Oil Recovery and Lecithin Production
Content
Journal of the Science of Food and Agriculture
J Sci Food Agric 88:2460–2466 (2008)
Oil recovery and lecithin production
using water degumming sludge of crude
soybean oils†
Liliana N Ceci,∗ Diana T Constenla and Guillermo H Crapiste
Planta Piloto de Ingenier´ıa Qu´ımica (PLAPIQUI), Universidad Nacional del Sur (UNS)-Consejo Nacional de Investigaciones Cient´ıficas y
´
Tecnicas
(CONICET), Camino Carrindanga Km 7, CC 717, 8000, Bah´ıa Blanca, Argentina
Abstract
BACKGROUND: Wet gums produced during aqueous degumming of crude soybean oils are currently processed
to produce lecithin or added to meals to increase their nutritive value for animal feed. Oils occluded in these
gums are generally not recovered or processed. In this work, three methods to recover occluded oil and obtain
lecithin from wet gums were assayed: direct extraction of oil with cold acetone (Method I), extraction after water
elimination under vacuum (Method II) and by solvent partition with hexane/ethanol (Method III).
RESULTS: Higher oil yields (up to 588 g kg−1 of occluded oil) were obtained when water was eliminated before
extraction (Methods II and III). No significant differences were observed in lecithin yields between three methods
(720–807 g kg−1 of dried gums). Recovered oils had acidity = 16.7–21.7 g kg−1 as oleic acid, TOTOX (total oxidation)
values ≤ 8.82, unsaponifiable matter = 9.0–12.1 g kg−1 , and Phosphorus = 87–330 mg kg−1 . Lecithins obtained by
Methods I, II and III hexane phase had the same purity level (610–691 g of total measured phospholipids kg−1 ).
CONCLUSIONS: The occluded oil in soybean wet gums can be recovered, with quality and stability indexes
compatible with their reinsertion in the productive process, by water elimination and extraction with acetone.
Lecithins can be obtained with different phospholipid composition and diverse application fields.
2008 Society of Chemical Industry
Keywords: soybean lecithin; soybean oil, gums; phospholipids
INTRODUCTION
World soybean production has been estimated for
2006/2007 as 218 million tonnes (Mt) and therefore
about 35 Mt of soybean oil will be produced. More
than 1 Mt of wet gums, which contain about 260 000 t
of lecithin, will be obtained in the processing of
soybean oil during water degumming step. These
gums or sludge are a complex mixture comprising
phospholipids or lecithin, oil, and minor amounts of
other constituents like phytoglycolipids, phytosterols,
tocopherols, and fatty acids. They have a high water
content that rapidly promotes damage if they are not
properly stored and processed. The composition and
molecular structure of this heterogeneous mixture
of compounds vary depending on the degumming
conditions of the oil.1
Some methods to produce, purify, and fractionate
lecithin from gums have been studied. In the
conventional industrial process wet gums are dried
under vacuum and then de-oiled with cold acetone, a
solvent in which phospholipids, glycolipids and related
compounds are almost insoluble.2 Partition methods
with solvents to enable water elimination1 and
∗
phospholipid fractionation3 have also been proposed.
Fractions with high phosphatidyl choline/phosphatidyl
ethanolamine (PC/PE) ratio and fractions with high
phosphatidyl inositol (PI) contents can be obtained
by fractionation with ethanol.4,5 The PC-enriched
fractions have unique anti-spattering properties as
oil/water (o/w) emulsifiers in margarines with low
or no salt content and they are suitable for systems
with hard water, salts and in the presence of
milk proteins, because PC is not flocculated by
calcium and magnesium ions. PI-enriched fractions
are used as water/oil (w/o) emulsifiers in the
confectionery industry. Pure PC can be obtained by
chromatographic procedures.3,6 Alternative methods
for extraction and partition of phospholipids using
supercritical fluids have been developed, although
they are not applied on an industrial scale.3 The
advantages of these processes are the absence of
oxygen, which would promote oxidation and solvent
residues with flammability risks and environmental
problems.
Lecithins can be modified by chemical reactions
such as acylation and hydroxylation to increase their
Correspondence to: Liliana N Ceci, Planta Piloto de Ingenier´ıa Qu´ımica (PLAPIQUI), Universidad Nacional del Sur (UNS)-Consejo Nacional de Investigaciones
´
Cient´ıficas y Tecnicas
(CONICET), Camino Carrindanga Km 7, CC 717, 8000, Bah´ıa Blanca, Argentina
E-mail: [email protected]
†
4th Euro Fed Lipid Congress, Madrid (Spain), 2006
(Received 11 March 2008; revised version received 3 July 2008; accepted 8 July 2008)
Published online 15 September 2008; DOI: 10.1002/jsfa.3363
2008 Society of Chemical Industry. J Sci Food Agric 0022–5142/2008/$30.00
Oil and lecithin from water degumming sludge
hydrophilic properties and improve their behaviour
characteristics as o/w emulsifiers.7,8 Special lecithins
can be produced by means of enzymatic modifications.
A2 -Phospholipase can be used to obtain lysophospholipids with o/w emulsification properties tremendously
enhanced.8
Soybeans represent 90% of total oleaginous crops
harvested in Argentina, which is the world’s third
highest producer of soybean behind the USA and
Brazil. However, only some hundreds of tonnes of
crude and purified lecithin are monthly produced.
No processed gums are fundamentally added to
pellets and meals to increase their nutritive value.
Argentinian soya is characterized by low protein
content, which is why gum addition to meals and
pellets reduces their protein availability and increases
their lipid content, making their commercialization
difficult. Gum processing should therefore provide a
solution to these drawbacks. In 2006, the installed
capacity to crush soybean in Argentina was 132 000
t per day and soon it will reach 160 000 t per day.
A major volume of wet gums will be produced as a
consequence of this increase.
On the other hand, in spite of extended information
about phospholipids obtained from gums, there are
no studies available on the recovery of the occluded
oil and the evaluation of its quality indexes. It can be
estimated that more than 200 000 t of occluded oil
will be lost with wet gums in 2006/2007. Moreover,
increases of $803.1 t−1 in the price of soybean oil
with future contracts on Chicago Board on Trade are
registered for December 2007.
The aim of the present study was the recovery
of occluded oil and lecithin production from wet
soybean gums. Direct extraction methods of oil with
cold acetone, and water elimination under vacuum
and by partition with hexane/ethanol, before acetone
extraction, were assayed and the oil and lecithin yields
were estimated. Quality indexes were evaluated for
the recovered oil and phospholipid composition was
determined for oils and lecithins.
EXPERIMENTAL
Materials
Five samples of wet gums from water degumming of
crude soybean oil were provided by two commercial
plants in the April–September period in 2004. The
samples were fractionated and stored at −20 ◦ C to
prevent damage before processing.
L-α-Phosphatidyl ethanolamine (PE), L-α-phosphatidyl inositol (PI), and L-α-phosphatidyl choline
(PC) from soybean, and L-α-phosphatidic acid (PA)
sodium salt from egg yolk lecithin with purity greater
than 98% were used as external standards for
phospholipid analysis. 1-Monopalmitoyl-rac-glycerol
(monopalmitin, purity 99%) was used as internal
standard for polar compound determination. Standards were provided by Sigma (St Louis, MO, USA).
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
Bakerbond SPE Diol (0.5 g) and Bakerbond SPE silica gel (1 g) disposable extraction columns (JT Baker,
Phillipsburg, NJ, USA) were used for phospholipids
and polar compound analyses, respectively. n-Hexane,
2-propanol, and tetrahydrofuran for high performance
liquid chromatography (HPLC) were provided by JT
Baker. Tetrahydrofuran was redistilled and stabilized.
All other chemicals and solvents were of analytical
grade.
Method I: oil recovery and lecithin production
by direct extraction
The first extraction was carried out with cold acetone
at 0 ◦ C in a wet gum–solvent ratio of 1:1.5 w/v,
with continuous shaking for 30 min. After resting
for 15 min the extract was separated by filtration.
The residue was extracted another two times with
cold acetone (1:1, w/v) under the same conditions.
The oil was recovered by elimination of the solvent
in a rotary evaporator under vacuum at 40 ◦ C and
with centrifugation (1500 × g, 5 min) to separate oil
and aqueous phase. Finally the oil was dried under
nitrogen. Lecithin was obtained after drying the
extraction residue in a vacuum oven (125 mmHg)
at room temperature. Yields for oil and lecithin were
gravimetrically determined.
