Braz. Arch. Biol. Technol. v.61: e18180055 2018 Food/Feed Science and Technology Vol.61: e18180055, 2018 http://dx.doi.org/10.1590/1678-4324-2018180055 ISSN 1678-4324 Online Edition BRAZILIAN ARCHIVES OF BIOLOGY AND TECHNOLOGY A N I N T E R N A T I O N A L J O U R N A L Evaluation of Oxidative Stability of Compound Oils under Accelerated Storage Conditions Geisa Pazzoti1, Camila Souza1, Carolina Veronezi1*, Débora Luzia2, Neuza Jorge1 1Universidade Estadual Paulista – UNESP – Departamento de Engenharia e Tecnologia de Alimentos -São José do Rio Preto , São Paulo, Brazil; 2Universidade do Estado de Minas Gerais, Frutal,Minas Gerais, Brazil. ABSTRACT The oxidative stability of linseed (L), cotton (A), and coconut (C) oils, as well as of linseed:cotton (LA), linseed:coconut (LC), and linseed:cotton:coconut (LAC) compound oils was evaluated under accelerated storage at 60°C/20 days. Coconut oil showed to be rather stable, mainly due to low levels of peroxides, conjugated dienes, ρ-anisidine, and long induction period. In addition, along with cotton oil, it improved the stability of linseed oil in the formulation of LAC compound oil. As to fatty acid profile, the compound oils showed to be composed mainly by unsaturated fatty acids. Cotton and coconut oils presented higher retention of total phytosterols, 78.87 and 76.16%, respectively, after 20 days of storage, when compared to linseed oil. The highest retention of total tocopherols at the end of storage was observed in LA (90.81%). In relation to antioxidant activity, by the DPPH method, with the increase in storage time, a reduction in the antioxidant substances of linseed, LC, and LAC oils was observed. Through the FRAP method, oscillations were observed, especially in linseed and compound oils. Although the oils were degraded over time, it was possible to verify that cotton and coconut oils contributed to increase the stability of linseed oil, which, in turn, raised the levels of coconut oil bioactive compounds. Keywords: linseed, cotton, coconut, lipid oxidation, vegetable oils * Author for correspondence: cveronezi@hotmail.com 2 Pazzoti, G. et al. Braz. Arch. Biol. Technol. v.61: e18180055 2018 INTRODUCTION Vegetable oils are widely consumed all over the world, being of great importance in human diet. These oils are sources of energy and essential fatty acids, carry liposoluble vitamins, and participate in the formation of steroid hormones1,2. In addition, they contribute to increased palatability, providing pleasant taste, aroma, and texture to foods3. However, vegetable oils are rather susceptible to lipid oxidation, due to their composition, especially the presence of unsaturated molecules, which can undergo degradation reactions, leading to nutritional losses and formation of compounds that are toxic to human health. Such degradation, although relatively slow, may occur during heating and/or storage of the final product, affecting its shelf life4. It is possible to evaluate the stability of oils by their storage under accelerated storage conditions, in which periodic analyses are performed to monitor chemical, physical, or sensorial changes. The Schaal Oven Test is one of the most widely used methods5. This test makes it possible to know the oil shelf life, since the results provided have a good correlation with the evaluation carried out in storage at room temperature6. In order to avoid oxidation problems and increase stability in the oils, it is necessary to eliminate traces of metals such as iron, copper, and chromium. Additionally, it is important to prevent contact with oxygen and high temperatures, as well as to eliminate the pro-oxidants and block the formation of free radicals through antioxidants, which, in small amounts, act to inhibit or retard the oxidation process of lipids7. The formulation of compound oils has emerged as an alternative to increase stability, in addition to raising the content of vegetable oil bioactive compounds. These oils are products obtained from the mixture of oils from two or more plant species, which