J Am Oil Chem Soc (2012) 89:1145–1153

DOI 10.1007/s11746-011-1993-7

ORIGINAL PAPER

Aqueous Extraction of Oil and Protein from Soybeans

with Subcritical Water
S. C. Ndlela • J. M. L. N. de Moura •

N. K. Olson • L. A. Johnson

Received: 29 July 2011 / Revised: 18 November 2011 / Accepted: 12 December 2011 / Published online: 18 January 2012
Ó AOCS 2012

Abstract Aqueous extraction using subcritical water is an Keywords Subcritical water
Aqueous extraction

environmentally friendly alternative to extracting oil and Oil extraction
Protein extraction
Flaking
Soybeans

protein from oilseeds with flammable organic solvents. The Extraction
effects of solids-to-liquid ratio (1:3.3–1:11.7), temperature
(66–234 °C), and extraction time (13–47 min) were eval-
uated on the extraction of oil and protein from soybean Introduction
flakes and from extruded soybeans flakes with subcritical
water. A central composite design (23) with three center Increasing worldwide soybean production is being driven
points and six axial points was used. Subcritical water by growing demand for high quality protein to feed live-
extractions were carried out in a 1-L high-pressure batch stock (primarily swine and poultry) and for vegetable oils
reactor with constant stirring (300 rpm) at 0.03–3.86 MPa. to supply food and fuel sectors [1]. Countercurrent hexane
In general, oil extraction was greater for extruded soybean extraction has long been used to extract most soybean oil
flakes than with soybean flakes. More complete oil extrac- [2]; however, increasing environmental regulations and
tion for extruded soybean flakes was achieved at around safety concerns regarding hexane use in oilseed-crushing
150 °C and extraction was not affected by solids-to-liquid units [3] are driving extensive research toward environ-
ratios over the range tested, while oil extraction from soy- mentally friendly extraction technologies [4].
bean flakes was more complete at 66 °C and low solids-to- Among the emerging technologies to extract oil and
liquid ratio (1:11.7). Protein extraction yields from flakes protein from oilseeds, enzyme-assisted aqueous extraction
were generally greater than from extruded flakes. Protein processing (EAEP) has been considered to be an effective
extraction yields from extruded flakes increased as tem- and environmentally friendly process in which oil and pro-
perature increased and solids-to-liquid ratio decreased, tein are simultaneously extracted from soybeans [4, 5]. This
while greater protein extraction yields from soybean flakes water and enzyme-based technology along with mechanical
were achieved when using low temperatures and low solids- treatments, such as flaking and extruding (expanding), has
to-liquid ratio. achieved similar levels of oil extraction as conventional
hexane extraction ([97%) [6]. Despite achieving high oil
extraction yields (removal from solids), however, overall free
S. C. Ndlela
N. K. Olson oil recovery in EAEP of soybeans ranges from 79 to 83% due
Iowa Energy Center-BECON, Nevada, IA 50201, USA to unrecovered emulsified oil in the skim (primarily sugar-
and protein-rich) and small residual amounts of unextracted
J. M. L. N. de Moura
L. A. Johnson (&)
Center for Crops Utilization Research, Iowa State University, oil in the insolubles (fiber-rich fraction) [7, 8]. The mild
1041 Food Sciences Building, Ames, IA 50011-1061, USA operating conditions used in EAEP of soybeans enables
e-mail: ljohnson@iastate.edu production of oil with good quality and protein with similar
nutritional compositions as proteins produced by conven-
J. M. L. N. de Moura
L. A. Johnson
Department of Food Science and Human Nutrition, tional extraction procedures such as soy protein concentrate
Iowa State University, Ames, IA 50011-1061, USA (SPC) or isolate (SPI) [9, 10].