Method II: oil recovery and lecithin production
by drying under vacuum and extraction
Wet gum was completely dried under vacuum
(70 mmHg) in a rotary evaporator at 60 ◦ C. The
eliminated water was periodically monitored by
weight. Oil and lecithin were obtained according to
the procedure in Method I. The centrifugation step
was eliminated because water was not present in the
extraction medium.
Method III: oil recovery and lecithin production
by solvent partition and extraction
Wet gum was dissolved in n-hexane (1:1 w/w) at 50 ◦ C
with magnetic stirring and then absolute ethanol was
added drop by drop until a steep colour change and
phase separation were observed. The hexane phase
(phase A), containing phospholipids, oil, pigments,
and other minor compounds, and alcohol/water phase
(phase B), with phospholipids and other components,
were recovered using a separating funnel. Phase A
was evaporated at 40 ◦ C under vacuum in a rotary
evaporator and treated as in Method I without
centrifugation, to recovery oil and to obtain lecithin
(phase A lecithin). Phase B was evaporated at 50 ◦ C
under vacuum and completely dried in a vacuum oven
at room temperature. Yields were calculated from the
weights of both lecithin fractions and recovered oil.
Wet gum analyses
Moisture content was determined by distillation
with an immiscible solvent (toluene) according
to AOCS Official Method Ja 2a-46.9 Acetoneinsoluble material was evaluated by AOCS Official
2461
LN Ceci, DT Constenla, GH Crapiste
Method Ja 4–46.9 This fraction basically includes
phosphatides in samples free from sand, meal, and
other petroleum ether-insoluble materials. AOCS
Official Method Ja 3–87 was used to measure hexaneinsoluble material.9 The percentage of occluded oil
was estimated by difference (100 – % moisture – %
acetone insoluble – % hexane insoluble).
Phospholipids were determined by HPLC (AOCS
Official Method Ja 7b-91)9 with a photodiode array
detector at 206 nm and n-hexane/2-propanol/acetate
buffer pH 4.2 (8:8:1, v/v/v) as mobile phase. The
HPLC system was equipped with a LiChrosorb Si-60
column (Polymer Laboratories Inc., Amherst, MA,
USA) (250 × 4 mm, particle size 5 µm) and Empower
2 software (Water Corporation, Milford, MA, USA).
Recovered oil analyses
Acidity, as oleic acid percentage, was measured
by titration with a standardized ethanolic solution
of potassium hydroxide and phenolphthalein as
indicator (IUPAC Standard Method 2.201).10 AOCS
Official Method Cd 8-539 for peroxide value (acetic
acid–chloroform method) was employed to measure
peroxides and other similar compounds that oxidize
potassium iodide as primary oxidation products.
p-Anisidine value was determined by AOCS Official
Method Cd 18–90,9 which measures the amount of
aldehydes (principally 2-alkenals and 2,4-dienals) as
secondary oxidation products. Unsaponifiable matter
was evaluated by AOCS Official Method Ca 6a-40.9
This fraction includes those substances frequently
dissolved in oils, such as higher aliphatic alcohols,
sterols, pigments, and hydrocarbons, which cannot
be saponified by the usual caustic treatment, but are
soluble in ordinary oil solvents.
Phosphorous content was determined by ashing
the oil in the presence of zinc oxide, followed by
spectrophotometric measurement of phosphorous as a
blue phosphomolybdic acid complex (AOCS Official
Method Ca 12–55).9
A partition procedure using SPE Diol extraction
columns as previously described was applied to phospholipid enrichment and their separation in oils.11
Briefly, this procedure included: (i) sorbent conditioning with 2 mL methanol, 2 mL chloroform, and
4 mL hexane; (ii) sample loading, in which 200 µL
of chloroform–oil solution containing 50 mg oil was
injected with a micropipette; (iii) triglyceride release
from the sorbent bed, accomplished by passing 2.5 mL
chloroform through; and (iv) phospholipid recovery by
elution with 7 mL of a 25% solution of ammonium
hydroxide–methanol (0.5% v/v). The phospholipids
were collected in a conical vial, evaporated to dryness
under nitrogen, and made up to 100 µL with mobile
phase. Phospholipids were analyzed following the procedure and conditions indicated for wet gum analyses.
Polar compounds, as products of oxidation, polymerization, and hydrolysis in heated oils, were
analyzed by high-performance size exclusion chromatography using monopalmitin as internal standard,
2462
after separation of non-polar compounds in SPE
extraction columns with silica gel phase.12 Columns
were first conditioned with 10 mL 40–60 ◦ C light
petroleum–diethyl ether (90:10, v/v). An aliquot of
2 mL light petroleum solution containing 50 mg oil
and 1 mg internal standard was injected in a silica bed (sample solution). Internal standard was
dissolved in diisopropyl ether (5 mg mL−1 ) before
adding to the sample solution. The non-polar fraction (triglycerides) was eluted by passing 15 mL light
petroleum–diethyl ether (90:10, v/v) through the silica bed. The polar fraction was collected in a conical
vial with 10 mL diethyl ether, evaporated to dryness
under nitrogen, and then diluted with tetrahydrofuran
mobile phase. Two PLgel columns connected in series
(300 × 7.5 mm, particle size 5 µm, pore size 500 and
˚ a refractive index detector, and a Millennium
100 A),
2010 chromatography manager were used.
Analysis of obtained lecithin
Phospholipids were determined by HPLC using the
analytical methodology indicated by wet gums.
Statistical analysis
Methods for oil recovery and lecithin production were
applied in triplicate. Results are expressed as mean
value ± standard deviation. The differences in mean
values between samples were assessed with Student’s
t-test, being statistically different at a significance level
of 5%.
RESULTS AND DISCUSSION
Composition of soybean degumming sludge
Wet gums showed high water content of almost
500 g kg−1 on average, indicating high sensitivity for
hydrolytic damage when are not stored and processed
under convenient conditions (Table 1). No changes
in general composition, acid, iodine, and peroxide
values were observed for rapeseed wet gum stored in
frozen state (−20 ◦ C) for 24 months.13 Gums used
in this study were frozen at −20 ◦ C and carefully
processed to minimize hydrolytic damage. As can
be observed in Table 1, the average phospholipid
content in wet gums was almost 300 g kg−1 and
the occluded oil was estimated as approximately
250 g kg−1 . Gums had a low content of hexaneinsoluble impurities (4 g kg−1 ) such as sand, meal,
and other insoluble materials. It can be observed that
the composition of wet gums is variable (Table 1).
The source of this variability may be genetic (plant
cultivar), seed quality (maturity, harvesting-induced
damage, and handling/storage conditions), and oilprocessing variables.14
Yields of recovered oil and obtained lecithin
The yields for recovered oil by methods II and III,
556 and 588 g kg−1 , respectively, were significantly
higher than the yield for the recovered oil by direct
extraction (Table 2). In both methods water was
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
Oil and lecithin from water degumming sludge
Table 1. Composition of soybean degumming sludges (wet gums)
Moisture (g kg−1 )
Acetone-insoluble material (g kg−1 )
Hexane-insoluble material (g kg−1 )
Occluded oil (g kg−1 )
462 ± 48
293 ± 62
4±1
241 ± 15
Phospholipids (g kg−1 )
PC
120 ± 31
% Relative
PE
PA
PI
PC
PE
PA
PI
68 ± 15
28 ± 0.5
43 ± 16
46 ± 2
28 ± 2
10 ± 1
16 ± 1
Table 2. Yields of recovered oil and obtained lecithin from wet gums
Recovered oil yield (g kg−1 )
Method
Wet gum
I
II
III
104 ± 18
135 ± 26
142 ± 14
Obtained lecithin yield (g kg−1 )
Occluded oil
429 ± 71 a
556 ± 96 b
588 ± 60 b
Wet gum
Dried gum
416 ± 82
437 ± 86
Total: 391 ± 87
Ph. A: 331 ± 76
Ph. B: 60 ± 12
767 ± 90a
807 ± 93a
Total: 720 ± 99a
Ph. A: 610 ± 89
Ph. B: 110 ± 13
Means within a column followed by the same letter are not significantly different (α = 0.05).
Table 3. Quality indexes for recovered oils from wet gums
Quality index
Acidity (g oleic acid kg−1 )
Peroxide value (mEq kg−1 )
p-Anisidine value
Unsaponifiable matter (g kg−1 )
Phosphorous content (mg kg−1 )
TOTOX valuea
Method I
Method II
Method III
16.9 ± 6.5a
1.62 ± 0.49a
1.86 ± 1.50a
9.0 ± 1.1a
87 ± 34a
5.09 ± 1.86a
21.7 ± 7.3a
1.02 ± 0.72a
4.03 ± 2.19a
12.1 ± 2.0b
330 ± 76b
5.77 ± 2.42a
16.7 ± 2.8a
3.24 ± 1.98a
2.34 ± 0.76a
11.5 ± 2.1ab
244 ± 73b
8.82 ± 3.91a
Total oxidation value = 2 × peroxide value + p-anisidine value.
