Low-molecular weight organic composition of acid water from coconut oil soapstock

Soapstock from alkaline refining of coconut oil was acidified, and the resulting acid water after neutralization was subjected to gas chromatography, electron-ionization and chemical-ionization mass spectroscopy, and high-performance liquid chromatography. The chief low-molecular weight organic components were C₄−C₁₈ fatty acids, hydroxylated acids, and sugar alcohols. The prevalence of acids and total absence of phosphate compounds make coconut acid water different in composition from the acid waters from other soapstocks.     

Production of FAME from acid oil,a by-product of vegetable oil refining

Simple alkyl FA esters have numerous uses, including serving as biodiesel, a fuel for compression ignition (diesel) engines. The use of acid-catalyzed esterification for the synthesis of FAME from acid oil, a by-product of edible vegetable oil refining that is produced from soapstock, was investigated. Soybean acid oil contained 59.3 wt% FFA, 28.0 wt% TAG, 4.4 wt% DAG, and less than 1% MAG. Maximum esterification occurred at 65°C and 26 h reaction at a molar ratio of total FA/methanol/sulfuric acid of 1∶15∶1.5. Residual unreacted species under these conditions, as a fraction of their content in unesterified acid oil, were FFA, 6.6%; TAG, 5.8%; and DAG, 2.6%. This corresponds to estimated concentrations of FFA, 3.2%; TAG, 1.3%; and DAG, 0.2%, on a mass basis, in the ester product. In an alternative approach, the acylglycerol species in soapstock were saponified prior to acidulation. High-acid (HA) acid oil made from this saponified soapstock had an FFA content of 96.2 wt% and no detectable TAG, DAG, or MAG. Optimal esterification conditions for HA acid oil at 65°C were a mole ratio of FFA/methanol/acid of 1∶1.8∶0.17, and 14 h incubation. FAME recovery under these conditions was 89% of theoretical, and the residual unesterified FFA content was approximately 20 mg/g. This was reduced to 3.5 mg/g, below the maximum FFA level allowed for biodiesel, by washing with NaCl, NaHCO₃, and Ca(OH)₂ solutions. Alternatively, by subjecting the unwashed ester layer to a second esterification, the FFA level was reduced to less than 2 mg/g. The acid value of this material exceeded the maximum allowed for biodiesel, but was reduced to an acceptable value by a brief wash with 0.5 N NaOH.     

Fatty acid and triacylglycerol compositions of seed oils of five Amaranthus accessions and their comparison to other oils

This paper reports the fatty acid and triacylglycerol (TAG) compositions of five Amaranthus accessions (RRC1011, R149, A.K343, A.K432, and A. K433) representing two species and a cross between one of these and a third species. Seed oils of these were analyzed by gas chromatography and reversed-phase high-performance liquid chromatography, and their compositional properties compared with buck-wheat (Fagopyrum esculentum), corn (Zea mays), rice bran (Oryza sativa), soybean (Glycine max L. Merr.), sesame (Sesamum indicum), quinoa (Chenopodium quinoa), and cottonseed (Gossypium hirsutum) oils. All Amaranthus accessions were relatively high in palmitic (21.4–23.8%) and low in oleic (22.8–31.5%) and linolenic (0.65–0.93%) acids when compared to most of the grain and seed oils. The fatty acid composition of Amaranthus accessions K343, K433, and K432 (group I) were different from R149 and RRC1011 (group II) in mono and polyunsaturated fatty acids, but the saturate/unsaturate (S/U) ratios were very similar. All Amaranthus accessions were similar in TAG type, but showed slight differences in percentage. High similarities in UUU, UUS, and USS composition were observed among Amaranthus K343, K433 and K432, and between R149 and RRC1011. The fatty acid compositions of Amaranthus oil (group I) and cottonseed oil were similar, but their TAG compositions were different. The grain and oilseed oils were different from each other and from the Amaranthus accessions oils in terms of fatty acid composition, S/U, and TAG ratios. The UUU, UUS, and USS percentages were very diverse in grain and seed oils. The percentages of squalene in the TAG sample from the Amaranthus accessions were 8.05% in K343, 11.10% in K433, 11.19% in K432, 9.96% in R149, and 9.16% in RRC1011. Squalene was also tentatively identified in quinoa and ricebran oils at levels of 3.39 and 3.10%, respectively.     

