carbon monoxide) was
always imposed.
The heat required by the SR reactor working at temperature T
SR
is supplied by the heat
exchanger H-B, where the stream coming from the burner is cooled to T
SR
+10°C. Model
library HEATER was used for this purpose. An additional stream of fuel (CH4-B) is sent to
the burner to satisfy the global heat demand of the system, when needed. The heat removed
for cooling the stream at the outlet of heat exchanger H-B, at the inlet of HTS, LTS and
PROX reactors and PEM fuel cell, as well as the heat for keeping the PROX at constant
temperature, is employed to preheat the SR inlet stream. On the other hand, the heat
removed for cooling the PEM fuel cell, is not recovered, since most of the times a simple air
fan is used to cool the stack.
As concerns the flow sheet of a conventional ATR-based fuel processor (FP.B) coupled with
a PEM fuel cell, for the sake of simplicity, the description of the flow sheet will be carried
out by indicating the differences with respect to the flow sheet of Fig. 4, which are
concentrated only in the reforming section. Indeed, in FP.B the reforming section is
constituted by an adiabatic reactor (ATR), modeled by using model library RGIBBS. The
heat exchanger H-B can be suppressed in this configuration, since the ATR reactor has no
heat requirement. The inlet temperature to the ATR reactor is fixed at 350°C, and is
regulated by means of the heat exchanger H-ATR.

3.2 Innovative fuel processor – PEMFC systems
The integrated membrane-reactors were simulated by discretizing the membrane reactor
with a series of N reactor-separator units. With this approximation, reactors are assumed to
reach equilibrium and the separators are modeled as ideal separators, SEP, whose output is
Energy effciency of Fuel Processor – PEM Fuel Cell systems 165

Each system configuration is investigated by varying operating parameters, such as steam to
methane and oxygen to methane inlet ratios, reforming temperature, as well as pressure; the
effect of the addition of steam as sweep gas on the permeate side of the membrane reactors
will be also presented and discussed.

3. Methodology
The simulations were performed in stationary conditions, by using the commercial package
Aspen Plus®. The selected property method was Peng-Robinson and the component list
was restricted to CH
4
, O
2
, N
2
, H
2
O, CO, H
2
and CO
2
.
Methane was considered as fuel, fed at 25°C and 1 atm, with a constant flow rate of 1
kmol/h. Feed to the system was completed with a liquid water stream (25°C and 1 atm)
both in SR and ATR-based FPs; an air stream (25°C and 1 atm) is also present in the ATR-
based FPs.
The configurations simulated (flow sheets) are presented in the following sections, where
the assumptions and the model libraries used to simulate the process are presented. Section
3.1 is dedicated to conventional fuel processors, whereas membrane-based fuel processors
are described in section 3.2. The quantities employed to calculate energy efficiency are
defined in section 3.4.

3.1 Conventional fuel processor – PEMFC systems
Fig. 4 reports the flow sheet of a conventional SR-based fuel processor coupled with a
PEMFC (FP.A). The fuel processor consists of a reforming and a CO clean-up section.


Fig. 4. Flowsheet of fuel processor FP.A and FP.B coupled with a PEMFC

The reforming section is an isothermal reactor (SR), modeled by using the model library
RGIBBS.
The CO clean-up section consists of a high (HTS) and low (LTS) temperature water gas shift
reactor followed by a PROX reactor. HTS and LTS were modeled by using model library
RGIBBS; the reactors were considered as adiabatic and methane was considered as an inert
in order to eliminate the undesired methanation reaction, kinetically suppressed on a real
catalytic system.
CH4
H2O
AI R
ATR
H-WGS H-ATR
R
FP.A:  R=SR
FP.B:  R=ATR
FUEL
H2O
AIR
CH4
H2
AOG
AIRFC
OUT-FC
AIRPROX
AIRAOG
CH4-B
EXHAUST
HTS LTS
ANODE
PROX
CATHODE
H-B
BURNER
H-LTS H-PROX
H-PEMFC
H-EX
H‐HTS
H

The inlet temperature to the HTS reactor was fixed at 350°C, while the inlet temperature to
the LTS one at 200°C. The PROX reactor was modeled as an adiabatic stoichiometric reactor,
RSTOIC; this kind of reactor models a stoichiometric reactor with specified reaction extent
or conversion; in the case of PROX, two reactions were considered: oxidation of CO to CO
2

with complete conversion of CO and oxidation of H
2
to H
2
O; the air fed to the PROX reactor
(AIR-PROX) was calculated in order to achieve a 50% oxygen excess with respect to the
stoichiometric amount required to convert all the CO to CO
2
. The RSTOIC specifics were
completed with the assignment of total conversion of CO and O
2
. The inlet temperature to
the PROX reactor was fixed at 90°C.
The PEM fuel cell section is simulated as the sequence of the anode, modeled as an ideal
separator, SEP, and the cathode, modeled as an isothermal stoichiometric reactor, RSTOIC.
The presence of the SEP unit allows to model a purge gas (anode off-gas, AOG) required for
mass balance reasons, whenever the hydrogen stream sent to the PEM fuel cell is not 100%
pure. In agreement with the literature, the hydrogen split fraction in the stream H2 at the
outlet of the SEP was fixed at 0.75 (Francesconi et al, 2007), whereas the split fractions of all
the other components were taken as 0. The RSTOIC unit models the hydrogen oxidation
reaction occurring in the fuel cell. The reactor specifics were completed by considering an
operating temperature of 80°C and pressure of 1 atm; the inlet air at the cathode (AIR-FC),
fed at 25°C and 1 atm, guarantees a 50% excess of oxygen in the RSTOIC reactor. In
agreement with Ratnamala et al (2005), these conditions were considered as sufficient to
assign total hydrogen conversion. The anode off-gas is sent to a burner, modeled as an
adiabatic RSTOIC, working at atmospheric pressure with 50% excess air (AIR-B); the
complete combustion of all fuels contained (i.e. hydrogen, methane, carbon monoxide) was
always imposed.
The heat required by the SR reactor working at temperature T
SR
is supplied by the heat
exchanger H-B, where the stream coming from the burner is cooled to T
SR
+10°C. Model
library HEATER was used for this purpose. An additional stream of fuel (CH4-B) is sent to
the burner to satisfy the global heat demand of the system, when needed. The heat removed
for cooling the stream at the outlet of heat exchanger H-B, at the inlet of HTS, LTS and
PROX reactors and PEM fuel cell, as well as the heat for keeping the PROX at constant
temperature, is employed to preheat the SR inlet stream. On the other hand, the heat
removed for cooling the PEM fuel cell, is not recovered, since most of the times a simple air
fan is used to cool the stack.
As concerns the flow sheet of a conventional ATR-based fuel processor (FP.B) coupled with
a PEM fuel cell, for the sake of simplicity, the description of the flow sheet will be carried
out by indicating the differences with respect to the flow sheet of Fig. 4, which are
concentrated only in the reforming section. Indeed, in FP.B the reforming section is
constituted by an adiabatic reactor (ATR), modeled by using model library RGIBBS. The
heat exchanger H-B can be suppressed in this configuration, since the ATR reactor has no
heat requirement. The inlet temperature to the ATR reactor is fixed at 350°C, and is
regulated by means of the heat exchanger H-ATR.

3.2 Innovative fuel processor – PEMFC systems
The integrated membrane-reactors were simulated by discretizing the membrane reactor
with a series of N reactor-separator units. With this approximation, reactors are assumed to
reach equilibrium and the separators are modeled as ideal separators, SEP, whose output is
Energy Effciency 166

given by a stream of pure hydrogen (permeate) and a stream containing the unseparated
hydrogen and all the balance (retentate). The amount of hydrogen separated (n
i
H2,P
) is
calculated assuming equilibrium between the partial pressure in the retentate and permeate
side, according to Eq.2:

i
P H2, i
P H2, i R,
i
P H2,
i
R H2,
R
P
n n
n n
P 




(2)

where P
R
is the pressure in the retentate side of the membrane, equals to reactor pressure;
n
i
H2,R
is the mole flow of hydrogen in the retentate stream; n
i
R
is the total mole flow of the
retentate stream; P
i
H2,P
is hydrogen partial pressure in the permeate side of the membrane,
calculated as:
P
SG
i
P H2,
i
P H2, i
P H2,
P
n n
n
P 



(3)

where P
P
is the pressure in the permeate side of the membrane, taken as 1 atm in all the
simulations, and n
SG
represents the molar flow rate of steam sweep gas (SG), which can be
introduced to increase the separation driving force in the membrane. When present, the
sweep gas is produced by liquid water, fed at 25°C and 1 atm to a heat exchanger and sent
to the membrane reactor in countercurrent flow mode.
The high hydrogen purity of the stream sent to the PEMFC allows taking as zero the anode
off-gas, simplifying the model of the PEMFC to the cathode side (RSTOIC) only.
Fig. 5 reports the flow sheet used to simulate a membrane-SR reactor (FP.C) and a membrane-
ATR reactor (FP.D) coupled with a PEMFC. The membrane-SR reactor was discretized with 30
units, whereas the membrane-ATR reactor was discretized with 20 units; the number of units
required to model each membrane-reactor was assessed by repeating the simulations with an
increasing number of reactor-separator units and was chosen as the minimum value above
which global efficiency remained constant within ± 0.1%.