Means within a row followed by the same letter are not significantly different (α = 0.05).
a
eliminated before extraction with acetone. Method
II was performed with total water elimination and in
Method III about 85% of the water was eliminated.
The direct extraction method showed lower oil
yields and, moreover, centrifugation was necessary
to separate oil and aqueous phase.
The lecithin yields (720–807 g kg−1 of dried gum)
were not significantly different in the three methods
(Table 2). Two fractions of lecithin can be obtained
by Method III, lecithin from Phase A being the most
abundant.
Quality indexes for recovered oils
Table 3 shows some quality indexes for recovered oils
using the three studied methods. The free fatty acids
in recovered oil ranged from 16.7 to 21.7 g kg−1 as
oleic acid and no significant differences were observed
between methods. The Codex Alimentarius establishes
a maximum level of 0.6 mg KOH g−1 (3 g kg−1 as
oleic acid) for refined oil.15 The National Oilseed
Processors Association (NOPA), in soybean oil trading
rules revised in 2007, fixes maximum levels of 7.5 g
kg−1 for crude degummed soybean oil, as oleic acid,
and applies allowances between 7.6 and 12.5 g kg−1 .16
However, crude soybean oil can have acid values of up
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
to 4.0 mg KOH g−1 or 20 g kg−1 as oleic acid,17 which
are easily reduced during refining process.
Peroxide values, as a measure of primary oxidation
products, were lower than 3.24 mEq kg−1 , and
anisidine values, which measure secondary oxidation
compounds, were lower than 4.03 (Table 3). No
significant differences were observed for these indexes
in recovered oils by the three methods. Total
oxidation values (TOTOX values), often used in
the industry, were lower than 8.82. These values
combine evidence about the past history of oil
(as reflected in the p-anisidine value) with its
present state (as evidenced in the peroxide value).
It is acceptable to maintain a peroxide value of
less than 4 and an anisidine value of less than
2 in the crude oil during storage.18 For crude
oils TOTOX values less than 10 correspond to
good-quality oils on an industrial scale. Oxidation
products are reduced during bleaching in the refining
process.
The Codex Alimentarius fixes a maximum value of
15 g kg−1 for crude soybean oils in unsaponifiable
matter.15 All recovered oils were adjusted to this norm
for characterization (Table 3).
2463
LN Ceci, DT Constenla, GH Crapiste
Phosphorous and phospholipids in recovered
oils
As shown in Table 3, the phosphorous content in
oils recovered by Method I averaged 87 mg kg−1 ,
in accordance with the values proposed for crude
degummed soybean oil (<200 mg kg−1 ).16 Shipments
up to 250 mg kg−1 are permitted with discounts.
Oils obtained by Methods II and III had higher
phosphorous contents than those obtained by direct
extraction (Table 3). Crude soybean oils can contain
more than 1000 mg phosphorous kg−1 , depending on
the extraction and preparation methods, and values
of 10–15 mg kg−1 are usual in refined soybean oils.19
Phosphorous contents of the order of 1–3 mg kg−1
can be observed in completely refined and packaged
soybean oils. The phosphorous content observed in
oils recovered by Methods II and III can be easily
reduced during the refining process by degumming.
To estimate the phospholipid contents in oils, a
factor of 30 is usually applied, to convert the
percentage of total phosphorous to the equivalent
content of phosphatides.20 However, in crude oils
this factor overestimates the phospholipids and lower
factors have been proposed.21 Crude oils contain
phosphorous from another sources, such as sand and
meal residues, including inorganic phosphorous that
is also determined by the spectrophotometric method.
The contents and relative composition of phospholipids in recovered oils determined by the chromatographic method are shown in Table 4. Although this
method has the advantage of identifying and quantifying phospholipids separately, other minor compounds
(e.g., lysophospholipids) are not determined. However, four major phospholipids in soybean oil can
be evaluated by chromatography and the remaining
fraction is relatively low. Oils recovered by Method
I had, on average, lower phosphorous and phospholipid contents than those obtained by Methods II and
III. These results suggest that during direct extraction
of oil water and hydratable phospholipids would be
pulled by the acetone and retained in aqueous phase
after centrifugation. Crude soybean oils have 15–30 g
phosphatides per kg−1 ; this content is reduced to 3–8 g
kg−1 in water-degummed soybean oil, and 0.03–0.45 g
kg−1 in refined oils.22,23 It is remarkable that all recovered oils had total phospholipid contents practically
in the range accepted for degummed products. A
large dispersion between samples was observed for
total phospholipid contents, demonstrating the strong
Table 4. Phospholipid composition for recovered oils (% relative)
Phospholipid
Method I
Method II
Method III
PE
PA
PI
PC
Total (g kg−1 )
17 ± 5a
7 ± 5ab
14 ± 4a
62 ± 11a
1.56 ± 0.64a
11 ± 6a
10 ± 4a
8 ± 2b
71 ± 10a
9.56 ± 3.43b
12 ± 5a
5 ± 3b
6 ± 6b
77 ± 12a
5.10 ± 4.16ab
Means within a row followed by the same letter are not significantly
different (α = 0.05).
2464
influence of wet gum composition on the phosphatide
contents in recovered oils.
As shown in Table 4, the recovered oils had a
high relative percentage of PC (62–77%), and low
percentages of PE and PI, when compared with
data provided in the literature for crude soybean
oils. Mounts et al.24 obtained the following relative
values, working on samples of crude soybean oils
extracted from standard and genetically modified
varieties: 20.7–35.8% for PE, 18.2–27.9% for PI,
2.1–35.0% for PA, and 25.4–49.7% for PC. An
additional reason for high relative PC content could be
that the reference phospholipid samples for the HPLCUV method have a different fatty acid pattern from the
soy phopholipids. PA is the most variable phospholipid
in crude soybean oil and its percentage in the recovered
oils was relatively low. A water degumming step during
the refining process could reduce the phospholipids
in the recovered oils without drawbacks, provided
that PC is the most easily hydratable compound,
whereas PA and PI are eliminated with difficulty by
hydratation. Only small significant differences in the
percentages of PA and PI were observed when the
methods for recovery of oil were compared (Table 4).
The oils recovered by direct extraction had more nonhydratable PI than those obtained by Methods II and
III. Evidently, hydratable phospholipids are carried
out to aqueous phase and non-hydratable PI is more
easily retained in recovered oils.
Polar compounds in recovered oils
No significant differences were observed in total and
individual polar compounds between the methods for
oil recovery from wet gums (Table 5). Polymerized,
dimerized, and oxidized triglycerides (PTG, DTG,
and OTG, respectively) are used as indicators of
thermal degradation (TD). Diglycerides (DG) and
free fatty acids (FFA) indicate hydrolytic degradation
(HD). All recovered oils showed a level of HD higher
than their TD (TD/HD < 1). These results suggest
that the storage of wet gums before processing, under
controlled conditions to reduce hydrolytic damage, is
crucial to obtain good-quality oils.
The content of total polar compounds in the
recovered oils from soybean wet gums ranged from
48.8 to 56.1 g kg−1 (Table 5). These contents are only
slightly higher than those observed in crude sunflower
oils, obtained by pressing (43.4 and 40.5 g kg−1 ) and
solvent extraction (45.3 and 38.9 g kg−1 ).25 These
results confirm the qualification of recovered oils as
good-quality products. Moreover, treatment by water
degumming could reduce polar compounds in the
recovered oils.25
No appreciable amounts of DTG and PTG were
detected in recovered oils – an expected result since
polymerization due to thermal degradation occurs at
temperatures higher than those used in the methods for
oil recovery. The most significant change is observed
in the concentration of OTG relating to oxidative
deterioration.