Rheology of vegetable oil analogs and triglycerides

The rheological properties of two complex mixtures of short-chain triglycerides were experimentally determined. Dynamic or absolute viscosities of the mixtures were measured for shear rates of 0.32 to 64.69 s⁻¹ at temperatures between 25 and 80°C. The compositions of the mixtures were based on the oil of the plant species Cuphea viscosissima VS-320, a natural source of short-chain triglycerides. The dynamic viscosities of these mixtures were compared to those of a traditional vegetable oil (peanut oil) and diesel fuel. The results of this comparison were used to make estimates of the performance of such triglyceride mixtures as diesel fuel substitutes, since viscosity can be a key indicator of fuel performance for possible substitute diesel fuels. The crystallization temperatures of these two mixtures were also determined experimentally, and the effects of crystallization on fuel performance were projected. Additionally, the dynamic viscosities of pure triglycerides from C6∶0 to C18∶0 at 75°C were plotted vs. chain length. These viscosities were measured at high shear rates (>6 s⁻¹) where dynamic viscosity is shear-independent. An obvious trend in the relationship between triglyceride chain length and viscosity was observed. A second-order regression was used to obtain an equation for this relationship. This equation was used as a model for composition dependence of viscosity. This model was applied to the viscosities of the triglyceride mixtures examined here. There was good agreement between the model and the actual, measured viscosity values determined in this study.     

Conversion of vegetable oil to biodiesel using immobilized Candida antarctica lipase

Biodiesel derived from vegetable oils has drawn considerable attention with increasing environmental consciousness. We attempted continuous methanolysis of vegetable oil by an enzymatic process. Immobilized Candida antarctica lipase was found to be the most effective for the methanolysis among lipases tested. The enzyme was inactivated by shaking in a mixture containing more than 1.5 molar equivalents of methanol against the oil. To fully convert the oil to its corresponding methyl esters, at least 3 molar equivalents of methanol are needed. Thus, the reaction was conducted by adding methanol stepwise to avoid lipase inactivation. The first step of the reaction was conducted at 30°C for 10 h in a mixture of oil/methanol (1:1, mol/mol) and 4% immobilized lipase with shaking at 130 oscillations/min. After more than 95% methanol was consumed in ester formation, a second molar equivalent of methanol was added and the reaction continued for 14 h. The third molar equivalent of methanol was finally added and the reaction continued for 24 h (total reaction time, 48 h). This three-step process converted 98.4% of the oil to its corresponding methyl esters. To investigate the stability of the lipase, the three-step methanolysis process was repeated by transferring the immobilized lipase to a fresh substrate mixture. As a result, more than 95% of the ester conversion was maintained even after 50 cycles of the reaction (100 d).     

Fast formation of high-purity methyl esters from vegetable oils

Experiments have confirmed that the base-catalyzed methanolysis of vegetable oils occurs much slower than butanolysis because of the two liquid phases initially present in the former reaction. For the same reason, second-order kinetics are not followed. The use of a cosolvent such as tetrahydrofuran or methyl tertiary butyl ether speeds up methanolysis considerably. However, like one-phase butanolysis, one-phase methanolysis initially exhibits a rapid formation of ester, but then slows drastically. Experiments show that the half-life of the hydroxide catalyst is too long to explain the sudden slowing of the reaction. Similarly, lower rate constants for the methylation of the mono- and diglycerides are not a reasonable explanation. Instead the cause has been identified as the fall in polarity which results from the mixing of the nonpolar oil with the methanol. This lowers the effectiveness of both hydroxide and alkoxide catalysts. Increasing the methanol/oil molar ratio to 27 in the one-phase system raises the polarity such that the methyl ester content of the ester product exceeds 99.4 wt% in 7 min. This has obvious implications for the size of new methyl ester plants as well as the capacity of existing facilities.     

Fast formation of high-purity methyl esters from vegetable oils

Experiments have confirmed that the base-catalyzed methanolysis of vegetable oils occurs much slower than butanolysis because of the two liquid phases initially present in the former reaction. For the same reason, second-order kinetics are not followed. The use of a cosolvent such as tetrahydrofuran or methyl tertiary butyl ether speeds up methanolysis considerably. However, like one-phase butanolysis, one-phase methanolysis initially exhibits a rapid formation of ester, but then slows drastically. Experiments show that the half-life of the hydroxide catalyst is too long to explain the sudden slowing of the reaction. Similarly, lower rate constants for the methylation of the mono- and diglycerides are not a reasonable explanation. Instead the cause has been identified as the fall in polarity which results from the mixing of the nonpolar oil with the methanol. This lowers the effectiveness of both hydroxide and alkoxide catalysts. Increasing the methanol/oil molar ratio to 27 in the one-phase system raises the polarity such that the methyl ester content of the ester product exceeds 99.4 wt% in 7 min. This has obvious implications for the size of new methyl ester plants as well as the capacity of existing facilities.     

