Article pubs.acs.org/JAFC
Novel Approach To Evaluate the Oxidation State of Vegetable Oils Using Characteristic Oxidation Indicators Jun Cao, Long Deng, Xue-Mei Zhu, Yawei Fan,* Jiang-Ning Hu, Jing Li, and Ze-Yuan Deng* State Key Laboratory of Food Science and Technology and Institute for Advanced Study, Nanchang University, Nanchang 330047, Jiangxi Province, China S Supporting Information *
ABSTRACT: Four vegetable oils with typical fatty acid compositions were chosen to determine their indicators of lipid oxidation under the conditions of accelerated oxidation. Good linear correlations were observed between the total nonpolar carbonyl amount and the total oxidation value (TOTOX, R2 = 0.89−0.97) or peroxide value (POV, R2 = 0.92−0.97) during 35 days of accelerated oxidation. Additionally, nonanal in camellia oil (oleic acid mainly) increased significantly, and correlated linearly with TOTOX (21.6 TOTOX − 595, R2 = 0.92); propanal increased significantly in perilla oil (linolenic acid mainly) and correlated linearly with TOTOX (8.10 TOTOX + 75.0, R2 = 0.90). Hexanal (9.56 TOTOX + 913, R2 = 0.90, and 7.10 TOTOX + 342, R2 = 0.78, respectively) and nonenal (10.5 TOTOX + 691, R2 = 0.95, and 6.65 TOTOX + 276, R2 = 0.84, respectively) in sunflower oil (linoleic acid mainly) and palm oil (palmitic and oleic acids mainly) also had good linear correlations with TOTOX. Considering the change patterns of these four aldehydes, it was found that the oxidation stability was in the order sunflower oil < camellia oil < perilla oil < palm oil, which was same as POV, TOTOX, and total nonpolar carbonyls. It was concluded that the four aldehydes nonanal, propanal, hexanal, and nonenal could be used as oxidation indicators for the four types of oils. KEYWORDS: nonpolar aldehydes/ketones, oxidation indicator, total oxidation value, fatty acid, oil oxidation, HPLC-QqQ-MS
■
molecules more easily than low temperature,13 autoxidation of lipid at room temperature actually is also the main cause of deterioration. The Schaal oven test at 60−70 °C could be used to simulate the oxidation process at room temperature.14 In addition, it was reported that there were some differences in the aldehyde compositions of oleic, linoleic, and linolenic acids during oxidation.15,16 The structure of the fatty acids, especially the position and number of double bonds in them, is the main cause. Thereby, oils having different fatty acid compositions may show different oxidation patterns. The main objectives of this study therefore were to (a) monitor the oxidation of vegetable oils with typical fatty acid compositions with limited air; (b) compare the types and amounts of nonpolar aldehyde/ketone products from these oils during accelerated oxidation; and (c) establish models from peroxide value, p-anisidine value, and aldehyde products to explore good and sensitive oxidation indicators for the evaluation of oil oxidation.
INTRODUCTION It is well-known that hydroperoxides are the primary products in lipid oxidation, but they are unstable and can be further decomposed into many secondary compounds.1 Beltran et al.2 detected 22 compounds including 7 nonpolar aldehydes from oxidized almond oils (100 °C for 20 days) by HS-SPME-GCMS. Using the same technique, Poyato et al.3 analyzed 21 nonpolar aldehydes from 7 vegetable oils (180 °C for 4 h), and Petersen et al.4 identified a total of 55 volatile oxidation compounds from rapeseed oil (40 °C for 26 days). Additionally, 32 toxic oxygenated α,β-unsaturated aldehydes in virgin olive, sunflower, and virgin linseed oils (190 °C for 20 h) were found by Guillen and Uriarte.5 Their team also analyzed many types of secondary products in corn oil6 and sunflower oil7 stored at room temperature with limited air and found that the most numerous group of volatiles was the nonpolar aldehydes. On the other hand, Seppanen and Saari Csallany8 determined simultaneously 17 nonpolar and 13 polar lipophilic aldehydes from soybean oil (185 °C for 8 