123
1146 J Am Oil Chem Soc (2012) 89:1145–1153

Over the past decade, there has been increasing interest in Materials and Methods
using subcritical water for extraction where hot water
(100 °C \ T \ 374 °C) under sufficient pressure (1–8 MPa) Soybeans
to maintain water in the condensed phase has been used to
extract protein, essential oils, and bioactive components from Full-fat soybean flakes were prepared from variety 92M91-
a wide variety of matrices [11–17]. At subcritical conditions, N201 soybeans (Pioneer, a DuPont Business, Johnston, IA,
the density, dielectric constant, dissociation constant, viscos- USA) harvested in 2008.
ity, diffusivity, electrical conductance, and solvency change
[18]. At subcritical conditions, water polarity decreases Processing Methods
thereby favoring extraction of organic bioactive components.
The water ionization constant (Kw) increases with increasing Soybean Flaking
temperature, making subcritical water a suitable media for
catalyzing hydrolytic reactions [12, 14, 16]. Changes in the The soybeans were cracked into 4–6 pieces by using a
dielectric constant of water, the outstanding feature of sub- corrugated roller mill (model 10X12SGL, Ferrell-Ross,
and supercritical water extraction, depends primarily on Oklahoma City, OK, USA) and the hulls were removed
temperature (changes from 90 to 20 in dielectric constant can from the meats (cotyledons) by aspirating with a multi-
be achieved by increasing the temperature from ambient to aspirator (Kice, Wichita, KS, USA). The meats were con-
300 °C) and much less on pressure. Increasing pressure causes ditioned at 60 °C to make them plastic for flaking by using
small increases in the dielectric constant, which could impair a triple-deck seed conditioner (French Oil Mill Machinery
the extraction of low-polar components. The most significant Co., Piqua, OH, USA) and were flaked to approximately
change in extraction conditions due to pressure is related to 0.25 mm thickness by using a smooth-surface roller mill
water density, determining whether the extraction is per- (Roskamp Mfg, Inc., Waterloo, IA, USA). The soybean
formed with liquid or with steam. Although pressure has little flakes contained 20.3% oil (wet-basis), 35.9% protein (wet-
effect on extraction efficiency, lower recovery of extracted basis), and 7.6% moisture.
components has been observed when using steam instead of
liquid water [17]. In addition to extracting organic compo- Extruding Full-Fat Soybean Flakes
nents, subcritical water has been used to hydrolyze triacyl-
glycerols into free fatty acids [19, 20] and to convert organic The moisture content of the flakes was increased to 15% by
waste (cellulose) into valuable products (glucose) [18]. spraying water onto the flakes while mixing in a Gilson
The use of subcritical water to simultaneously extract oil mixer (model 59016A, St. Joseph, MO, USA). The
and protein from oilseeds has not been extensively evalu- moistened full-fat soybean flakes were extruded/expanded
ated. Most research has focused on extraction of oil and/or by using a twin-screw extruder (ZSE 27 mm diameter;
protein from rice bran and to a lesser extent on extracting American Leistritz Extruders, Somerville, NJ, USA). High-
protein from defatted soybean meal and full-fat soy flour shear geometry screws were used in co-rotational orienta-
[12–16]. Protein extraction yields from deoiled rice bran tion at 90-rpm screw speed. The extruder barrel (1,080 mm
were greater when using subcritical water conditions length) was composed of ten heating blocks that were set
compared with the alkali extraction [12]. The effects of for the temperature profile 30-70-100-100-100-100-100-
temperature (200–220 °C), reaction time (10–30 min), and 100-100-100 °C. The extruder was manually fed to achieve
solids-to-liquid ratio (1:5–2:5) were evaluated when using a 10.5-kg/h output rate of extruded flakes. The collets were
subcritical water to extract protein from defatted soybean cooled to room temperature, placed in polyethylene bags,
meal and full-fat soy flour [14]. About 50% protein and stored in a cold room at 4 °C until extracted. The
recovery was achieved relative to starting material when extruded soybean flakes contained 22.7% oil (wet-basis),
extractions were performed at 200–210 °C and 1:5 solids- 88.7% solids (wet-basis), 35.3% protein (wet-basis), and
to-liquid ratio for 30 min. Although full-fat soy flour has 11.3% moisture.
been used, no oil extraction data were reported.
Oil and protein extraction yields from soybeans when Subcritical Water Extraction
using aqueous extraction systems are greatly affected by
the extent of cell wall disruption, the solids-to-liquid ratio, Oil and protein extractions from flakes and extruded soy-
the presence or lack of enzyme during extraction, and the bean flakes were carried out in a high-pressure, 316
extraction time and temperature [5, 21–24]. The present stainless-steel, batch reactor as shown in Fig. 1 (High-
study was undertaken to gain a better understanding of how Pressure Equipment Co., Erie, PA, USA). The reactor,
these parameters affect oil and protein extractions from pressure rated for 241 MPa, consisted of 1 L internal vol-
soybeans under subcritical extraction conditions. ume (7.6 cm internal diameter and 23.1 cm length) heated

123
J Am Oil Chem Soc (2012) 89:1145–1153 1147

Fig. 1 Schematic for the pilot- MagneDrive agitator assembly

plant subcritical extraction unit
(TC1 internal reactor
thermocouple, TC2 external Vent line/Purge line Pa P
reactor thermocouple, Pa
analog pressure digital gauge
(0–241 MPa), P digital pressure
gauge (0–45 MPa), F 250
micron filter; high pressure
valve (414 MPa)