Fig. 5. Flowsheet of fuel processors FP.C and FP.D coupled with a PEMFC
AIR- FC
OUT-FC
FUEL
H2O
PERMEATE
SG
AIR-B
FUEL- B
RETENT
EXHAUST
H
H
H
H‐B H
CATHODE
BURNER
R‐1 R‐2 R‐(N‐1) R‐N
SEP‐N
SEP‐(N‐1) SEP‐2
SEP‐1
FROM SEP3
TO R‐3
FROM 
SEP‐(N‐2)
TO SEP‐(N‐2)
80°C
600°C
AIR
FP.C:   R=SR;     N=30
FP.D:  R=ATR;  N=20
CH4
0
0
0
0
CH4-B

As for the case of the conventional system, the heat eventually required by the reforming
reactor is supplied by the heat exchanger H-B and an additional stream of fuel (CH4-B) is
sent to the burner to satisfy the global heat demand of the system, when needed. The heat
removed for cooling the streams at the outlet of heat exchanger H-B and at the inlet of the
PEMFC is recovered to preheat SR inlet stream and eventually to produce sweep gas.
Fig. 6 reports the flow sheet used to simulate a SR-based FP coupled with a PEMFC, where
the SR reactor is followed by a membrane WGS reactor (FP.E) and a ATR-based FP coupled
with a PEMFC, where the ATR reactor is followed by a membrane WGS reactor (FP.F). With
respect to FP.A and FP.B, in this case only one Water Gas Shift reactor is present, with an
inlet temperature of 300°C; the membrane WGS reactor was discredited into four units.
As for the case of described above, the heat eventually required by the reforming reactor is
supplied by H-B and an additional stream of fuel (CH4-B) is sent to the burner to satisfy the
global heat demand of the system, when needed. The heat removed for cooling the streams
at the outlet of heat exchanger H-B and at the inlet of the PEM fuel cell is recovered to
preheat SR inlet stream and eventually to produce sweep gas.


Fig. 6. Flow sheet of fuel processor FP.E and FP.F coupled with a PEMFC

All the reactors were considered as operating at the same pressure.
Auxiliary power units for compression of the reactants were considered in the
configurations where pressure was explored as an operation variable, i.e. FP.C and FP.D.
100°C was chosen as the minimum exhaust gas temperature (T
ex
), when compatible with the
constraint of a positive driving force in the heat exchangers present in the plant.
Finally, it is worth mentioning that the assumptions made to model the system are the same
for all the configurations investigated and do not affect the conclusions drawn in this
comparative analysis.

3.3 System Efficiency
Energy efficiency, , was defined according to the following Eq.4:

CH4 B CH4, F CH4,
a e
LHV ) n (n
P P
η
 



(4)

AIR-FC
OUT-FC
PERMEATE
SG
AIR- B
FUEL-B
RETENT
EXHAUST
H
H
H
H‐B H
CATHODE
BURNER
WGS‐1 WGS‐2 WGS‐(N‐1) WGS‐N
SEP‐N
SEP‐(N‐1) SEP‐2
SEP‐1
FROM SEP3
TO R‐3
FROM 
SEP‐(N‐2)
TO SEP‐(N‐2)
80°C
600°C
0
0
0
0
CH4-B
CH4
H2O
AI R
ATR
H-WGS H-ATR
R‐1
H
FP.E:   R=SR;     N=10
FP.F:  R=ATR;  N=10
FUEL
H2O
AIR
CH4
Energy effciency of Fuel Processor – PEM Fuel Cell systems 167

given by a stream of pure hydrogen (permeate) and a stream containing the unseparated
hydrogen and all the balance (retentate). The amount of hydrogen separated (n
i
H2,P
) is
calculated assuming equilibrium between the partial pressure in the retentate and permeate
side, according to Eq.2:

i
P H2, i
P H2, i R,
i
P H2,
i
R H2,
R
P
n n
n n
P 




(2)

where P
R
is the pressure in the retentate side of the membrane, equals to reactor pressure;
n
i
H2,R
is the mole flow of hydrogen in the retentate stream; n
i
R
is the total mole flow of the
retentate stream; P
i
H2,P
is hydrogen partial pressure in the permeate side of the membrane,
calculated as:
P
SG
i
P H2,
i
P H2, i
P H2,
P
n n
n
P 



(3)

where P
P
is the pressure in the permeate side of the membrane, taken as 1 atm in all the
simulations, and n
SG
represents the molar flow rate of steam sweep gas (SG), which can be
introduced to increase the separation driving force in the membrane. When present, the
sweep gas is produced by liquid water, fed at 25°C and 1 atm to a heat exchanger and sent
to the membrane reactor in countercurrent flow mode.
The high hydrogen purity of the stream sent to the PEMFC allows taking as zero the anode
off-gas, simplifying the model of the PEMFC to the cathode side (RSTOIC) only.
Fig. 5 reports the flow sheet used to simulate a membrane-SR reactor (FP.C) and a membrane-
ATR reactor (FP.D) coupled with a PEMFC. The membrane-SR reactor was discretized with 30
units, whereas the membrane-ATR reactor was discretized with 20 units; the number of units
required to model each membrane-reactor was assessed by repeating the simulations with an
increasing number of reactor-separator units and was chosen as the minimum value above
which global efficiency remained constant within ± 0.1%.


Fig. 5. Flowsheet of fuel processors FP.C and FP.D coupled with a PEMFC
AIR- FC
OUT-FC
FUEL
H2O
PERMEATE
SG
AIR-B
FUEL- B
RETENT
EXHAUST
H
H
H
H‐B H
CATHODE
BURNER
R‐1 R‐2 R‐(N‐1) R‐N
SEP‐N
SEP‐(N‐1) SEP‐2
SEP‐1
FROM SEP3
TO R‐3
FROM 
SEP‐(N‐2)
TO SEP‐(N‐2)
80°C
600°C
AIR
FP.C:   R=SR;     N=30
FP.D:  R=ATR;  N=20
CH4
0
0
0
0
CH4-B

As for the case of the conventional system, the heat eventually required by the reforming
reactor is supplied by the heat exchanger H-B and an additional stream of fuel (CH4-B) is
sent to the burner to satisfy the global heat demand of the system, when needed. The heat
removed for cooling the streams at the outlet of heat exchanger H-B and at the inlet of the
PEMFC is recovered to preheat SR inlet stream and eventually to produce sweep gas.
Fig. 6 reports the flow sheet used to simulate a SR-based FP coupled with a PEMFC, where
the SR reactor is followed by a membrane WGS reactor (FP.E) and a ATR-based FP coupled
with a PEMFC, where the ATR reactor is followed by a membrane WGS reactor (FP.F). With
respect to FP.A and FP.B, in this case only one Water Gas Shift reactor is present, with an
inlet temperature of 300°C; the membrane WGS reactor was discredited into four units.
As for the case of described above, the heat eventually required by the reforming reactor is
supplied by H-B and an additional stream of fuel (CH4-B) is sent to the burner to satisfy the
global heat demand of the system, when needed. The heat removed for cooling the streams
at the outlet of heat exchanger H-B and at the inlet of the PEM fuel cell is recovered to
preheat SR inlet stream and eventually to produce sweep gas.


Fig. 6. Flow sheet of fuel processor FP.E and FP.F coupled with a PEMFC

All the reactors were considered as operating at the same pressure.
Auxiliary power units for compression of the reactants were considered in the
configurations where pressure was explored as an operation variable, i.e. FP.C and FP.D.
100°C was chosen as the minimum exhaust gas temperature (T
ex
), when compatible with the
constraint of a positive driving force in the heat exchangers present in the plant.
Finally, it is worth mentioning that the assumptions made to model the system are the same
for all the configurations investigated and do not affect the conclusions drawn in this
comparative analysis.

3.3 System Efficiency
Energy efficiency, , was defined according to the following Eq.4:

CH4 B CH4, F CH4,
a e
LHV ) n (n
P P
η
 



(4)

AIR-FC
OUT-FC
PERMEATE
SG
AIR- B
FUEL-B
RETENT
EXHAUST
H
H
H
H‐B H
CATHODE
BURNER
WGS‐1 WGS‐2 WGS‐(N‐1) WGS‐N
SEP‐N
SEP‐(N‐1) SEP‐2
SEP‐1
FROM SEP3
TO R‐3
FROM 
SEP‐(N‐2)
TO SEP‐(N‐2)
80°C
600°C
0
0
0
0
CH4-B
CH4
H2O
AI R
ATR
H-WGS H-ATR
R‐1
H
FP.E:   R=SR;     N=10
FP.F:  R=ATR;  N=10
FUEL
H2O
AIR
CH4
Energy Effciency 168

where P
a
is the electric power required by the auxiliary units for compression of methane,
air and water, n
CH4,F
is the inlet molar flow rate of methane to the reactor, n
CH4,B
is the molar
flow rate of methane fed to the burner, LHV
CH4
is the lower heating value of methane and P
e

is the electric power generated by the fuel cell, calculated as:

FC H2 H2 e
η LHV n P · · =
(5)

where n
H2
is the molar flow rate of hydrogen that reacts in the fuel cell, LHV
H2
is the lower
heating value of hydrogen, η
FC
is the electrochemical efficiency of the cell, taken as 0.6 (Hou
et al, 2007). In the membrane-based fuel cell systems, an important parameter is the global
hydrogen recovery (HR), defined as:

¿
¿
=
=
+
=
N
1 i
i
P H2, R H2,
N
1 i
i
P H2,
n n
n
HR


(6)

where n
i
H2,P
is the molar flow rate of hydrogen separated by the i-th membrane unit, n
H2,R
is
the molar flow rate of hydrogen in the RETENT stream at the exit of the last separator and N
is the number of separators.
According to the definitions given above, q can be expressed as it follows:

) 1 ( ) f η η HR ( η
a FC R
o ÷ · ÷ · · = (7)

where f
a
is the fraction of inlet methane required to run the auxiliary units, defined by Eq. 8:

CH4 F CH4,
a
a
LHV n
P
f
·
=

(8)

α is the ratio between methane flow rate fed to the burner and total methane flow rate fed to
the system, defined by Eq. 9:

B CH4, F CH4,
B CH4,
n n
n
+
= o

(9)

f
R
is the reforming factor, defined by Eq. 10:

CH4 F CH4,
H2
N
1 i
i
P H2, R H2,
R
LHV n
LHV n n
f
·
·
|
|
.
|


\
|
+
=
¿
=



(10)


This factor is related to the global amount of hydrogen produced in the fuel processor per
moles of methane fed to the reforming reactor; therefore it does not depend on the heat
requirement of the system.