have been studied as an economical way to modify fatty acid composition and physicochemical characteristics2. The investigation of compound oils, which are formulated by the combination of vegetable oils, is a field of emerging research that has not yet been explored. In this context, the objective of this study was to evaluate the oxidative stability and antioxidant activity of compound oils formulated with linseed, cotton, and coconut oils stored at 60°C for 20 days. MATERIAL AND METHODS Oils Linseed (L) oil was extracted by cold continuous pressing (< 60°C) in Scott Tech Equipamentos industries, located in Vinhedo, São Paulo, Brazil. The seeds were previously submitted to a dryer (model SMR 610-G, Scott Tech, Vinhedo, SP, Brazil) with a rotary system and LPG gas at 50°C for 25 minutes, to reduce moisture. Subsequently, they were taken to a vegetable oil extractor (model ERT 60 III, Scott Tech, Vinhedo, SP, Brazil) with a system of tubular radial extraction. Then, the oil went through filter press (model FP 240-N2, Scott Tech, Vinhedo, SP, Brazil), was bottled in amber bottles, and kept under freezing temperature (-18°C) until analysis. Cotton (A) and coconut (C) refined oils from Triângulo Alimentos industries, located in Itápolis, São Paulo, Brazil, were used to formulate the compound oils: linseed and cotton (LA), 1:1 (v:v); linseed and coconut (LC), 1:1 (v:v); and linseed, cotton, and coconut (LAC), 2:1:1 (v:v:v). Evaluation of compound oils 3 Braz. Arch. Biol. Technol. v.61: e18180055 2018 Accelerated storage test The accelerated storage test was performed at 60ºC for 20 days, using beakers containing 30 mL of each oil with a surface/volume ratio of 0.3/cm. The samples were collected in 0, 10, and 20 days, inertized with nitrogen gas, and stored at -18°C until analysis. Physicochemical properties Physicochemical analysis were performed using the compound oils The following parameters were analysed: peroxide value, conjugated dienes, and ρ-anisidine indexes, according to the AOCS method8. The total oxidation value (Totox) was calculated by using the equation: Totox = 2 (peroxide index) + (ρ-anisidine value)9. Oxidative stability was performed using Rancimat (Model 743, Metrohm Ltd., Herisau, Switzerland) at 110°C, with 20 L/h air flow8. In the sequence, the fatty acid profile was determined by means of gas chromatography of the esterified oils according to the AOCS method8. A gas chromatograph (Model 3900, Varian, Walnut Creek, CA, USA) was used with a flame ionization detector, split injection system, and fused silica capillary column (CP-Sil 88, Microsorb, Varian, Walnut Creek, CA, USA). The initial oven temperature was 90°C for 4 min, heated at 10°C/min up to 195°C, then maintained at the same temperature for 16 min. The temperatures used in the injector and the detector were 230 and 250ºC, respectively. Hydrogen was used as carrier gas. Fatty acids were identified according to their retention times, comparing them to the standard composed of 37 fatty acid methyl esters of C4:0 to C24:1, with purity between 99.1 and 99.9% (Supelco, Bellefonte, USA). The phytosterol profile was determined by gas chromatography from the unsaponifiable matter. Saponification was performed according to Duchateau10. For the determination, a gas chromatograph (Model 2010 Plus-Shimadzu, Chiyoda-ku, Tokyo, Japan) was used with flame ionization detector, split injector and fused silica capillary column (Restek RTX 5, Shimadzu, Chiyoda-ku, Tokyo, Japan). The column temperature was maintained at 260°C for 35 min. The temperatures used in the injector and the detector were 280 and 320°C, respectively. Hydrogen was used as carrier gas. The quantification of each isomer was performed by internal standardization (cholestane-5α-3β-ol), based on peak areas, and expressed as mg/kg. The analysis of tocopherols was conducted in a high performance liquid chromatograph (Model 210-263 Varian Inc., Walnut Creek, CA, USA), with fluorescence detector, silica packed stainless steel column (100 Si, Microsorb, Varian, Walnut