Article pubs.acs.org/JAFC
Soy Sauce Residue Oil Extracted by a Novel Continuous Phase Transition Extraction under Low Temperature and Its Refining Process Lichao Zhao,† Yong Zhang,† Liping He,‡ Weijie Dai,† Yingyi Lai,† Xueyi Yao,† and Yong Cao*,† †
College of Food Science and ‡Instrumental Analysis & Reasearch Center, South China Agricultural University, NO. 483 Wushan Road, Guangzhou 510642, People’s Republic of China ABSTRACT: On the basis of previous single-factor experiments, extraction parameters of soy sauce residue (SSR) oil extracted using a self-developed continuous phase transition extraction method at low temperature was optimized using the response surface methodology. The established optimal conditions for maximum oil yield were n-butane solvent, 0.5 MPa extraction pressure, 45 °C temperature, 62 min extraction time, and 45 mesh raw material granularity. Under these conditions, the actual yield was 28.43% ± 0.17%, which is relatively close to the predicted yield. Meanwhile, isoflavone was extracted from defatted SSR using the same method, but the parameters and solvent used were altered. The new solvent was 95% (v/v) ethanol, and extraction was performed under 1.0 MPa at 60 °C for 90 min. The extracted isoflavones, with 0.18% ± 0.012% yield, mainly comprised daidzein and genistein, two kinds of aglycones. The novel continuous phase transition extraction under low temperature could provide favorable conditions for the extraction of nonpolar or strongly polar substances. The oil physicochemical properties and fatty acids compositions were analyzed. Results showed that the main drawback of the crude oil was the excess of acid value (AV, 63.9 ± 0.1 mg KOH/g) and peroxide value (POV, 9.05 ± 0.3 mmol/kg), compared with that of normal soybean oil. However, through molecular distillation, AV and POV dropped to 1.78 ± 0.12 mg KOH/g and 5.9 ± 0.08 mmol/kg, respectively. This refined oil may be used as feedstuff oil. KEYWORDS: soy sauce residue, oil, isoflavone, continuous phase transition extraction under low temperature, refinement
■
INTRODUCTION Soy sauce, a famous seasoning in Asia, is essential for improving the food taste. Soy sauce is mainly produced from soybean, wheat, and salt using a traditional fermentation technique. Soy sauce residues (SSRs) are the main byproducts of soy sauce fermentation.1 According to the capacity and growth rate of the soy sauce market in China,2 SSR yield was estimated to almost reach 1.63 million tons (dry basic) in 2013. These residues are mainly disposed using low-cost methods, such as for feeds and fertilizers. These residues are even buried directly, which results in pollution because of their high salt content.3 During soy sauce fermentation, the protein and starch of raw materials are actually only partially depleted. A large amount of valuable components, such as oil, fibers, and isoflavones, remained.4 Given their high moisture, SSRs are easily deteriorated and affected by mildew. Effective reuse of these residues will result in considerable economic value. The chemical structures of soybeans were altered after soy sauce fermentation. The oil residue is tightly combined with fiber.2 This process results in difficult oil removal, thereby restricting SSR utilization. Only a few studies have analyzed SSR utilization as substrates for the production of polysaccharides, lactic acid, xylitol, and ethanol.5−9 Mechanical press extraction, organic solvent extraction, and supercritical fluid extraction (SFE) are conventional processes. However, these processes have drawbacks. The yield in the mechanical press extraction is lower than the desired value. The organic solvent extraction also leaves traces of solvent, which may cause a safety problem.10 SFE is very costly and requires high pressure limiting its industrial application.11 © 2014 American Chemical Society
Thus, a novel, continuous phase transition extraction under low temperature is explored. At different pressures (from 0.5 to 1.2 MPa) and temperatures (from 40 to 60 °C), the phase of the extractant is varied between gas and liquid, coupled with strong penetrability. The gaseous extractant is first compressed into liquid using a high-pressure pump and runs through the extracter heated by a water jacket. In the separation pot, the extractant is converted into gas through vacuum evaporation and separated from the extract. Afterward, the extractant is condensed, and the extraction continues under pressure by the pump, which circulates repeatedly until the extraction is completed. Finally, the extractant is recycled in the storage pot with low loss. Compared with subcritical fluid extraction, the most important advantage of this new technique is its continuity. The cyclic and fresh solvent can produce the oil solution and ensure sufficient and efficient extraction. The entire process is safe, economical, and efficient. It can also protect the heat-sensitive components and does not form a hazardous substance. After oil extraction, isoflavone is extracted using defatted SSR, and by the same method, the solvent was replaced by ethanol. The novel continuous phase transition extraction under low temperature can both provide favorable conditions for the extraction of nonpolar to strongly polar substances. The proposed process is very much suitable for industrial application. Received: Revised: Accepted: Published: 3230
December March 18, March 20, March 20,
11, 2013 2014 2014 2014
dx.doi.org/10.1021/jf405459v | J. Agric. Food Chem. 2014, 62, 3230−3235
Journal of Agricultural and Food Chemistry
Article
All experiments were performed randomly to minimize the effect of extraneous factors. A quadratic polynomial regression was assumed in the prediction of oil yield (Y1) as a function of the independent variables, as shown in eq 1:
To remedy the issue of environmental hazards, waste should be recycled, and new technology should be promoted. The oil and isoflavones extraction from SSR using the proposed technology was investigated. The physicochemical properties and fatty acid composition of crude oil, as well as the yield of extracted isoflavones and their components, were analyzed. Results showed that the crude oil is refined through molecular distillation and may be used as feedstuff oil. The isoflavones remaining in SSR mainly exists in the form of daidzein and genistein. This study may be referential to the comprehensive utilization of SSR.