4. Results
Simulation where performed by varying the main operating parameters for each system.
The parameters investigated and the ranges explored are reported in Table 1. For
conventional systems (FP.A and FP.B) pressure was fixed at 1 atm since reforming processes
are inhibited by pressure increase, whereas the WGS and PROX processes are independent
of pressure. The operating ranges of H
2
O/CH
4
and T
SR
for the system with membrane SR
reactor (FP.C) are chosen in order to guarantee thermal stability of the membrane and to
avoid coke formation. The pressure range investigated for the innovative systems was
chosen in order to guarantee the mechanical resistance of the membrane. The operating
ranges of H
2
O/CH
4
and of O
2
/CH
4
for the ATR systems are chosen in order to avoid coke
formation and to guarantee the autothermicity of the process (Seo et al, 2002).

Case H2O/CH4 O2/CH4 TSR [°C] SG/CH4 P [atm]
SR
FP.A 2.0 – 6.0 - 600 - 800 - 1
FP.C 2.5 – 6.0 - 500 - 600 0 – 3.0 3 - 15
FP.E 2.0 – 6.0 - 600 - 800 0 – 3.0 3 - 15
ATR
FP.B 1.2 – 4.0 0.3 – 1.0 - - 1
FP.D 1.2 – 4.0 0.3 – 1.0 - 0 – 3.0 3 - 15
FP.F 1.2 – 4.0 0.3 – 1.0 - 0 – 3.0 3 - 15
Table 1. Range of operating parameters investigated

4.1 Conventional Fuel Processors
Fig. 7 shows the trend of energy efficiency , methane conversion x
CH4
, reforming factor f
R

and the fraction of total inlet methane that is sent to the burner α as a function of H
2
O/CH
4
,
parametric in the steam reforming reactor temperature.
For all the temperatures investigated, an increase of water content in the feed has a positive
effect on methane conversion x
CH4
and on the reforming factor f
R
. This well note trend is due
to the fact that water is a reactant of reforming reactions.
For each temperature and until a certain value of H
2
O/CH
4
, the value of α is equal to zero.
For higher H
2
O/CH
4
, the increase of this ratio leads to an increase of α; indeed, the increase
of H
2
O/CH
4
causes an increase of the heat required to sustain the reforming process,
moreover the improvement of reforming reactor performance with H
2
O/CH
4
causes a
reduction of the heating value of the AOG stream, thus an increase of the quantity of
methane that needs to be sent to the burner for sustaining the endothermicity of the process.
As described in the System efficiency Section, the energy efficiency is a combination of f
R

and of α; indeed,  shows a non monotone trend as a function of H
2
O/CH
4
because,
although an increase of water content causes a continuous increase of reforming reactor
performance, the amount of methane sent to the burner also increases with H
2
O/CH
4
.
For all the H
2
O/CH
4
investigated, the increase of reforming reactor temperature (T
SR
) causes
an increase of x
CH4
, f
R
and α. Energy efficiency  shows a different trend on the basis of the
Energy effciency of Fuel Processor – PEM Fuel Cell systems 169

where P
a
is the electric power required by the auxiliary units for compression of methane,
air and water, n
CH4,F
is the inlet molar flow rate of methane to the reactor, n
CH4,B
is the molar
flow rate of methane fed to the burner, LHV
CH4
is the lower heating value of methane and P
e

is the electric power generated by the fuel cell, calculated as:

FC H2 H2 e
η LHV n P · · =
(5)

where n
H2
is the molar flow rate of hydrogen that reacts in the fuel cell, LHV
H2
is the lower
heating value of hydrogen, η
FC
is the electrochemical efficiency of the cell, taken as 0.6 (Hou
et al, 2007). In the membrane-based fuel cell systems, an important parameter is the global
hydrogen recovery (HR), defined as:

¿
¿
=
=
+
=
N
1 i
i
P H2, R H2,
N
1 i
i
P H2,
n n
n
HR


(6)

where n
i
H2,P
is the molar flow rate of hydrogen separated by the i-th membrane unit, n
H2,R
is
the molar flow rate of hydrogen in the RETENT stream at the exit of the last separator and N
is the number of separators.
According to the definitions given above, q can be expressed as it follows:

) 1 ( ) f η η HR ( η
a FC R
o ÷ · ÷ · · = (7)

where f
a
is the fraction of inlet methane required to run the auxiliary units, defined by Eq. 8:

CH4 F CH4,
a
a
LHV n
P
f
·
=

(8)

α is the ratio between methane flow rate fed to the burner and total methane flow rate fed to
the system, defined by Eq. 9:

B CH4, F CH4,
B CH4,
n n
n
+
= o

(9)

f
R
is the reforming factor, defined by Eq. 10:

CH4 F CH4,
H2
N
1 i
i
P H2, R H2,
R
LHV n
LHV n n
f
·
·
|
|
.
|


\
|
+
=
¿
=



(10)


This factor is related to the global amount of hydrogen produced in the fuel processor per
moles of methane fed to the reforming reactor; therefore it does not depend on the heat
requirement of the system.

4. Results
Simulation where performed by varying the main operating parameters for each system.
The parameters investigated and the ranges explored are reported in Table 1. For
conventional systems (FP.A and FP.B) pressure was fixed at 1 atm since reforming processes
are inhibited by pressure increase, whereas the WGS and PROX processes are independent
of pressure. The operating ranges of H
2
O/CH
4
and T
SR
for the system with membrane SR
reactor (FP.C) are chosen in order to guarantee thermal stability of the membrane and to
avoid coke formation. The pressure range investigated for the innovative systems was
chosen in order to guarantee the mechanical resistance of the membrane. The operating
ranges of H
2
O/CH
4
and of O
2
/CH
4
for the ATR systems are chosen in order to avoid coke
formation and to guarantee the autothermicity of the process (Seo et al, 2002).

Case H2O/CH4 O2/CH4 TSR [°C] SG/CH4 P [atm]
SR
FP.A 2.0 – 6.0 - 600 - 800 - 1
FP.C 2.5 – 6.0 - 500 - 600 0 – 3.0 3 - 15
FP.E 2.0 – 6.0 - 600 - 800 0 – 3.0 3 - 15
ATR
FP.B 1.2 – 4.0 0.3 – 1.0 - - 1
FP.D 1.2 – 4.0 0.3 – 1.0 - 0 – 3.0 3 - 15
FP.F 1.2 – 4.0 0.3 – 1.0 - 0 – 3.0 3 - 15
Table 1. Range of operating parameters investigated

4.1 Conventional Fuel Processors
Fig. 7 shows the trend of energy efficiency , methane conversion x
CH4
, reforming factor f
R

and the fraction of total inlet methane that is sent to the burner α as a function of H
2
O/CH
4
,
parametric in the steam reforming reactor temperature.
For all the temperatures investigated, an increase of water content in the feed has a positive
effect on methane conversion x
CH4
and on the reforming factor f
R
. This well note trend is due
to the fact that water is a reactant of reforming reactions.
For each temperature and until a certain value of H
2
O/CH
4
, the value of α is equal to zero.
For higher H
2
O/CH
4
, the increase of this ratio leads to an increase of α; indeed, the increase
of H
2
O/CH
4
causes an increase of the heat required to sustain the reforming process,
moreover the improvement of reforming reactor performance with H
2
O/CH
4
causes a
reduction of the heating value of the AOG stream, thus an increase of the quantity of
methane that needs to be sent to the burner for sustaining the endothermicity of the process.
As described in the System efficiency Section, the energy efficiency is a combination of f
R

and of α; indeed,  shows a non monotone trend as a function of H
2
O/CH
4
because,
although an increase of water content causes a continuous increase of reforming reactor
performance, the amount of methane sent to the burner also increases with H
2
O/CH
4
.
For all the H
2
O/CH
4
investigated, the increase of reforming reactor temperature (T
SR
) causes
an increase of x
CH4
, f
R
and α. Energy efficiency  shows a different trend on the basis of the
Energy Effciency 170

weight of these factors: for low H
2
O/CH
4
,  shows a continuous increase with T
SR
in the
range investigated, whereas, for high H
2
O/CH
4
,  shows a non monotone trend with T
SR.



Fig. 7.  (a), x
CH4
(b), f
R
(c) and α (d) as a function of H
2
O/CH
4
parametric in T
SR


Fig. 8 shows the trend of energy efficiency , methane conversion x
CH4
, reforming factor f
R

for conventional ATR-based fuel processor – PEMFC systems (systems with FP.B), as a
function of O
2
/CH
4
parametric in H
2
O/CH
4
.
Methane conversion shows a monotone increase as a function of O
2
/CH
4
. The effect of
water addition on methane conversion is positive in case x
CH4
is far lower than unity,
whereas this effect can be considered as negligible when the conversion approaches to unity.
Reforming factor shows a non monotone trend as a function of O
2
/CH
4
; indeed, for low
O
2
/CH
4
values the process cannot reach the temperature values that favor the reforming
reactions, whereas for high O
2
/CH
4
values, although the reforming temperature results to
be strongly increased, the hydrogen and methane oxidation reactions are favorite, with
subsequent reduction of the amount of hydrogen produced and, thus, of the f
R
.
The addition of water leads to an increase of f
R
, being water a reactant of the reforming
reactions; this increase becomes negligible for H
2
O/CH
4
values higher than 2.
For all the O
2
/CH
4
and H
2
O/CH
4
values investigated, α remains equal to zero, therefore,
the trend of energy efficiency results to be the same of the reforming factor; moreover, there
is a waste of heat from the system, related to the autothermic nature of the process, which
hinders the possibility of recovering the energy content of the AOG.
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0
10
20
30
40
50
550
600
650
700


(
%
)
(a)
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
20
40
60
80
100
x
C
H
4

(
%
)
(b)
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
20
40
60
80
100
120
550
600
650
700
f
R

(
%
)
(c)
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0
5
10
15
20
25


(
%
)
(d)
550
600
650
700
550
600
650
700
T
SR
[°C]
T
SR
[°C]
T
SR
[°C]
T
SR
[°C]


Fig. 8.  (a), x
CH4
(b) and f
R
(c) as a function of O
2
/CH
4
parametric in H
2
O/CH
4
.