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
Oil and lecithin from water degumming sludge
Table 5. Polar compounds (g kg−1 ) for recovered oils from wet gums
Compound
PTG
DTG
OTG
DG
FFA
Total
TD
HD
TD/HD
Method I
Method II
Method III
0.6 ± 0.1a
0.0 ± 0.0a
22.5 ± 3.6a
8.4 ± 3.8a
24.6 ± 9.3a
56.1 ± 13.7a
23.1 ± 3.7a
33.0 ± 13.0a
0.70 ± 0.24a
0.5 ± 0.3a
0.0 ± 0.0a
21.2 ± 5.9a
6.8 ± 1.5a
26.6 ± 7.9a
55.1 ± 10.7a
21.7 ± 6.1a
33.4 ± 9.4a
0.65 ± 0.24a
0.6 ± 0.1a
0.0 ± 0.0a
17.2 ± 2.1a
6.1 ± 1.3a
24.9 ± 9.3a
48.8 ± 10.7a
17.8 ± 2.4a
31.0 ± 10.4a
0.57 ± 0.16a
PTG, polymerized triglycerides; DTG, dimerized triglycerides; OTG,
oxidized triglycerides; DG, diglycerides; FFA, free fatty acids; TD, thermal degradation = PTG + DTG + OTG; HD, hydrolytic degradation =
DG + FFA.
Means within a row followed by the same letter are not significantly
different (α = 0.05).
Phospholipids in obtained lecithins
The content and relative composition of phospholipids
in the lecithins obtained by the three methods
are shown in Table 6. No significant differences
were detected in total phospholipid content between
Methods I, II and III (hexane phase), with values
ranging from 610 to 691 g kg−1 . These contents
include the four phospholipids more relevant in
soybean: PC, PE, PI and PA. More abundant
phospholipids were PC and PE, in the range
203–319 g kg−1 and 185–218 g kg−1 , respectively
(Table 6). The following values have been obtained in
the literature for soybean lecithin with an intermediate
range of composition: PC = 290–390 g kg−1 and
PE = 200–263 g kg−1 . However, soybean lecithin with
a low range of composition contains 120–210 g kg−1
of PC and 80–95 g kg−1 of PE.14
Lecithin obtained by Method I had a higher
PC content than those obtained by extraction after
water elimination. PC is extracted more efficiently
by acetone when no water is disposable. Lecithin
Table 6. Phospholipid composition for obtained lecithins
Phospholipid
Method
I
Method
II
Method
Method
III Phase A III Phase B
PE
(g kg−1 )
185 ± 28a 214 ± 21a 218 ± 30a
(% Relative) 27 ± 3a
31 ± 3b
36 ± 2c
55 ± 10b
33 ± 1b
PA
(g kg−1 )
(% Relative)
74 ± 14a
12 ± 1ab
22 ± 5c
13 ± 1b
151 ± 5b
22 ± 1b
107 ± 11a
18 ± 1c
33 ± 6c
20 ± 2bc
PC
319 ± 15a 203 ± 4b
(g kg−1 )
(% Relative) 46 ± 2a
29 ± 1b
211 ± 20b
35 ± 1c
58 ± 12c
35 ± 2c
PI
74 ± 5a
11 ± 1a
113 ± 7a
(g kg−1 )
(% Relative) 16 ± 1a
94 ± 11b
13 ± 2b
Total (g kg−1 ) 691 ± 53a 662 ± 31a 610 ± 70a 168 ± 32b
Means within a row followed by the same letter are not significantly
different (α = 0.05).
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
obtained by extraction of oil after drying under vacuum
(Method II) had slightly higher PA and PI contents
than those obtained by Methods I and III Phase
A. Lecithin obtained by Method III from hexane
phase had the highest content of PE. The results
show that lecithin obtained by Methods I, II and III
hexane phase have the same purity level but different
relative compositions and could be used in different
applications.
From the alcoholic phase lecithin was obtained with
low total phospholipid content (168 g kg−1 , Table 6).
Moreover, the relative composition of phospholipids
in this lecithin is not significantly different from
that obtained from hexane phase. More studies are
required to analyse the composition of the colourless
product with a crystalline appearance obtained from
the alcoholic phase. Other components such as
glycolipids and complex carbohydrates could be
profitably recovered.
ACKNOWLEDGEMENTS
Thanks are given for financial support to the
´ Argentina de Grasas y Aceites (ASAGA),
Asociacion
the Concejo Nacional de Investigaciones Cient´ıficas y
T´ecnicas (CONICET) and the Universidad Nacional
del Sur (UNS). Thanks are also due to Oleaginosa
Moreno Hermanos (OMHSA) for providing samples.
REFERENCES
1 List GR, Avellaneda JM and Mounts TL, Efectos de las
´ y la calidad de la
condiciones de desgomado en la extraccion
lecitina de soja. Aceites y Grasas 43:207–218 (2001).
2 List GR, Commercial manufacture of lecithin, in Lecithins:
Sources, Manufacture and Uses, ed. by Szuhaj BF. AOCS,
Champaign, IL, pp. 145–162 (1989).
3 Van Nieuwenhuyzen W, Fractionation of lecithins. Eur Food
Drink Rev, Process Technology 27–32 (1999).
4 Wu Y and Wang T, Soybean lecithin fractionation and
functionality. J Am Oil Chem Soc 80:319–326 (2003).
5 Wu Y and Wang T, Fractionation of crude soybean lecithin with
aqueous ethanol. J Am Oil Chem Soc 81:697–704 (2004).
6 Schneider M, Fractionation and purification of lecithin, in
Lecithins: Sources, Manufacture and Uuses, ed. by Szuhaj BF.
AOCS, Champaign, IL, pp. 109–130 (1989).
7 Ghyczy M, Synthesis and modification of phospholipids, in
Lecithins: Sources, Manufacture and Uses, ed. by Szuhaj BF.
AOCS, Champaign, IL, pp. 131–144 (1989).
8 Dashiell GL, Fuentes, m´etodos de proceso y usos comerciales
de la lecitina. Aceites y Grasas 43:197–204 (2001).
9 American Oil Chemists’ Society, Official Methods and Recommended Practices, (5th edn), ed. by Firestone D. AOCS,
Champaign, IL (1998).
10 International Union of Pure and Applied Chemistry (IUPAC),
Standard Methods for the Analysis of Oils, Fats and Derivatives
(7th edn),ed. by Paquot C and Hautfenne A. Blackwell
Scientific, Oxford (1992).
11 Carelli AA, Brevedan MIV and Crapiste GH, Quantitative
determination of phospholipids in sunflower oil. J Am Oil
Chem Soc 74:511–514 (1997).
12 Dobarganes MC, Velasco J and Dieffenbacher A, Determination of polar compounds, polymerized and oxidized triacylglycerols, and diacylglycerols in oils and fats. Pure Appl Chem
72:1563–1575 (2000).
2465
LN Ceci, DT Constenla, GH Crapiste
13 Sosada M, Studies on stability of rapeseed wet gum as a source
of pharmaceutical lecithin. J Am Oil Chem Soc 73:367–370
(1996).
14 Cherry JP and Kramer WH, Plant sources of lecithin, in
Lecithins: Sources, Manufacture and Uses, ed. by Szuhaj BF.
AOCS, Champaign, IL, pp. 16–31 (1989).
15 Codex Alimentarius, Codex standard for named vegetable oils,
CODEX-STAN 210 (2005).
16 National Oilseed Processors Association, Trading rules for the
purchase and sale of soybean oil. NOPA, Washington DC
(2007).
17 USDA Foreign Agricultural Service, GB1535-2003 Soybean
Oil Standard – G/TBT/N/CHN/25. Gain Report number
CH4022 (2004).
18 Gupta MK, Fundamental quality control in vegetable oil
refining, Oil Mill Gaz 108:6–9 (2003).
19 Lence SH and Agarwal S, Assessing the feasibility of processing
and marketing niche soy oil. Research Paper 03-MRP
6, Midwest Agribusiness Trade Research and Information
Center (MATRIC) (2003).
2466
20 Chapman GW, A conversion factor to determine phospholipid
content in soybean and sunflower crude oils. J Am Oil Chem
Soc 57:299–302 (1980).
21 Carelli AA, Ceci LN and Crapiste GH, Phosphorous-tophospholipid conversion factors for crude and degummed
sunflower oils. J Am Oil Chem Soc 79:1177–1180 (2002).
22 Buchold H, Enzymax: the enzyme-catalyzed degumming process of vegetable oils, 48. DGF (German Finance Association), Jahrestagung, Essen (1992).
23 Debruyne I, Soybean oil processing: quality criteria and flavour
reversion. Oil Mill Gaz 110:10–11 (2004).
24 Mounts TL, Abidi SL and Rennick KA, Effect of genetic
modification on the content and composition of bioactive
constituents in soybean oil. J Am Oil Chem Soc 73:581–586
(1996).
25 Brevedan MIV, Carelli AA and Crapiste GH, Changes in
composition and quality of sunflower oils during extraction
and degumming, Grasas y Aceites 51:417–423 (2000).