Vent Line

Reactor

Helium Tank Chamber Tc2

Electrically heated jacket

Rupture disk
103 MPa
Tc1

Table 1 Variables and levels evaluated in the experimental design to model oil and protein extraction from extruded soybean flakes and soybean
flakes
Treatments Coded levels Uncoded levels
Solids-to-liquid Temperature Time (min) Solids-to-liquid Temperature Time (min)
ratio (X1) (°C) (X2) (X3) ratio (X1) (°C) (X2) (X3)

1 1 1 1 1:10 200 40
2 -1 1 1 1:5 200 40
3 1 -1 1 1:10 100 40
4 1 1 -1 1:10 200 20
5 -1 -1 1 1:5 100 40
6 -1 1 -1 1:5 200 20
7 1 -1 -1 1:10 100 20
8 -1 -1 -1 1:5 100 20
9 0 0 0 1:7.5 150 30
10 0 0 0 1:7.5 150 30
11 0 0 0 1:7.5 150 30
12 1.68 0 0 1:11.7 150 30
13 -1.68 0 0 1:3.3 150 30
14 0 1.68 0 1:7.5 234 30
15 0 -1.68 0 1:7.5 66 30
16 0 0 1.68 1:7.5 150 47
17 0 0 -1.68 1:7.5 150 13
3
Complete 2 factorial design parameters, with three independent variables in two levels, three repetitions in the central point and six repetitions in
the axial points

by an electrical jacket, two J thermocouples (TC1 and Soybean flakes and extruded soybeans flakes were dis-
TC2), an analog pressure gauge (Pa), digital pressure gauge persed into deionized water to achieve solids-to-liquid
(Pb), and a MagneDrive stirring assembly (Autoclave ratios ranging from 1:3.3 to 1:11.7 (Table 1). The amount
Engineers, Supercritical Fluid Technologies Inc., Newark, of solids per 100 g of slurry for solids-to-liquid ratios of
DE, USA). 1:3.3, 1:5, 1:7.5, 1:10, and 1:11.7 were 22.6, 16.3, 12.3,

123
1148 J Am Oil Chem Soc (2012) 89:1145–1153

8.9, and 7.8 g, respectively. The amount of slurry loaded Soybeans

into the reactor was determined based on the minimum
amount needed to achieve adequate stirring (400 g) and the Cracking
maximum amount that would fit into the reactor without
plugging the vent lines located just above the slurry level. Aspirating Hulls
Slurries of approximately 675 g for soybean flakes and 775
g for extruded soybean flakes were loaded into the reactor Conditioning (60°C)
at room temperature (extruded flakes are denser than
flakes). A leak test was performed with helium at 0.69
Flaking
MPa. In order to degas the mixture prior to starting the
experiments, the reactor was purged with helium at 0.69
Moistening (15%)
MPa and maintained for 6 min at 0.69 MPa after which it
was vented. The degassing cycle was repeated five times
while stirring the slurry at 300 rpm. Extraction experiments Extruding
were performed at temperatures ranging from 66 to 234 °C
and extraction times ranging from 13.2 to 46.8 min
Sub-critical water
(Table 1). The slurries were stirred at 300 rpm and the extraction
resulting pressure varied from 0.14 to 3.96 MPa. The
reactor heating cycle ranged from 40 min (66 °C) to 240
min (234 °C) while the cooling time from the target tem- 2-Phase
Liquid phase
Centrifuging
perature to 70 °C ranged from 120 min (100 °C) to 480
min (234 °C). The system was cooled to 70 °C to ensure
Insoluble fraction Funnel
safe handling of the product before opening the reactor to Cream
Separation
remove the slurry.
Following extraction, the slurry was centrifuged at
Skim
3,000g (20 min at 25 °C) to separate insolubles from the
liquid phase (Fig. 2). Three phase layers were observed Fig. 2 Process flow diagram for subcritical extraction from flaked
after centrifuging: an insoluble fraction (fiber-rich and extruded soybeans
fraction), a skim fraction (protein- and sugar-rich frac-
tion), and a cream fraction (oil-rich emulsion). After
removing the insoluble fraction, the liquid phase was Experimental Design and Statistical Analysis
placed into a separatory funnel (2 L) and allowed to
settle overnight at 4 °C. During settling, the liquid phase In order to optimize for the best combination of solids-to-
separated into two fractions (skim fraction and cream ? liquid ratio, temperature and reaction time for oil and
free oil fraction). Since oil and protein extractabilities protein extraction, a complete 23 factorial design of the
were the focus of this study, only the insoluble fraction central rotational type was established, with three central
was analyzed to determine mass balances for oil and points and six axial points, based on Response Surface
protein. Methodology [22]. The effects of solids-to-liquid ratio
(1:3.3–1:11.7), temperature (66–234 °C), and extraction
Oil, Protein, and Solids Recoveries time (13.2–46.8 min) on the extraction of oil and protein
from soybean flakes and extruded soybeans flakes by using
Oil and protein analyses were carried out on the insoluble subcritical water treatment were evaluated. Pressures were
and starting materials (soybean flakes and extruded a consequence of temperature and slurry composition, not
soybean flakes). Total oil content was determined by being one of the independent variables in the experiment.
using the Mojonnier acid hydrolysis method (AOCS The independent variables (solids-to-liquid ratio, temper-
method 922.06) and protein content was determined by ature, and extraction time) were evaluated according to
using the Dumas combustion method and the N conver- coded levels (a, -1, 0, 1, ?a). The variable levels used in
sion factor of 6.25 (vario MAXCN Elementary Analyses the experimental design were selected in order to achieve
system Gmbh, Hanau, Germany). Extraction yields were slightly lower and/or higher values than the common range
expressed as percentages of each component in the of values reported in the literature. Central points are the
insoluble fraction relative to the initial amount in the average of levels -1 and ?1, while axial points were
starting material. All chemical analyses were performed determined by interpolation (a = ±1.68). Coded and
in duplicate. uncoded levels and their corresponding independent