Table 2 reports the simulation results and the value of the operative parameters given as
simulation input that maximize the energy efficiency , for FP.A and for FP.B, respectively.
FP.A shows the highest global efficiency (48.0%) at T
SR
=670°C and H
2
O/CH
4
=2.5. It should
be noticed that, in the optimal conditions, methane conversion (x
CH4
) is lower than unity;
however, the non converted methane is not energetically wasted, since it contributes to the
energy content of the AOG, used to sustain the endothermicity of the SR reactor. In this
conditions, no addition of methane to the burner is needed (α=0). According to the flow
sheet of FP.A, the minimum exhaust gas temperature achievable is 226°C. Further heat
recovery is hindered by temperature cross-over in the heat exchangers.

Simulation results
xCH4 α  TEX (°C)
FP.A (SR) 91.0 0.0 48.0 226
FP.B (ATR) 98.8 0.0 38.5 444
Simulation Input
P (atm) H2O/CH4 O2/CH4 TSR (°C)
FP.A (SR) 1 2.5 - 670
FP.B (ATR) 1 4.0 0.56 -
Table 2. Conventional Fuel Processor – PEMFC systems

O
2
/CH
4
0.2 0.4 0.6 0.8 1.0 1.2 1.4
15
20
25
30
35
40
1.0
1.5
2.0
2.5
3.0
3.5


(
%
)
O
2
/CH
4
0.2 0.4 0.6 0.8 1.0 1.2 1.4
f
R

(
%
)
30
40
50
60
70
80
90
1.0
1.5
2.0
2.5
3.0
3.5
(a)
(b)
(c)
O
2
/CH
4
0.2 0.4 0.6 0.8 1.0 1.2 1.4
x
C
H
4

(
%
)
40
50
60
70
80
90
100
1.0
1.5
2.0
2.5
3.0
3.5
(b)
H2O/CH
4
H2O/CH
4
H2O/CH
4
Energy effciency of Fuel Processor – PEM Fuel Cell systems 171

weight of these factors: for low H
2
O/CH
4
,  shows a continuous increase with T
SR
in the
range investigated, whereas, for high H
2
O/CH
4
,  shows a non monotone trend with T
SR.



Fig. 7.  (a), x
CH4
(b), f
R
(c) and α (d) as a function of H
2
O/CH
4
parametric in T
SR


Fig. 8 shows the trend of energy efficiency , methane conversion x
CH4
, reforming factor f
R

for conventional ATR-based fuel processor – PEMFC systems (systems with FP.B), as a
function of O
2
/CH
4
parametric in H
2
O/CH
4
.
Methane conversion shows a monotone increase as a function of O
2
/CH
4
. The effect of
water addition on methane conversion is positive in case x
CH4
is far lower than unity,
whereas this effect can be considered as negligible when the conversion approaches to unity.
Reforming factor shows a non monotone trend as a function of O
2
/CH
4
; indeed, for low
O
2
/CH
4
values the process cannot reach the temperature values that favor the reforming
reactions, whereas for high O
2
/CH
4
values, although the reforming temperature results to
be strongly increased, the hydrogen and methane oxidation reactions are favorite, with
subsequent reduction of the amount of hydrogen produced and, thus, of the f
R
.
The addition of water leads to an increase of f
R
, being water a reactant of the reforming
reactions; this increase becomes negligible for H
2
O/CH
4
values higher than 2.
For all the O
2
/CH
4
and H
2
O/CH
4
values investigated, α remains equal to zero, therefore,
the trend of energy efficiency results to be the same of the reforming factor; moreover, there
is a waste of heat from the system, related to the autothermic nature of the process, which
hinders the possibility of recovering the energy content of the AOG.
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0
10
20
30
40
50
550
600
650
700


(
%
)
(a)
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
20
40
60
80
100
x
C
H
4

(
%
)
(b)
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
20
40
60
80
100
120
550
600
650
700
f
R

(
%
)
(c)
H
2
O/CH
4
1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0
0
5
10
15
20
25


(
%
)
(d)
550
600
650
700
550
600
650
700
T
SR
[°C]
T
SR
[°C]
T
SR
[°C]
T
SR
[°C]


Fig. 8.  (a), x
CH4
(b) and f
R
(c) as a function of O
2
/CH
4
parametric in H
2
O/CH
4
.

Table 2 reports the simulation results and the value of the operative parameters given as
simulation input that maximize the energy efficiency , for FP.A and for FP.B, respectively.
FP.A shows the highest global efficiency (48.0%) at T
SR
=670°C and H
2
O/CH
4
=2.5. It should
be noticed that, in the optimal conditions, methane conversion (x
CH4
) is lower than unity;
however, the non converted methane is not energetically wasted, since it contributes to the
energy content of the AOG, used to sustain the endothermicity of the SR reactor. In this
conditions, no addition of methane to the burner is needed (α=0). According to the flow
sheet of FP.A, the minimum exhaust gas temperature achievable is 226°C. Further heat
recovery is hindered by temperature cross-over in the heat exchangers.

Simulation results
xCH4 α  TEX (°C)
FP.A (SR) 91.0 0.0 48.0 226
FP.B (ATR) 98.8 0.0 38.5 444
Simulation Input
P (atm) H2O/CH4 O2/CH4 TSR (°C)
FP.A (SR) 1 2.5 - 670
FP.B (ATR) 1 4.0 0.56 -
Table 2. Conventional Fuel Processor – PEMFC systems

O
2
/CH
4
0.2 0.4 0.6 0.8 1.0 1.2 1.4
15
20
25
30
35
40
1.0
1.5
2.0
2.5
3.0
3.5


(
%
)
O
2
/CH
4
0.2 0.4 0.6 0.8 1.0 1.2 1.4
f
R

(
%
)
30
40
50
60
70
80
90
1.0
1.5
2.0
2.5
3.0
3.5
(a)
(b)
(c)
O
2
/CH
4
0.2 0.4 0.6 0.8 1.0 1.2 1.4
x
C
H
4

(
%
)
40
50
60
70
80
90
100
1.0
1.5
2.0
2.5
3.0
3.5
(b)
H2O/CH
4
H2O/CH
4
H2O/CH
4
Energy Effciency 172

FP.B shows the highest global efficiency (38.5%) at O
2
/CH
4
=0.56 and H
2
O/CH
4
=4.0; the
value of  is significantly lower than what achieved with FP.A, mainly due to the
autothermal nature of the ATR process, that limits the possibility to recover the energy
content of the AOG. This reflects into a higher exhaust gas temperature in FP.B (444°C) than
in FP.A (226°C).

4.2 Innovative Fuel Processors
Fuel Processors based on membrane reforming reactor
Fig. 9 reports the energy efficiency of system with FP.C as a function of pressure.
Energy efficiency rapidly increases with pressure in the range 3-5 atm, where no methane
addition to the burner is required to sustain the endothermic steam reforming reaction.


Fig. 9.  as a function of pressure for system with FP.C. Operating conditions: T
SR
=600°C,
H
2
O/CH
4
=2.5, SG/CH
4
=0

As pressure increases above 5 atm ca,  continues to grows with pressure, but at a lower
rate, because methane addition to the burner becomes necessary. The dotted line,
superimposed to Fig. 9 as an aid to this discussion, represents the value of  that would be
calculated if the methane sent to the burner was not factored in the computation.
The trend of  vs P is the combined effect of hydrogen recovery (HR), reforming factor (f
R
),
the power of the auxiliary units (related to f
a
), whose values are reported in Table 3 together
with the value of methane conversion (x
CH4
) and fraction of methane sent to the burner (α. 

P (atm) xCH4 α TEX (°C) HR fa fR 
3 70.6 0 803.8 58 0.5 80.4 27.5
5 86.3 1.8 100 85.9 0.7 100.5 50.2
7 91.8 12.8 100 91.9 0.9 108.0 51.2
9 94.5 17.2 100 94.4 1.1 111.8 51.5
12 96.6 20.4 100 96.2 1.3 114.9 51.8
15 97.6 22 100 97.1 1.4 116.7 51.9
Table 3. System with FP.C. Operating conditions: T
SR
=600°C, H
2
O/CH
4
=2.5, SG/CH
4
=0

3 5 7 9 11 13 15


(
%
)
20
30
40
50
60
70
SR
SR no CH4-B
P (atm)

In particular, HR increases with pressure due to the increase of hydrogen separation driving
force through the membrane; f
R
increases with pressure because it is positively influenced
by the trend of HR with pressure, due to the positive effect on reaction equilibrium of
increasing hydrogen separation. f
a
increases with pressure, due to increasing compression
ratios. To complete the picture, it should be kept in mind that the heating value of the
retentate decreases with pressure, as a consequence of higher x
CH4
and HR. This, in turn,
influences the quantity of methane sent to the burner to sustain the endothermic steam
reforming reaction.
In the low pressure range, the positive effect of HR and f
R
on energy efficiency overrules the
negative effect of f
a
and αThe plateau value reached at higher pressure indicates that the
drawback of f
a
and αcompensates the positive effect of HR and f
R

Fig. 10 reports the effects of SG/CH
4
on system efficiency of FP.C parametric in pressure, at
a fixed outlet exhaust gases temperature of 100°C. Simulation details for P = 10 atm are
reported in Table 4.