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
J Sci Food Agric 88:2460–2466 (2008)
Oil recovery and lecithin production
using water degumming sludge of crude
soybean oils†
Liliana N Ceci,∗ Diana T Constenla and Guillermo H Crapiste
Planta Piloto de Ingenier´ıa Qu´ımica (PLAPIQUI), Universidad Nacional del Sur (UNS)-Consejo Nacional de Investigaciones Cient´ıficas y
´
Tecnicas
(CONICET), Camino Carrindanga Km 7, CC 717, 8000, Bah´ıa Blanca, Argentina
Abstract
BACKGROUND: Wet gums produced during aqueous degumming of crude soybean oils are currently processed
to produce lecithin or added to meals to increase their nutritive value for animal feed. Oils occluded in these
gums are generally not recovered or processed. In this work, three methods to recover occluded oil and obtain
lecithin from wet gums were assayed: direct extraction of oil with cold acetone (Method I), extraction after water
elimination under vacuum (Method II) and by solvent partition with hexane/ethanol (Method III).
RESULTS: Higher oil yields (up to 588 g kg−1 of occluded oil) were obtained when water was eliminated before
extraction (Methods II and III). No significant differences were observed in lecithin yields between three methods
(720–807 g kg−1 of dried gums). Recovered oils had acidity = 16.7–21.7 g kg−1 as oleic acid, TOTOX (total oxidation)
values ≤ 8.82, unsaponifiable matter = 9.0–12.1 g kg−1 , and Phosphorus = 87–330 mg kg−1 . Lecithins obtained by
Methods I, II and III hexane phase had the same purity level (610–691 g of total measured phospholipids kg−1 ).
CONCLUSIONS: The occluded oil in soybean wet gums can be recovered, with quality and stability indexes
compatible with their reinsertion in the productive process, by water elimination and extraction with acetone.
Lecithins can be obtained with different phospholipid composition and diverse application fields.
2008 Society of Chemical Industry
Keywords: soybean lecithin; soybean oil, gums; phospholipids
INTRODUCTION
World soybean production has been estimated for
2006/2007 as 218 million tonnes (Mt) and therefore
about 35 Mt of soybean oil will be produced. More
than 1 Mt of wet gums, which contain about 260 000 t
of lecithin, will be obtained in the processing of
soybean oil during water degumming step. These
gums or sludge are a complex mixture comprising
phospholipids or lecithin, oil, and minor amounts of
other constituents like phytoglycolipids, phytosterols,
tocopherols, and fatty acids. They have a high water
content that rapidly promotes damage if they are not
properly stored and processed. The composition and
molecular structure of this heterogeneous mixture
of compounds vary depending on the degumming
conditions of the oil.1
Some methods to produce, purify, and fractionate
lecithin from gums have been studied. In the
conventional industrial process wet gums are dried
under vacuum and then de-oiled with cold acetone, a
solvent in which phospholipids, glycolipids and related
compounds are almost insoluble.2 Partition methods
with solvents to enable water elimination1 and
∗
phospholipid fractionation3 have also been proposed.
Fractions with high phosphatidyl choline/phosphatidyl
ethanolamine (PC/PE) ratio and fractions with high
phosphatidyl inositol (PI) contents can be obtained
by fractionation with ethanol.4,5 The PC-enriched
fractions have unique anti-spattering properties as
oil/water (o/w) emulsifiers in margarines with low
or no salt content and they are suitable for systems
with hard water, salts and in the presence of
milk proteins, because PC is not flocculated by
calcium and magnesium ions. PI-enriched fractions
are used as water/oil (w/o) emulsifiers in the
confectionery industry. Pure PC can be obtained by
chromatographic procedures.3,6 Alternative methods
for extraction and partition of phospholipids using
supercritical fluids have been developed, although
they are not applied on an industrial scale.3 The
advantages of these processes are the absence of
oxygen, which would promote oxidation and solvent
residues with flammability risks and environmental
problems.
Lecithins can be modified by chemical reactions
such as acylation and hydroxylation to increase their
Correspondence to: Liliana N Ceci, Planta Piloto de Ingenier´ıa Qu´ımica (PLAPIQUI), Universidad Nacional del Sur (UNS)-Consejo Nacional de Investigaciones
´
Cient´ıficas y Tecnicas
(CONICET), Camino Carrindanga Km 7, CC 717, 8000, Bah´ıa Blanca, Argentina
E-mail: [email protected]
†
4th Euro Fed Lipid Congress, Madrid (Spain), 2006
(Received 11 March 2008; revised version received 3 July 2008; accepted 8 July 2008)
Published online 15 September 2008; DOI: 10.1002/jsfa.3363
2008 Society of Chemical Industry. J Sci Food Agric 0022–5142/2008/$30.00
Oil and lecithin from water degumming sludge
hydrophilic properties and improve their behaviour
characteristics as o/w emulsifiers.7,8 Special lecithins
can be produced by means of enzymatic modifications.
A2 -Phospholipase can be used to obtain lysophospholipids with o/w emulsification properties tremendously
enhanced.8
Soybeans represent 90% of total oleaginous crops
harvested in Argentina, which is the world’s third
highest producer of soybean behind the USA and
Brazil. However, only some hundreds of tonnes of
crude and purified lecithin are monthly produced.
No processed gums are fundamentally added to
pellets and meals to increase their nutritive value.
Argentinian soya is characterized by low protein
content, which is why gum addition to meals and
pellets reduces their protein availability and increases
their lipid content, making their commercialization
difficult. Gum processing should therefore provide a
solution to these drawbacks. In 2006, the installed
capacity to crush soybean in Argentina was 132 000
t per day and soon it will reach 160 000 t per day.
A major volume of wet gums will be produced as a
consequence of this increase.
On the other hand, in spite of extended information
about phospholipids obtained from gums, there are
no studies available on the recovery of the occluded
oil and the evaluation of its quality indexes. It can be
estimated that more than 200 000 t of occluded oil
will be lost with wet gums in 2006/2007. Moreover,
increases of $803.1 t−1 in the price of soybean oil
with future contracts on Chicago Board on Trade are
registered for December 2007.
The aim of the present study was the recovery
of occluded oil and lecithin production from wet
soybean gums. Direct extraction methods of oil with
cold acetone, and water elimination under vacuum
and by partition with hexane/ethanol, before acetone
extraction, were assayed and the oil and lecithin yields
were estimated. Quality indexes were evaluated for
the recovered oil and phospholipid composition was
determined for oils and lecithins.
EXPERIMENTAL
Materials
Five samples of wet gums from water degumming of
crude soybean oil were provided by two commercial
plants in the April–September period in 2004. The
samples were fractionated and stored at −20 ◦ C to
prevent damage before processing.
L-α-Phosphatidyl ethanolamine (PE), L-α-phosphatidyl inositol (PI), and L-α-phosphatidyl choline
(PC) from soybean, and L-α-phosphatidic acid (PA)
sodium salt from egg yolk lecithin with purity greater
than 98% were used as external standards for
phospholipid analysis. 1-Monopalmitoyl-rac-glycerol
(monopalmitin, purity 99%) was used as internal
standard for polar compound determination. Standards were provided by Sigma (St Louis, MO, USA).
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
Bakerbond SPE Diol (0.5 g) and Bakerbond SPE silica gel (1 g) disposable extraction columns (JT Baker,
Phillipsburg, NJ, USA) were used for phospholipids
and polar compound analyses, respectively. n-Hexane,
2-propanol, and tetrahydrofuran for high performance
liquid chromatography (HPLC) were provided by JT
Baker. Tetrahydrofuran was redistilled and stabilized.
All other chemicals and solvents were of analytical
grade.
Method I: oil recovery and lecithin production
by direct extraction
The first extraction was carried out with cold acetone
at 0 ◦ C in a wet gum–solvent ratio of 1:1.5 w/v,
with continuous shaking for 30 min. After resting
for 15 min the extract was separated by filtration.
The residue was extracted another two times with
cold acetone (1:1, w/v) under the same conditions.
The oil was recovered by elimination of the solvent
in a rotary evaporator under vacuum at 40 ◦ C and
with centrifugation (1500 × g, 5 min) to separate oil
and aqueous phase. Finally the oil was dried under
nitrogen. Lecithin was obtained after drying the
extraction residue in a vacuum oven (125 mmHg)
at room temperature. Yields for oil and lecithin were
gravimetrically determined.
Method II: oil recovery and lecithin production
by drying under vacuum and extraction
Wet gum was completely dried under vacuum
(70 mmHg) in a rotary evaporator at 60 ◦ C. The
eliminated water was periodically monitored by
weight. Oil and lecithin were obtained according to
the procedure in Method I. The centrifugation step
was eliminated because water was not present in the
extraction medium.
