A Review of Essential Oil Extr Action Technologies

Mint Oil Energy Consumption, Energy Use Efficiency and Distillation Processes A Re v i e w o f E s s e n t i a l O i l E x t r a c t i o n Te c h n o l o g i e s
Prepared by: Scott Sanford August 12, 2011 (revised) University of Wisconsin-Madison Biological Systems Engineering 460 Henry Mall Madison, WI 53706 Phone: 608-262-5062 E-mail: [email protected]

1

Project funded by a grant from the Mint Industry Research Council.

2

Review of Essential Oil Extraction Technologies
The common method for extraction of mint oil used by mint growers in the U.S. is steam distillation. The mint hay is chopped directly into a movable tub (wagon) and then moved to the distillation area where it is connected to the steam source and condensate recovery system. A study by Hackleman et al. in 2006 found many things that could be done to improve energy efficiency of current distillation systems. However the technology being used today is about 50 years old and newer more energy efficient methods are available for the extraction of mint oil. Steam distillation or water distillation have been used since antiquity. Energy for distillation ranges from 2.5 kWh/kg to 4 kWh/kg (3868—6190 Btu/lb) of raw product (plant material) (Berka-Zougali 2010). This paper will look at the different process that can be used to extract essential oils from plant matter. Water distillation Water distillation is generally used for extraction oils from dried or powdered plant materials such as spice powders, ground woody plants such as cinnamon bark, some flowers like rose or tough materials such as roots, wood or nuts. Plant materials are immersed in water, allowed to soak and then boiled with direct heat. The pressure on the still can be reduced or increased to change the distillation products. The volatile components are mostly extracted at a temperature just below 100°C by diffusion mechanism. The volatiles are transported with the steam to a condenser and oil is separated using a Florentine Vessel. The remaining water in the vessel after distillation can be transferred to another vessel hot and redistilled reducing energy and water use. An advantage of water distillation is the plant material is always in contact with boiling water. The vessel should be stirred to prevent material from clumping or settling to the bottom of the vessel. Some disadvantages included low water levels causing overheating or charring resulting in off-notes (off-flavors), lower quality oil, slower process resulting in higher energy use, incomplete release of essential oil, and requires more stills because of lower density in the still. Water-Steam Distillation This process uses similar equipment to water distillation but is a combination of water and steam distillation processes. It can be used for mint or other leafy plants. Plant material is loaded in a vessel on a grate with water below. The water is boiled creating a low-pressure wet steam. The plant material must be uniformly loaded into the vessel otherwise steam will bypass some of the material. This method can cause overheating of the material on the vessel walls through conduction if the still is direct fired resulting in off-notes. The wet steam can saturate the plant material on the grates and slow down the distillation process. The method does decrease processing time and improves energy use compared to water distillation along with high yields and better quality oil. Steam Distillation This is the most commonly used method for commercial scale extraction of essential oils including all types of mint. This method is similar to the other methods except steam is supplied by an external source. Dry steam is injected under the plant material that is resting on a perforated floor. The material needs to be distributed evenly on the floor so the steam will flow evenly through the material. The oil is extracted by diffusion as the steam passes through the material. The steam-oil mixture then passes through the condenser and the oil is decanted in a Florentine vessel. This process is more energy efficient, cost effective, better process control, produces more consistent oil and less likely to damage oils. A disadvantage of steam distillation
3

is channeling of the steam through the plant material resulting in lower extraction rates. This can be reduced by chopping the material which is a common practice but this can result in losing considerable amounts of oil (Öztekin 2007) during wilting and chopping. Mixing the material in the still (such as an auger) during processing can also reduce channeling of the steam. Even though steam distillation is the most energy efficient of the distillation processes it still requires a large amount of energy per pound of oil and large amounts of water for cooling. Base on supplied by MIRC, it requires about 79,500 Btu per pound of oil recovered. Distillation has long processing times, about 2+ hours, and is energy intensive compared to other methods of oil extraction. Lawrence (1995) estimated the cost for a 4 tub distillation system in the U.S. would cost $218,000 ($1992) ($329,000—$2009) and have the capacity to support 400 acres of mint. Hydro Distillation This method uses steam at atmospheric pressure passed into the plant material from overhead. The advantage to this version is the steam saturates the plant materials more evenly in less time than with steam distillation. The condenser is located under a basket or perforate floor that the material is placed on. This method results in higher quality oil that smells more like the original plant. Continuous Steam Distillation A continuous steam distillation process has been used in Russia since the late 1970’s. In the mid 1980’s, Bouchard et.al. (1986) and Bouchard and Serth (1991) describe a continuous process for extracting oil from cedarwood by Texarome (http://www.texarome.com/) in Texas. The logs were pulverized and pneumatically conveyed with steam as the carrier into a distillation chamber which is approximately a ½ mile of pipe. The conditions in the chamber were controlled to vaporize the volatile with the highest boiling point that was desired. The residence time in the distillation chamber was only 25-30 seconds. This commercial system is capable of processing 13 tons per day. The spent wood could be used for boiler fuel after extraction. Based on the process conditions, the energy per pound of material processed would be approximately 1154 Btu per pound of input plant material which would translate to approximately 23,000 Btu per pound of mint oil based on the same processing parameters as used for cedarwood. If mint could be processed in this type of system it would reduce energy costs by approximately 60 to 70%. The capacity of this plant would be too low for mint. Assuming the same capacity per day, the 13 ton plant could only process 3.25 acres per day based on 4 tons per acre. No patent was found. Texarome does offer consulting services for distillation construction. Carle and Fiedler describe the use of a cylindrical continuous distillation system under the brand name of Padovan (De Silva 1995—pg 96). It appears that Padovan no longer makes distillation units (www.padovan.com). Arnaudo describes the plant material being fed through the distillation chamber using an Archimedian screw (auger) and steam passage in countercurrent flow. This system works best with powdered materials and is being used in France for fennel oil production. Arnaudo also describes a continuous distillation system, named DCF, that cascades material through a series of auger modules. The augers configuration creates plugs of material to prevent the escape of steam. This type of system uses less steam and is ideal for low boiling point oils such as peppermint. Another continuous distillation system is described by Arnaudo (1991) and Vacchiano (1992) is known as the Biolandes process. Biolandes is a company based in France that specializes in
4

essential oils and natural extracts. They developed a continuous distillation process and hold several patents on the equipment developed for the process. A unit comprised of two 265-cubic foot (7.5 cu. meters) vessels can process 3 tons of pine needles per hour. Figures 1 show illustrations from the patent for the distillation vessel. Figure 2 is an illustration of the overall system. The steam and oil vapor would traditionally be condensed using lots of cold water but Biolandes route the steam mixture through an aerothemic radiator (air cooled), item 42 in FigFigure 1: Biolanders Patent illustrations (Source: US Patent 5,024,820)

Figure 2: Biolanders Continuous Steam Distillation System Schematic (Source: US Patent 4,935,104)

5

ure 2, with a high thermal transfer capacity to transform the 100°C steam mixture to 100°C water. The hot air is used to dry the distillation residues that are immediately feed into the boiler to create the steam. This reduces water demand and energy expenditures. U.S. Patent 4,935,104 covers the process described above and expired June of 2010. A second U.S. Patent 5,024,820 covers the loading and unloading system used for the continuous still which uses the plant material to form a plug to prevent the steam from escaping, Figure 1. This patent will expire in June 2011. Figure 3: Patent Illustration for Rathbun and Thalheimer