Fig. 10.  as a function of SG/CH
4
parametric in pressure for system with FP.C. Operating
conditions: T
SR
=600°C, H
2
O/CH
4
=2.5

SG/CH4 xCH4 α TEX (°C) HR fa fR 
0.0 95.4 18.6 100.0 95.1 1.1 113.1 51.6
0.1 99.8 25.7 100.0 99.1 1.1 119.8 52.1
0.5 100.0 28.5 100.0 100.0 1.1 121.4 51.3
1.0 100.0 30.0 100.0 100.0 1.1 121.3 50.2
1.5 100.0 31.5 100.0 100.0 1.2 121.3 49.1
2.0 100.0 33.0 100.0 100.0 1.2 121.5 48.1
Table 4. System with FP.C. Operating conditions: T
SR
=600°C, H
2
O/CH
4
=2.5, P=10 atm

It is possible to observe that 

shows a maximum as a function of SG/CH
4
ratio, which shifts
leftwards and upwards as pressure increases. For each pressure value investigated,
hydrogen recovery is enhanced by the presence of the sweep gas, as a consequence of
reduced hydrogen partial pressure in the permeate side; this leads also to an increase of f
R

thanks to the positive effect of hydrogen removal on reactions equilibrium.
However, the production of sweep gas is always coupled with addition of methane to the
burner, with an increment of α that can overrule the increment of HR and f
R
. For this reason,
SG/CH
4
0.0 0.5 1.0 1.5 2.0
0
10
20
30
40
50
3.0
5.0
10.0
15.0


(
%
)
P [atm]
Energy effciency of Fuel Processor – PEM Fuel Cell systems 173

FP.B shows the highest global efficiency (38.5%) at O
2
/CH
4
=0.56 and H
2
O/CH
4
=4.0; the
value of  is significantly lower than what achieved with FP.A, mainly due to the
autothermal nature of the ATR process, that limits the possibility to recover the energy
content of the AOG. This reflects into a higher exhaust gas temperature in FP.B (444°C) than
in FP.A (226°C).

4.2 Innovative Fuel Processors
Fuel Processors based on membrane reforming reactor
Fig. 9 reports the energy efficiency of system with FP.C as a function of pressure.
Energy efficiency rapidly increases with pressure in the range 3-5 atm, where no methane
addition to the burner is required to sustain the endothermic steam reforming reaction.


Fig. 9.  as a function of pressure for system with FP.C. Operating conditions: T
SR
=600°C,
H
2
O/CH
4
=2.5, SG/CH
4
=0

As pressure increases above 5 atm ca,  continues to grows with pressure, but at a lower
rate, because methane addition to the burner becomes necessary. The dotted line,
superimposed to Fig. 9 as an aid to this discussion, represents the value of  that would be
calculated if the methane sent to the burner was not factored in the computation.
The trend of  vs P is the combined effect of hydrogen recovery (HR), reforming factor (f
R
),
the power of the auxiliary units (related to f
a
), whose values are reported in Table 3 together
with the value of methane conversion (x
CH4
) and fraction of methane sent to the burner (α. 

P (atm) xCH4 α TEX (°C) HR fa fR 
3 70.6 0 803.8 58 0.5 80.4 27.5
5 86.3 1.8 100 85.9 0.7 100.5 50.2
7 91.8 12.8 100 91.9 0.9 108.0 51.2
9 94.5 17.2 100 94.4 1.1 111.8 51.5
12 96.6 20.4 100 96.2 1.3 114.9 51.8
15 97.6 22 100 97.1 1.4 116.7 51.9
Table 3. System with FP.C. Operating conditions: T
SR
=600°C, H
2
O/CH
4
=2.5, SG/CH
4
=0

3 5 7 9 11 13 15


(
%
)
20
30
40
50
60
70
SR
SR no CH4-B
P (atm)

In particular, HR increases with pressure due to the increase of hydrogen separation driving
force through the membrane; f
R
increases with pressure because it is positively influenced
by the trend of HR with pressure, due to the positive effect on reaction equilibrium of
increasing hydrogen separation. f
a
increases with pressure, due to increasing compression
ratios. To complete the picture, it should be kept in mind that the heating value of the
retentate decreases with pressure, as a consequence of higher x
CH4
and HR. This, in turn,
influences the quantity of methane sent to the burner to sustain the endothermic steam
reforming reaction.
In the low pressure range, the positive effect of HR and f
R
on energy efficiency overrules the
negative effect of f
a
and αThe plateau value reached at higher pressure indicates that the
drawback of f
a
and αcompensates the positive effect of HR and f
R

Fig. 10 reports the effects of SG/CH
4
on system efficiency of FP.C parametric in pressure, at
a fixed outlet exhaust gases temperature of 100°C. Simulation details for P = 10 atm are
reported in Table 4.


Fig. 10.  as a function of SG/CH
4
parametric in pressure for system with FP.C. Operating
conditions: T
SR
=600°C, H
2
O/CH
4
=2.5

SG/CH4 xCH4 α TEX (°C) HR fa fR 
0.0 95.4 18.6 100.0 95.1 1.1 113.1 51.6
0.1 99.8 25.7 100.0 99.1 1.1 119.8 52.1
0.5 100.0 28.5 100.0 100.0 1.1 121.4 51.3
1.0 100.0 30.0 100.0 100.0 1.1 121.3 50.2
1.5 100.0 31.5 100.0 100.0 1.2 121.3 49.1
2.0 100.0 33.0 100.0 100.0 1.2 121.5 48.1
Table 4. System with FP.C. Operating conditions: T
SR
=600°C, H
2
O/CH
4
=2.5, P=10 atm

It is possible to observe that 

shows a maximum as a function of SG/CH
4
ratio, which shifts
leftwards and upwards as pressure increases. For each pressure value investigated,
hydrogen recovery is enhanced by the presence of the sweep gas, as a consequence of
reduced hydrogen partial pressure in the permeate side; this leads also to an increase of f
R

thanks to the positive effect of hydrogen removal on reactions equilibrium.
However, the production of sweep gas is always coupled with addition of methane to the
burner, with an increment of α that can overrule the increment of HR and f
R
. For this reason,
SG/CH
4
0.0 0.5 1.0 1.5 2.0
0
10
20
30
40
50
3.0
5.0
10.0
15.0


(
%
)
P [atm]
Energy Effciency 174

being  combination of f
R
, HR and α, after an initial small increment, it decreases with
addition of sweep gas.
The effect of pressure depends on the SG/CH
4
value. For low SG/CH
4
, an increase of
pressure causes an increase of , whereas a decreasing trend of the  with pressure is
observed at high SG/CH
4
. This is due to the fact that the increment of pressure increases
both HR and f
a
; when SG/CH
4
is high, HR becomes close to 100% already at low pressure
values, therefore an increase of pressure only causes an increase of f
a
, lowering .
Table 5 report the detail of the simulation results and value of the operating parameters
given as simulation input that maximize the energy efficiency η, for FP.C.
The best way to operate a membrane SR system is to increase the pressure without addition
of sweep gas.
It is possible to observe that the energy efficiency of a SR-based system is increased if a
membrane reactor is used (FP.C), in place of a conventional reactor. This is due to the
possibility to recover a higher amount of heat in FP.C than in FP.A. Indeed the heat
exchanger network needed in FP.A has to satisfy the temperature requirements of the Shift
and PROX reactors resulting in a higher exhaust gas temperature (226°C), while in FP.C the
heat exchanger network allows to cool the exhaust gas to 100°C (as chosen in the
methodology), without any temperature cross over.

Simulation results
fR α HR fa  TEX (°C)
FP.C (SR) 120.0 25.6 99.2 1.3 52.2 100.0
Simulation Input
P (atm) H2O/CH4 TSR (°C) SG/CH4
FP.C (SR) 15 2.5 600 0.1
Table 5. Membrane SR – PEMFC system based on membrane reforming reactor

Fig. 11 reports energy efficiency of system with FP.D as a function of pressure. As for the
case of FP.C,  shows a continuous increase with pressure, but the values are significantly
lower, due to limited recovery of the energy contained in the retentate stream and to the
negative contribution of the compressor (see T
ex
and f
a
in Table 6).


Fig. 11.  as a function of pressure for system with FP.D. Operating conditions:
O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15, SG/CH
4
=0
3 5 7 9 11 13 15


(
%
)
0
10
20
30
40
P (atm)

It should be noted that, in FP.D, the maximum value of  (37.2%) is even lower than what is
obtained with the conventional ATR reactor ( = 38.5%). This should be attributed to the fact
that, notwithstanding the absence of the AOG stream, the dilution of the reacting mixture
with nitrogen reduces HR (affecting, in turn, also x
CH4
) leading to a retentate with relatively
high amount of methane and hydrogen, whose heating value cannot be totally recovered.
It should be pointed out that, due to the exothermic nature of the reactions, no additional
methane to the burner is required, i.e. α=0, and the exhaust gas stream leaves the plant at
quite high temperatures. Data are reported in Table 6.

P (atm) xCH4 α TEX (°C) HR fa fR 
3.0 85.2 0.0 1369.1 5.8 1.6 60.1 0.5
5.0 88.4 0.0 1248.1 60.2 2.6 67.6 21.8
7.0 90.0 0.0 1178.1 75.6 3.3 71.4 29.1
9.0 90.9 0.0 1132.7 82.7 3.9 73.8 32.7
12.0 91.8 0.0 1089.8 88.0 4.6 76.2 35.6
15.0 92.4 0.0 1063.6 90.9 5.2 77.8 37.2
Table 6. System with FP.C. Operating conditions: O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15, SG/CH
4
=0

Fig. 12 reports the energy efficiency of system with FP.D as a function of SG/CH
4

parametric in pressure. Simulation details for P = 10 atm are reported in Table 7.