Creek, CA, USA) with a 290 nm excitation wavelength and a 330 nm emission wavelength. The concentration values were calculated based on the peak excitation areas and expressed as values for each separate isomer. A standard curve of α-, β-, γ- and δ-tocopherols (Supelco, Bellefonte, USA) was prepared with a high purity level to express the tocopherol contents in mg/kg. Vitamin E was calculated according to the method described by McLaughlin and Weihrauch11. The conversion values were: α-tocopherol x 1.0; β-tocopherol x 0.40; γ-tocopherol x 0.10; and δ- tocopherol x 0.01. Phenolic compounds and antioxidant activity The extraction of total phenolic compounds was performed by the procedure described by Parry12 and they were quantified using the Folin-Ciocalteu reagent, according to a 4 Pazzoti, G. et al. Braz. Arch. Biol. Technol. v.61: e18180055 2018 methodology described by Singleton and Rossi13. Total phenolic compounds were detected at λ = 765 nm (UV-VIS mini model 1240, Shimadzu, Chiyoda-ku, Tokyo, Japan). Gallic acid was used to plot the standard curve (R2 = 0.9999), and the results were expressed as mg EAG/kg. The evaluation of antioxidant activity was carried out by two different spectrophotometer methodologies (Model UV-VIS mini 1240, Shimadzu, Chiyoda-ku, Tokyo, Japan). The DPPH analysis, which consists of evaluating the scavenging activity of the free radical 2,2-diphenyl-1-picrylhydrazyl, was carried out according to the method of Kalantzakis14, through which the sample absorbance was measured at λ = 517 nm and the result expressed as percentage. The antioxidant activity analysis was also conducted through the FRAP method, which is based on the ability of phenols to reduce Fe+ TPTZ-3 (tripyridil-s-triazine ferric iron) complex into Fe+ TPTZ-2 (tripyridil iron-s-ferrous triazine) complex at pH 3.6. This methodology has been described by Szydlowska-Czerniak15 and the results were expressed as μM trolox/100 g. Statistical analysis The results obtained from the analytical determinations, in triplicate, were submitted to analysis of variance and the differences between means were tested at 5% probability by the Tukey test16, through the ESTAT program, version 2.0. RESULTS AND DISCUSSION Physicochemical properties According to Codex Alimentarius17, refined and crude vegetable oils should have a maximum of 10 and 15 meq/kg of peroxides, respectively, in order to be considered good quality oils. Thus, it was possible to observe that, initially, all the oils presented peroxide value within the established limits (Table 1). Evaluation of compound oils 5 Braz. Arch. Biol. Technol. v.61: e18180055 2018 Table 1-Physicochemical properties of oils under storage. Properties Oils Days of storage 0 10 20 Peroxide (meq/kg) L 1.80 ± 0.04cA 31.22 ± 3.80bA 47.71 ± 2.45aA A 3.30 ± 0.27bA 4.06 ± 0.02abC 6.73 ± 0.06aE C 0.66 ± 0.04aA 1.20 ± 0.04aC 1.75 ± 0.21aF LA 1.17 ± 0.06cA 4.87 ± 0.43bC 21.94 ± 1.91aD LC 0.70 ± 0.03cA 14.89 ± 0.06bB 40.50 ± 0.04aB LAC 1.11 ± 0.06bA 2.70 ± 0.05bC 28.59 ± 1.97aC Conjugated dienes (%) L 0.14 ± 0.01cD 0.40 ± 0.01bB 0.61 ± 0.002aA A 0.42 ± 0.01cA 0.54 ± 0.01aA 0.52 ± 0.002bB C 0.09 ± 0.001bE 0.11 ± 0.002aF 0.12 ± 0.01aE LA 0.25 ± 0.01cB 0.34 ± 0.01bC 0.46 ± 0.01aC LC 0.14 ± 0.01cD 0.17 ± 0.002bE 0.42 ± 0.004aD LAC 0.23 ± 0.01cC 0.22 ± 0.003bD 0.45 ± 0.003aC ρ-anisidine L 0.33 ± 0.03cD 11.81 ± 0.41bB 28.19 ± 0.16aA A 5.77 ± 0.18bA 6.33 ± 0.11bC 7.33 ± 0.06aD C 2.99 ± 0.11bC 3.20 ± 0.19bD 5.73 ± 0.05aE LA 4.40 ± 0.33cB 5.71 ± 0.03bC 6.84 ± 0.44aDE LC 0.73 ± 0.01cD 13.76 ± 0.03bA 24.06 ± 1.06aB LAC 1.42 ± 0.01cD 11.43 ± 0.84bB 21.88 ± 0.50aC Totox L 3.93 74.25 123.61 A 12.37 14.45 20.79 C 4.31 5.60 9.23 LA 6.74 15.45 50.72 LC 2.13 43.54 105.06 LAC 3.64 16.83 79.06 Oxidative stability (h) L 0.9 ± 0.03aD 0.71 ± 0.03bE 0.77 ± 0.03bC A 15.40 ± 0.07aB 12.94 ± 0.25bB 9.85 ± 0.17cB C 53.79 ± 0.02aA 45.49 ± 0.69bA 45.66 ± 0.98bA LA 5.80 ± 0.14aC 5.31 ± 0.05aC 2.06 ± 0.13bC LC 1.01 ± 0.01aD 2.28 ± 0.04bD 0.76 ± 0.01cC LAC 5.84 ± 0.16aC 6.16 ± 0.01aC 1.13 ± 0.03bC The mean ± standard