6

The plant material plug to prevent steam from escaping was first used in U.S. Patent 4,495,033, Figure 3, which uses a series of augers in channels with steam emitted into the plant material as it is moved down the channel. The system can be setup with a series of augers that cascades the plant material from one auger to the next to increase distillation time and mixing of material. This was developed by Albert Thalheimer, a mint grower, and Robert Rathbun, a machine shop owner, both from Toppenish, WA. This still was never commercialized. The patent expired in 2005. The patent application claims the unit was more energy efficient, a 65% energy savings, but no data was published or is available to substantiate the claim. The process time is claimed to be 5 to 8 minutes and could be setup to operate automatically 24 hours per day. No processing capacity was stated. U.S. Patent 5,891,501 (1999) describes the use of a surfactant to improve oil yields from steam distillation. The surfactant can be applied at the time of mowing, to the windrow as the mint hay is being feed into a chopper or as it is transferred to the distillation tub. Based on data provided in the patent from 11 tubs with surfactant and 11 tubs without, the surfactant resulted in a 7.6% increase in oil recovered. The inventor was contacted several times to find out if this was commercially available but never returned calls. There is concern if the surfactant were to end up in the oil or affect the separation of the oil from the water. Cold Pressing / Expression Cold Pressing is a method used for citrus oil extraction and oilseeds but is not practical for mint oil. Solvent Extraction Solvent extraction is the most widely used extraction process for extracting oils from plants. The plant material is immersed into the solvent and the oils diffuse into the solvent. The solvent are removed from the oils by distillation or evaporation. There are various materials that can be used for solvents depending on the material being extracted. There are four classes of solvents. Low boiling point organic solvents which would include hexane, propane, butane, methanol, ethanol and others. Water can be used as a solvent along with fats or waxes. There are also liquefied gaseous solvents such as carbon dioxide and freons. Each solvent has drawbacks, there isn’t a perfect solvent. The properties of a perfect solvent would be as follows: doesn’t dissolve water, has low viscosity, high solution capacity, low latent heat of vaporization, low boiling point, stable and inert, non-toxic food grade, readily available, recoverable, doesn’t leave a residue, nonflammable, inexpensive and environmentally friendly. Hexane appears to be the most widely used, predominantly for oil seeds. Types of solvent extractors include Static extractors and Rotational extractors. A static extractor is typically a cylindrical vessel in which the plant material is placed and the solvent is successively circulated. The solvent is removed by draining, followed by a steam cycle to strip the solvent from the plant material. The essential oils are extracted from the solvent by evaporation of the solvent. The solvent is generally condensed and reused for environmental and cost reasons. A rotational extractor is similar to a drum washing machine. The process is similar to the static extractor but agitation aids in increasing contact between the plant material and the solvent. Biolandes (Vacchiano - 1992) has a continuous solvent extraction process that works on the same principle as the continuous steam extraction system but the steam is replaced with a solvent. The plant material is introduced at the top and distributed with a spreading mechanism and the solvent is introduced at the bottom of the vessel and flows up through the plant material. The exhausted material is de-solventized by compression and then run through a vessel that strips the remaining solvent with steam.
7

A Rotocel or Carrousel is a continuous extractor used in the oil seed industry which consists of 18 pie shaped cells located in a circle, Figure 4. The cells are loaded and then rotate around the vessel while solvent is sprayed on the plant matter and allowed to percolate through. The cell bottoms are perforated to allow the solvent to drain through and be re-circulated. One rotation is about an hour but varies with the material being extracted. At the end of one rotation the plant matter exits the cell as the cell rotates over an open bottom cell. This type of equipment has a low energy requirement. There are several other types of extractors including the loop extractor (Crown Iron Works Company – Minneapolis, MN) and sliding bed extractor (Lurgi AG Frankfort, Germany), see Figures 5 & 6. These extractors work similar to the carrousel extractor except the material path is linear. The mixture of solvent and essential oil is called miscella. The solvent is removed from the miscella in a two-step process: evaporation followed by steam stripping. The vapor from evaporation is condensed and the solvent separated from the water. The solvent is then recycled. Dai et.al. (2010) reported that solvent extraction with hexane was 180 times faster than using steam distillation. Microwave Assisted Solvent extraction Microwave-assisted process (MAP) applies microwave energy to selectively heat components in a solvent solution. Materials have different dielectric Figure 6: Sliding Cell Solvent Extractor Cut-away view Courtesy of Lurgi AG

Figure 4: Carrousel Extractor cut-away view (source: Harburg-Freudenberger)

Figure 5: Sliding Cell Solvent Extractor

constants and in general the higher the absolute value, the higher the level of absorption of microwave energy. The absorption level does vary with temperature of the substrate and the microwave frequency used. Because of the variation in the absorption of microwave energy, it is possible to selectively heat substances. Free water molecules have a high dielectric constant (~80) and hence absorb energy and heat quickly. Some solvents such as hexane have a low dielectric constant (1.9) and essentially allow the microwave energy to pass through without absorbing energy. For the extraction of mint oil, the microwaves will interact with the free water in the plant structure causing localized heating. More heating will occur in areas with higher
8

water contents such as glandular and vascular systems of the plant material. The water will increase in temperature to the boiling point or higher and cause rapid expansion and rupture cells. This releases the oil and other components into the solvent and produces pathways for the solvent into the plant materials. The soluble components are dissolved by the solvent and then recovered by evaporating the solvent. This process requires smaller volumes of solvent, can used less toxic solvents and reduces energy consumption over more commonly used extraction methods. Paré et.al. (1994) reported exposure to 20-30 seconds of microwave (at 625 watt) showed more glandular disruption than six hours of soxhlet extraction or 2 hours of steam distillation. They also found that oil yield on a per weigh basis was greater for the soxhlet extractor (lab scale solvent extraction method) but the oil quality was superior for the MAP extracted oil because in contained little chlorophyll and less pulegone (a monoterpene which is clear colorless oily liquid and has a pleasant odor). In their U.S. Patent, 5,002,784, Paré et.al., indicated that MAP would reduce processing costs by 45% and labor costs by 50% compared to steam distillation which could increase net revenue from 18% to 35%. Extraction lab tests of peppermint produced 25% more oil using MAP than traditional steam distillation. This patented process was developed by J.R. Jocelyn Paré et.al. at the Minister of the Environment in Canada and is covered by U.S. Patents: 5,002,784 (1991), 5,338,557 (1994), 5,377,426 (1995), 5,458,897 (1995), 5,519,947 (1996), 5,675,909 (1997), 5,884,417 (1999) and 6,061,926 (2000). The application of microwave to flammable solvents can be hazardous if not carefully controlled. A study by Dai, et.al. (2010) found the highest yield of menthone, menthofuran, and menthol from peppermint was achieved in lab testing using microwave assisted solvent extraction, with a 30:70 mixture of ethanol and hexane, an extraction time of about 30 minutes and a sample-tosolvent ratio of 2g to 80 mL. Using 100% ethanol or a 70:30 mixture of ethanol and hexane resulted in about a 10% reduction in oil yields. Oil yield for sample-to-solvent ratio down to 2g to 20 mL also gave an acceptable yield. The extraction method had the largest affect on yield. Ultrasonic Assisted Extraction Sonication or ultrasonically assisted extraction involves immersing the plant material in a solvent in a vessel and then subjecting it to ultrasonic sound waves. Figure 7 is an illustration of the process. Solvent and plant material are placed in a temperature controlled vessel and subjected to ultrasonic vibrations to free the oil. After 10 to 15 minutes the solution is filtered to separate the plant solids from the solvent and extracted oil. The liquid fraction is distilled to separate the solvent from the oil and the solid fraction is dried to recover the solvent. The solvent is condensed and reused. If a low boiling point solvent is used and the temperature is kept below its boiling point, ultrasound can increase oil Figure 7: Ultrasonic-Assited Solvent extraction yield. Solvent extraction is often done with cold sol- process illustration (Vinatoru—1997) vent. Using sonication resulted in similar yields Solvent Plant compared with conventional extraction methods but Material in less time. Ultrasonic works by breaking the thin cell walls of the oil glands and rinsing out the oil Solvent Transducers Extraction once the cell walls are broken. If the plant material Unit has been dried, the solvent diffuses into the cells causing swelling and hydration. Sonication increases Solvent the rate of swelling and hydration. Lower frequenDistillation Filtration cies (20 kHz) result in more cell damage and faster Oil release of oil than high frequency (500 kHz) which Drying Solvent Recovery left the leaf undamaged (Vinatoru 2001). Stirring of Dry Exhausted the solvent also aided in decreasing the processing Plant Material
9