Continuous production of biodiesel fuel from vegetable oil using immobilized Candida antarctica lipase

Candida antarctica lipase is inactivated in a mixture of vegetable oil and more than 1∶2 molar equivalent of methanol against the total fatty acids. We have revealed that the inactivation was eliminated by three successive additions of 1∶3 molar equivalent of methanol and have developed a three-step methanolysis by which over 95% of the oil triacylglycerols (TAG) were converted to their corresponding methyl esters (ME). In this study, the lipase was not inactivated even though 2∶3 molar equivalent of methanol was present in a mixture of acylglycerols (AG) and 33% ME (AG/ME33). This finding led to a two-step methanolysis of the oil TAG: The first-step was conducted at 30°C for 12 h with shaking in a mixture of the oil, 1∶3 molar equivalent of methanol, and 4% immobilized lipase; the second-step reaction was done for 24 h after adding 2∶3 molar equivalent of methanol (36 h in total). The two-step methanolysis achieved more than 95% of conversion. When two-step reaction was repeated by transferring the immobilized lipase to a fresh substrate mixture, the enzyme could be used 70 cycles (105 d) without any decrease in the conversion. From the viewpoint of the industrial production of biodiesel fuel production, the two-step reaction was conducted using a reactor with impeller. However, the enzyme carrier was easily destroyed, and the lipase could be used only several times. Thus, we attempted flow reaction using a column packed with immobilized Candida lipase. Because the lipase packed in the column was drastically inactivated by feeding a mixture of AG/ME33 and 2∶3 molar equivalent of methanol, three-step flow reaction was performed using three columns packed with 3.0 g immobilized lipase. A mixture of vegetable oil and 1∶3 molar equivalent of methanol was fed into the first column at a constant flow rate of 6.0 mL/h. The eluate and 1∶3 molar equivalent of methanol were mixed and then fed into the second column at the same flow rate. The final step reaction was done by feeding a mixture of eluate from the second column and 1∶3 molar equivalent of methanol at the same flow rate. The ME content in the final-step eluate reached 93%, and the lipase could be used for 100 d without any decrease in the conversion.     

Physical and chemical properties of trans-free fats produced by chemical interesterification of vegetable oil blends

Fat blends, formulated by mixing a highly saturated fat (palm stearin or fully hydrogenated soybean oil) with a native vegetable oil (soybean oil) in different ratios from 10:90 to 75:25 (wt%), were subjected to chemical interesterification reactions on laboratory scale (0.2% sodium methoxide catalyst, time=90 min, temperature=90°C). Starting and interesterified blends were investigated for triglyceride composition, solid fat content, free fatty acid content, and trans fatty acid (TFA) levels. Obtained values were compared to those of low- and high-trans commercial food fats. The interesterified blends with 30–50% of hard stock had plasticity curves in the range of commercial shortenings and stick-type margarines, while interesterified blends with 20% hard stock were suitable for use in soft tubtype margarines. Confectionery fat basestocks could be prepared from interesterified fat blends with 40% palm stearin or 25% fully hydrogenated soybean oil. TFA levels of interesterified blends were low (0.1%) compared to 1.3–12.1% in commercial food fats. Presented at the 88th AOCS Annual Meeting and Expo, May 11–14, 1997, Seattle, Washington.     

Isolation of erucic acid from rapeseed oil by lipase-catalyzed hydrolysis

Three lipases were compared for their ability to hydrolyze high erucic acid rapeseed oil, with the objective of concentrating the erucic acid in a single glyceride fraction. Lipase fromPseudomonas cepacia released all fatty acids rapidly and did not result in selective distribution of erucic acid.Geotrichum candidum lipase released C20 and C22 fatty acids extremely slowly, resulting in their accumulation in the di- and triglyceride fractions. Less than 2% of the total erucic acid was found in the free fatty acid (FFA) fraction. Lipase fromCandida rugosa released erucic acid more slowly than C20 and C18 fatty acids at 35°C but only resulted in a limited accumulation of the erucic acid in the di- and triglyceride fractions. However, when hydrolysis catalyzed byC. rugosa lipase was carried out below 20°C, the reaction mixture solidified and was composed solely of FFAs and diglycerides. The diglyceride fraction contained approximately 95% erucic acid while about 20% of the total erucic acid was found in the FFA fraction. It is concluded that hydrolysis at low temperature withC. rugosa lipase results in a higher purity of erucic acid in the glyceride fraction than can be obtained withG. candidum lipase, but with considerable loss of erucic acid to the FFA fraction.     