deviation followed by lowercase letters in the lines do not differ by Tukey test (p > 0.05). Means ± standard deviation followed by upper case letters in the columns do not differ by Tukey test (p > 0.05). However, as the storage time increased, linseed oil showed high oxidation, reaching 47.71 meq/kg of peroxides in 20 days of storage. This increase may have influenced the compound oils peroxide content, as cotton (6.73 meq/kg) and coconut (1.75 meq/kg) oils remained stable, contributing to a lower peroxide formation compared to linseed oil. Conjugated dienes are primary oxidation products of polyunsaturated fatty acids formed by the displacement of double bonds18. The oils showed low values of conjugated dienes at the beginning of storage. However, there was significant increase during storage at 60ºC, except in coconut oil, which was stable after 10 days of storage. It was also possible to verify, when comparing the compound oils, that LC showed lower formation of conjugated dienes, possibly due to the influence of coconut oil. According to Guillén and Cabo19 and Marina et al.20, oils of good quality must have ρ- anisidine index below 10. Thus, it is possible to infer that, initially, all the oils presented good quality. However, the values of ρ-anisidine increased during storage. In 20 days, only cotton, coconut, and LA oils remained below 10, while linseed oil 6 Pazzoti, G. et al. Braz. Arch. Biol. Technol. v.61: e18180055 2018 Cont. showed the lowest quality, 28.19 of ρ-anisidine. This fact is possibly due to the fact that linseed oil is crude and constituted by a large amount of α-linolenic acid. It is also considered that well-preserved fat must have a Totox value below 1021. According to Table 1, initially, all oils presented good conservation status, except for cotton oil (12.37). However, after 10 days of storage, only coconut oil remained below this limit, indicating that it is a stable oil. In the oxidative stability test, coconut (53.79 h) and cotton (15.40 h) oils showed the highest rates. According to Michotte et al.22, the high content of unsaturated fatty acids of linseed oil makes it extremely sensitive to oxidative reactions. This information is confirmed in this study, since linseed oil presented the lowest oxidative stability during storage, in relation to the other oils. When comparing the compound oils, it was found that LA and LAC remained stable up to 10 days of storage, probably due to the synergism between linseed and cotton oils. Ten different types of fatty acids were identified (Table 2). Initially, coconut oil showed higher value of α-linolenic acid (47.9%) than linseed oil (47.1%). With heating, lauric, palmitic, and stearic fatty acids were found to remain stable for up to 20 days of storage. However, α-linolenic acid showed significant reduction in compound oils, especially in LC, in which it was reduced by 47.41%. Table 2-Fatty acid composition of oils under storage. Fatty acids (%) Oils Days of storage 0 20 Caproic (6:0) L nd nd A nd nd C 2.07 ± 0.02bA 2.24 ± 0.01aA LA nd Nd LC 0.58 ± 0.01bB 1.36 ± 0.01aB LAC 0.32 ± 0.01bC 0.55 ± 0.01aC Caprylic (8:0) L nd nd A nd nd C 2.44 ± 0.01bA 2.50 ± 0.01aA LA nd nd LC 0.70 ± 0.01bB 1.56 ± 0.02aB LAC 0.36 ± 0.01bC 0.61 ± 0.01aC Lauric (12:0) L nd nd A 0.27 ± 0.02aD 0.29 ± 0.02aD C 47.10 ± 0.01aA 47.11 ± 0.01aA LA 0.18 ± 0.0aE 0.19 ± 0.01aE LC 13.62 ± 0.02bB 29.87 ± 0.03aB LAC 7.14 ± 0.01bC 11.56 ± 0.02aC Myrístic (14:0) L nd nd A 0.83 ± 0.01aD 0.83 ± 0.02aD C 13.33 ± 0.01bA 14.92 ± 0.02aA LA 0.24 ± 0.01bE 0.54 ± 0.01aE LC 4.44 ± 0.01bB 9.62 ± 0.03aB LAC 2.59 ± 0.01bC 4.05 ± 0.01aC Palmitic (16:0) L 11.44 ± 0.01aF 11.64 ± 0.01aF A 40.31 ± 0.01aA 40.04 ± 0.01aA C 12.75 ± 0.01aD 12.54 ± 0.01aD LA 25.19 ± 0.02bB 29.15 ± 0.01aB LC 11.93 ± 0.03aE 12.31 ± 0.01aE LAC 20.51 ± 0.01aC 20.13 ± 0.01aC Stearic (18:0) L 5.34 ± 0.02aA 5.09 ± 0.01aB A 3.21 ± 0.01aE 3.22 ± 0.02aF Evaluation of compound oils 7 Braz. Arch. Biol. Technol. v.61: e18180055 2018 Cont. C 3.76 ± 0.01aD 3.66 ± 0.02aE LA 3.22 ± 0.03aE 3.99 ± 0.02aD LC 4.92 ± 0.02aB 