time. This can be a low temperature process when used with vacuum distillation to preserve the thermal sensitive components of the oil. Da Porto (2009) reported increases of oxygenated monoterpenes by up to 8 times more using ultrasound-assisted solvent extraction with 70% ethanol as a solvent compared to hydrodistillation. Carvone was the component that increased the most. Ultrasound does not increase oil yield with water distillation, it only produces more rapid boiling. Solvent-Free Microwave Extraction (SFME) Microwave for cooking was discovered by Percy Spencer at Raytheon Corporation while developing radar systems in 1945. In 1947 the first microwave ovens for use in restaurants and commercial kitchens were introduced. The first units weighed 750 pounds and stood five foot, six inches and require water for cooling the magnetron. It took decades before it was refined enough for the average consumer. Today, industrial applications for microwaves included food processing, laboratory analysis, preheating and vulcanization of rubber, drying, baking sand core molds for metal casting, drying resins in paper production, curing fiberboard and chipboard, and many other applications. Microwave essential oil extraction can be done in a batch or continuous flow method. The green plant material is exposed to microwave energy which causes the in-situ water of the plant material to heat and boil causing the plant glandular membranes to rupture and release the oil. The oil and water is condensed to recover the oil and the water can be return to the microwave oven to facilitate additional vaporization of the essential oil or disposed. Testing by Lucchesi et al. (2004) compared solvent-free microwave extraction with hydro-distillation and found 30 minutes of processing time recover the same amount of oil as after 4.5 hours of hydrodistillation (HD). The first droplets of oil were recovered in 5 minutes compared to 90 minutes with distillation. The oil composition for the SFME had higher amounts of oxygenated compounds and lower amounts of monoterpenes than the HD extracted oil. The oxygenated compounds are highly odoriferous and more valuable than the monoterpenes according to Lucchesi. The mint oil contained 65% and 52% carvone, an oxygenated compound, for SFME and HD, respectively, while limonene content, a monoterpene, was 9.7% and 20% for SFME and HD, respectively. The difference in composition is not likely that they weren’t extracted but the reduction in extraction times and reduced water volumes with SFME results is less degradation of the oils by hydrolysis, trans-esterification or oxidation. The energy requirement was also greatly reduced from 4.5 kWh for HD to 0.50 kWh for SFME using a 500g sample (Lucchesi 2004). The oil yields of 0.095% for crispate mint which is substantially lower that is reported in most articles on mint oil yields. The equivalent energy consumption is 14,664 Btu/lb of oil for hydro-distillation and 1,629 Btu/lb for microwave extraction. SFME has the advantage over solvent extraction methods of not having to be concerned with toxic solvent residue remaining in the oil or secondary operations to remove it. Vian (2008) published information on a new method that microwaves the plant matter in an upside down flask, Figure 8. The concept is that the vapors will fall by gravity through a condensing unit into a Florentine flask. The oil yield when compared to hydro-distillation was comparable, 0.6% versus 0.59% but the extraction time was reduced from 90 minutes to 20 minutes. This concept would not be practical for commercial production. Microwave steam distillation (MSD) passes steam through the plant material while at the same time irradiating the plant material. The steam is not heated by the microwaves because of a low dielectric constant (1.0) but the water in the plant material is heated by the microwaves result10

Figure 8: Microwave hydro-diffusion and gravity apparatus (Vian—2008)

ing in rupturing of the cell walls and vaporization of the water and essential oils. The steam passing through the plant material acts as a carrier to move the essential oils to the condenser and Florentine flask. Figure 9 shows an illustration of the process. The microwave power setting is important for fast extraction but excessive power can cause the loss of volatile compounds. Lavender flowers were used in the study. The sample was processed until no more essential oils were obtained. It took less than 10 minutes to completely extract the oil. Comparing MSD to steam distillation, the oil yield was the same but it only took 6 minutes to obtain the same oil yield as 30 minutes of Figure 9: Microwave steam distillation (Sahraoui—2008) steam distillation. Using the same process, Sahraoui (2008) found almost 100% of the extractable oil was recovered in 5 minutes compared to about 30 minutes using steam distillation alone for lavender flowers. Mengal and Mompon (2006) were issued a patent (U.S. Patent 7,001,629 B1 (2006)) for a microwave extraction system that cycles the extraction vessel from atmospheric pressure 100 kPa to -25 kPa (14.5 to -3.6 psi) vacuum three times in a 15 minute cycle while keeping the temperature below 75°C . The lower pressure would cause lower boiling points. Water vapor from the extraction vessel is condensed, the oil is separated and the water is routed back to the chamber. A stirrer in the vessel increases the contact with the irradiation and reduces hot spots. The oil yield reported appears to be nearly equal to steam distillation. This method would likely be batch process.
11

MIRC has supported work on microwave extraction at Oregon State University. Velasco (2007) studied different power settings and combinations of power settings and time exposures. Her study found that the most effective power settings was a combination of 1120 W for 2 minutes and 518 W for 1.25 minutes extracted the greatest oil yield. The energy consumption was calculated at 0.22 kWh per pound of mint hay. The energy cost was $1.22 per pound of oil extracted. The process time was 3.25 minutes compared with 120 minutes for steam distillation. The peppermint oil composition is less than the ideal for Menthol and Menthone, about the same for Cineole, and higher for Limonene, Pulegone and Furon. The mint oil composition was acceptable based on the acceptable industry ranges. Further work was done in 2009 by Hackleman (2010) setting up and running a pilot scale study using an industrial planar microwave system. Two trials were done, one at a grower’s facility and one using frozen mint at the microwave supplier’s facility. They had trouble getting the oil to be transported to the condenser unit and oil passing through the condenser unit. A blower was used to draw the steam and oil vapors from the microwave chamber to the condenser since there wasn’t steam pressure to move the vapor. They reported problems getting the oil to not condense in the pipe leading to the condenser unit. This may explain the reason why Mengal and Mompon (2006) in their patent are routing the condensed water back to the microwave chamber. An increase in water would provide more vapor to help carry the oil to the condenser unit. No oil was collected in the Florentine vessel. The 100 kW microwave unit tested should be capable of processing 8 cubic feet per minute and Hackleman concluded that if it was operated 24 hours per day using an automated feeding system and the harvest season was extended, it would replace 2 or 3 still tubs. Based on those assumptions, a complete system was estimated to cost $400,000 in capital and have maintenance of $2000 per year for cathode tube replacement (approximately every 8 years) plus maintenance for belts, motors, etc. This cost includes a closed loop condensation cooling system to reduce water use and discharge. Based on assumptions, Hackleman estimated the energy cost would be $0.40 per pound of oil or approximately a 93% energy cost savings over petroleum fired steam distillation which cost $5-7 per pound of oil. However, in this trial only 11% of the oil was extracted from that mint hay based on samples taken before and after so much work needs to be done if this is to be a viable extraction method. The lack of success of Hackleman to extract oil can likely be explained by the difference in dielectric constants between the water and the oil components. The dielectric constant of a material is related to the ability of the material to be heated by microwave energy. A high dielectric constant such as water results in high absorption of microwaves and heating . Water has a dielectric constant of between 55 and 80 depending on the temperature although once it turns to steam the dielectric constant drops to 1 (no absorption of heat). The volatile components of mint oil have dielectric constants ranging from 2.3 to 11, see Table 1. Hexane which is used in microwave assisted solvent extraction has a dielectric constant of 1.9 at 20°C and is considered to be transparent to microwave (very little absorption). Low dielectric constants means it will take a long time to heat, therefore without high water content in the mint plant material the volatilization of mint oil will take a long time. A grower at the Midwest Mint Growers conference in Wisconsin in Feb 2011 told me he’d tried using microwave on green-chopped mint (no wilting) and got 100 percent extraction. This would have resulted from the high water content in the mint turning to steam and volatilizing the oil components. If the mint can be successfully green-chopped, the setup that Hackleman used might be successful.
12