Fig. 12.  as a function SG/CH
4
parametric in pressure for system with FP.D. Operating
conditions: O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15

SG/CH4 xCH4 α TEX (°C) HR fa fR 
0.0 91.3 0.0 1115.9 84.9 4.1 74.8 34.0
0.1 95.1 0.0 894.7 95 4.1 81.5 42.4
0.5 99.1 0.0 463.8 99.1 4.1 88.6 48.6
1.0 100 0.4 100.0 99.9 4.1 91.2 50.3
1.5 100 4.0 100.0 100.0 4.1 91.8 48.9
2.0 100 7.0 100.0 100.0 4.1 91.9 47.5
Table 7. System with FP.D. Operating conditions: O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15, P=10 atm

SG/CH
4
0.0 0.5 1.0 1.5 2.0
0
10
20
30
40
50
3.0
5.0
10.0
15.0


(
%
)
P [atm]
Energy effciency of Fuel Processor – PEM Fuel Cell systems 175

being  combination of f
R
, HR and α, after an initial small increment, it decreases with
addition of sweep gas.
The effect of pressure depends on the SG/CH
4
value. For low SG/CH
4
, an increase of
pressure causes an increase of , whereas a decreasing trend of the  with pressure is
observed at high SG/CH
4
. This is due to the fact that the increment of pressure increases
both HR and f
a
; when SG/CH
4
is high, HR becomes close to 100% already at low pressure
values, therefore an increase of pressure only causes an increase of f
a
, lowering .
Table 5 report the detail of the simulation results and value of the operating parameters
given as simulation input that maximize the energy efficiency η, for FP.C.
The best way to operate a membrane SR system is to increase the pressure without addition
of sweep gas.
It is possible to observe that the energy efficiency of a SR-based system is increased if a
membrane reactor is used (FP.C), in place of a conventional reactor. This is due to the
possibility to recover a higher amount of heat in FP.C than in FP.A. Indeed the heat
exchanger network needed in FP.A has to satisfy the temperature requirements of the Shift
and PROX reactors resulting in a higher exhaust gas temperature (226°C), while in FP.C the
heat exchanger network allows to cool the exhaust gas to 100°C (as chosen in the
methodology), without any temperature cross over.

Simulation results
fR α HR fa  TEX (°C)
FP.C (SR) 120.0 25.6 99.2 1.3 52.2 100.0
Simulation Input
P (atm) H2O/CH4 TSR (°C) SG/CH4
FP.C (SR) 15 2.5 600 0.1
Table 5. Membrane SR – PEMFC system based on membrane reforming reactor

Fig. 11 reports energy efficiency of system with FP.D as a function of pressure. As for the
case of FP.C,  shows a continuous increase with pressure, but the values are significantly
lower, due to limited recovery of the energy contained in the retentate stream and to the
negative contribution of the compressor (see T
ex
and f
a
in Table 6).


Fig. 11.  as a function of pressure for system with FP.D. Operating conditions:
O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15, SG/CH
4
=0
3 5 7 9 11 13 15


(
%
)
0
10
20
30
40
P (atm)

It should be noted that, in FP.D, the maximum value of  (37.2%) is even lower than what is
obtained with the conventional ATR reactor ( = 38.5%). This should be attributed to the fact
that, notwithstanding the absence of the AOG stream, the dilution of the reacting mixture
with nitrogen reduces HR (affecting, in turn, also x
CH4
) leading to a retentate with relatively
high amount of methane and hydrogen, whose heating value cannot be totally recovered.
It should be pointed out that, due to the exothermic nature of the reactions, no additional
methane to the burner is required, i.e. α=0, and the exhaust gas stream leaves the plant at
quite high temperatures. Data are reported in Table 6.

P (atm) xCH4 α TEX (°C) HR fa fR 
3.0 85.2 0.0 1369.1 5.8 1.6 60.1 0.5
5.0 88.4 0.0 1248.1 60.2 2.6 67.6 21.8
7.0 90.0 0.0 1178.1 75.6 3.3 71.4 29.1
9.0 90.9 0.0 1132.7 82.7 3.9 73.8 32.7
12.0 91.8 0.0 1089.8 88.0 4.6 76.2 35.6
15.0 92.4 0.0 1063.6 90.9 5.2 77.8 37.2
Table 6. System with FP.C. Operating conditions: O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15, SG/CH
4
=0

Fig. 12 reports the energy efficiency of system with FP.D as a function of SG/CH
4

parametric in pressure. Simulation details for P = 10 atm are reported in Table 7.


Fig. 12.  as a function SG/CH
4
parametric in pressure for system with FP.D. Operating
conditions: O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15

SG/CH4 xCH4 α TEX (°C) HR fa fR 
0.0 91.3 0.0 1115.9 84.9 4.1 74.8 34.0
0.1 95.1 0.0 894.7 95 4.1 81.5 42.4
0.5 99.1 0.0 463.8 99.1 4.1 88.6 48.6
1.0 100 0.4 100.0 99.9 4.1 91.2 50.3
1.5 100 4.0 100.0 100.0 4.1 91.8 48.9
2.0 100 7.0 100.0 100.0 4.1 91.9 47.5
Table 7. System with FP.D. Operating conditions: O
2
/CH
4
=0.48, H
2
O/CH
4
=1.15, P=10 atm

SG/CH
4
0.0 0.5 1.0 1.5 2.0
0
10
20
30
40
50
3.0
5.0
10.0
15.0


(
%
)
P [atm]
Energy Effciency 176

The trend of  with SG/CH
4
and pressure is similar to the one observed for the system
based on SR. However, it is important to note that for each pressure value investigated, the
SG/CH
4
value that maximizes energy efficiency is higher than the corresponding one in the
SR-based fuel processor.
This is due to the fact that in an ATR-based system, there is an excess energy due to the
autothermic nature of the process, that allows a consistent sweep gas production without
methane addition to the burner, i.e. α=0.
Moreover, it is worth noting that energy efficiency of FP.D is highly improved by adding
sweep gas, increasing from 34.0% (SG/CH
4
=0) to 50.3% (SG/CH
4
=1.0).
Table 8 report the detail of the simulation results and value of the operating parameters
given as simulation input that maximize the energy efficiency η, for FP.D.

Simulation results
fR α HR fa  TEX (°C)
FP.D 90.2 0.0 99.6 3.3 50.6 100.0
Simulation Input
P (atm) H2O/CH4 O2/CH4 SG/CH4
FP.D (ATR) 7 1.2 0.5 1.0
Table 8. Membrane ATR – PEMFC system based on membrane reforming reactor.

The best way to operate an autothermal reforming membrane system is to moderately
increase pressure and to employ some sweep gas to improve HR (the maximum  is reached
for P=7 atm and SG/CH
4
=1.0, as reported in Table 7).
The lower value of pressure that maximize  with respect to SR system is due to the higher
power required by the auxiliary units, needed essentially to compress the air in the feed.
Finally, it should be noted that the addition of sweep gas in system with FP.D allows
reaching energy efficiency values significantly higher than the optimum value of the
conventional system (38.5%) and similar to the energy efficiency of SR based systems.
It should be kept in mind that, due to limited thermal stability of the highly selective
membranes, membrane units should not be exposed to temperatures higher than 600°C.
While FP.C always meets this constraint (since reactor temperature is fixed at 600°C), FP. D
does not. Indeed, in the optimal conditions, the first reactors reach temperatures as high as
720°C. Therefore, the actual realization of an integrated membrane reactor would require
significant improvements of membrane compatibility with high temperatures. A more
realistic configuration of an ATR based membrane reactor should consider a first ATR
reactor, where most of the methane oxidation takes place, followed by a membrane reactor,
interposing between the two units a heat exchanger to cool down the temperature before
entrance into the membrane reactor, so that the membranes are never exposed to
temperatures higher than 600°C. With this configuration, energy efficiency becomes 48.5%
and the best operating conditions are P=7 atm; O
2
/CH
4
=0.5; H
2
O/CH
4
=1.7; SG/CH
4
=1.0.

Fuel Processors based on membrane WGS reactor
Optimization performed for systems based on membrane WGS reactors (FP.E for SR and
FP.F for ATR) followed the same criteria of what reported for systems based on membrane
reforming reactors. Although quantitatively different, the trend of performance with
operating parameters were similar to what reported for the systems with membrane
reforming reactors, therefore data are not reported for the sake of brevity.

Table 9 reports the simulation results and the value of the operating parameters given as
simulation input that maximize the energy efficiency , for FP.E and for FP.F, respectively.
It is possible to observe that the introduction of the membrane in the WGS reactor allows
obtaining higher energy efficiencies than what achieved in the conventional systems.

Simulation results
fR Α HR fa  TEX (°C)
FP.E (SR) 110.9 18.4 96.8 0.5 52.2 141.5
FP.F (ATR) 83.0 0.0 99.4 1.9 47.6 100.0
Simulation Input
P (atm) H2O/CH4 O2/CH4 SG/CH4 TSR (°C) TWGS (°C)
FP.E (SR) 3 2.0 - 0.2 800 300
FP.F (ATR) 3 1.2 0.6 1.9 - 300
Table 9. Innovative Fuel Processor – PEMFC systems based on membrane WGS reactor

As far as system with FP.E is concerned, the temperature value required for system
optimization corresponds to the highest value investigated; this is due to the positive effect
of temperature on the SR reactor, and thus on the membrane WGS reactor, that overcomes
the negative effect of temperature increase on α.
The maximum efficiency value is limited by the problem of a not complete heat recovery of
the exhaust gases (T
EX
>100°C); this is due to the problem of temperature cross-over that can
arise in the heat exchangers when the system works at high SR temperatures.
Since the endothermic nature of the process imposes the necessity of operating with
additional methane to the burner, the amount of sweep gas required to optimize the system
is small (SG/CH
4
=0.2).
It is also possible to observe that the pressure value required for system optimization
corresponds to the lowest value investigated; this is due to the negative effect of pressure on
the SR reactor, which overcomes the positive effect of pressure increase on the membrane
WGS reactor. This one, indeed, allows reaching a high HR, notwithstanding the low
pressure value, thanks to the high hydrogen concentration achieved at the outlet of the SR
reactor, which positively acts on the driving force.
As far as system with FP.F is concerned, it is possible to observe that the value of H
2
O/CH
4

that maximizes the energy efficiency is by far lower than what required for the conventional
case. For the ATR systems, indeed, the autothermal nature of the process allows to have an
excess heat in the system that can be used to produce steam. In the conventional system, the
steam can be used only as reactant, with only moderate improvement of energy efficiency
for H
2
O/CH
4
>3, thus making further steam production useless. In the innovative system,
the steam can be used as reactant as well as sweep gas and the energy efficiency resulted to
be favored more by an increase of SG/CH
4
than by an increase of H
2
O/CH
4
.
The autothermal nature of the process allows operating with no additional methane to the
burner and the high amount of sweep gas allows the system to operate at low pressure
values, favoring the conditions in the ATR reactor.
Although working at the same pressure, the fraction of inlet methane required to run the
auxiliary unit is higher in the ATR case than in the SR case, for the presence of air in the feed
(f
a
=0.5 for FP.E and 1.9 for FP.F).
It is also possible to note that the introduction of the membrane in the WGS reactor not only
allows to reach efficiency values higher than what achieved in the conventional systems, but
Energy effciency of Fuel Processor – PEM Fuel Cell systems 177

The trend of  with SG/CH
4
and pressure is similar to the one observed for the system
based on SR. However, it is important to note that for each pressure value investigated, the
SG/CH
4
value that maximizes energy efficiency is higher than the corresponding one in the
SR-based fuel processor.
This is due to the fact that in an ATR-based system, there is an excess energy due to the
autothermic nature of the process, that allows a consistent sweep gas production without
methane addition to the burner, i.e. α=0.
Moreover, it is worth noting that energy efficiency of FP.D is highly improved by adding
sweep gas, increasing from 34.0% (SG/CH
4
=0) to 50.3% (SG/CH
4
=1.0).
Table 8 report the detail of the simulation results and value of the operating parameters
given as simulation input that maximize the energy efficiency η, for FP.D.