123
J Am Oil Chem Soc (2012) 89:1145–1153 1149

variables are shown in Table 1. Dependent variables (i.e., when comminuted by extrusion, in relation to milling, flak-
evaluated responses) were oil and protein extraction yields ing, and flaking followed by milling. Our results are in
for soybean flakes and extruded soybean flakes. Data were agreement with those of Lamsal et al. [20] in which oil
analyzed by using Statistica version 8.0 software. The extractability improved from 46 to 71% when combining
significance of each model was tested by Analysis of flaking and extruding of soybeans compared to flaking alone.
Variance (ANOVA). Higher oil extraction yields from soybean flakes (50%)
and from extruded soybean flakes (84%) were achieved
when extracting at 1:10 solids-to-liquid ratio and 100 °C
Results and Discussion for 20 min and at 1:11.7 solids-to-liquid ratio and 150 °C
for 30 min, respectively. Lower oil extraction yields for
Oil and Protein Extraction Yields both soybean flakes (4%) and extruded soybean flakes
(38%) were observed at 1:7.5 solids-to-liquid ratio and
Oil and protein extractabilities from soybeans are signifi- 234 °C for 30 min. Absolute pressures and water densities
cantly affected by mechanical treatments such as grinding, are presented in Table 3 as functions of extraction tem-
flaking, extruding, and combinations of these treatments perature. Water density was slightly reduced when moving
[21–24]. The combination of flaking and extrusion is more from 66 to 150 °C extraction temperature (from 980 to 917
effective in extracting oil when using aqueous extraction of kg/m3), being more pronounced from 150 to 234 °C (from
soybeans due to more complete cell disruption that facili- 917 to 823 kg/m3). In all cases, however, water was present
tates water penetration into the matrix, releasing soluble in the liquid phase. Although no water was observed to
and insoluble components into the external environment change from liquid to vapor, which would affect the
[21–24]. In Table 2, the effects of extent of cell wall dis- recovery of extracted components [17], the increase in
ruption on oil and protein extraction from soybean flakes pressure from 0.14 to 3.96 MPa when moving from 66 to
and extruded soybean flakes under different subcritical 234 °C extraction temperature could have reduced
water extraction conditions are presented. Regardless of extractability of oil from both extruded soybean flakes and
treatment applied, oil extraction yields were significantly soybean flakes at 234 °C. A slight increase in the dielectric
improved when using extruded soybean flakes (38–84%) constant due to increasing pressure could affect extraction
compared with soybean flakes (4–50%). of low polar components [17].
Campbell and Glatz [24] reported that nearly complete Temperature can affect extraction due to enhanced solute
cellular disruption of soybeans cotyledons was achieved solubility and diffusion into the solvent bulk [25]. In some

Table 2 Experimental design for optimizing oil and protein extraction from soybean flakes and extruded soybean flakes
Treatments Solids-to-liquid Temperature Time (min) Oil extraction Oil extraction Protein extraction Protein extraction
ratio (X1) (°C) (X2) (X3) (%) EF* (%) F** (%) EF* (%) F**