Supercritical Fluid Extraction Supercritical fluid extraction (SFE) is a form of solvent extraction using a supercritical fluid as a solvent. The most commonly used fluids are carbon dioxide, ammonia, ethane, ethylene, propane, pentane and water but they are sometimes mixed with ethanol or methanol. A substance
Table 1: Dielectric constants and Boiling oils of mint oil components Compound d-Limonene Cineole Menthone Menthol Pulegone Menthyl acetate Dihydrocarvone Carvone Water Dielectric Constant 2.3 4.57 8.8 4.0 9.5 7.07 8.5 11 80 @ 68F, 55.3 @ 212F; 1.00 steam Boiling Pt. C 175-178 176-177 210 217 224 227-229 221.5 227-231 100

Source: Lange’s Handbook of Chemistry 15th ed, McGraw Hill 1999

becomes a supercritical fluid at a temperature and pressure above its critical point, the point at which it changes from a liquid to a vapor, Figure 10. In the supercritical state a substance has properties of both a gas and liquid state. The densities are close to that of the liquid form while viscosity is near that of the gaseous form. Supercritical fluids can flow through a solid like a gas and dissolves materials like a liquid. The solvent properties can be adjusted by changing the temperature and pressures. Carbon dioxide is most commonly used for food applications such as decaffeination of coffee beans or the extraction of hops for beer production. It is also used for the extraction of essential oils and pharmaceutical products from plants. The critical temperature for CO2 extraction is 31.1°C (88°F) while the critical pressure is 7.3 MPa (1060 psi). For traditional steam distillation, the temperatures will be greater than 100°C (212°F) with pressures a little above atmospheric. The higher temperatures of steam distillation can cause degradation of thermally sensitive oils during extraction. Supercritical fluid extraction using CO2 has many advantages. CO2 is relatively inexpensive, readily available and can be easily recycled Figure 10: Definition of Supercritical State and recovered from the extraction leaving no (Brunner—2005) harmful solvent residues. It is non-flammable, non-explosive, non-toxic, colorless and odorless. SFE reduces processing times, results in a better quality product with longer self life and process conditions are easily achievable. The process involves loading plant material into a pressure vessel and pressurizing it with carbon dioxide until the pressure and temperature is above 7.4 MPa and 90°F, respectively. Above critical pressure and temperatures, the CO2 will become supercritical and act like a solvent, dissolving oils, pigments and resins from the plant material. The CO2 is usually circulated through the vessel to extract the sub13

strates. To remove the oil, the CO2 is de-pressurized and allowed to evaporate leaving the oil and other soluble components behind. Ideally the CO2 can be evaporated leaving oil without any residues but many of the highly volatile components will pass though a dry separator therefore the CO2 is typically passed through a solvent such as ethanol to trap the volatiles. This will require secondary processes but results in higher yields. Supercritical CO2 solvent properties can be modified by changing the temperature and pressure to be selective for the components that are desired to be extracted. It is possible to extract more constituents than steam distillation and may require additional processing if using a single stage system. Extraction process can be batch or continuous flow with extraction times of 15 to 120 minutes depending on conditions. Figure 11 is an illustration of the flow schematic of a single stage processing with a supercritical fluid. Continuous Flow SFE General Foods developed a continuous process for SCF extraction of caffeine from coffee and received a patent for the process in 1989. Figure 12 is a flow schematic for the process. The desired amount of product flows through a valve and enters a small pressure vessel that is used to
Figure 11: Flow Scheme of single stage processing with supercritical fluids (Egger & Jaeger, 2003) Figure 12: Continuous Supercritical extraction process flow (adapted from Katz, 1989)
Product in Valve In Charging vessel Valve Out

Throttling Valve

Evaporator

Feed Extractor

Separator
CO2 In

CO2 Out to separator

Extractor

Heater

Pump

Chiller Oil Product

Valve In Discharge vessel Valve Out Spent product Out

charge the extraction vessel. The charging vessel has a valve on both the inlet and outlet so it can be isolated. During filling the outlet valve is closed and the inlet is open. Once the desired quantity of product is in charge vessel the inlet valve is closed and the pressure is increased using the extraction solvent to the same pressure as the extraction vessel. The charging vessel outlet valve is then opened allowing the product charge to flow into the extraction vessel. At the same time as the charging outlet valve is open, the valve at the bottom of the extraction vessel is opened to the discharge vessel, allowing an equal amount of spent material to exit the extraction vessel. The discharge vessel is also pressurized before the valve is opened to the extraction vessel. The valve at the top and bottom of the extraction vessel are closed. The supercritical fluids in the charging and discharge vessel are vented and the discharge vessel is emptied. The supercritical fluids are typically vented to a holding vessel and recycled. The discharge vessel can be vented to the charging vessel to conserve supercritical fluids. The valves and charging / discharge vessel act as air locks so product can be loaded and unloaded from the extraction vessel without disrupting the process. Rotary locks can be used in place of the two valves and charging / discharge vessels but are more complex, cost more and generally require more maintenance. The residence time for extraction can be adjusted by the charging frequency and quantity
14

and the vessel size. The flow of supercritical fluids is continuous during charging and discharging and flows counter-current. This aids in maximizing extraction because supercritical fluids free of dissolved oils are passing through product that is at the bottom of the vessel that have the lowest amounts of oils left to extract. The supercritical fluids are routed to a separator after exiting the extraction vessel to separate the essential oil from the solvent. The CO2 is repressurized and recycled back to the extractor. Hughes Aircraft Company also has a similar patent issued in May 1994 (US Patent No. 5,313,965). The temperature and pressure of extraction will have an effect on the oil yield from spearmint. Al-Marzouqi (2007) found in lab scale experiments that increasing the pressure from 15 MPa (2175 psi) to 35 MPa (5076 psi) increased the yield by 48.5% at 30°C, 18.6% at 40°C and 17.6% at 50°C. Increasing the temperature at a constant pressure also provided similar results; 44.4% increase in yield raising the temperature from 30°C to 40°C and 13.1% increase in yield raising the temperature from 40°C to 50°C. The project sourced mint from 5 different countries and found a difference of up to 27% between the lowest and highest oil yields based on the source. The quality of the extract was considered better with respect to oil composition extracted with supercritical CO2 extraction at operating conditions of 30°C and 15 MPa. Using the major flavor components as quality markers, carvone and limonene, the oil produced with supercritical CO2 extraction contained 11443 and 2453 µg/g of dry material, respectively, compared to 10780 and 510 µg/g of dry material, respectively, using steam distillation. Table 1 shows the principal components for steam and supercritical CO2 extraction at 30°C and 15 MPa. Özer, et.al. (1996) reported oil extraction yields from 23 to 80% from spearmint using supercritical fluid extraction (SFE) with CO2. The highest yields were at 40°C at 11 MPa (1595 psi) with an extraction time of 4 hours. The lowest yields were at 60°C at 8 MPa (1160 psi) with a 1 hour extraction time. In this study lower yields were extracted as temperatures increase or procTable 2: Mint oil composition (µg/g on dry basis) (from Al-Marzouqi (2007)) a-pinene Steam SCFE* 0.0 48.2 Limonene 510 2453 Cineole 54.1 95.6 Linalool 35.0 33.4 Menthol 28.2 22.5 Dihydrocarvone 21.4 27.8 Carvone 10780 11443