4.35 ± 0.01aC LAC 4.33 ± 0.04bC 5.36 ± 0.03aA Fatty acids (%) Oils Days of storage 0 20 Oleic (18:1n9c) L 26.98 ± 0.01aA 27.37 ± 0.02aA A 18.72 ± 0.02aE 18.87 ± 0.01aE C 15.40 ± 0.01aF 15.11 ± 0.01aF LA 19.29 ± 0.04bD 22.06 ± 0.01aB LC 23.56 ± 0.01aB 19.02 ± 0.03bD LAC 23.07 ± 0.04aC 21.54 ± 0.02bC Linoleic (18:2n6c) L 13.31 ± 0.01aD 13.34 ± 0.01aD A 36.23 ± 0.01aA 36.23 ± 0.04aA C 1.29 ± 0.01aF 1.28 ± 0.02aF LA 23.50 ± 0.03bB 26.71 ± 0.01aB LC 9.81 ± 0.01aE 5.77 ± 0.02bE LAC 18.49 ± 0.01aC 16.33 ± 0.04bC Arachidic (20:0) L nd nd A 0.15 ± 0.01aC 0.15 ± 0.01aB C 1.89 ± 0.02aA 0.67 ± 0.04bA LA nd nd LC 0.57 ± 0.02aB 0.13 ± 0.03bB LAC 0.14 ± 0.04aC nd α-Linolenic (18:3n3) L 42.93 ± 0.01aA 42.56 ± 0.01aA A 0.30 ± 0.01aE 0.37 ± 0.01aE C nd nd LA 28.39 ± 0.02aC 17.34 ± 0.05bC LC 30.48 ± 0.06aB 16.03 ± 0.11bD LAC 23.06 ± 0.03aD 19.89 ± 0.09bB Ʃ Saturated L 16.78 ± 0.03aF 16.73 ± 0.02aF A 44.62 ± 0.01aB 44.38 ± 0.01aC C 81.45 ± 0.01aA 82.97 ± 0.01aA LA 28.83 ± 0.03aE 33.87 ± 0.02aE LC 36.19 ± 0.05bC 59.07 ± 0.03aB LAC 35.25 ± 0.04bD 42.26 ± 0.03aD ∑ Monounsaturated L 26.98 ± 0.07aA 27.37 ± 0.02aA A 18.72 ± 0.02aE 18.87 ± 0.01aE C 15.40 ± 0.01aF 15.11 ± 0.01aF LA 19.29 ± 0.03bD 22.06 ± 0.01aB LC 23.56 ± 0.01aB 19.02 ± 0.03bD LAC 23.07 ± 0.04aC 21.54 ± 0.02bC Ʃ Polyunsaturated L 56.24 ± 0.03aA 55.90 ± 0.04aA A 36.68 ± 0.01aE 36.75 ± 0.01aC C 3.18 ± 0.01aF 1.95 ± 0.02bF LA 51.89 ± 0.01aB 44.05 ± 0.04bB LC 40.86 ± 0.04aD 21.93 ± 0.06bE LAC 41.69 ± 0.01aC 36.22 ± 0.06bD The mean ± standard deviation followed by lowercase letters in the lines do not differ by Tukey test (p > 0.05). Means ± standard deviation followed by upper case letters in the columns do not differ by Tukey test (p > 0.05). nd: not detected. The compound oils showed to be constituted mainly of unsaturated fatty acids, representing 64 to 71% of the totality (Table 2). These oils can help lower cholesterol and triglyceride levels, regulate blood pressure, and reduce chronic inflammation and the development of cancer, heart diseases, and stroke23,24. During storage, saturated 8 Pazzoti, G. et al. Braz. Arch. Biol. Technol. v.61: e18180055 2018 fatty acids did not change significantly, except in LC and LAC, probably due to the influence of linseed and coconut oils. On the other hand, the compound oils showed reduction in relation to unsaturated fatty acids, especially in LC (46.32%). As for the phytosterols profile, the oils showed four different isomers: campesterol, stigmasterol, β-sitosterol, and stigmastanol (Table 3). The presence of β-sitosterol was higher than the other isomers in all oils, and stigmastanol was not detected in cotton and coconut oils. Initially, linseed oil had higher amounts of campesterol (6.51 mg/kg) and stigmastanol (56.09 mg/kg), while cotton had higher amount of β-sitosterol (90.27 mg/kg), and coconut oil had higher stigmasterol content (8.50 mg/kg). According to Ferrari et al.25, phytosterols are susceptible to oxidation by reaction with oxygen, light, metal ions, and high temperature, depending on their degree of unsaturation. Thus, it was observed that all phytosterol isomers decreased during storage. However, linseed oil presented greater degradation of total phytosterols than cotton and coconut oils, which remained stable after 10 days of storage. When comparing the compound oils, it was verified that LC retained greater amount of total phytosterols, 95.1%, in 20 days. Table 3-Phytosterol profile of oils under storage. Phytosterol (mg/kg) Oils Days of storage 0 10 20 Campesterol L 6.51 ± 0.04aA 5.38 ± 0.04bA 4.86 ± 0.14cA A 4.81 ± 0.06aD 3.53 ± 0.04bE 3.31 ± 0.27bB C 5.71 ± 0.03aB 4.94 ± 0.05bB 4.89 ± 0.04bA LA 4.13 ± 0.04aF 3.93 ± 0.04aD 3.59 ± 0.12bB LC 5.54 ± 0.02aC 5.46 ± 0.09abA 5.23 ± 0.04bA LAC 4.61 ± 0.01aE 4.25 ± 0.06bC 3.54 ± 0.06cB β-sitosterol L 58.13 ± 0.03aC 55.44 ± 0.05bB 43.25 ± 0.07cE A 90.27 ± 0.05aA 72.11 ± 0.02bA 71.72 ± 0.59bA C 27.07 ± 0.04aF 19.75 ± 0.11bF 19.82 ± 0.04bF LA 58.42 ± 0.04aB 54.07 ± 0.04bC 53.57 ± 0.04cB LC 49.37 ± 0.04aE 49.33 ± 0.33aD 47.37 ± 0.04bC LAC 50.49 ± 0.04aD 