Method III: oil recovery and lecithin production
by solvent partition and extraction
Wet gum was dissolved in n-hexane (1:1 w/w) at 50 ◦ C
with magnetic stirring and then absolute ethanol was
added drop by drop until a steep colour change and
phase separation were observed. The hexane phase
(phase A), containing phospholipids, oil, pigments,
and other minor compounds, and alcohol/water phase
(phase B), with phospholipids and other components,
were recovered using a separating funnel. Phase A
was evaporated at 40 ◦ C under vacuum in a rotary
evaporator and treated as in Method I without
centrifugation, to recovery oil and to obtain lecithin
(phase A lecithin). Phase B was evaporated at 50 ◦ C
under vacuum and completely dried in a vacuum oven
at room temperature. Yields were calculated from the
weights of both lecithin fractions and recovered oil.
Wet gum analyses
Moisture content was determined by distillation
with an immiscible solvent (toluene) according
to AOCS Official Method Ja 2a-46.9 Acetoneinsoluble material was evaluated by AOCS Official
2461
LN Ceci, DT Constenla, GH Crapiste
Method Ja 4–46.9 This fraction basically includes
phosphatides in samples free from sand, meal, and
other petroleum ether-insoluble materials. AOCS
Official Method Ja 3–87 was used to measure hexaneinsoluble material.9 The percentage of occluded oil
was estimated by difference (100 – % moisture – %
acetone insoluble – % hexane insoluble).
Phospholipids were determined by HPLC (AOCS
Official Method Ja 7b-91)9 with a photodiode array
detector at 206 nm and n-hexane/2-propanol/acetate
buffer pH 4.2 (8:8:1, v/v/v) as mobile phase. The
HPLC system was equipped with a LiChrosorb Si-60
column (Polymer Laboratories Inc., Amherst, MA,
USA) (250 × 4 mm, particle size 5 µm) and Empower
2 software (Water Corporation, Milford, MA, USA).
Recovered oil analyses
Acidity, as oleic acid percentage, was measured
by titration with a standardized ethanolic solution
of potassium hydroxide and phenolphthalein as
indicator (IUPAC Standard Method 2.201).10 AOCS
Official Method Cd 8-539 for peroxide value (acetic
acid–chloroform method) was employed to measure
peroxides and other similar compounds that oxidize
potassium iodide as primary oxidation products.
p-Anisidine value was determined by AOCS Official
Method Cd 18–90,9 which measures the amount of
aldehydes (principally 2-alkenals and 2,4-dienals) as
secondary oxidation products. Unsaponifiable matter
was evaluated by AOCS Official Method Ca 6a-40.9
This fraction includes those substances frequently
dissolved in oils, such as higher aliphatic alcohols,
sterols, pigments, and hydrocarbons, which cannot
be saponified by the usual caustic treatment, but are
soluble in ordinary oil solvents.
Phosphorous content was determined by ashing
the oil in the presence of zinc oxide, followed by
spectrophotometric measurement of phosphorous as a
blue phosphomolybdic acid complex (AOCS Official
Method Ca 12–55).9
A partition procedure using SPE Diol extraction
columns as previously described was applied to phospholipid enrichment and their separation in oils.11
Briefly, this procedure included: (i) sorbent conditioning with 2 mL methanol, 2 mL chloroform, and
4 mL hexane; (ii) sample loading, in which 200 µL
of chloroform–oil solution containing 50 mg oil was
injected with a micropipette; (iii) triglyceride release
from the sorbent bed, accomplished by passing 2.5 mL
chloroform through; and (iv) phospholipid recovery by
elution with 7 mL of a 25% solution of ammonium
hydroxide–methanol (0.5% v/v). The phospholipids
were collected in a conical vial, evaporated to dryness
under nitrogen, and made up to 100 µL with mobile
phase. Phospholipids were analyzed following the procedure and conditions indicated for wet gum analyses.
Polar compounds, as products of oxidation, polymerization, and hydrolysis in heated oils, were
analyzed by high-performance size exclusion chromatography using monopalmitin as internal standard,
2462
after separation of non-polar compounds in SPE
extraction columns with silica gel phase.12 Columns
were first conditioned with 10 mL 40–60 ◦ C light
petroleum–diethyl ether (90:10, v/v). An aliquot of
2 mL light petroleum solution containing 50 mg oil
and 1 mg internal standard was injected in a silica bed (sample solution). Internal standard was
dissolved in diisopropyl ether (5 mg mL−1 ) before
adding to the sample solution. The non-polar fraction (triglycerides) was eluted by passing 15 mL light
petroleum–diethyl ether (90:10, v/v) through the silica bed. The polar fraction was collected in a conical
vial with 10 mL diethyl ether, evaporated to dryness
under nitrogen, and then diluted with tetrahydrofuran
mobile phase. Two PLgel columns connected in series
(300 × 7.5 mm, particle size 5 µm, pore size 500 and
˚ a refractive index detector, and a Millennium
100 A),
2010 chromatography manager were used.
Analysis of obtained lecithin
Phospholipids were determined by HPLC using the
analytical methodology indicated by wet gums.
Statistical analysis
Methods for oil recovery and lecithin production were
applied in triplicate. Results are expressed as mean
value ± standard deviation. The differences in mean
values between samples were assessed with Student’s
t-test, being statistically different at a significance level
of 5%.
RESULTS AND DISCUSSION
Composition of soybean degumming sludge
Wet gums showed high water content of almost
500 g kg−1 on average, indicating high sensitivity for
hydrolytic damage when are not stored and processed
under convenient conditions (Table 1). No changes
in general composition, acid, iodine, and peroxide
values were observed for rapeseed wet gum stored in
frozen state (−20 ◦ C) for 24 months.13 Gums used
in this study were frozen at −20 ◦ C and carefully
processed to minimize hydrolytic damage. As can
be observed in Table 1, the average phospholipid
content in wet gums was almost 300 g kg−1 and
the occluded oil was estimated as approximately
250 g kg−1 . Gums had a low content of hexaneinsoluble impurities (4 g kg−1 ) such as sand, meal,
and other insoluble materials. It can be observed that
the composition of wet gums is variable (Table 1).
The source of this variability may be genetic (plant
cultivar), seed quality (maturity, harvesting-induced
damage, and handling/storage conditions), and oilprocessing variables.14
Yields of recovered oil and obtained lecithin
The yields for recovered oil by methods II and III,
556 and 588 g kg−1 , respectively, were significantly
higher than the yield for the recovered oil by direct
extraction (Table 2). In both methods water was
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
Oil and lecithin from water degumming sludge
Table 1. Composition of soybean degumming sludges (wet gums)
Moisture (g kg−1 )
Acetone-insoluble material (g kg−1 )
Hexane-insoluble material (g kg−1 )
Occluded oil (g kg−1 )
462 ± 48
293 ± 62
4±1
241 ± 15
Phospholipids (g kg−1 )
PC
120 ± 31
% Relative
PE
PA
PI
PC
PE
PA
PI
68 ± 15
28 ± 0.5
43 ± 16
46 ± 2
28 ± 2
10 ± 1
16 ± 1
Table 2. Yields of recovered oil and obtained lecithin from wet gums
Recovered oil yield (g kg−1 )
Method
Wet gum
I
II
III
104 ± 18
135 ± 26
142 ± 14
Obtained lecithin yield (g kg−1 )
Occluded oil
429 ± 71 a
556 ± 96 b
588 ± 60 b
Wet gum
Dried gum
416 ± 82
437 ± 86
Total: 391 ± 87
Ph. A: 331 ± 76
Ph. B: 60 ± 12
767 ± 90a
807 ± 93a
Total: 720 ± 99a
Ph. A: 610 ± 89
Ph. B: 110 ± 13
Means within a column followed by the same letter are not significantly different (α = 0.05).
Table 3. Quality indexes for recovered oils from wet gums
Quality index
Acidity (g oleic acid kg−1 )
Peroxide value (mEq kg−1 )
p-Anisidine value
Unsaponifiable matter (g kg−1 )
Phosphorous content (mg kg−1 )
TOTOX valuea
Method I
Method II
Method III
16.9 ± 6.5a
1.62 ± 0.49a
1.86 ± 1.50a
9.0 ± 1.1a
87 ± 34a
5.09 ± 1.86a
21.7 ± 7.3a
1.02 ± 0.72a
4.03 ± 2.19a
12.1 ± 2.0b
330 ± 76b
5.77 ± 2.42a
16.7 ± 2.8a
3.24 ± 1.98a
2.34 ± 0.76a
11.5 ± 2.1ab
244 ± 73b
8.82 ± 3.91a
Total oxidation value = 2 × peroxide value + p-anisidine value.