1 1:10 200 40 74.7 13.4 65.7 64.8

2 1:5 200 40 72.4 37.9 53.6 56.1
3 1:10 100 40 69.5 42.7 37.6 71.9
4 1:10 200 20 79.6 22.0 61.7 65.3
5 1:5 100 40 53.4 16.3 34.1 33.1
6 1:5 200 20 80.3 31.3 56.6 57.1
7 1:10 100 20 61.3 50.2 36.4 65.0
8 1:5 100 20 56.8 21.2 29.7 30.8
9 1:7.5 150 30 64.6 7.1 49.2 45.5
10 1:7.5 150 30 68.1 12.3 49.6 47.7
11 1:7.5 150 30 83.1 7.08 51.7 46.3
12 1:11.7 150 30 83.9 17.9 59.9 55.5
13 1:3.3 150 30 62.5 45.1 40.9 55.0
14 1:7.5 234 30 38.2 4.2 73.1 67.8
15 1:7.5 66 30 49.0 44.0 26.6 53.2
16 1:7.5 150 47 64.2 39.9 52.5 52.8
17 1:7.5 150 13 66.0 19.8 49.2 46.4
EF* Extruded soybean flakes, F** soybean flakes

123
1150 J Am Oil Chem Soc (2012) 89:1145–1153

Table 3 Effects of reaction temperature on pressure and water suggested some oil binding by unextracted protein and/or
density reduced oil extractability due to pressure increase at 234 °C
Reaction Absolute pressure Density Water (Table 3). The use of subcritical water to extract protein
temperature (°C) (MPa) (Kg/m3)a phasea from raw and deoiled soybean meal was previously reported
by Watchararuji et al. [14]. In that study, 50% protein
66 0.14 980.02 Liquid
recovery was achieved when extracting at 210 °C, 1:5 sol-
100 0.22 958.4 Liquid
ids-to-liquid ratio, and 30 min residence time although oil
150 0.51 917.03 Liquid
extractability was not reported.
200 2.08 865.06 Liquid
234 3.96 822.65 Liquid
a
Statistical Analysis
Webbook NIST—http://webbook.nist.gov/chemistry/fluid

Estimated regression models and coefficients of determi-

cases, however, temperature has been associated with nation for oil and protein extractions from soybean flakes
reduced oil and protein extraction during aqueous extraction and extruded soybean flakes are shown in Table 4. Only
of soybeans [14]. Reduced extractability is likely a conse- parameters significant at P \ 0.05 were used in the
quence of protein thermal denaturation [26], which affects regression models. Coefficients of determination (R2) for
both oil and protein extraction to different extents. Heating models of oil extraction from soybean flakes and extruded
favors protein-lipid interactions in which denatured protein soybean flakes were 0.27 and 0.38, respectively; while
is likely to sequester oil by exposing hydrophobic amino coefficients for models of protein extraction from soybean
acids [22]. Since soybean flakes and extruded soybean flakes flakes and extruded soybean flakes were 0.63 and 0.96,
contain proteins with different solubilities [27] due to pro- respectively. While satisfactory coefficients of determina-
tein denaturation during extrusion, one would expect lower tion were achieved for the models predicting protein
protein extractability when using extruded soybean flakes extraction, low coefficients for oil extraction suggested that
compared with soybean flakes. Protein extraction yields other sources of variation were not accounted for.
from soybean flakes and from extruded soybean flakes The effects of soybean moisture content before flaking
ranged from 31 to 72 and 27 to 73%, respectively. In gen- and storage time of the beans and flakes before extruding
eral, extruded soybeans flakes had similar or slightly higher have been associated with phospholipase D activity [7, 28–
protein extractability when using water temperatures 30], which could increase emulsion stability and affect the
[150 °C. However, an opposite effect was observed when separation of extracted material before and after centrifu-
using temperatures \100 °C, where protein denaturation gation. Our material was prepared and used over a 2- to
reduced protein solubility adversely affecting extraction 3-month period, which might have allowed phospholipase
yields. These results suggested that using temperatures D activity, a factor not accounted for by the models.
[150 °C likely compensated for the reduced protein solu- Another source of variation that may have impacted the
bility due to extrusion. High protein extraction yields of 68 oil extraction models was the heating/cooling system of the
and 73% were achieved when extracting soybean flakes and reactor. Typically, fast heating and fast cooling are pre-
extruded soybean flakes at 234 °C, respectively, which was ferred for these types of experiments under subcritical
the highest temperature evaluated and produced the lowest conditions to avoid thermal degradation of the reactants
oil extraction yields for both soybean flakes and extruded and for better evaluation of reaction time necessary to
soybean flakes. Although parameters that favor protein achieve the desired extraction yields. In order to follow our
extractability also usually favor oil extraction, an opposite experimental design parameters (reaction time and tem-
trend was observed at the highest temperature, which perature), heating the reactor required 0.67 h (25-66 °C)