Simulation results
fR α HR fa  TEX (°C)
FP.D 90.2 0.0 99.6 3.3 50.6 100.0
Simulation Input
P (atm) H2O/CH4 O2/CH4 SG/CH4
FP.D (ATR) 7 1.2 0.5 1.0
Table 8. Membrane ATR – PEMFC system based on membrane reforming reactor.

The best way to operate an autothermal reforming membrane system is to moderately
increase pressure and to employ some sweep gas to improve HR (the maximum  is reached
for P=7 atm and SG/CH
4
=1.0, as reported in Table 7).
The lower value of pressure that maximize  with respect to SR system is due to the higher
power required by the auxiliary units, needed essentially to compress the air in the feed.
Finally, it should be noted that the addition of sweep gas in system with FP.D allows
reaching energy efficiency values significantly higher than the optimum value of the
conventional system (38.5%) and similar to the energy efficiency of SR based systems.
It should be kept in mind that, due to limited thermal stability of the highly selective
membranes, membrane units should not be exposed to temperatures higher than 600°C.
While FP.C always meets this constraint (since reactor temperature is fixed at 600°C), FP. D
does not. Indeed, in the optimal conditions, the first reactors reach temperatures as high as
720°C. Therefore, the actual realization of an integrated membrane reactor would require
significant improvements of membrane compatibility with high temperatures. A more
realistic configuration of an ATR based membrane reactor should consider a first ATR
reactor, where most of the methane oxidation takes place, followed by a membrane reactor,
interposing between the two units a heat exchanger to cool down the temperature before
entrance into the membrane reactor, so that the membranes are never exposed to
temperatures higher than 600°C. With this configuration, energy efficiency becomes 48.5%
and the best operating conditions are P=7 atm; O
2
/CH
4
=0.5; H
2
O/CH
4
=1.7; SG/CH
4
=1.0.

Fuel Processors based on membrane WGS reactor
Optimization performed for systems based on membrane WGS reactors (FP.E for SR and
FP.F for ATR) followed the same criteria of what reported for systems based on membrane
reforming reactors. Although quantitatively different, the trend of performance with
operating parameters were similar to what reported for the systems with membrane
reforming reactors, therefore data are not reported for the sake of brevity.

Table 9 reports the simulation results and the value of the operating parameters given as
simulation input that maximize the energy efficiency , for FP.E and for FP.F, respectively.
It is possible to observe that the introduction of the membrane in the WGS reactor allows
obtaining higher energy efficiencies than what achieved in the conventional systems.

Simulation results
fR Α HR fa  TEX (°C)
FP.E (SR) 110.9 18.4 96.8 0.5 52.2 141.5
FP.F (ATR) 83.0 0.0 99.4 1.9 47.6 100.0
Simulation Input
P (atm) H2O/CH4 O2/CH4 SG/CH4 TSR (°C) TWGS (°C)
FP.E (SR) 3 2.0 - 0.2 800 300
FP.F (ATR) 3 1.2 0.6 1.9 - 300
Table 9. Innovative Fuel Processor – PEMFC systems based on membrane WGS reactor

As far as system with FP.E is concerned, the temperature value required for system
optimization corresponds to the highest value investigated; this is due to the positive effect
of temperature on the SR reactor, and thus on the membrane WGS reactor, that overcomes
the negative effect of temperature increase on α.
The maximum efficiency value is limited by the problem of a not complete heat recovery of
the exhaust gases (T
EX
>100°C); this is due to the problem of temperature cross-over that can
arise in the heat exchangers when the system works at high SR temperatures.
Since the endothermic nature of the process imposes the necessity of operating with
additional methane to the burner, the amount of sweep gas required to optimize the system
is small (SG/CH
4
=0.2).
It is also possible to observe that the pressure value required for system optimization
corresponds to the lowest value investigated; this is due to the negative effect of pressure on
the SR reactor, which overcomes the positive effect of pressure increase on the membrane
WGS reactor. This one, indeed, allows reaching a high HR, notwithstanding the low
pressure value, thanks to the high hydrogen concentration achieved at the outlet of the SR
reactor, which positively acts on the driving force.
As far as system with FP.F is concerned, it is possible to observe that the value of H
2
O/CH
4

that maximizes the energy efficiency is by far lower than what required for the conventional
case. For the ATR systems, indeed, the autothermal nature of the process allows to have an
excess heat in the system that can be used to produce steam. In the conventional system, the
steam can be used only as reactant, with only moderate improvement of energy efficiency
for H
2
O/CH
4
>3, thus making further steam production useless. In the innovative system,
the steam can be used as reactant as well as sweep gas and the energy efficiency resulted to
be favored more by an increase of SG/CH
4
than by an increase of H
2
O/CH
4
.
The autothermal nature of the process allows operating with no additional methane to the
burner and the high amount of sweep gas allows the system to operate at low pressure
values, favoring the conditions in the ATR reactor.
Although working at the same pressure, the fraction of inlet methane required to run the
auxiliary unit is higher in the ATR case than in the SR case, for the presence of air in the feed
(f
a
=0.5 for FP.E and 1.9 for FP.F).
It is also possible to note that the introduction of the membrane in the WGS reactor not only
allows to reach efficiency values higher than what achieved in the conventional systems, but
Energy Effciency 178

also makes the SR and ATR based systems similar in terms of energy efficiency (the
difference between SR and ATR in the conventional case is ca. 20%, whereas in this case it is
only ca. 8%).

5. Conclusions
As a general conclusion on system analysis, the optimum of each fuel processor – PEMFC
system and the corresponding operating parameters are reported in Table 9.
It is possible to observe that the SR-based processes always show higher energy efficiency
than the corresponding ATR-based processes, with a marked difference in the case of
conventional systems (FP.A and FP.B have a difference of about 21% in the energy efficiency
value). However, the introduction of the membrane allows to obtain energy efficiency
values of the ATR system closer to the efficiency levels reached in the SR ones (differences
between SR and ATR based systems of ca. 7% when the membrane is introduced in the
reforming reactor and of ca. 9% when the membrane is introduced in the WGS reactor).

Case H2O/CH4 O2/CH4 TSR [°C] SG/CH4 P [atm]  %
SR
FP.A 2.5 - 670 - 1 48.0
FP.C 2.5 - 600 0.1 15 52.1
FP.E 2.0 - 800 0.2 3 52.2
ATR
FP.B 4.0 0.56 - - 1 38.5
FP.D 1.7 0.5 - 1.0 7 48.5
FP.F 1.2 0.6 - 1.9 3 47.6
Table 10. Comparison of various FP – PEMFC systems in correspondence of operating
conditions that maximize system performance

The comparison between the steam reforming based systems (innovative systems with FP.C
and FP.E vs conventional system with FP.A) showed that the employment of a membrane
reactor can increase system efficiency from 48.0% to values above 52.0%. Such an efficiency
increase requires almost no addition of sweep gas due to the endothermic nature of the
process.
The pressure that optimizes the energy efficiency of the two membrane-based system is
different; the system with integrated reforming reactor (FP.C) requires to operate at high
pressure value (15 atm), whereas the system with membrane WGS reactor (FP.E) at low
pressure value (3 atm). This is due to the fact that the SR reactor is negatively influenced by
the pressure increase; therefore the system is optimized by increasing the hydrogen
recovery in the membrane WGS reactor by increasing hydrogen concentration at the inlet of
the WGS reactor more than by increasing pressure.
As regards temperature, all systems require to operate at the highest possible temperature
compatible with material stability.
However, although the limit on temperature imposed to the system with membrane
reforming reactor is more tighten, energy efficiency results to be as high as the value
reached in the system with membrane WGS reactor, that operates at high SR temperature.
This is due to the fact than the hydrogen removal from the reaction environment allows to
achieve higher performance at lower temperature.
The comparison between the autothermal reforming systems (innovative systems with FP.D
and FP.F vs conventional system FP.B) shows that energy efficiency can be improved from

38.5% to values around 48%, if a membrane reactor is employed. To obtain such an energy
efficiency improvement, sweep gas addition is required.
The considerations on pressure are the same of what reported for the SR case, although the
system with membrane reforming reactor is optimized at pressure values lower that the SR
case (7 atm instead of 15 atm) due to the higher value of power required to run the auxiliary
units.
It is possible to observe that the value of H
2
O/CH
4
that maximizes the energy efficiency of
the innovative ATR systems is far lower than what required for the conventional case.
Indeed, in the innovative systems, the steam can be used as reactant and as sweep gas and
the energy efficiency resulted to be favored more by an increase of SG/CH
4
than by an
increase of H
2
O/CH
4
.