Enzymatic synthesis of trierucin from high-erucic acid rapeseed oil

The lipase fromCandida rugosa has been shown to discriminate against erucic acid. Advantage of this property has been taken to produce trierucin from high-erucic acid rapeseed (HEAR) oil. A method has been developed for extracting erucic acid from the oil as dierucin and subsequently enzymatically converting it to trierucin. Unrefined HEAR oil was hydrolyzed with lipase fromC. rugosa to produce a mixture of free fatty acids and dierucin. Precipitation and filtration from cold ethanol gave 73% pure dierucin, free of fatty acids. This dierucin was treated in two ways to produce trierucin. First, in the presence of an immobilized lipase and a known amount of water, some trierucin is produced by interesterification. Second, a more efficient route to trierucin utilizedRhizopus arrhizus lipase to completely hydrolyze dierucin to erucic acid, which was then combined with an appropriate amount of dierucin in the presence of an immobilized lipase to produce trierucin in a quantitative yield. Partly presented at the AOCS Annual Meeting held in Toronto, May 10–14, 1992.     

Determination of seed oil content and fatty acid composition in sunflower through the analysis of intact seeds, husked seeds, meal and oil by near-infrared reflectance spectroscopy

A methodological study was conducted to test the potential of near-infrared reflectance spectroscopy (NIRS) to estimate the oil content and fatty acid composition of sunflower seeds. A set of 387 intact-seed samples, each from a single plant, were scanned by NIRS, and 120 of them were selected and further scanned as husked seed, meal, and oil. All samples were analyzed for oil content (nuclear magnetic resonance) and fatty acid composition (gas chromatography), and calibration equations for oil content and individual fatty acids (C_16:0, C_16:1, C_18:0, C_18:1, and C_18:2) were developed for intact seed, husked seed, meal, and oil. For intact seed, the performance of the calibration equations was evaluated through both cross- and external validation, while cross-validation was used in the rest. The results showed that NIRS is a reliable and accurate technique to estimate these traits in sunflower oil (validation r ² ranged from 0.97 to 0.99), meal (r ² from 0.92 to 0.98), and husked seeds (r ² from 0.90 to 0.97). According to these results, there is no need to grind the seeds to scan the meal; similarly accurate results are obtained by analyzing husked seeds. The analysis of intact seeds was less accurate (r ² from 0.76 to 0.85), although it is reliable enough to use for pre-screening purposes to identify variants with significantly different fatty acid compositions from standard phenotypes. Screening of intact sunflower seeds by NIRS represents a rapid, simple, and cost-effective alternative that may be of great utility for users who need to analyze a large number of samples.     

Comparison between two procedures for stereospecific analysis of triacylglycerols from vegetable oils—I: Olive oil

Two methods for stereospecific analysis of triacylglycerols are compared. Procedure A, based on stereospecific phosphorylation ofsn-1,2-diacylglycerols to phosphatidic acids, and procedure B, based on separation of the diastereomeric 1,2(2,3)-diacylglycerol urethane derivatives by high-performance liquid chromatography on silica, were applied to olive oil triacyl-sn-glycerols. Statistical evaluation of the results showed good reproducibility, and Student'st-test indicates no statistical differences between the two considered procedures, although some small differences were observed and discussed. Fifteen samples of extra-virgin olive oil, produced in the same region (Umbria, Italy), were analyzed with the two considered procedures.     

Fatty acid selectivity of lipases: γ-linolenic acid from borage oil

The γ-linolenic acid (Z,Z,Z-6,9,12-octadecatrienoic acid, GLA) present in borage oil free fatty acids was concentrated in esterification reactions that were catalyzed by several preparations of the acyl-specific lipase ofGeotrichum candidum. In this manner, a 95% recovery of the GLA originally present in borage oil (25% GLA) was obtained as a highly enriched fatty acid fraction with a GLA content of >70%. Other fatty acids concentrated in this fraction were the monounsaturated fatty acids with chainlengths of C-20 and longer that were present in the oil. An immobilized preparation ofG. candidum on silica gel also was used for the enrichment of GLA in borage oil. In this instance, a 75% recovery of GLA was obtained, and the supported lipase was reusable (three cycles) with minimal loss in activity. Presented in part at the 84th Annual Meeting of the American Oil Chemists’ Society, Anaheim, California, May 1993.     