* Supercritical Fluid Extraction

ess time decreased. This report also compared the composition of the extracted oil compared to steam distillation. Limonene percentages were 31% less with SFE (5.26%) than using steam distillation (7.63%) but carvone components were 5.2% higher at 81.15% with SFE compared to steam distillation at 77.13%. This contradicts more recent work by Al-Marzouqi (2007) who reported a 79% increase in limonene extraction but similar increases in carvone at 6.1%. AlMarzouqi study used much higher pressures which might be the reason for higher recovery percentages. Barton (1992) compared peppermint and spearmint oil extraction from green plants and fielddried hay with supercritical fluid extraction with CO2 at various rates and time and steam distillation. Barton’s results indicated that the CO2 extracted oil are “strongly colored, dark yellow to greenish yellow” but differed only slightly in specific gravity and refractive index. The yields of spearmint oil from field dried hay using SFE-CO2 match the yield from steam distillation with process temperatures of about 34°C and pressures of 10 MPa (1450 psi). An extraction time of 4 hours was used for the experiment. The minimum extraction time was not determined.
15

Barton found that the CO2 flow needed to be 7.5 g CO2 per g dry hay using a single pass down-flow extraction. Economics of Supercritical Fluid Extraction Öztekin and Martinov (2007) reported that a three 500 liter (132 gallon) vessels supercritical fluid extraction system using CO2 for processing cloves, cumin, cardamom and ginger was estimated to cost $2.2 million for capital and $4 to $6 million/year for operating costs. A comparison by Pereira and Meireles (2007) for rosemary, fennel and anise estimated the manufacturing cost in Brazil using SFE compared to steam distillation to be 44%, 55% and 58% less, respectively. A student project at Rowan University in New Jersey, looked at the extraction of peanut oil using supercritical CO2 extraction and compared it to a hexane solvent process. They made energy and cost estimates for a 10 million pound per year processing facility with a 30% oil yield (3 million pound per year). Referring to Table 3, extracting peanut oil using supercritical CO2 extraction reduced energy use and costs by 60% and 55%, respectively. Brunner (2005) reported the cost for a 1000 ton per year batch system is in the range of $1.36 per pound of feed stock but economies of scale could reduce cost to about $0.25 per pound. Continuous flow systems could reduce costs further although continuous SFE extraction has only been carried out at the lab scale so far. Lack and Seiditz (2001) reported cost estimates for
Table 3: Costs for peanut oil processing with supercritical CO2 extraction (Gifford, 2001) Solvent type Solvent cost Solvent use Energy input Operating cost Energy per lb of oil Cost per lb of oil CO2 $0.07 / lb. 87 million lbs per year 1.8 GWh/yr $ 6,200,000 0.6 kWh/lb $ 2.07 / lb Hexane $0.07 / lb. 38 million lbs per year 4.6 GWh/yr 14,000,000 1.53 kWh/lb $ 4.67 / lb

the decaffeination of coffee beans from $0.50 to $0.65 per pound for a 3500 ton per year capacity system to $0.33 to $0.45 per pound for a 7000 ton per year capacity system but warns that cost estimates can vary by ±30%. The cost breakdown excluding raw material costs are as follows: Interest and depreciation = 36.1%; labor = 24.5%; utilities = 17.2%; taxes = 20.5%; administration = 1%. Capital costs are about one third of the total processing cost. Pressurized Fluid Extraction Many terms are used to describe this technique: pressurized fluid extraction (PFE), accelerated solvent extraction (ASE – trademarked by Dionex), pressurized liquid extraction (PLE), pressurized solvent extraction (PSE) or enhanced solvent extraction (ESE) (Camel 2001). This method is similar to supercritical fluid extraction in that it maintains a solvent in a liquid state at an elevated temperature with pressure. Using this method, temperatures between 200 – 300°C may be used with common organic solvents. The decrease in viscosity at elevated temperatures aids in the solvent diffusing into materials and increases the solubility of compounds into the solvent. The high pressure also aids in the solvent penetration in to compounds. Extraction time is 5 – 10 minutes and has been used as a lab bench test for recovery of environmental contaminates. Higher levels of moisture in soil samples inhibited the extraction of pesticides which would indicate the need for dry hay if used for mint oil extraction but high temperature may degrade oil.

16

Subcritical Water Extraction Subcritical Water Extraction uses superheated water at elevated temperatures (100°C to 374°C) and pressures high enough to maintain liquid state. Under these conditions water is a low polarity extraction solvent. Advantages reported by research: - Higher extraction of polar compounds - avoids extraction of waxes and lipids - faster, cheaper, cleaner (solvent free) - environmentally friendly - nontoxic - lower pressures than supercritical fluid extraction - cheaper equipment than supercritical fluid extraction - more efficient extraction – qualitative compounds contain higher oxygenated compounds and fewer terpene fractions - better representation of natural aroma - eliminates drying stage - selective extraction based on temperature selection Subcritical water extraction uses hot water under pressure to maintain the water in a liquid state to act as a solvent. It may not be suitable for thermally sensitive oils. Kubátová et.al. (2001) reported nearly complete decomposition of linalool and γ-terpinene at temperature of 175°C during oil extraction from peppermint. Water is corrosive under supercritical conditions and can damage equipment but can be prevented by using ultra pure and degasified water. This requires additional equipment and cost. Kubátová did extraction rates at different temperatures and showed 90% or higher extraction rates for all major components except menthol acetate at 125°C. At 150°C all components were nearly at 100% extraction in less than 25 minutes with the exception of menthol acetate which was only 30% extraction. Higher temperatures and longer times would be needed to obtain higher yields of menthol acetate. For peppermint, 30 minutes at 150°C or 12 minutes at 175°C resulted in similar extraction quantities of carvone, pulegone, piperitone, eucalyptol, menthone, neomenthol and menthol compared to 1 hour of supercritical fluid extraction and 4 hours of hydro-distillation. Instant controlled pressure drop technology This method is based on a patent by a French group. A U.S. Patent of the process, No. 5,855,941, was granted in 1999. The process involves heating dry plant material (11% moisture) with steam for a short period, followed by an abrupt pressure drop to a slight vacuum (5 kPa). This sudden pressure drop causes vaporization of fluids and the breaking of cell walls which aids in releasing essential oils. This process is repeated 2 to 6 times at saturated steam pressures of up to 0.6 MPa (85 psi) with heating time of 0.5 to 20 minutes. Higher pressures and the number of cycles resulted in increased oil recovered from Cananga flowers. A 4 minute cycle resulted in 2.74% oil yield versus 2.60% for a 24 hour steam distillation process, a 5% yield increase. This work has also been done for other crops and typically results in higher oil yields with shorter processing times. The processing of Myrtle leaves resulted in 10% higher oil yields with a 2 minute process time compared to 180 minutes with hydrodistillation (BerkaZougali 2010), processing Lavandin resulted in an 84% increased oil yield in 8 minutes (Besombes 2010) and Rosemary oil was processed in 10 minutes, recovering 91 to 97% of the oil compared to more than an hour with hydro-distillation (Rezzoug 1998). This process has
17

the advantage of fast processing compared to hydro-distillation, selective extraction compared to supercritical extraction, no solvent residue, low energy and water consumption (662 kWh and 42 kg per ton of raw product, respectively) and produces high quality oil. The disadvantage is that it has not yet been commercialized and is a batch process. Using multiple vessels could simulate a continuous process. Another concern is fatigue stress on vessels from the constant cycling. No published data was found on the extraction of mint oil. Operation – The plant matter is placed in vessel 1 and heated with direct saturated steam (F1) and steam through a heating jacket (F2) on pressure vessel to increase the temperature and pressure. This step can last from several seconds to minutes but preferable not more than one minute. Valve V2 is closed during loading and heating. Vessel 2 is evacuated to a vacuum of 0.05 MPa (-7.5 psi) with vacuum pump (3). Valve 7 is closed, valve V2 is opened allowing the pressure to drop by 0.6 to 0.9 MPa (90 to 135 psi) in approximately 0.5 seconds. This causes the water and oil in the plant matter to vaporize. Valve V2 is closed and the heating and pressurization step is repeated followed by valve V2 opening to depressurize the system. This is repeated the number of cycles specified for the product, typically two to six cycles. Cooling water flows (F3) into a water jacket on vessel 2 to condense the water and oil vapor, Valve V4 is opened to allow the water and oil to flow into a Florentine type vessel to separate the water and oil. The water is removed through V6 and the oil recovered through valve V5.