■
Y1 = K 0 + K1X1 + K 2X 2 + K3X3 + K11X12 + K 22X 2 2 + K33X32 + K12X1X 2 + K13X1X3 + K 23X 2X3
where K0 is a constant. Ki, Kii, and Kij are linear, quadratic, and interactive regression coefficients, respectively. The statistical significance of the model and the goodness of fit were evaluated using analysis of variance (ANOVA). The RSM was used to investigate the effect of the process variables on the response and to determine the optimal conditions for maximum oil yield. The experimental data were statistically analyzed using Design-Expert version 8.0.0 software (StatEase Inc., Minneapolis, MN, USA). Oil Extraction. Prior to extraction, the SSRs were dried at 60 °C to ensure that moisture content was less than 5%, and the samples were grounded into powder. Extractions were conducted by continuous nbutane in a 3 L extraction pot. The solvent and extracted oil were separated through low-temperature vacuum evaporation. The oil yield was calculated using the following equation:
MATERIALS AND METHODS
Materials. SSRs were donated by the PRB Biotech Co., Ltd. (Zhongshan, Guangdong, China), and the main components are shown in Table 1. Samples were stored at 4 °C inside a polythene
Table 1. Main Components of Fresh and Defatted SSR (Dry Basic, %) content (%) main component
fresh SSR
crude fat crude protein crude fiber ash
29.2 21.8 18.1 29.5
± ± ± ±
defatted SSR
0.68 0.57 0.82 1.02
0.62 29.3 24.8 43.3
± ± ± ±
oil yield (%) =
0.18 0.84 0.32 0.69
Table 2. Experimental Design and Data for the Yield of SSR Oil from the CCDa
run no. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
coded value 0 −1 0 −1 1 0 0 1 0 0 −1 1 0 0 1 −1 0
−1 0 1 0 0 1 0 −1 −1 0 −1 1 0 0 0 1 0
1 1 −1 −1 1 1 0 0 −1 0 0 0 0 0 −1 0 0
response
X1
X2
X3
45 40 45 40 50 45 45 50 45 45 40 50 45 45 50 40 45
45 60 75 60 60 75 60 45 45 60 45 75 60 60 60 75 60
80 80 40 40 80 80 60 60 40 60 60 60 60 60 40 60 60
Y1 21.60 18.38 28.03 24.78 20.76 23.96 27.80 22.91 27.92 27.30 21.08 23.54 27.60 26.90 23.20 22.50 27.90
± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ± ±
the weight of extracted oil (g) × 100 the weight of raw materials (g)
(2)
Physicochemical Analyses of Crude Oil. Assays were performed for the determination of acid value (AV), peroxide value (POV), iodine value, saponification value, and refractive index, in accordance with GB 1535-2003 (an officially recognized Chinese criterion for physicochemical property detection and quality standard of soybean oil).13 Fatty Acid Analysis. Fatty acid composition was determined by conversion of the oil to fatty acid methyl esters (FAMEs) and analyzed by gas chromatography−mass spectrometry (GC-MS) (Agilent Co., Santa Clara, CA, USA) equipped with a flame ionization detector (FID). FAMEs were prepared by adding 50 mg of extracted oil, which was dissolved with 4 mL of isooctane; afterward, 200 μL of sodium methoxide was mixed with the solution.14 The mixtures were allowed to settle for 5 min, and the supernatant (0.2 μL) was injected into the gas chromatograph−mass spectrometer. The components were identified by comparing their retention time with those of the standards in the mass spectra library. In addition, their relative content was calculated from the GC peak area. The GC-MS conditions were as follows: a polar capillary column HP-INNOWAX (30 m × 0.25 mm, 0.25 μm film thickness) was used at a column head pressure of 0.04 MPa. The oven temperature was from 120 to 220 °C at heating rate of 8 °C/min, which was held for 1 min, continuously increased to 250 °C at 4 °C/min, and held for 5 min. The detector and injector temperatures were 220 and 230 °C, respectively.15 Helium was used as the carrier gas at approximately 40 mL/min. Deacidification by Molecular Distillation. Crude oil deacidification was performed using a laboratory-scale molecular distillation unit (model DCH70, Meichengaoxin Separation Technology Co., Ltd., Guangzhou, Guangdong, China). The distillation unit consisted of an evaporator with an internal condenser. For feeding, a jacketed vessel equipped with a flow regulation valve and level indication was used. The vacuum was achieved using a mechanical pump and a diffusion pump. The samples were first fully melted by circulating