Table 4 Estimated regression models and coefficients of determination (R2) for oil and protein extraction yields from soybean flakes and
extruded soybean flakes
Estimated regression modelsa R2

% Estimated oil extraction from extruded soybean flakes = 73.18-8.54 X22 0.38
% Estimated oil extraction from soybean flakes = 25.43-11.15 X1 9 X2 0.27
% Estimated protein extraction from extruded soybean flakes = 48.70 ? 4.33 X1 ? 13.02 X2 0.96
% Estimated protein extraction from soybean flakes = 53.78 ? 6.65 X1 ? 4.91 X2 - 7.01 X1 9 X2 0.63
a
Only parameters with a confidence level above 95% (P \ 0.05) were considered significant
X1 Coded level corresponding to solids-to-liquid ratio; X2 coded level corresponding to temperature

123
J Am Oil Chem Soc (2012) 89:1145–1153 1151

Table 5 Analysis of variances of estimated regression models

Source of Sum of Degrees of Mean F test
variation squares freedom squares

Oil extraction from extruded soybean flakes

Regression 948 1 948 9.30a
Residual 1,531 15 102
Lack of fit 1,337 13 103 1.06b
Pure error 193 2 97
Total 2,479
Oil extraction from soybean flakes
Regression 995 1 995 5.55a
Residual 2,690 15 179
Lack of fit 2,672 13 206 22.71b
Pure error 18 2 9
Total 3,685
Protein extraction from extruded soybean flakes
Regression 2,573 2 1,287 162a
Residual 111 14 7.90
Lack of fit 107 12 8.90 4.73b
Pure error 3.77 2 1.88
Total 2,684
Protein extraction from soybean flakes
Regression 1,326 3 442 7.44a
Residual 773 13 59
Lack of fit 770 11 70 56.46b
Pure error 2.48 2 1.24
Total 2,099
Values in bold are statistically meaningful (P \ 0.05). Coefficient of
determination: R2 = 0.38; F0.95–1.15 = 4.54; F0.95–13.2 = 19.42
Values in italic are statistically meaningful (P \ 0.05). Coefficient of Fig. 3 Effects of temperature and solids-to-liquid ratio on oil extrac-
determination: R2 = 0.27; F0.95–1.15 = 4.54; F0.95–13.2 = 19.42 tion from extruded soybean flakes (a) and from soybean flakes (b)
Values in bold italic are statistically meaningful (P \ 0.05). Coeffi-
cient of determination: R2 = 0.96; F0.95–2.14 = 3.74; F0.95–12.2 = 19.41
Values in underline are statistically meaningful (P \ 0.05). Coeffi-
cient of determination: R2 = 0.63; F0.95–3.13 = 3.26; F0.95–11.2 = Table 5 shows the ANOVAs of the models. For all cases,
19.405 regression was significant (Fcalculated/Ftable). Significance of
a
F:ratio (regression/residual) the results was determined by comparing the Fcalculated
b
F:ratio (lack of fit/pure error) (verifying the equality of the means) with the theoretical
value (Ftable). Except for the models for oil and protein
to 4 h (25-234 °C) followed by cooling 2 h (100-70 °C) extraction from soybean flakes, the F tests for the lack of fit
to 8 h (234-70 °C), which could have confounded mea- were not statistically significant (Fcalculated/Ftable) indicating
suring effects of reaction time on extraction yields. In some that the models do not show lack of fit and thus can be used
cases, the long cooling times surpassed the time to conduct for prediction in the range of the parameters evaluated. The
the extraction. Reaction time was not statistically signifi- F test for lack of fit was statistically significant for oil and
cant, likely a consequence of the long residence time due to protein extraction from soybean flakes (Fcalculated [ Ftable),
the heating/cooling system that was used. These challenges which was associated with the pure error (extremely low)
with heating and cooling could be eliminated when tran- due to the low degrees of freedom (2). Based on the esti-
sitioning from the current batch system to a continuous mated regression models, response surfaces were built to
reactor. Although the real effect of residence time was not express oil and protein extractions from soybean flakes and
determined, the remaining parameters (temperature and extruded soybean flakes (Figs. 3, 4).
solids-to-liquid ratio) provided useful information on their According to the estimated regression model and
effects on extraction yields. Fig. 3a regardless of the solids-to-liquid ratio used, oil

123
1152 J Am Oil Chem Soc (2012) 89:1145–1153

low temperature (66 °C) and low solids-to-liquid ratio

(1:11.7). The possibility of achieving high extractability of
protein from soybean flakes at low temperature indicates
the higher solubility of protein in soybean flakes compared
with protein in extruded soybean flakes.