Figure 12: Schematic diagram of the instant controlled pressure drop DIC apparatus: (1) autoclave with heating jacket; (V2) rapid valve; (2) vacuum tank with cooling water jacket; (3) vacuum pump; (4) extract container; F1 & F2 steam flow; F3 cooling water flow. (From Besombes – 2010)

Moderate Electric Field Extraction (MEF) MEF processing applies a voltage across a food material to rupture cell membranes and increase permeability. Sensoy and Sastry (2004) experimented with black tea and dry and fresh mint. Samples were placed in breaker with a salt solution and electrodes on opposite sides. The salt solution was used to improve the electrical conductivity and was varied depending on the voltage used. Voltages ranged from 200 to 1000 volts. Three different frequencies were used: 50, 500, 5000 Hz. The method was compared with hot water extraction and found that there wasn’t a
18

difference when using dried tissue but the MEF process on fresh mint resulted in an 80% increase in total solids extracted. The extracted solids would include oils along with other substances. There was no significant difference in solids extracted as voltage was increased and a decrease in solid extract at frequencies of 500 and 5000 Hz compared to 50 Hz. Heating occurs while the samples are subjected to MEF. The process temperatures started at room temperature (25°C) and the test was concluded when the solution temperature reached 80°C. The process could be done in a continuous process by pumping a slurry of the plant tissue and brine solution through a pipe past a power source. The technique needs further study to determine its efficiency at extracting mint oil compared with current methods. High-Voltage Electrical Discharge (HVED) High-Voltage Electrical Discharge (HVED) as described by Grémy-Gros, et.al. (2008) passes an electrical charge through plant materials that is submerged in an aqueous solution. The electrical charge travels through the plant material causing cells to rupture, pressure pulses and produce oxidative chemistry. The electrical pulse is applied for only a few microseconds which avoids overheating. HVED has been used in experiments to enhance extraction of mucilage extraction from whole linseed, solutes from tea leaves and moonflower roots and oil from linseed cake. Extraction times for fennel gratings with HVED were about 20 minutes compared with 40 minutes with MEF and 200 minutes for ultrasonic assisted extraction. All methods resulted in a extraction rate of about 97% compared to extraction without treatment resulted in 60% extraction in 20 minutes. Similar work was done by Dobreva (2010) on rose pedals. Dobreva called the method Pulsed Electric Fields (PEF) which subjected rose pedals to electric fields of 1-5 kV/cm at a specific energy of 5-20 kJ/kg. The results increased essential oil yield by 13-33% over distillation alone and reduced distillation times from 2.5 to 1.5 hours. No undesirable changes in the properties of the oil were observed. Oil Composition One of the criteria for using a different extraction method is that the oil quantity as well as quality of oil must meet the requirements of producers and consumers. The tables 4 and 5 compare the MIRC criteria for the mint oil components reported in different studies sited in this paper. The orange highlighted cells in the tables indicate study values that are below or above the MIRC criteria (first row of the table). If a study published the oil composition for a control distillation method (steam or hydro-distillation), it is listed above the alternative extraction method. In Table 4 – Peppermint oil composition, all of the values for the three studies sited are within the MIRC criteria except for Menthol which is 1-2% points less than the minimum for all studies except for the microwave extraction study by Hackleman (2009). Hackleman’s Microwave study was also below the MIRC criteria levels for Limonene and Cineole but this was likely because the study resulted in incomplete oil extraction (11%). Menthyl acetate was low by less than 1% points for the study by Barton for both the control using steam and the supercritical fluid extraction method which may reflect the plant material used. In Table 5 –Spearmint oil composition, the ultrasonic study by Da Port (2009) reported significantly low values for Limonene but higher values for Cineole than the control or the MIRC criteria. The hydro-distillation control in Da Port’s study was 8% points lower than the MIRC criteria for Carvone but the ultrasonic extraction method was 21% points higher and within the MIRC criteria. The Carvone content for Lucchesi (2004) control (hydro-distillation) was 6%
19

Table 4: Peppermint oil composition of select studies Peppermint Limonene Mentha piperita MIRC Critera 1-3% SFME Velasco-2007 Steam Barton - 1992 SFE-CO2 Barton - 1992 Steam Hackleman 2009 SFME Hackleman 2009 1.1 Cineole Menthone Menthofuran Menthol Pulegone Menthyl acetate 3-8%

3-6% 3

15 - 35 % 15

0.10 - 8 % 5.2

39 - 60 % 38

< 2.5 % 1.8

1.78

5.41

22.49

4.01

38.24

1.24

2.19

1.63

4.36

22.58

4.79

37.27

1.46

2.28

1.16

4.92

25.83

1.3

37.3

1.4

3.77

.1

.72

15.9

1.96

54.9

1.5

3.35

Table 5: Spearmint oil composition of select studies Spearmint Mentha spicata MIRC Critera HD Da Port 2009 Ultrasonic Da Port 2009 HD * Lucchesi 2004 SFME * Lucchesi 2004 Steam – Farwest Barton - 1992 SFE-CO2 Farwest Barton - 1992 Steam - Midwest Barton 1992 SFE-CO2 Midwest Barton - 1992 Limonene 7 - 22 % 14.7 Cineole 0.5 – 3 % 1.6 Menthone <1 – 2.0 % Dihydrocarvone <3% Carvone 58 - 74 % 50.5

0.8

9.0

71.4

20.2

52.3

9.7

1.5

64.9

17.69

1.38

1.07

.83

67.22

15.82

.92

.93

.83

69.01

30.01

1.52

1.44

.77

54.51

14.65

.86

1.22

.81

71.77

MIRC—Mint Industry Research Council SFME—Solvent-Free Microwave Extraction SFE-CO2—Supercritical Fluid Extraction using CO2 HD—Hydro-distillation Ultrasonic—Ultrasonic-Assisted solvent extraction 20

points lower than the MIRC criteria but the results for the solvent-free microwave extraction were within the MIRC criteria. Mint Oil Distillation Energy Survey Summary MRIC collected energy data from a small sample of growers on two occasions and supplied the data pertaining to the mint oil distillation. The first set contained 6 surveys of which 4 were usable and the second set contained 17 surveys of which 15 were usable. Based on surveys provided, the oil yield in pounds per acre ranged from 45 to 180 with an average of 82.4 pounds per acre while the energy use for steam distillation ranged from 41,325 Btu per pound of oil to 265,574 Btu per pound of oil, see Appendix A. Three farms reported value for spearmint and the rest distilled peppermint. The high energy data point was an outlier in the data set being almost twice the next highest value and was omitted from the data summary. This grower was burning used motor oil, the only farm that reported the use of used motor oil. The grower distilled spearmint and reported lower than average yields. Excluding the one farm from the survey, the range for distillation energy use would be 41,325 Btu per pound of oil to 145,635 Btu per pound of oil. The low distillation energy use system was from a farm with average yield per acre using natural gas. Three farms that reported multiple fuel sources. Two of the growers switch between diesel fuel and natural gas during the season. The other farm has two distillation systems. The high value of 145,635 Btu per pound of oil was from a farm in the first survey using diesel fuel. The grower reported distillation energy use for peppermint and spearmint at 143,556 and 145,635 Btu per pound of oil, respectively. The average distillation energy use was 79,509 Btu per pound of oil. One might expect that as oil yield per acre increased the distillation energy use might decline but based on Figure 13 there doesn’t appear to be any relationship (trend line is nearly flat). There was a difference in efficiency depending on the fuel type used. The average of the two growers who use Propane was 68,079 Btu per pound of oil, the lowest of the four fuels. Eight systems used natural gas fired distillation systems with and average efficiency of 73,295 Btu per pound of oil, 7.6% higher and eleven systems used diesel fuel with an
Figure 13 - Oil Yield per Acre versus Distillation Energy Use
200 180 160 Oil Yield per Acre (lbs) 140 120 100 80 60 40 20 0 20,000