h) using HPLC combined with 2,4-dinitrophenylhydrazine (DNPH) derivatization. Zhu et al.9 identified nine characteristic nonpolar aldehydes/ketones in virgin olive oil (heated to 45 °C) by UHPLC-MS and found this method provided comparable linearity and repeatability with SPME-GC. Reasonably, GC/HPLC-MS has become the common method for analyzing carbonyl compounds in oil samples, also in other lipid matrices such as ham,10 seafood,11 and biological material.12 Generally, the above studies mainly focused on the separation and identification of secondary lipid oxidation products under high-temperature condition. Although high temperature can provide energy to break double bonds in oil © 2014 American Chemical Society
■
MATERIALS AND METHODS
Oil Samples. The oils chosen for this study were palm oil, camellia oil, sunflower oil, and perilla oil with different fatty acid compositions. All four were purchased from local shops (Nanchang, China), and they complied with the legal requirements in China. Reagents and Standards. Standard solutions of the following carbonyl-DNPHs were obtained as reference solutions in acetonitrile Received: Revised: Accepted: Published: 12545
October 1, 2014 December 8, 2014 December 8, 2014 December 8, 2014 dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552
Journal of Agricultural and Food Chemistry
Article
Table 1. Initial Fatty Acid Composition (Area %), Tocopherol Content (mg/kg), Acid Value (mg KOH/g), Peroxide Value (mequiv O2/kg), and p-Anisidine Value of Vegetable Oila sample palmitic acid stearic acid oleic acid linoleic acid α-linolenic acid total SFA total MUFA total PUFA α-tocopherol β-tocopherol γ-tocopherol δ-tocopherol total tocopherol AV POV p-AV
palm oil 38.23 4.49 43.05 10.33 0.14 44.52 43.41 10.51 191.1 nd 2.26 nd 193.4 0.06 5.11 4.07
± ± ± ± ± ± ± ± ±
1.18a 0.07a 0.88b 0.35c 0.01b 1.25a 0.88b 0.35c 2.8b
± 0.07c ± ± ± ±
2.8b 0.02b 0.57a 0.69a
camellia oil 7.80 2.42 80.14 6.98 0.22 10.36 80.89 7.22 199.0 nd nd nd 198.7 0.04 2.89 2.15
± ± ± ± ± ± ± ± ±
± ± ± ±
0.01b 0.07d 0.08a 0.01d 0.01b 0.05bc 0.09a 0.03d 3.5b
3.5b 0.02b 0.65b 0.01b
sunflower oil 6.27 3.94 29.50 55.24 0.07 11.46 29.75 55.55 1106 nd 7.63 nd 1113 0.04 2.16 3.81
± ± ± ± ± ± ± ± ±
0.02c 0.07b 0.03c 0.11a 0.01b 0.07b 0.05c 0.11b 39a
± 0.23b ± ± ± ±
39a 0.01b 0.38b 0.05a
perilla oil 6.02 2.76 21.97 12.73 54.37 9.17 22.28 67.18 37.22 nd 128.3 19.51 185.0 0.72 0.41 0.22
± ± ± ± ± ± ± ± ±
0.10c 0.07c 0.25d 0.12b 0.29a 0.25c 0.14d 0.41a 4.98c
± ± ± ± ± ±
2.3a 0.20 3.8b 0.04a 0.06c 0.01c
Data are mean values ± SD, n = 3. nd, peak not detected under this analysis condition; SFA, saturated fatty acid; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; AV, acid value; POV, peroxide value; p-AV, p-anisidine value. Values followed by the same letter in the same row are not significantly different (p < 0.05). a
(15 μg/mL of each carbonyl) from Supelco Inc. (TO11/IP-6A Mix, Supelco Inc., Bellefonte, PA, USA): formaldehyde, acetaldehyde, acrolein, acetone, propionaldehyde, crotonaldehyde, butyraldehyde, benzaldehyde, isovaleraldehyde, valeraldehyde, o-tolualdehyde, mtolualdehyde, p-tolualdehyde, hexaldehyde, and 2,5-dimethylbenzaldehyde. Standard fatty acid methyl esters (FAME, GLC-463) were obtained from Nu-Chek Prep Inc. (Elysian, MN, USA). Tocopherol standards (α, β, γ, and δ) were purchased from Sigma (Sigma Chemicals, Shanghai, China). Methanol, n-hexane, acetonitrile, and isopropanol (HPLC grade) were purchased from Merck (Darmstadt, Germany). Water was purified using a Milli-Q system from Millipore (Bedford, MA, USA). All other solvents were of analytical reagent grade. Potassium hydroxide, phenolphthalein, ethyl alcohol, acetic acid, isooctane, soluble starch, potassium iodide, sodium thiosulfate, and panisidine were of analytical grade. Measurements of Acid Value (AV), Peroxide Value (POV), and p-Anisidine Value (p-AV). AOCS official methods were used to determine acid value (method Cd 3d-63), peroxide value (method Cd 8-53), and p-anisidine value (method Cd 18-90) of oil samples. Total oxidation value (TOTOX) was calculated as follows:17