Oil content and fatty acid composition of seed oil from guayule plants with different chromosome numbers

Guayule, a perennial desert plant, is being developed for domestic production of natural rubber, a strategic commodity for which the United States presently depends totally on foreign sources. At present, rubber alone is not sufficient to make guayule a commercial crop, and additional revenues are being sought from by-products. Because guayule flowers profusely during several years of growth before it is harvested for rubber, seed may also contribute to the economics of guayule production. Seed from 120 plants, including 20 genotypes with 36, 37, 54 and 72 chromosomes, were analyzed for oil content and fatty acid composition. Oil content ranged from 17.1 to 30.5%. On average, seed from diploid and aneuploid plants (with 36 and 37 chromosomes) contained 40.4% more oil than the seed from polyploid plants. The oil consisted of four fatty acids—palmitic (8.7–11.5%), stearic (3.7–6.2%), oleic (6.5–13.9%) and linoleic (69.1–80.2%)—at all ploidy levels. Guayule seed oil was similar to the seed oil from high-linoleic safflower varieties. The use of genetic variation to increase seed yield and seed oil will depend on the absence of negative correlation between oil and rubber production.     

Purification of steryl esters from soybean oil deodorizer distillate

Soybean oil deodorizer distillate (SODD) contains steryl esters in addition to tocopherols and sterols. Tocopherols and sterols have been industrially purified from SODD but no purification process for steryl esters has been developed. SODD was efficiently separated to low b.p. substances (including tocopherols and sterols) and high b.p. substances (including 11.2 wt% DAG, 32.1 wt% TAG, and 45.4 wt% steryl esters) by molecular distillation. The high b.p. fraction is referred to as soybean oil deodorizer distillate steryl ester concentrate (SODDSEC). We attempted to purify steryl esters after a lipase-catalyzed hydrolysis of acylglycerols in SODDSEC. Screening of industrially available lipases indicated that Candida rugosa lipase was most effective. Based on the study of several factors affecting hydrolysis, the reaction conditions were determined as follows: ratio of SODDSEC/water, 1∶1 (w/w); lipase amount, 15 U/g reaction mixture; temperature, 30°C. When SODDSEC was agitated for 24 h under these conditions, acylglycerols were almost completely hydrolyzed and the content of steryl esters did not change. However, study with a mixture of steryl oleate/trilinolein (1∶1, w/w) indicated that about 20% of constituent FA in steryl esters were exchanged with constituent FA in acylglycerols. Steryl esters in the oil layer obtained by the SODDSEC treatment with lipase were successfully purified by molecular distillation (purity, 97.3%; recovery, 87.7%).     

Production of FAME from acid oil model using immobilized <Emphasis Type="Italic">Candida antarctica</Emphasis> lipase

Acid oil is a by-product in the neutralization step of vegetable oil refining and is an alternative source of biodiesel fuel. A model substrate of acid oil, which is composed of TAG and FFA, was used in experiments on the conversion to FAME by immobilized Candida antarctica lipase. FFA in the mixture of TAG/FFA were efficiently esterified with methanol (MeOH), but the water generated by the esterification significantly inhibited methanolysis of TAG. We thus attempted to convert a mixture of TAG/FFA to FAME by a two-step process comprising methyl esterification of FFA and methanolysis of TAG by immobilized C. antarctica lipase. The first reaction was conducted at 30°C in a mixture of TAG/FFA (1∶1, wt/wt) and 10 wt% MeOH using 0.5 wt% immobilized lipase, resulting in efficient esterification of FFA. The reaction mixture after 24 h was composed of 49.1 wt% TAG, 1.3 wt% FFA, 49.1 wt% FAME, and negligible amounts of DAG and MAG (<0.5 wt%). The reaction mixture was then dehydrated and used as a substrate for the second reaction, which was conducted at 30°C in a solution of the dehydrated mixture and 5.5 wt% MeOH using 6 wt% immobilized lipase. The activity of the lipase increased gradually when the reaction was repeated by transferring the enzyme to a fresh substrate mixture. The activity reached a maximum after 6 cycles, and the content of FAME achieved was >98.5 wt% after a 24-h reaction. The immobilized lipase was very stable in the first-and second-step reactions and could be used for >100 d without significant loss of activity.     