40,000

60,000

80,000

100,000

120,000

140,000

Distillation Energy Use (Btu/ ac)

21

average efficiency of 98,809 Btu per pound of oil or 45% more energy. The decrease in efficiency might be due in some part to boiler maintenance and the tendency for fuel oil to foul heat exchanger surfaces. Natural gas and propane generally burn very clean. If a distillation system is operating at 100 to 120 psi steam pressure and the boiler efficiency is 80% (typical value for well maintained boiler) then it will require approximately 1500 Btu to produce a pound of wet steam. If the heat exchanger is fouled with soot or the boiler is poorly adjusted, the efficiency may drop to 60% and require 2000 Btu to produce the same pound of steam. The difference in energy use could also be the result of running the distillation system longer than necessary to harvest the mint oil from a loaded tub. How the tub is loaded and possibly the length of cut could possible effect the distillation time as well. When comparing different fuel types of energy sources it is necessary to convert all to the same unit of measure. For energy in the English measurement system, British Thermal Units (Btu) are use. The definition for a Btu is the amount of energy required to increase the temperature of a pound of water by one degree Fahrenheit. Different fuel types contain different quantities of energy. A gallon of propane or liquid petroleum gas contains about 91,500 Btu per gallon while diesel fuel contains 138,700 Btu per gallon. Therefore it takes 1.5 gallons of propane to equal one gallon of diesel fuel on an energy basis. Natural gas is sold in units of Therms or cubic feet. A Therm is defined as 100,000 Btu and a cubic foot of natural gas generally contains about 1030 Btu but can vary from about 950 to 1100 Btu per cubic foot. Thus it takes 1.39 Therms of natural to provide the same energy as 1 gallon of diesel fuel. For a future survey it would be advantageous to find out the number of cutting per year and the oil yield and energy use per cutting. From the wide range of efficiency reported, there is likely substantial energy savings from doing routine boiler maintenance / tuning and using best management practices for majority of growers.

22

Summary: Energy data for the different method of oil extraction are sketchy in the literature so side-byside energy comparison could not be made for most methods. Most of the methods discuss would likely reduce the energy input but may not necessary be economical because of higher capital costs, larger use of electricity and limited number of hours of operation per year. Some of the extraction methods have been develop mainly for lab analysis and are not practical for large production volumes. Some of the methods would not be conducive for on-farm processing because of the specialized equipment, flammable solvents, or high pressure vessels. A cooperative could be formed to share the capital costs and employ the technical talent to operate the equipment but capacity would be an issue since all of the mint would be ready to harvest at approximately the same time. The two methods that appear to have the best chance of success at the farm level would be a continuous steam distillation system or the solvent-free microwave extraction. Both are continuous processing methods that would lend themselves to better process control, heat recovery, un-attended/automatic operation and lower capital costs. A continuous steam distillation system would need to be developed based on the existing patents and additional test runs would be needed to develop and refine the process for a microwave system. Based on processing parameters published by Bouchard and Serth (1991) for cedarwood and assuming mint could be processed in this system with the same parameters, the energy to extract a pound of mint oil could be reduced by 60 to 70% with a 25 second process time. The drawback is the capacity would need to be increased to meet the needs of an average farm. Based on a 10 day, 10 hours per day harvest window and 4 tons per acre, the processing capacity would need to be 20 tons per hour versus the current capacity of 13 tons per day. The microwave system would have a longer process time than the continuous steam system describe by Bouchard and Serth (1991) ranging in the 1 to 3 minute processing time based on the demonstration work by Hackleman (2009). The energy cost between batch steam distillation and microwave may not be any less expensive base on Velasco (2007) who reported steam distillation energy costs of $1.26/lb of oil compare to $1.22/lb of oil for microwave at optimal conditions based on bench scale testing. If settings weren’t optimal, energy costs for microwave could be much higher. The oil composition does not seem to be adversely affected by the method of extraction. More work will need to be done to work out the process parameters for microwave. Table 6 lists the different extraction methods review and lists the process times and relative capital cost. The methods with the shortest process times are the continuous steam, instantaneous pressure drop technology and solvent-free microwave. There is little information on the capital costs for the different methods in the literature. Capital costs for commercial scale systems are few. Hackleman (2009) estimated the cost of a microwave system equal to a two or three station conventional steam system at $400,000 while a supercritical CO2 extraction system is estimated at $2 million for a system with two 132 gallon extractors, likely not large enough to process an acre per day.

23

Table 6: Process type, extraction method, process times Process type Steam / Hydro distillation Continuous steam Solvent Extraction Microwave Assisted Solvent Extraction Ultrasonic Assisted Solvent Extraction Solvent-Free Microwave Microwave Steam Distillation Supercritical Fluid Extraction Subcritical Water Extraction Pressurized Fluid Extraction Instant controlled pressure drop technology Moderate Electric Field Extraction (MEF) High-Voltage Electrical Discharge Method type Batch Continuous Batch or continuous Batch or continuous Batch (possible continuous) Continuous Batch Batch or Continuous Batch Batch Batch Batch Batch or continuous Batch or continuous 20 minutes 10-15 minutes 1 to 3 minutes 6 minutes 60 minutes + 12 – 30 minutes 5-10 minutes 0.5 to 20 minutes 2 to 10 minutes 20 minutes Process time 2+ hours 25 sec – 8 min.

24

References: Al-Marzouqi, A.H., Madduri V. Rao, Baboucarr Jobe, (2007) Comparative Evaluation of SFE and Steam Distillation Methods on the Yield and Composition of Essential Oil Extracted from Spearmint (Mentha Spicata), Journal of Liquid Chromatography & Related Technologies, Vol. 30, pg 463-475. Anonymous, Solvent Extraction from Oilseeds, Bulletin 263e/03.06/10, Lurgi AG Frankfurt, Germany. Accessed on Dec 29 at http://www.lurgi.com/website/fileadmin/pdfs/brochures/ Br_solventExtr.pdf Anonymous, Wikipedia, http://en.wikipedia.org/, used for definitions of various terms. Arnaudo, J. F., (1991). Le Gout du Naturel. Booklet, Biolandes Aromes Laboratories, Mougins Cedex, France Berka-Zougali, B., A. Hassani, C. Besombes, K. Allaf (2010), Extraction of essential oils from Algerian myrtle leaves using instant controlled pressure drop technology, Journal of Chromatography A, Vol. 1217, pg 6134-42. Besombes, C., B. Berka-Zougali, K. Allaf (2010), Instant controlled pressure drop extraction of lavandin essential oils: Fundamentals and experimental studies, Journal of Chromatography A, Vol. 1217, pg 6807-6815. Brunner, G. (2005), Supercritical Fluids: technology and applications to food processing, Journal of Food Engineering, Vol. 67, pg 21-33. Camel, V. (2001), Recent extraction techniques for solid matrices – supercritical fluid extraction, pressure fluid extraction and microwave-assisted extraction: their potential and pitfalls, Analyst – The Royal Society of Chemistry, Vol. 126, pg. 1182-1193. Carle, R., G. Feilder (1990), Über ein kontinuierliches Verfahren zur gewinnung ätherischer Öle (Study of a continuous process for the manufacture of essential oils), Pharmazeutische Industrie, Vol. 52, No 9, pg 1142-1146. Dai, J., V. Orsat, G.S.V. Raghavan, V Yaylayan (2010), Investigation of various factors for the extraction of peppermint (Mentha piperita L.) leaves, Journal of Food Engineering, Vol. 96, pg 540-543. Da Porto, C., D. Decorti (2009), Ultrasound-assisted extraction coupled with under vacuum distillation of flavour compounds from spearmint (carvone-rich) plants: Comparison with conventional hydrodistillation, Ultrasonics Sonochemistry, Vol. 16, pg 795-799. De Silva, K.T. (1995), A manual on the Essential Oil Industry, U.N. Industrial Developemtn Organization, Vienna, Austria. Dobreva, A., F. Tintchev, V. Heinz, H. Schulz, S. Toepfl (2010), Effects of pulsed electric fields (PEF) on oil yield and quality during distillation of white oil-bearing rose (Rosa alba L.), Z Arznei– Gewurzpfla, Vol. 15, No. 3, pg 127-132. Gifford, M., Elizabeth Biancani, William Kearsley, Walter Maluchnik, Stephanie Farrell, Mariano J. Savelski, and Robert P. Hesketh (2001) Economic Feasibility Study on the Supercritical Fluid Extraction of Edible Oils, Green Engineering Poster Competition, AICHE, Reno, NV (PowerPoint presentation slides). Grémy-Gros, C., J.L. Lanoisellé, E. Vorobiev (2008), Application of high-Voltage Electrical Discharge for Aqueous Extraction from Oilseeds and Other Plants, Chapter in Electrotechnologies fro Extraction from Food Plants and Biomaterials, Springer Science+Business Media LLC, New York, pg 217-235.
25