ionization detector. Hydrogen was used as the carrier gas at a flow rate of 30 mL/min. The oven temperature program (86 min in total) was as follows: The initial temperature was set at 45 °C for 4 min, increased to 175 °C at a rate of 13 °C/min, maintained for 27 min, further raised to 215 °C at a rate of 4 °C/min, and finally kept at 215 °C for 35 min.20 The injector and detector were kept at 250 °C. Analysis of all peaks was accomplished by comparison of their retention time with FAME standards. Area under each fatty acid peak relative to the total area of all fatty acid peaks was used to quantify the fatty acids identified,21,22 and the obtained results are reported as percentage of fatty acid. Accelerated Oxidation Assay. A Schaal oven test was used to accelerate the oxidation process in different vegetable oils.23 The modified protocol followed AOCS (1998) Recommended Practice Cg 5-97 (AOCS cg 5-97). In total, 500 g of oils was weighed into 500 mL glass bottles that were loosely capped. Bottles were kept randomly inside an oven maintained at 62 ± 1 °C without light to accelerate the oxidation (WGS201, Hongzhan Products Inc., Shanghai, China). After intervals of 7 days of heating, samples were removed into bottles. Then they were topped immediately with nitrogen, capped tightly, and kept frozen (−20 °C) until analysis. Analyses were conducted on the same day for some sensitive indices such as POV, p-AV, and secondary oxidative products or for other indices within 2 days of removal from the oven. Synthesis of DNPH Derivatives. DNPH derivatives of aldehyde/ ketone products in oxidized oils were synthesized as follows: Oil sample (0.1−1.0 g) was put into a 10 mL volumetric flask, and the flask was filled to volume with isopropanol containing DNPH (3 g/L) and hydrochloric acid (3%).8,24 The test tube was shaken and kept at 40 °C for 1 h to accelerate derivatization reaction. Then it was cooled in water and centrifuged at 4200 rpm for 5 min. The sample was then filtered through a 0.22 μm membrane syringe filter and injected into the HPLC column. Isolation and Identification of DNPH Derivatives by HPLCESI-QqQ-MS. An Agilent 1260 series HPLC system equipped with a diode array detector (DAD) was used to analyze samples. Separation of DNPH derivatives was achieved on a Zorbax Edipse plus-C18 column (4.6 mm × 25 cm, 5 μm, Agilent Technologies, Shanghai, China). The column was thermostatically controlled at 30 °C. The flow rate was set to 1 mL/min, and the injection volume was 5 μL. The mobile phase consisted of two solvents: methanol (A, 100%) and Milli-Q water (B, 100%). The solvent gradient in volumetric ratios was set as follows: 0−15 min, 70−75% A; 15−45 min, 75−100% A; and held at 100% A for an additional 2 min. There was a 3 min postrun,
TOTOX = 2POV + p‐AV Tocopherols Analysis by HPLC. The concentration of tocopherols in oil samples was determined according to the method reported by Hu et al.18 Approximately 1 g of oil sample was put into a volumetric flask and made up to 10 mL with HPLC grade hexane. After filtering through a 0.45 μm filter, 3 μL of sample was directly injected into an Agilent 1100 series HPLC system equipped with a fluorometric detector. The excitation wavelength was set at 295 nm, and the emission wavelength was set at 325 nm. The column was a Hypersil ODS2 C18 column (4.6 mm × 15 cm, 5 μm, Elite, Dalian, China). The mobile phase was methanol/water (98:2, v/v), and the flow rate was 0.8 mL/min. The corresponding standards (α-, β-, γ-, and δ-tocopherol) for identification and quantification were prepared separately in a 100 mL volumetric flask with hexane, and the final concentrations of α-, β-, γ-, and δ-tocopherol were 100, 20, 15, and 80 μg/mL, respectively. Then they were serially diluted to different concentration gradients for generating the standard curves. Fatty Acids Analysis by GC. Fatty acid compositions of vegetable oils were analyzed by gas chromatography. Samples were methylated as described by Cruz-Hernandez et al.19 The FAMEs were identified using a capillary column of fused silica (100 m × 0.25 mm × 0.20 μm) coated with 100% cyanopropyl polysiloxane (CP-Sil 88, Chrompack, Agilent Technologies, Shanghai, China) equipped with a flame 12546
dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552
Journal of Agricultural and Food Chemistry
Article
which brings it back to the starting condition. The ultraviolet (UV) absorbance values of the peaks were collected between 200 and 620 nm using a DAD and monitored at wavelengths of 360 nm. A triple-quadrupole mass spectrometer (6430 QqQ LC/MS system; Agilent Technologies) equipped with an orthogonal electrospray ionization (ESI) source was used to identify the derivatives. Negative ion mode was selected for data collection.25,26 MS parameters were set as follows: sheath gas was nitrogen and collision gas was helium; drying gas flow rate, 11.0 L/min; drying gas temperature, 300 °C; nebulizing gas pressure, 15 psi; capillary voltage, 4 kV. Some of the ketones, alkanals, alkenals, and alkadienals were identified by using standards. Others were tentatively identified according to their retention times, mass spectra, and matching with mass spectra by the software MassHunter Acquisition B.03.01, Qualitative Analysis B.03.01 and Quantitative Analysis B.03.02 (Agilent Technologies). Quantification of Carbonyls. To have a more complete picture of the carbonyls present in the four oils and of their evolution with heating time, their quantification was expressed as peak area per gram of oil acquired by HPLC chromatogram. Statistical Analysis. Data are expressed as the mean value ± standard deviation. One-way ANOVA was used to compare the differences between mean values. A level of probability at p < 0.05 was set as statistically significant. Regression analysis was used to determine the correlation of two data sets. All of the data were performed with SPSS 13.0 software for Windows (SPSS Inc., Chicago, IL, USA).
Figure 1. Initial and final total oxidation value (TOTOX) for four oils heated at 62 °C with limited air for 35 days. TOTOX = 2POV + p-AV. Data are mean values ± SD, n = 3. Values followed by the same letter in each set are not significantly different (p < 0.05).
perilla oil (14 types) lacked some types of compounds. Twenty carbonyls identified in camellia oil were ranked as retention time of peak: ethanal, 2-propenal, acetone, propanal, pentanal, 2-hexenal, hexanal, 2-hexanone, 2,5-dimethylbenzaldehyde, 2heptenal, heptanal, 2-octenal, octanal, octanone, 2-nonenal, nonanal, 2-nonanone, 2-decenal, decanal, and decanone, respectively. The above 19 types of nonpolar carbonyls, except 2-decenal, were also detected in palm oil. Sunflower oil contained the most similar components of nonpolar carbonyls to palm oil, but acetone was not detected in sunflower oil and 2,4-decadienal was not detected in palm oil. Fourteen types of carbonyls identified in perilla oil were ethanal, acetone, propanal, butanal, 2-pentenal, 2-hexenal, hexanal, 2,4-heptadienal, 2,5-dimethylbenzaldehyde, 2,4-nonadienal, 2-octenal, octanal, 2-nonenal, and nonanal. These compounds represented groups of secondary oxidation products resulting mainly from the oxidative deterioration of unsaturated fatty acids.3 Although the four oil samples contained the most similar components of nonpolar carbonyls, 2-decenal was detected only in camellia oil, and 2,4-decadienal was found only in sunflower oil. Butanal, 2-pentenal, 2,4-heptadienal, and 2,4nonadienal were observed only in perilla oil. Guillén and Uriarte5 reported that aldehydes and relative carbonyls generated in lipids during oxidation were substantially different in quantity and type depending on the composition of the lipid matrix. Hence, it was estimated that 2-decenal might be derived mainly from oleic acid, and 2,4-decadienal might