6. References
Ahmed, S. Krumpelt, M. (2001). Hydrogen from hydrocarbon fuels for fuel cells,
International Journal of Hydrogen Energy, 26, 291–230, ISSN 0360-3199
Aspen Technology Inc. http://www.aspentech.com
Basile, A. Chiappetta, G. Tosti, S. Violante, V. (2001). Experimental and simulation of both
Pd and Pd/Ag for a water gas shift membrane reactor, Separation and purification
technology, 25, 549-571, ISSN 1383-5866
Campanari, S. Macchi, E. Manzolini, G. (2008). Innovative membrane reformer for hydrogen
production applied to PEM micro-cogeneration: Simulation model and
thermodynamic analysis, International Journal of Hydrogen Energy, 33, 1361-1373,
ISSN 0360-3199
Ersoz, A. Olgun, H. Ozdogan, S. (2006). Reforming options for hydrogen production from
fossil fuels for PEM fuel cells, Journal of Power Sources, 154, 67-73, ISSN 0378-7753
Francesconi, JA. Mussati, MC. Mato, RO. Aguirre, PA. (2007). Analysis of the Energy
efficiency of a integrated ethanol processor for PEM fuel cell systems, Journal of
Power Sources, 167, 151–161, ISSN 0378-7753
Hou, J. Zhuang, M. Wag, G. (2007). The analysis for the efficiency properties of the fuel cell
engine, Renewable Energy, 32, 1175-1186, ISSN 0960-1481
Larminie, J. & Dicks, A. (2004). Fuel Cell Systems Explained, Second Edition, John Wiley & Sons
Ltd, 0-470-84857-X, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ,
England
Lampert, J. (2004). Selective catalytic oxidation: a new catalytic approach to the
desulfurization of natural gas and liquid petroleum gas for fuel cell reformer
applications, Journal of Power Sorces, 131, 27-34, ISSN 0378-7753
Lattner, R. Harold, MP. (2004). Comparison of methanol-based fuel processors for PEM fuel
cell systems. International Journal of Hydrogen Energy, 29, 393-417, ISSN 0360-3199
Lattner, R. Harold, MP. (2005). Comparison of conventional and membrane reactor fuel
processors for hydrocarbon-based PEM fuel cell systems, Applied Catalysis B:
Environmental, 56, 149-169, ISSN 0926-3373
Lyubovsky, M. Walsh, D. (2006). Reforming system for co-generation of hydrogen and
mechanical work, Journal of Power Sources, 157, 430–437, ISSN 0378-7753
Energy effciency of Fuel Processor – PEM Fuel Cell systems 179

also makes the SR and ATR based systems similar in terms of energy efficiency (the
difference between SR and ATR in the conventional case is ca. 20%, whereas in this case it is
only ca. 8%).

5. Conclusions
As a general conclusion on system analysis, the optimum of each fuel processor – PEMFC
system and the corresponding operating parameters are reported in Table 9.
It is possible to observe that the SR-based processes always show higher energy efficiency
than the corresponding ATR-based processes, with a marked difference in the case of
conventional systems (FP.A and FP.B have a difference of about 21% in the energy efficiency
value). However, the introduction of the membrane allows to obtain energy efficiency
values of the ATR system closer to the efficiency levels reached in the SR ones (differences
between SR and ATR based systems of ca. 7% when the membrane is introduced in the
reforming reactor and of ca. 9% when the membrane is introduced in the WGS reactor).

Case H2O/CH4 O2/CH4 TSR [°C] SG/CH4 P [atm]  %
SR
FP.A 2.5 - 670 - 1 48.0
FP.C 2.5 - 600 0.1 15 52.1
FP.E 2.0 - 800 0.2 3 52.2
ATR
FP.B 4.0 0.56 - - 1 38.5
FP.D 1.7 0.5 - 1.0 7 48.5
FP.F 1.2 0.6 - 1.9 3 47.6
Table 10. Comparison of various FP – PEMFC systems in correspondence of operating
conditions that maximize system performance

The comparison between the steam reforming based systems (innovative systems with FP.C
and FP.E vs conventional system with FP.A) showed that the employment of a membrane
reactor can increase system efficiency from 48.0% to values above 52.0%. Such an efficiency
increase requires almost no addition of sweep gas due to the endothermic nature of the
process.
The pressure that optimizes the energy efficiency of the two membrane-based system is
different; the system with integrated reforming reactor (FP.C) requires to operate at high
pressure value (15 atm), whereas the system with membrane WGS reactor (FP.E) at low
pressure value (3 atm). This is due to the fact that the SR reactor is negatively influenced by
the pressure increase; therefore the system is optimized by increasing the hydrogen
recovery in the membrane WGS reactor by increasing hydrogen concentration at the inlet of
the WGS reactor more than by increasing pressure.
As regards temperature, all systems require to operate at the highest possible temperature
compatible with material stability.
However, although the limit on temperature imposed to the system with membrane
reforming reactor is more tighten, energy efficiency results to be as high as the value
reached in the system with membrane WGS reactor, that operates at high SR temperature.
This is due to the fact than the hydrogen removal from the reaction environment allows to
achieve higher performance at lower temperature.
The comparison between the autothermal reforming systems (innovative systems with FP.D
and FP.F vs conventional system FP.B) shows that energy efficiency can be improved from

38.5% to values around 48%, if a membrane reactor is employed. To obtain such an energy
efficiency improvement, sweep gas addition is required.
The considerations on pressure are the same of what reported for the SR case, although the
system with membrane reforming reactor is optimized at pressure values lower that the SR
case (7 atm instead of 15 atm) due to the higher value of power required to run the auxiliary
units.
It is possible to observe that the value of H
2
O/CH
4
that maximizes the energy efficiency of
the innovative ATR systems is far lower than what required for the conventional case.
Indeed, in the innovative systems, the steam can be used as reactant and as sweep gas and
the energy efficiency resulted to be favored more by an increase of SG/CH
4
than by an
increase of H
2
O/CH
4
.

6. References
Ahmed, S. Krumpelt, M. (2001). Hydrogen from hydrocarbon fuels for fuel cells,
International Journal of Hydrogen Energy, 26, 291–230, ISSN 0360-3199
Aspen Technology Inc. http://www.aspentech.com
Basile, A. Chiappetta, G. Tosti, S. Violante, V. (2001). Experimental and simulation of both
Pd and Pd/Ag for a water gas shift membrane reactor, Separation and purification
technology, 25, 549-571, ISSN 1383-5866
Campanari, S. Macchi, E. Manzolini, G. (2008). Innovative membrane reformer for hydrogen
production applied to PEM micro-cogeneration: Simulation model and
thermodynamic analysis, International Journal of Hydrogen Energy, 33, 1361-1373,
ISSN 0360-3199
Ersoz, A. Olgun, H. Ozdogan, S. (2006). Reforming options for hydrogen production from
fossil fuels for PEM fuel cells, Journal of Power Sources, 154, 67-73, ISSN 0378-7753
Francesconi, JA. Mussati, MC. Mato, RO. Aguirre, PA. (2007). Analysis of the Energy
efficiency of a integrated ethanol processor for PEM fuel cell systems, Journal of
Power Sources, 167, 151–161, ISSN 0378-7753
Hou, J. Zhuang, M. Wag, G. (2007). The analysis for the efficiency properties of the fuel cell
engine, Renewable Energy, 32, 1175-1186, ISSN 0960-1481
Larminie, J. & Dicks, A. (2004). Fuel Cell Systems Explained, Second Edition, John Wiley & Sons
Ltd, 0-470-84857-X, The Atrium, Southern Gate, Chichester, West Sussex PO19 8SQ,
England
Lampert, J. (2004). Selective catalytic oxidation: a new catalytic approach to the
desulfurization of natural gas and liquid petroleum gas for fuel cell reformer
applications, Journal of Power Sorces, 131, 27-34, ISSN 0378-7753
Lattner, R. Harold, MP. (2004). Comparison of methanol-based fuel processors for PEM fuel
cell systems. International Journal of Hydrogen Energy, 29, 393-417, ISSN 0360-3199
Lattner, R. Harold, MP. (2005). Comparison of conventional and membrane reactor fuel
processors for hydrocarbon-based PEM fuel cell systems, Applied Catalysis B:
Environmental, 56, 149-169, ISSN 0926-3373
Lyubovsky, M. Walsh, D. (2006). Reforming system for co-generation of hydrogen and
mechanical work, Journal of Power Sources, 157, 430–437, ISSN 0378-7753
Energy Effciency 180

Manzolini, G. Tosti, S. (2008). Hydrogen production from ethanol steam reforming: energy
efficiency analysis of traditional and membrane processes, International Journal of
Hydrogen Energy, 33, 5571-5582, ISSN 0360-3199
Pearlman, B. Bhargav, A. Shields, EB. Jackson, GS. Hearn, PL. (2008). Modeling efficiency
and water balance in PEM fuel cell systems with liquid fuel processing and
hydrogen membranes, Journal of Power Sources, 185, 1056–1065, ISSN 0378-7753
Ratnamala, GM. Shah, N. Mehta, V. Rao, PV. (2005). Integrated Fuel Cell Processor for a 5-
kW Proton-Exchange Membrane Fuel Cell, Industrial & Engineering Chemistry
Research, 44, 1535-1541, ISSN 0888-5885
Semelsberger, TA. Brown, LF. Borup, RL. Inbody, MA. (2004). Equilibrium products from
autothermal processes for generating hydrogen-rich fuel-cell feeds, International
Journal of Hydrogen Energy, 29, 1047-1064, ISSN 0360-3199
Seo, YS. Shirley, A. Kolaczkowski, ST. (2002). Evaluation of thermodynamically favourable
operating conditions for production of hydrogen in three different reforming
technologies, Journal of Power Sources, 108, 213–225, ISSN 0378-7753
Shu, J. Grandjean, BPA. van Neste, A. Kaliaguine, S. (1991). Catalytic palladium-based
membrane reactors: a review, Canadian Journal of Chemical Engineering, 69, 1036-
1060, ISSN 0008-4034
Song, C. (2002). Fuel processing for low-temperature and high-temperature fuel cells:
Challenges, and opportunities for sustainable development in the 21st century,
Catalysis Today, 77, 17–49, ISSN 0920-5861
Xu, J. Froment, GF. (1989). Methane Steam Reforming, Methanation and Water-Gas Shift.
Intrinsic Kinetics, AIChE Journal, 35, 88-96, ISSN 0001-1541