Purification of docosahexaenoic acid from tuna oil by a two-step enzymatic method: Hydrolysis and selective esterification

Purification of docosahexaenoic acid (DHA) was attempted by a two-step enzymatic method that consisted of hydrolysis of tuna oil and selective esterification of the resulting free fatty acids (FFA). When more than 60% of tuna oil was hydrolyzed with Pseudomonas sp. lipase (Lipase-AK), the DHA content in the FFA fraction coincided with its content in the original tuna oil. This lipase showed stronger activity on the DHA ester than on the eicosapentaenoic acid ester and was suitable for preparation of FFA rich in DHA. When a mixture of 2.5 g tuna oil, 2.5 g water, and 500 units (U) of Lipase-AK per 1 g of the reaction mixture was stirred at 40°C for 48 h, 83% of DHA in tuna oil was recovered in the FFA fraction at 79% hydrolysis. These fatty acids were named tuna-FFA-Ps. Selective esterification was then conducted at 30°C for 20 h by stirring a mixture of 4.0 g of tuna-FFA-Ps/lauryl alcohol (1:2, mol/mol), 1.0 g water, and 1,000 U of Rhizopus delemar lipase. As a result, the DHA content in the unesterified FFA fraction could be raised from 24 to 72 wt% in an 83% yield. To elevate the DHA content further, the FFA were extracted from the reaction mixture with n-hexane and esterified again under the same conditions. The DHA content was raised to 91 wt% in 88% yield by the repeated esterification. Because selective esterification of fatty acids with lauryl alcohol proceeded most efficiently in a mixture that contained 20% water, simultaneous reactions during the esterification were analyzed qualitatively. The fatty acid lauryl esters (L-FA) generated by the esterification were not hydrolyzed. In addition, L-FA were acidolyzed with linoleic acid, but not with DHA. These results suggest that lauryl DHA was generated only by esterification.     

Methyl esters from a partially hydrogenated vegetable oil is a better infrared external standard than methyl elaidate for the measurement of totaltrans content

An infrared spectrophotometric procedure, based on the fatty acid methyl ester mixture derived from a partially hydrogenated vegetable oil as the calibration standard, has been developed for accurate analysis of the totaltrans content of hydrogenated fats. This procedure produces more accurate results than the current official methods of Association of Official Analytical Chemists and American Oil Chemists’ Society, both of which use methyl elaidate as the external standard. The results obtained with this procedure were in close agreement to those determined by the combined procedure of silver-nitrate thin-layer chromatography and capillary gas-liquid chromatography. The improved results, obtained with the partially hydrogenated vegetable oil methyl esters as the calibration standard, may be attributable to its assortment oftrans isomers, which may have different absorptivities relative to methyl elaidate.     

Increase in the γ-linolenic acid content by solvent winterization of fungal oil extracted fromMortierella genus

The fungal oil extracted fromMortierella ramanniana var.angulispora (IFO 8187) was solvent winterized in order to raise the content of γ-linolenic acid (GLA). Effects of winterization conditions (solvent, oil concentration in the solvent and temperature) and changes of glyceride compositions were discussed. The fungal oil was separated into four diglycerides and 17 triglycerides (TG) with high performance liquid chromatography. The predominant species were POO, POP and LOP, whose contents were 24.4, 22.9 and 9.4% of the total TG, respectively. Ethanol at 4°C gave the highest GLA content of 10.5% in spite of lower yield than with acetone at −20°C. The highest separation efficiency for GLA (η_GLA) was 0.27 with acetone at −20°C and 10% oil concentration, resulting in 8.3% of GLA from the fungal oil at 5.7% LGA. In case of lower oil concentration at 5–20%, η_GLA showed higher in the following order: acetone (−20°C)>n-hexane (−20°C)>acetone (4°C)>petroleum ether (−20°C). The winterization process also proved to be effective for the separation of TG type, Sa₂U (Sa; saturated fatty acid; U, unsaturated fatty acid) into the crystallized fraction and SaU₂ into the liquid fraction. Acetone at −20°C showed higher separation efficiency for triunsaturated TG than the other solvents.