Conclusions

The temperature and the solids-to-liquid ratio significantly

affected oil and protein extraction yields with subcritical
water from soybean flakes and extruded soybean flakes.
The economical viability of aqueous extraction of soybeans
depends on both oil and protein extractabilities, therefore,
conditions that favor both oil and protein extractions must
be considered. Although oil extraction yields from extru-
ded soybean flakes was not significantly affected by the
solids-to-liquid ratio, conditions which favor both oil and
protein extraction yields are: temperature around 150 °C
and low solids-to-liquid ratio (1:11.7). For soybean flakes,
low temperature (66 °C) and low solids-to-liquid ratio
(1:11.7) favored both oil and protein extraction yields.

Acknowledgments This work was supported by funds provided by

the U.S. Department of Agriculture, Cooperative Research, Educa-
tion, and Extension Service, Grant #2009-34432-20057; and the U.S.
Department of Energy, Grant DE-FG3606GO86014.

References
Fig. 4 Effects of temperature and solids-to-liquid ratio on protein
extraction from extruded soybean flakes (a) and from soybean
1. USDA-FAS (2007) Oilseeds: World Markets and Trade. Circular
flakes (b)
Series FOP 8-07, FOP 10-07 http://www.fas.usda.gov/oilseeds/
circular/2007/October/oilseedsfull1007.pdf, http://www.fas.usda.
gov/oilseeds/circular/2007/August/oilseedsfinal0807.pdf
extraction from extruded soybean flakes decreases when 2. Johnson LA (2008) Oil recovery from soybeans. In: Johnson LA,
White PJ, Galloway R (eds) Soybeans: chemistry, production
temperature gets further away from the center point
processing and utilization. AOCS Press, Urbana, pp 331–375
(150 °C) with lower extraction values at the axial points 3. Galvin JB (1997) Toxicity data for commercial hexane and
(±1.68). According to the estimated regression model and hexane isomers. In: Wan PJ, Wakelyn PJ (eds) Technology and
Fig. 3b, highest oil extraction yields from soybean flakes solvents for extracting oilseeds and nonpetroleum oils. AOCS
Press, Champaign, pp 75–85
are achieved at high temperature values (?1.68, 234 °C)
4. de Moura JMLN, Campbell K, Mahfuz A, Jung S, Glatz CE,
and high solids to liquid ratio (-1.68, 1:3.3) and at low Johnson LA (2008) Enzyme-assisted aqueous extraction of oil
temperature values (-1.68, 66 °C) and low solids-to-liquid and protein from soybeans and cream de-emulsification. J Am Oil
ratio (?1.68, 1:11.7). Based on the estimated regression Chem Soc 85:985–995
5. de Moura JMLN, Johnson LA (2009) Two-stage countercurrent
model and Fig. 4a, protein extraction from extruded soy-
enzyme-assisted aqueous extraction processing of oil and protein
bean flakes is favored by increased temperature and low from soybeans. J Am Oil Chem Soc 86:283–289
solids-to-liquid ratio, with higher extraction values at 6. de Moura JMLN, Almeida NM, de Johnson LA (2009) Scale-up
?1.68 for both variables (234 °C and 1:11.7 solids-to- of enzyme-assisted aqueous extraction processing of soybeans.
J Am Oil Chem Soc 86:809–815
liquid ratio). Although protein extraction yields from
7. de Moura JMLN, Almeida NM, de Jung S, Johnson LA (2010)
extruded flakes increased as temperature increased, the Flaking as a pretreatment for enzyme-assisted aqueous extraction
nutritional quality of the extracted protein might be processing of soybeans. J Am Oil Chem Soc 87:1507–1515
adversely affected by thermal degradation. An opposite 8. de Moura JMLN, Maurer D, Jung S, Johnson LA. Proof-of-concept
of countercurrent two-stage enzyme-assisted aqueous extraction
trend was observed for protein extraction from soybean
processing of soybeans. J Am Oil Chem Soc 88:1649–1658
flakes. According to the estimated regression model and 9. Maurer D, Mahfuz A, Johnson LA, Jung S (2008) Enzymatic
Fig. 4b, highest protein extraction can also be achieved at destabilization of natural occurring soy emulsion and oil quality.