be from linoleic acid. Linolenic acid might be the parent fatty acid of butanal, 2-pentenal, 2,4-heptadienal, and 2,4-nonadienal. Similar results have been reported by other researchers.11,27 For example, there were high levels of 2-decenal in high-oleic rapeseed oil4 and high-oleic sunflower oil.28 The distribution of nonpolar carbonyls in four vegetable oils is shown in Figure 3. Ketones, alkanals, and alkenals were identified in camellia and palm oils during the whole experimental period. Alkanals were the major compounds in camellia oil on the 35th day (72% of total carbonyls), followed by ketones (15%). Both alkanals and alkenals were the main carbonyls in palm oil (49 and 24%) and sunflower oil (45 and 41%). For perilla oil, the major products were alkanals (44%), followed by alkenals (19%) and alkadienals (16%). These carbonyl compounds with different double bonds in samples
■
RESULTS AND DISCUSSION Initial Properties of Oils. Initial fatty acid composition, tocopherol content, acid value, peroxide value, and p-anisidine value of oil samples are shown in Table 1. These oils contained the typical proportions of fatty acids: palm oil contained higher percentages of palmitic acid (38.23%) and oleic acid (43.05%); camellia oil was rich in oleic acid (80.14%), whereas sunflower oil and perilla oil were rich in linoleic acid (55.24%) and linolenic acid (54.37%), respectively. The analysis of tocopherols showed that sunflower oil contained the highest concentration of total tocopherols (the sum of α-, β-, γ-, and δ-tocopherols), at 1113 mg/kg. Palm, camellia, and perilla oils contained a similar amount of total tocopherols (193.4, 198.7, and 185.0 mg/kg, respectively). Additionally, palm, camellia, and sunflower oils had high proportions of α-tocopherol (>98%), whereas perilla oil had 69% of γ-tocopherol and only 20% of α-tocopherol. Before accelerated oxidation assay, analysis of AV, POV, and p-AV showed that the four oil samples were so-called fresh oils. All complied with the legal dietary requirements of China. The maximal AV was 0.72 mg KOH/g in perilla oil, and maximal POV and p-AV were 5.11 mequiv O2/kg and 4.07 in palm oil. Changes of Total Oxidation Value during Oxidation. The TOTOX is often used to estimate oxidative deterioration of lipids, as it has the advantage of combining the amounts of primary oxidation products (hydroperoxides) with secondary products (principally alkenals and alkadienals) in fats or oils.17 As shown in Figure 1, the highest increment of TOTOX at 62 °C with limited air for 35 days was found in sunflower oil (814), rising from 8.12 to 822, followed by camellia oil (247) and perilla oil (223). Palm oil gave rise to the lowest increment of TOTOX, only 155. That means the order of oxidation stability was sunflower oil < camellia oil < perilla oil < palm oil. Comparison of Carbonyl Products Generated in Oils. Identification of carbonyls in these oils heated at 62 °C with limited air for 35 days (Figure 2) showed that camellia oil (20 types), sunflower oil (19 types), and palm oil (19 types) contained numerous nonpolar carbonyl compounds, whereas 12547
dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552
Journal of Agricultural and Food Chemistry
Article
Figure 2. Initial and final nonpolar aldehydes/ketones (peak area/g oil) identified for four oils heated at 62 °C with limited air by HPLC-QqQ-MS. 1, ethanal; 2, 2-propenal; 3, acetone; 4, propanal; 5, butanal; 6, 2-pentenal; 7, pentanal; 8, 2-hexenal; 9, hexanal; 10, 2,4-heptadienal; 11, 2-hexanone; 12, 2,5-dimethylbenzaldehyde; 13, 2-heptenal; 14, heptanal; 15, 2,4-nonadienal; 16, 2-octenal; 17, octanal; 18, octanone; 19, 2-nonenal; 20, nonanal; 21, 2,4-decadienal; 22, 2-nonanone; 23, 2-decenal; 24, decanal; 25, decanone. They were ranked as retention time of peak. The identification or quantification was done according to the retention times and the m/z ratio reported in the author’s previous paper.41 Data are mean values ± SD, n = 3.