46.44 ± 0.06bE 44.64 ± 0.06cD Stigmasterol L 4.28 ± 0.04bC 4.10 ± 0.03cC 4.87 ± 0.04aB A 6.57 ± 0.04aB 5.27 ± 0.05bB 5.15 ± 0.22bB C 8.50 ± 0.03aA 6.75 ± 0.04bA 6.74 ± 0.03bA LA 3.68 ± 0.04aD 3.65 ± 0.01aD 3.12 ± 0.02bD LC 4.31 ± 0.04aC 4.22 ± 0.03aC 3.83 ± 0.06bC LAC 3.50 ± 0.06aE 2.80 ± 0.15bE 2.15 ± 0.07cE Stigmastanol L 56.09 ± 0.02aA 48.08 ± 0.03bA 24.39 ± 0.16cD A nd nd nd C nd nd nd LA 37.04 ± 0.03aC 34.19 ± 0.02bC 31.32 ± 0.12cB LC 47.97 ± 0.04aB 46.13 ± 0.16bB 45.44 ± 0.09cA LAC 36.89 ± 0.04aD 30.74 ± 0.23bD 27.45 ± 0.64cC Total L 124.99 ± 0.01aA 112.99 ± 0.14bA 77.37 ± 0.13cD A 101.65 ± 0.05aD 80.91 ± 0.03bE 80.17 ± 1.07bC C 41.28 ± 0.03aF 31.44 ± 0.11bF 31.44 ± 0.03bE LA 103.26 ± 0.01aC 95.83 ± 0.01bC 91.66 ± 0.06cB LC 107.17 ± 0.01aB 105.14 ± 0.04bB 101.87 ± 0.04cA LAC 95.49 ± 0.04aE 84.22 ± 0.26bD 77.78 ± 0.83cD The mean ± standard deviation followed by lowercase letters in the lines do not differ by Tukey test (p > 0.05). Means ± standard deviation followed by upper case letters in the columns do not differ by Tukey test (p > 0.05). nd: not detected. Limits of detection: campesterol ≤ 52 mg/kg; stigmasterol ≤ 56 mg/kg e stigmastanol ≤ 42,5 mg/kg. Evaluation of compound oils 9 Braz. Arch. Biol. Technol. v.61: e18180055 2018 The oils presented α-, γ-, and δ-tocopherol isomers (Table 4). Initially, regarding α- tocopherol, cotton oil (86.05 mg/kg) stood out, whereas in LA and linseed oils, γ- and δ-tocopherol were higher, 79.60 and 17.65 mg/kg, respectively. Table 4-Quantification of tocopherols and vitamin E of oils under storage. Tocopherols (mg/kg) Oils Days of storage 0 10 20 α-tocol L 29.53 ± 0.11aC 27.50 ± 0.14bC 8.20 ± 0.14cD A 86.05 ± 0.36aA 85.35 ± 0.35aA 83.90 ± 0.28bA C nd nd nd LA 55.00 ± 0.14aB 48.25 ± 0.07bB 47.35 ± 0.07cB LC 24.70 ± 0.57D nd nd LAC 21.80 ± 0.14aE 18.45 ± 0.07bD 17.45 ± 0.07cC γ-tocol L 76.05 ± 0.49aB 71.20 ± 0.14bB 39.65 ± 0.49cD A 59.85 ± 0.07aE 59.50 ± 0.28aC 53.25 ± 0.21bB C nd nd nd LA 79.60 ± 0.14aA 78.90 ± 0.14aA 75.35 ± 0.77bA LC 63.15 ± 0.35aD 9.25 ± 0.21bD nd LAC 65.15 ± 0.21aC 59.85 ± 0.21bC 42.45 ± 0.21cC δ-tocol L 17.65 ± 0.07aA 16.55 ± 0.35bA nd A 12.55 ± 0.07aD 12.15 ± 0.21aB 12.45 ± 0.07aA C nd nd nd LA 11.25 ± 0.07aE 10.40 ± 0.14bC 9.75 ± 0.21cB LC 16.60 ± 0.28B nd nd LAC 14.50 ± 0.14aC 9.45 ± 0.21bC 9.05 ± 0.07bC Total L 123.22 ± 0.67aC 115.25 ± 0.35bC 47.85 ± 0.35cD A 158.44 ± 0.50aA 157.00 ± 0.42aA 149.60 ± 0.01bA C nd nd nd LA 145.85 ± 0.21aB 137.55 ± 0.07bB 132.45 ± 0.49cB LC 104.45 ± 0.64aD 9.25 ± 0.21bE nd LAC 101.45 ± 0.07aE 87.75 ± 0.35bD 68.95 ± 0.07cC Vitamin E* L 39.98 ± 0.28aC 37.36 ± 0.13bC 13.61 ± 0.07cD A 94.16 ± 0.44aA 93.47 ± 0.39aA 91.16 ± 0.26bA C nd nd nd LA 65.96 ± 0.16aB 59.11 ± 0.09bB 57.57 ± 0.03cB LC 33.46 ± 0.61aD 1.26 ± 0.03bE nd LAC 30.82 ± 0.11aE 26.70 ± 0.04bD 23.32 ± 0.04cC The mean ± standard deviation followed by lowercase letters in the lines do not differ by Tukey test (p > 0.05). Means ± standard deviation followed by upper case letters in the columns do not differ by Tukey test (p > 0.05). nd: not detected. Limits of detection: δ-tocol < 2,30 mg/kg.*Expressed as α-tocol. It can be observed that γ-tocopherol isomer was the one in highest amount in all oils. Tocopherols and vitamin E were not detected in coconut oil. Concerning vitamin E, the highest amount was found in cotton (94.16 mg/kg) and LA (65.96 mg/kg) oils, mainly due to the amount of α-tocopherol present in cotton oil. It was observed that the isomers of tocopherols and vitamin E decreased during storage. According to Lampi et al.26, oxidizing agents, especially in the presence of heat, light, metals, and alkali, easily oxidize tocopherols. In relation to total tocopherols, cotton oil remained stable, with retention of 94.42% at the end of the process. Among compound oils, LA retained higher amount of total tocopherols (90.81%) and vitamin E (87.28%) at the end of storage. Phenolic compounds and antioxidant activity In relation to phenolic compounds (Table 5), at the beginning of storage, coconut oil stood out with 220.40 mg/kg. 10 