hot water (45 °C). Then, the melted samples were placed into the circulating hot oil-heated evaporator (220 °C), with a feed flow rate from 1.5 to 2.5 mL/min under 0.05 mbar pressure.16 Thin-film samples were achieved using a roller wiper system (475 rpm), and the condensation temperature was 25 °C. Free fatty acids were collected as the distillate, with the refined oil as the residue. Isoflavone Extraction. The SSRs were defatted prior to isoflavone extraction. Extractions were conducted by continuous 95% (v/v) ethanol in a 3 L extraction pot under 1.0 MPa at 60 °C for 90 min. The solvent and extracted isoflavones were separated through low-
drum until extraction. n-Butane (99.99%) was purchased from Shenzheng Shenyan Gas Co., Ltd. (Shenzhen, Guangdong, China). All other reagents were purchased from Sigma Chemical Co. (St. Louis, MO, USA) and of analytical grade. Experimental Design. On the basis of previous single-factor experiments, a central composite design (CCD) was applied to evaluate the effects of extraction temperature, time, and raw material granularity (RMG) on oil yield. The coded and uncoded independent variables used in the response surface methodology (RSM) design, as well as their respective levels, are listed in Table 2.12
decoded value
(1)
0.07 0.12 0.03 0.17 0.09 0.14 0.16 0.32 0.06 0.23 0.19 0.08 0.12 0.18 0.04 0.31 0.03
a
X1 = temperature (°C), X2 = time (min), X3 = RMG (mesh), Y1 = oil yield (%). 3231
dx.doi.org/10.1021/jf405459v | J. Agric. Food Chem. 2014, 62, 3230−3235
Journal of Agricultural and Food Chemistry
Article
temperature vacuum evaporation. The isoflavone yield was calculated using the following equation:
Analysis of Response Surfaces. The relationship between independent variables and responses in a three-dimensional response surface curve generated by the models is shown in Figure 1.
isoflavone yield (%) =
the weight of extracted isoflavone (g) × 100 the weight of defatted SSR (g)
(3)
Isoflavone High-Performance Liquid Chromatography (HPLC) Analysis. HPLC analysis was performed on a LC-10A system (Shimadzu, Kyoto, Japan) equipped with an UV detector (200−600 nm) and a Diamonsil C18 column (5 μm, 250 mm × 4.6 mm) at room temperature. Extracted isoflavones and standard substances were dissolved in ethanol. The injection volume was 20 μL, and the absorption was measured at 260 nm. The mobile phase was deionized water (A) and acetonitrile (B). Gradient elution was operated at a flow rate of 1.0 mL/min according to the following gradient procedure: 0.0−10.0 min, 10.0−45.0% B; 10.0−25.0 min, 45.0% B; 25.0−45.0 min, 85.0% B; 45.0−60.0 min, 10.0% B.17 The components were identified by comparing their retention times with those of the standards, and their contents were calculated from the peak area. Statistical Analysis. All assays were performed in triplicate, and the results were expressed as mean values ± standard deviations. ANOVA was performed, and the differences were considered statistically significant at P < 0.05.
■
RESULTS AND DISCUSSION Model Fitting. SSR oil yield obtained for each experiment is listed in Table 2. The response and variables were fitted to each other by multiple regressions. A good fit was obtained, and the secondorder polynomial equation is shown below by eq 4: Y1 = 27.5 + 0.46*X1 + 0.44*X 2 − 2.53*X3 − 0.20*X1*X 2 + 0.99*X1*X3 + 0.31*X 2*X3 − 4.17*X12 − 0.82*X 2 2 − 1.55*X32
(4)
Results from ANOVA showed that the polynomial model had a satisfactory P value (P < 0.0001), which indicated significant results at 1% (Table 3). The P value of lack-of-fit was 0.172 (P > 0.05), which means that each lack-of-fit was not significantly related to pure error. The large value of the multiple determination coefficient (R2 = 0.9886) revealed that the model adequately represented the experimental results. Table 3. ANOVA for Response Surface Model Fittinga source
ss
df
ms
f value
P value
model X1 X2 X3 X1X2 X1X3 X2X3 X12 X22 X32 residual lack-of-fit pure error the total correction
150.8 1.68 1.55 51.16 0.16 3.92 0.39 73.32 2.85 10.12 2.05 1.39 0.66 152.85
9 1 1 1 1 1 1 1 1 1 7 3 4 16
16.76 1.68 1.55 51.16 0.16 3.92 0.39 73.32 2.85 10.12 0.29 0.46 0.17
57.20 5.75 5.29 174.64 0.53 13.38 1.33 249.95 9.72 34.53