123
J Am Oil Chem Soc (2012) 89:1145–1153 1153

99th AOCS Annual Meeting & Expo. Seattle, Washington, USA, 21. Rosenthal A, Pyle DL, Niranjan K (1996) Aqueous and enzy-
p 124 matic process for edible oil extraction. Enz Microb Tech
10. de Almeida, NM, de Moura JMLN, and Johnson LA (2010) 19:402–420
Functional properties of protein produced by two-stage aqueous 22. Lamsal BP, Murphy PA, Johnson LA (2006) Flaking and extru-
countercurrent enzyme-assisted aqueous extraction. Abstracts sion as mechanical treatments for enzyme-assisted aqueous
pp 130. 101st Annual Meeting and Exposition of the American extraction of oil from soybeans. J Am Oil Chem Soc 83:973–979
Oil Chemists Society, May 16–19, Phoenix, AZ 23. Rodrigues MI, Iemma AF (2005) Planejamento de experimentos
11. Teo CC, Tan SN, Yong JWH, Hew CS, Ong ES (2010) Pres- e otimizacao de processos: uma estrategia sequencial de plane-
surized hot water extraction (PHWE). J Chromatography A jamentos. Casa do Pao, Campinas, Brazil, pp 95–123
1217:2484–2494 24. Campbell KA, Glatz CE (2009) Mechanisms of aqueous extrac-
12. Sereewatthanawut I, Prapintip S, Watchiraruji K, Goto M, Sasaki tion of soybean oil. J Ag Food Chem 57:10904–10912
M, Shotipruk A (2008) Extraction of protein and amino acids 25. Takeuchi TM, Pereira CG, Braga MEM, Marostica MR, Leal PF,
from deoiled rice bran by subcritical water hydrolysis. Bioresour Meireles MAA (2009) Low-pressure solvent extraction (solid-
Tech 99:555–561 liquid extraction, microwave assisted, and ultrasound assisted)
13. Pourali O, Asghri FS, Yoshida H (2009) Simultaneous rice bran from condimentary plants. In: Meireles MAA (ed) Extracting
oil stabilization and extraction using sub-critical water medium. bioactive compounds for food products: theory and applications.
J Food Eng 95:510–516 CRC Press, Boca Raton
14. Watchararuji K, Goto M, Sasaki M, Shotipruck A (2008) Value- 26. Kinsella JE (1979) Functional properties of soy proteins. J Am
added subcritical water hydrolysate from rice bran and soybean Oil Chem Soc 56:242–258
meal. Bioresource Tech 99:6207–6213 27. de Almeida NM, JMLN Moura, L Johnson (2010) Functional
15. Pourali O, Asghari FS, Yoshida H (2009) Sub-critical water properties of protein produced by two-stage aqueous counter-
treatment of rice bran to produce valuable materials. Food Chem current enzyme-assisted aqueous extraction. Abstracts pp 130.
115:1–7 101st Annual Meeting and Exposition of the American Oil
16. Basile A, Jimenez-Carmona MM, Clifford AA (1998) Extraction of Chemists Society, May 16-19, Phoenix, AZ
rosemary by superheated water. J Agric Food Chem 46:5205–5209 28. Yao L, Jung S (2010) 31P NMR phospholipids profiling of soy-
17. Hawthorne SB, Yang Y, Miller DJ (1994) Extraction of organic bean emulsion recovered from aqueous extraction. J Agric Food
pollutants from environmental solids with sub- and supercritical Chem 58:4866–4872
water. Analytical Chem 66:2912–2920 29. List RG, Mounts TL, Lanser AC, Holloway RK (1990) Effect of
18. Goto M, Obuchi R, Hirose T, Sakaki T, Shibata M (2004) moisture, microwave heating, and live steam treatment on
Hydrothermal conversion of municipal organic waste into phospholipase D activity in soybeans and soy flakes. J Am Oil
resources. Bioresource Tech 93:279–284 Chem Soc 67:867–871
19. Holliday RL, King JW, List GR (1997) Hydrolysis of vegetable 30. Jung S, Maurer D, Johnson LA (2009) Factors affecting emulsion
oils in sub- and supercritical water. Ind Eng Chem Res 36:932–935 stability and quality of oil recovered from enzyme-assisted
20. King JW, Holliday RL, List GR (1997) Hydrolysis of soybean oil aqueous extraction of soybeans. Bioresource Tech 100:5340–
in subcritical water flow reactor. Green Chem 1(6):261–264 5534

123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.