Pazzoti, G. et al. Braz. Arch. Biol. Technol. v.61: e18180055 2018 Table 5-Phenolic compounds and antioxidant activity of oils under storage. Oils Days of storage 0 10 20 Phenolic compounds (mg/kg) L 214.14±15.41bA 235.77 ± 16.32bD 218,.7 ± 10.01aCD A 142.84 ± 4.44cB 371.07 ± 19.23aA 213.51 ± 19.02bE C 220.40 ± 2.00cA 378.85 ± 4.02bA 515.29 ± 4.33aB LA 213.95 ± 20.00bA 278.62 ± 1.54aC 238.84±28.07abDE LC 150.40 ± 1.76cB 371.29 ± 5.67bA 667.29 ± 9.05aA LAC 211.51 ± 1.68bA 318.85 ± 15.67aB 330.84 ± 24.24aC DPPH (%) L 57.20 ± 1.19aB 55.79 ± 1.69aC 47.07 ± 0.73bC A 92.76 ± 1.13aA 88.05 ± 2.26aA 88.64 ± 2.19aA C 26.01 ± 0.67aD 26.66 ± 0.52aD 25.0 ± 0.18aDE LA 59.18 ± 7.12cB 80.91 ± 1.98aB 71.75 ± 2.12bB LC 42.42 ± 7.20aC 27.99 ± 1.66bD 29.46 ± 0.67bD LAC 38.71 ± 1.77aC 19.51 ± 0.96bE 19.51 ± 0.41bE FRAP (µM/100 g) L 97.23 ± 6.69bA 107.95 ± 0.39aA 103.05 ± 0.13abA A 102.59 ± 2.44aA 74.09 ± 0.45bC 71.27 ± 3.15bB C 89.73 ± 2.51aB 66.68 ± 3.21bC 57.0 ± 0.32cC LA 75.55 ± 0.32bC 90.14 ± 1.67aB 76.55 ± 1.61bB LC 99.09 ± 1.22bA 109.41 ± 0.51aA 109.86 ± 0.64aA LAC 74.32 ± 3.34cC 84.64 ± 0.19bB 109.0 ± 1.74aA The mean ± standard deviation followed by lowercase letters in the lines do not differ by Tukey test (p > 0.05). Means ± standard deviation followed by upper case letters in the columns do not differ by Tukey test (p > 0.05). Throughout storage, there was an increase in the levels of phenolic compounds in all oils. Such increase may have occurred due to the reaction of Folin-Ciocalteu reagent with phenol groups resulting from the degradation of tocopherols, since the presence of these easily oxidizable compounds results in the formation of blue complexes, causing overestimation of the total phenolic compounds27. In the test for the reduction of DPPH radical, cotton oil (92.76%), followed by LA (59.18%) and linseed (57.20%) oils, stood out showing higher antioxidant activity at the beginning of storage. This may be due to the greater presence of natural antioxidants in these oils. Cotton and coconut oils showed stable antioxidant activity up to 20 days of storage. Yet, it was possible to detect a decrease in antioxidant substances in the other oils after 10 days of storage. Regarding the FRAP method, cotton (102.59 μM/100 g), LC (99.09 μM/100 g), and linseed (97.23 μM/100 g) oils showed higher antioxidant power. However, oscillations occurred during storage, possibly due to the presence of pro-oxidant compounds from crude linseed oil. CONCLUSIONS The oils underwent degradation during storage at 60°C/20 days, as peroxide, conjugated dienes, ρ-anisidine, and Totox values increased. However, it was found that the formation of the degradation compounds was lower in the compound oils, especially in LA. In the compound oils, unsaturated fatty acids predominated, and α- linolenic acid stood out. The oils showed reduction of phytosterols and tocopherols during storage, with higher retention in the compound oils, especially LC, with 95.1% of phytosterols, and LA, with 90.81% of tocopherols, at the end of storage. The oils showed significant antioxidant activity, probably due to the presence of phenolic compounds, phytosterols, and tocopherols. It is possible to conclude that cotton and Evaluation of compound oils 11 Braz. Arch. Biol. Technol. v.61: e18180055 2018 coconut oils can be used in the formulation of compound oils, adding oxidative stability to linseed oil and, consequently, helping in the retention of compounds such as phytosterols and tocopherols. REFERENCES 1-Iqbal S, Bhanger MI. Stabilization of sunflower oil by garlic extract during accelerated storage. Food Chem. 2007; 100(1): 246-254. 2-O’Brien RD. Fats and oils: formulating and processing for applications. 3 ed. Boca Raton: CRC Press; 2008. 3-Ho CT, Mussinan C, Shahidi F, Contis ET. Nutrition, functional and sensory properties of foods. London: Royal Society of Chemistry; 2013. 4-Casal S, Malheiro R, Sendas A, Oliveira PB, Pereira, JA. Olive oil stability under deep-frying conditions. Food Chem. Toxicol. 2010; 48(1): 2972 - 2979. 