Gunstone, F., F. Padley (1997), Lipid Technologies and Applications, pg. 127-133, Marcel Dekker, New York. Hackleman, D. (2009), Microwave Field Test Project-Field Trial Report, Oregon State University, Funded by MIRC. Hackleman, D., K. Levien, S. Kimura, C. Velasco (2006), Research Progress Report—Energy Efficiency of Mint Extraction, Oregon State University, Funded by MIRC. Katz, S.N., (1989), Method for decaffeinating Coffee with supercritical fluid, U.S. Patent 4,820,537. Kristiawan, M., V. Sobolik, K. Allaf (2008), Isolation of Indonesian Cananga Oil by Instantaneous Controlled Pressure Drop, Journal of Essential Oil Research, Vol 20, No. 2, pg 135-146. Kubátová, A., Arnaud J. M. Lagadec, David J. Miller, Steven B. Hawthorne (2001), Selective extraction of oxygenates from savory and peppermint using subcritical water, Flavour and Fragrance Journal, Vol. 16, pg 64-73. Lack, E., H. Seidlitz (2001), Economics of high pressure processes, In A.Bertucco & G. Vetter (Eds.), High Pressure process technology, pg 437-452. Amsterdam, The Netherlands: Elsevier. Lucchesi, M.E., F. Chemat, J. Smadja (2004), Solvent-free microwave extraction of essential oil from aromatic herbs: comparison with conventional hydro-distillation, Journal of Chromatography, Vol. 1043, pg 323-327. Mengal, P., B. Mompon (2006), Method and plant for solvent-free microwave extraction of natural products, U.S. Patent 7,001,629 B1; Dated Feb. 21, 2006. Özer, E.Ö., Sevan Platïn, Uğar Akman, Öner Hortaçsu, (1996) Supercritical Carbon Dioxide Extraction of Spearmint Oil from Mint-Plant Leaves, The Canadian Journal of Chemical Engineering, Vol. 74, pg 920-928. Öztekin, S., Milan Martinov, (2007) Medicinal and Aromatic Crops: Harvesting, Drying and Processing, Haworth Food & Agricultural Products Press, New York. Palen, E.J., (1994), Continuous Operation Supercritical Fluid Treatment Process and System, U.S. Patent 5,313,965. Paré, J.R.J., J.M.R. Bélanger, S.S. Stafford (1994), Microwave-Assisted Process (MAP): a new tool for the analytical Laboratory, Trends in Analytical Chemistry, Vol. 13, pg 176-184. Paré, J.R.J., M. Sigouin, J. Lapointe, (1991), Microwave-Assisted Natural Products Extraction, U.S. Patent 5,002,784. Pereira, C.G., M. Angela A. Meiriles, (2007) Economic analysis of rosemary, fennel and anise essential oils obtained by supercritical fluid extraction, Flavour and Fragrance Journal, Vol. 22, pg 407- 413. Rezzoug, S.A., M.W. Baghdadi, N. Louka, C. Boutekedjiret, K. Allaf (1998), Study of a new extraction process: controlled instantaneous decompression. Application to the extraction of essential oil from rosemary leaves, Flavour and Fragrance Journal, Vol. 13, 251-258. Sahraoui, N., M.A. Vain, I. Bornard, C. Boutekedjiret, F. Chemat (2008), Improved microwave steam distillation apparatus for isolation of essential oils: Comparison with conventional steam distillation, Journal of Chromatography A, Vol.1210, pg 229-233.

26

Tzia, C., G. Liadakis (2003), Extraction Optimization in Food Engineering, Marcel Dekker, New York. Velasco, C.A. (2007), Microwave Extraction of Peppermint Oil and Comparison to the Current Practice of Steam Extraction, M.S. Thesis, Oregon State University. Vian, M.A., X. Fernandez, F. Visinoni, F. Chemat (2008), Microwave hydrodiffusion and gravity, a new technique for extraction of essential oils, Journal of Chromatography A, Vol. 1190, pg 14-17. Vinatoru, M. (2001), An Overview of the ultrasonically assisted extraction of bioactive principles from herbs, Ultrasonics Sonochemistry, Vol. 8, pg 303-313. Vinatoru, M., M. Toma, O. Radu, P.I. Filip, D. Lazurca, T.J. Mason (1997), The use of ultrasound for the extraction of bioactive principles from plant materials, Ultrasonics Sonochemistry, Vol. 4, pg 135-139.

27

Appendix A—Mint distillation energy estimates September 2010 PDF document Fuel Surveys # of mint oil use 1 27536 28500 1 40000 42000 2 60392 3 4

8/12/2011 S.Sanford Mint type P S P

Type fuel D D NG NG NG

Fuel/# 1.0350 1.0500 0.5670 0.4738 0.6117

Btu/# 143,556 145,635 56,700 47,380 61,171

Comments

^ Cu.Ft to Therm correction

32630 19960 2 surveys could not be used

P

May 2011 PDF document Survey # #/acre oil Fuel/acre 1* 80 20 89 2 56.3 35 3* 80 20 110 4 85 5** 6 7 8 9 10 11 12 13 15 16 17 45 45 67.9 90 75 100 85 86 83 51.3 73 100 180 43.26 25 50 54 57.5 95 63.894 40 34.3 33.166 40 68

NG D D NG D Custom Distilled off-site NG D p D D D p D NG

P P P P P P

Fuel/# 0.8989 0.2247 0.6217 0.7273 0.1818

Btu/# 89,888 31,169 121,056 86,226 72,727 25,218 97,945 ^ Cu.Ft to Therm correction

P P P P P P P P P

0.9613 0.5556 0.7364 0.6000 0.7667 0.9500 0.7517 0.4651 0.4133

96,133 77,056 67,378 83,220 106,337 131,765 68,780 64,512 41,325 64,651 76,000 52,398 79,509

NG P 0.6465 D P 0.5479 Can't use gave $ not amount D S 0.3778

^ Cu.Ft to Therm correction 104 #/acre Average

Outlier 14 61 120 UMO * Growers switched fuels during season. ^ Give as cu-ft/ac used 1.03 correction to estimate Therms Mint Types P - Peppermint S - Spearmint

S 1.9672 265,574 ** Grower has two distillation systems

Assumed Energy Values Fuel Abbreviation Type D Diesel NG Natural Gas p Propane UMO Used Motor Oil 28

Btu/unit 138700 100000 91500 135000

Fuel units gallons Therms gallons gallons

29