Means within a row followed by the same letter are not significantly different (α = 0.05).
a
eliminated before extraction with acetone. Method
II was performed with total water elimination and in
Method III about 85% of the water was eliminated.
The direct extraction method showed lower oil
yields and, moreover, centrifugation was necessary
to separate oil and aqueous phase.
The lecithin yields (720–807 g kg−1 of dried gum)
were not significantly different in the three methods
(Table 2). Two fractions of lecithin can be obtained
by Method III, lecithin from Phase A being the most
abundant.
Quality indexes for recovered oils
Table 3 shows some quality indexes for recovered oils
using the three studied methods. The free fatty acids
in recovered oil ranged from 16.7 to 21.7 g kg−1 as
oleic acid and no significant differences were observed
between methods. The Codex Alimentarius establishes
a maximum level of 0.6 mg KOH g−1 (3 g kg−1 as
oleic acid) for refined oil.15 The National Oilseed
Processors Association (NOPA), in soybean oil trading
rules revised in 2007, fixes maximum levels of 7.5 g
kg−1 for crude degummed soybean oil, as oleic acid,
and applies allowances between 7.6 and 12.5 g kg−1 .16
However, crude soybean oil can have acid values of up
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
to 4.0 mg KOH g−1 or 20 g kg−1 as oleic acid,17 which
are easily reduced during refining process.
Peroxide values, as a measure of primary oxidation
products, were lower than 3.24 mEq kg−1 , and
anisidine values, which measure secondary oxidation
compounds, were lower than 4.03 (Table 3). No
significant differences were observed for these indexes
in recovered oils by the three methods. Total
oxidation values (TOTOX values), often used in
the industry, were lower than 8.82. These values
combine evidence about the past history of oil
(as reflected in the p-anisidine value) with its
present state (as evidenced in the peroxide value).
It is acceptable to maintain a peroxide value of
less than 4 and an anisidine value of less than
2 in the crude oil during storage.18 For crude
oils TOTOX values less than 10 correspond to
good-quality oils on an industrial scale. Oxidation
products are reduced during bleaching in the refining
process.
The Codex Alimentarius fixes a maximum value of
15 g kg−1 for crude soybean oils in unsaponifiable
matter.15 All recovered oils were adjusted to this norm
for characterization (Table 3).
2463
LN Ceci, DT Constenla, GH Crapiste
Phosphorous and phospholipids in recovered
oils
As shown in Table 3, the phosphorous content in
oils recovered by Method I averaged 87 mg kg−1 ,
in accordance with the values proposed for crude
degummed soybean oil (<200 mg kg−1 ).16 Shipments
up to 250 mg kg−1 are permitted with discounts.
Oils obtained by Methods II and III had higher
phosphorous contents than those obtained by direct
extraction (Table 3). Crude soybean oils can contain
more than 1000 mg phosphorous kg−1 , depending on
the extraction and preparation methods, and values
of 10–15 mg kg−1 are usual in refined soybean oils.19
Phosphorous contents of the order of 1–3 mg kg−1
can be observed in completely refined and packaged
soybean oils. The phosphorous content observed in
oils recovered by Methods II and III can be easily
reduced during the refining process by degumming.
To estimate the phospholipid contents in oils, a
factor of 30 is usually applied, to convert the
percentage of total phosphorous to the equivalent
content of phosphatides.20 However, in crude oils
this factor overestimates the phospholipids and lower
factors have been proposed.21 Crude oils contain
phosphorous from another sources, such as sand and
meal residues, including inorganic phosphorous that
is also determined by the spectrophotometric method.
The contents and relative composition of phospholipids in recovered oils determined by the chromatographic method are shown in Table 4. Although this
method has the advantage of identifying and quantifying phospholipids separately, other minor compounds
(e.g., lysophospholipids) are not determined. However, four major phospholipids in soybean oil can
be evaluated by chromatography and the remaining
fraction is relatively low. Oils recovered by Method
I had, on average, lower phosphorous and phospholipid contents than those obtained by Methods II and
III. These results suggest that during direct extraction
of oil water and hydratable phospholipids would be
pulled by the acetone and retained in aqueous phase
after centrifugation. Crude soybean oils have 15–30 g
phosphatides per kg−1 ; this content is reduced to 3–8 g
kg−1 in water-degummed soybean oil, and 0.03–0.45 g
kg−1 in refined oils.22,23 It is remarkable that all recovered oils had total phospholipid contents practically
in the range accepted for degummed products. A
large dispersion between samples was observed for
total phospholipid contents, demonstrating the strong
Table 4. Phospholipid composition for recovered oils (% relative)
Phospholipid
Method I
Method II
Method III
PE
PA
PI
PC
Total (g kg−1 )
17 ± 5a
7 ± 5ab
14 ± 4a
62 ± 11a
1.56 ± 0.64a
11 ± 6a
10 ± 4a
8 ± 2b
71 ± 10a
9.56 ± 3.43b
12 ± 5a
5 ± 3b
6 ± 6b
77 ± 12a
5.10 ± 4.16ab
Means within a row followed by the same letter are not significantly
different (α = 0.05).
2464
influence of wet gum composition on the phosphatide
contents in recovered oils.
As shown in Table 4, the recovered oils had a
high relative percentage of PC (62–77%), and low
percentages of PE and PI, when compared with
data provided in the literature for crude soybean
oils. Mounts et al.24 obtained the following relative
values, working on samples of crude soybean oils
extracted from standard and genetically modified
varieties: 20.7–35.8% for PE, 18.2–27.9% for PI,
2.1–35.0% for PA, and 25.4–49.7% for PC. An
additional reason for high relative PC content could be
that the reference phospholipid samples for the HPLCUV method have a different fatty acid pattern from the
soy phopholipids. PA is the most variable phospholipid
in crude soybean oil and its percentage in the recovered
oils was relatively low. A water degumming step during
the refining process could reduce the phospholipids
in the recovered oils without drawbacks, provided
that PC is the most easily hydratable compound,
whereas PA and PI are eliminated with difficulty by
hydratation. Only small significant differences in the
percentages of PA and PI were observed when the
methods for recovery of oil were compared (Table 4).
The oils recovered by direct extraction had more nonhydratable PI than those obtained by Methods II and
III. Evidently, hydratable phospholipids are carried
out to aqueous phase and non-hydratable PI is more
easily retained in recovered oils.
Polar compounds in recovered oils
No significant differences were observed in total and
individual polar compounds between the methods for
oil recovery from wet gums (Table 5). Polymerized,
dimerized, and oxidized triglycerides (PTG, DTG,
and OTG, respectively) are used as indicators of
thermal degradation (TD). Diglycerides (DG) and
free fatty acids (FFA) indicate hydrolytic degradation
(HD). All recovered oils showed a level of HD higher
than their TD (TD/HD < 1). These results suggest
that the storage of wet gums before processing, under
controlled conditions to reduce hydrolytic damage, is
crucial to obtain good-quality oils.
The content of total polar compounds in the
recovered oils from soybean wet gums ranged from
48.8 to 56.1 g kg−1 (Table 5). These contents are only
slightly higher than those observed in crude sunflower
oils, obtained by pressing (43.4 and 40.5 g kg−1 ) and
solvent extraction (45.3 and 38.9 g kg−1 ).25 These
results confirm the qualification of recovered oils as
good-quality products. Moreover, treatment by water
degumming could reduce polar compounds in the
recovered oils.25
No appreciable amounts of DTG and PTG were
detected in recovered oils – an expected result since
polymerization due to thermal degradation occurs at
temperatures higher than those used in the methods for
oil recovery. The most significant change is observed
in the concentration of OTG relating to oxidative
deterioration.