5-Dutton HJ. Analysis of fats and oils. J. Am. Oil Chem. Soc. 1978; 55(5): 806-808. 6-Gómez-Alonso S, Mancebo-Campos V, Salvador MD, Fregapane G. Evolution of major and minor components and oxidation indices of virgen olive oil during 21 months storage at room temperature. Food Chem. 2007; 100(1): 36-42. 7-Choe EO, Lee J, Min DB. Chemistry for oxidative stability of edible oils. In: Akoh CC, Lai OM, editors. Healthful lipids. Champaign: AOCS Press; 2005. p 569-574. 8-AOCS. Official methods and recommended practices of the American Oil Chemists’ Society. 6. ed. Champaing: American Oil Chemists’ Society, 2009. 9-Abreu DAP, Losada PP, Maroto J, Cruz JM. Evaluation of the effectiveness of a antioxidants (from barley husks) that retard lipid damage in frozen Atlantic Salmon (Salmo salar L.). Food Res. Int. 2010; 43(5): 1277-1282. 10-Duchateau GSMJE, Bauer-Plank CG, Louter AJH, Van Der Ham M, Boerma JA, Van Rooijen JJ M, Zandbelt PA. Fast and accurate method for total 4-desmethy sterol(s) content in spreads, fat-blends a raw material. J. Am. Oil Chem. Soc. 2002; 79(3): 273-278. 11-McLaughlin PJ, Weihrauch JL. Vitamin E content of foods. J. Am. Diet Assoc. 1979; 75:647-665. 12-Parry JW, Su L, Luther M, Zhou K., Yurawecz MP, Whittaker P, Yu L. Fatty acid composition and antioxidant properties of cold-pressed marionberry, boysenberry, red raspberry, and blueberry seed oils. J. Agric. Food Chem. 2005; 53(3): 566-573. 13-Singleton VL, Rossi JA. Colorimetry of total phenolics with phosphomolybdic and phosphotungstic acid reagents. Am. J. Enol. Vitic. 1965; 16(3): 144-158. 14-Kalantzakis G, Blekas G, Pegklidou K, Boskou D. Stability and radical-scavenging activity of heated olive oil and other vegetable oils. Eur. J. Lipid Sci. Tech. 2006; 108(4): 329-335. 15-Szydlowska-Czerniak A, Karlovits G, Dianoczki C, Recseg K, Szłyk E. Comparison of two analytical methods for assessing antioxidant capacity of rapessed and olive oils. J. Am. Oil Chem. Soc. 2008; 85(2): 141-149. 16-Gacula JM, Singh J, Bi S. Statistical methods in food and consumer research. London: Academic Press; 2008. 17-Codex alimentarius commision. Codex-Stan 210: codex standard for named vegetable oils. Rome: Codex alimentarius commission; 2009. 18-Frankel EN. In search of better methods to evaluate natural antioxidants and oxidative stability in food lipids. Trends Food Sci.Tech. 1993; 4(7): 220-225. 19-Guillén MD, Cabo N. Fourier transform infrared spectra data versus peroxide and anisidine values to determine oxidative stability of edible oils. Food Chem. 2002; 77(4): 503-510. 20-Marina AM, Che Man YB, Nazimah SAH, Amin I. Chemical properties of virgin coconut oil. J. Am. Oil Chem. Soc. 2009; 86(4): 301-307. 12 Pazzoti, G. et al. Braz. Arch. Biol. Technol. v.61: e18180055 2018 21-Berset C, Cuvelier ME. Methods of estimating the degree of lipid oxidation and of measuring antioxidizing power. Sci.Aliments 1996; 16(3): 219-245. 22-Michotte D, Rogez H, Chirinos R, Mignolet E, Campos D, Larondelle Y. Linseed oil stabilisation with pure natural phenolic compounds. Food Chem. 2011; 129(3): 1228-1231. 23-Fredman G, Serhan CN. Specialized proresolving mediator targets for RvE1 and RvD1 in peripheral blood and mechanisms of resolution. Bioch. J. 2011; 437(2): 185- 197. 24-Pinazo-Duran MD, Boscá-Gomar L. Anti-inflammatory properties of polyunsaturated fatty acid omega 3 indications in ophthalmology. Arch. Soc. Esp. Oftalmol. 2012; 87(8): 203-220. 25-Ferrari RA, Schulte E, Esteves W, Bruhl L, Mukherjee KD. Minor constituents of vegetable oils during industrial processing. J. Am. Oil Chem. Soc. 1996; 73(1): 587- 592. 26-Lampi AM, kamal-Eldin A, Piironen V. Tocopherols and tocotrienols from oil and cereal grains. In: Shi J, Mazza G, Maguer ML, editors. Functional Foods, Biochemical and Processing Aspects. Boca Raton: CRC Press; 2002. 27-Prior RL, Wu X, Schaich K.. Standardized methods for de determination of antioxidant capacity and phenolics in foods and dietary supplements. J. Agric. Food Chem. 2005; 53(10): 4290-4302. Received: February 03, 2018 Accepted: September 27, 2018