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
Oil and lecithin from water degumming sludge
Table 5. Polar compounds (g kg−1 ) for recovered oils from wet gums
Compound
PTG
DTG
OTG
DG
FFA
Total
TD
HD
TD/HD
Method I
Method II
Method III
0.6 ± 0.1a
0.0 ± 0.0a
22.5 ± 3.6a
8.4 ± 3.8a
24.6 ± 9.3a
56.1 ± 13.7a
23.1 ± 3.7a
33.0 ± 13.0a
0.70 ± 0.24a
0.5 ± 0.3a
0.0 ± 0.0a
21.2 ± 5.9a
6.8 ± 1.5a
26.6 ± 7.9a
55.1 ± 10.7a
21.7 ± 6.1a
33.4 ± 9.4a
0.65 ± 0.24a
0.6 ± 0.1a
0.0 ± 0.0a
17.2 ± 2.1a
6.1 ± 1.3a
24.9 ± 9.3a
48.8 ± 10.7a
17.8 ± 2.4a
31.0 ± 10.4a
0.57 ± 0.16a
PTG, polymerized triglycerides; DTG, dimerized triglycerides; OTG,
oxidized triglycerides; DG, diglycerides; FFA, free fatty acids; TD, thermal degradation = PTG + DTG + OTG; HD, hydrolytic degradation =
DG + FFA.
Means within a row followed by the same letter are not significantly
different (α = 0.05).
Phospholipids in obtained lecithins
The content and relative composition of phospholipids
in the lecithins obtained by the three methods
are shown in Table 6. No significant differences
were detected in total phospholipid content between
Methods I, II and III (hexane phase), with values
ranging from 610 to 691 g kg−1 . These contents
include the four phospholipids more relevant in
soybean: PC, PE, PI and PA. More abundant
phospholipids were PC and PE, in the range
203–319 g kg−1 and 185–218 g kg−1 , respectively
(Table 6). The following values have been obtained in
the literature for soybean lecithin with an intermediate
range of composition: PC = 290–390 g kg−1 and
PE = 200–263 g kg−1 . However, soybean lecithin with
a low range of composition contains 120–210 g kg−1
of PC and 80–95 g kg−1 of PE.14
Lecithin obtained by Method I had a higher
PC content than those obtained by extraction after
water elimination. PC is extracted more efficiently
by acetone when no water is disposable. Lecithin
Table 6. Phospholipid composition for obtained lecithins
Phospholipid
Method
I
Method
II
Method
Method
III Phase A III Phase B
PE
(g kg−1 )
185 ± 28a 214 ± 21a 218 ± 30a
(% Relative) 27 ± 3a
31 ± 3b
36 ± 2c
55 ± 10b
33 ± 1b
PA
(g kg−1 )
(% Relative)
74 ± 14a
12 ± 1ab
22 ± 5c
13 ± 1b
151 ± 5b
22 ± 1b
107 ± 11a
18 ± 1c
33 ± 6c
20 ± 2bc
PC
319 ± 15a 203 ± 4b
(g kg−1 )
(% Relative) 46 ± 2a
29 ± 1b
211 ± 20b
35 ± 1c
58 ± 12c
35 ± 2c
PI
74 ± 5a
11 ± 1a
113 ± 7a
(g kg−1 )
(% Relative) 16 ± 1a
94 ± 11b
13 ± 2b
Total (g kg−1 ) 691 ± 53a 662 ± 31a 610 ± 70a 168 ± 32b
Means within a row followed by the same letter are not significantly
different (α = 0.05).
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
obtained by extraction of oil after drying under vacuum
(Method II) had slightly higher PA and PI contents
than those obtained by Methods I and III Phase
A. Lecithin obtained by Method III from hexane
phase had the highest content of PE. The results
show that lecithin obtained by Methods I, II and III
hexane phase have the same purity level but different
relative compositions and could be used in different
applications.
From the alcoholic phase lecithin was obtained with
low total phospholipid content (168 g kg−1 , Table 6).
Moreover, the relative composition of phospholipids
in this lecithin is not significantly different from
that obtained from hexane phase. More studies are
required to analyse the composition of the colourless
product with a crystalline appearance obtained from
the alcoholic phase. Other components such as
glycolipids and complex carbohydrates could be
profitably recovered.
ACKNOWLEDGEMENTS
Thanks are given for financial support to the
´ Argentina de Grasas y Aceites (ASAGA),
Asociacion
the Concejo Nacional de Investigaciones Cient´ıficas y
T´ecnicas (CONICET) and the Universidad Nacional
del Sur (UNS). Thanks are also due to Oleaginosa
Moreno Hermanos (OMHSA) for providing samples.
REFERENCES
1 List GR, Avellaneda JM and Mounts TL, Efectos de las
´ y la calidad de la
condiciones de desgomado en la extraccion
lecitina de soja. Aceites y Grasas 43:207–218 (2001).
2 List GR, Commercial manufacture of lecithin, in Lecithins:
Sources, Manufacture and Uses, ed. by Szuhaj BF. AOCS,
Champaign, IL, pp. 145–162 (1989).
3 Van Nieuwenhuyzen W, Fractionation of lecithins. Eur Food
Drink Rev, Process Technology 27–32 (1999).
4 Wu Y and Wang T, Soybean lecithin fractionation and
functionality. J Am Oil Chem Soc 80:319–326 (2003).
5 Wu Y and Wang T, Fractionation of crude soybean lecithin with
aqueous ethanol. J Am Oil Chem Soc 81:697–704 (2004).
6 Schneider M, Fractionation and purification of lecithin, in
Lecithins: Sources, Manufacture and Uuses, ed. by Szuhaj BF.
AOCS, Champaign, IL, pp. 109–130 (1989).
7 Ghyczy M, Synthesis and modification of phospholipids, in
Lecithins: Sources, Manufacture and Uses, ed. by Szuhaj BF.
AOCS, Champaign, IL, pp. 131–144 (1989).
8 Dashiell GL, Fuentes, m´etodos de proceso y usos comerciales
de la lecitina. Aceites y Grasas 43:197–204 (2001).
9 American Oil Chemists’ Society, Official Methods and Recommended Practices, (5th edn), ed. by Firestone D. AOCS,
Champaign, IL (1998).
10 International Union of Pure and Applied Chemistry (IUPAC),
Standard Methods for the Analysis of Oils, Fats and Derivatives
(7th edn),ed. by Paquot C and Hautfenne A. Blackwell
Scientific, Oxford (1992).
11 Carelli AA, Brevedan MIV and Crapiste GH, Quantitative
determination of phospholipids in sunflower oil. J Am Oil
Chem Soc 74:511–514 (1997).
12 Dobarganes MC, Velasco J and Dieffenbacher A, Determination of polar compounds, polymerized and oxidized triacylglycerols, and diacylglycerols in oils and fats. Pure Appl Chem
72:1563–1575 (2000).
2465
LN Ceci, DT Constenla, GH Crapiste
13 Sosada M, Studies on stability of rapeseed wet gum as a source
of pharmaceutical lecithin. J Am Oil Chem Soc 73:367–370
(1996).
14 Cherry JP and Kramer WH, Plant sources of lecithin, in
Lecithins: Sources, Manufacture and Uses, ed. by Szuhaj BF.
AOCS, Champaign, IL, pp. 16–31 (1989).
15 Codex Alimentarius, Codex standard for named vegetable oils,
CODEX-STAN 210 (2005).
16 National Oilseed Processors Association, Trading rules for the
purchase and sale of soybean oil. NOPA, Washington DC
(2007).
17 USDA Foreign Agricultural Service, GB1535-2003 Soybean
Oil Standard – G/TBT/N/CHN/25. Gain Report number
CH4022 (2004).
18 Gupta MK, Fundamental quality control in vegetable oil
refining, Oil Mill Gaz 108:6–9 (2003).
19 Lence SH and Agarwal S, Assessing the feasibility of processing
and marketing niche soy oil. Research Paper 03-MRP
6, Midwest Agribusiness Trade Research and Information
Center (MATRIC) (2003).
2466
20 Chapman GW, A conversion factor to determine phospholipid
content in soybean and sunflower crude oils. J Am Oil Chem
Soc 57:299–302 (1980).
21 Carelli AA, Ceci LN and Crapiste GH, Phosphorous-tophospholipid conversion factors for crude and degummed
sunflower oils. J Am Oil Chem Soc 79:1177–1180 (2002).
22 Buchold H, Enzymax: the enzyme-catalyzed degumming process of vegetable oils, 48. DGF (German Finance Association), Jahrestagung, Essen (1992).
23 Debruyne I, Soybean oil processing: quality criteria and flavour
reversion. Oil Mill Gaz 110:10–11 (2004).
24 Mounts TL, Abidi SL and Rennick KA, Effect of genetic
modification on the content and composition of bioactive
constituents in soybean oil. J Am Oil Chem Soc 73:581–586
(1996).
25 Brevedan MIV, Carelli AA and Crapiste GH, Changes in
composition and quality of sunflower oils during extraction
and degumming, Grasas y Aceites 51:417–423 (2000).
J Sci Food Agric 88:2460–2466 (2008)
DOI: 10.1002/jsfa