Figure 3. Distribution of nonpolar aldehydes/ketones (peak area/g oil) in four oils heated at 62 °C with limited air. Data are mean values ± SD, n = 3.
initiation, propagation, and termination.1 Hydroperoxides as the primary oxidation products are first formed. However, they are unstable and rapidly decomposed to a variety of volatile compounds, including aldehydes, ketones, hydrocarbons, and other secondary products.29,30 Aldehydes, either saturated or
were mainly owing to the fatty acid chains with different double bonds.27 Possible Pathways of Fatty Acid Decomposition and Aldehyde/Ketone Formation. The autoxidation of lipid or the reaction of free radical chain mainly includes three steps: 12548
dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552
Journal of Agricultural and Food Chemistry
Article
Figure 4. Nonanal, hexanal, nonenal, and propanal, from oleic, linoleic, and linolenic acyl group oxidation, identified in the HPLC-QqQ-MS after derivatization.
Figure 5. Oxidative volatile profiles obtained for four oils by plotting the content of five major carbonyl compounds (chosen from Figure 2) versus time of thermal treatment at 62 °C with limited air. Data are mean values ± SD, n = 3. Values followed by the same letter in the same assay are not significantly different (p < 0.05).
unsaturated, are considered as the main volatile secondary compounds formed in any oil oxidation process.6 Vegetable oils are rich in unsaturated fatty acids, especially oleic, linoleic, and linolenic acids, which are the main causes of deterioration. It was reported that the relative oxidation rates of methyl oleate, linoleate, and linolenate were 1:10.3:21.6.31 Different types of unsaturated fatty acids would generate various types of hydroperoxides and aldehydes/ketones.32,33 Possible formation pathways of nonanal, propanal, hexanal, and 2-nonenal are shown in Figure 4. Nonanal is produced probably by
breakdown of oleate 10-OOH; hexanal and 2-nonenal are probably from linoleate 13- and 10-OOH; propanal is probably by decomposition of linolenate 16-OOH. Additionally, decanal, 2-decenal, and octanal from oleate 8-, 9-, and 11-OOH, heptanal, 2-octenal, and pentanal originated by decomposition of linoleate 11- and 13-OOH; 2-hexenal and 2,4-heptadienal from linolenate 13- or 12-OOH were also produced.32 The breakdown of acyl group chains in unsaturated fatty acids was the main formation pathway of aldehydes/ketones. According to the distribution of aldehydes/ketones shown in 12549
dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552
Journal of Agricultural and Food Chemistry
Article
Figure 6. Linear correlations of TOTOX versus carbonyl compounds (peak area/g oil) during 35 days of accelerated oxidation at 62 °C.
°C with limited air. It is very obvious that oil samples with different fatty acid compositions had their characteristic carbonyl compounds. For camellia oil (oleic acid mainly), nonanal was the most abundant carbonyl on the 35th day, followed by octanal and decanal. For sunflower oil (linoleic acid mainly), 2-nonenal and hexanal (measured as area count) showed the highest increments during 35 days of oxidation assay, increasing by 8.93 × 103 and 7.97 × 103, respectively. Hence, the two carbonyls might be good oxidation indicators for evaluating the oxidation state of sunflower oil. For perilla oil (linolenic acid mainly), propanal showed a more remarkable growth compared with other products, which might be also the key oxidation marker. Although 2-nonenal and hexanal were the most abundant carbonyls in palm oil, they showed slow increases after 14 days of oxidation. On the other hand, the concentration of nonanal was sharply increased after 14 days of oxidation and showed a high level on the 35th day. The increment tendency of nonanal, octanal, and decanal in palm oil had the same patterns as that in camellia oil. These results indicated linoleic acid was first oxidized and generated 2nonenal and hexanal and then oleic acid was oxidized and formed nonanal (octanal and decanal) in palm oil. The changes of the above characteristic carbonyls also played an important role in the evaluation of oil deterioration. By choosing 2-nonenal or hexanal as the oxidation indicators for palm and sunflower oils, nonanal for camellia oil, and propanal for perilla oil, it was found that sunflower oil had the highest indicator increment (8.93 × 103 or 7.97 × 103), followed by camellia oil (5.36 × 103), perilla oil (1.65 × 103), and palm oil
Figures 2 and 3, these carbonyls derived from linoleic acyl groups showed the highest abundances, followed by those derived from oleic acyl and much lower proportions of those derived from linolenic acyl. Similar results were also found in the studies of many other studies on oil oxidation.6,34 Additionally, the formation of all secondary products depends not only on the intrinsic property of lipid but also on the conditions under which the process takes place, such as temperature, amount of air, and type of bottle.35,36 Changes of Carbonyls during Oxidation. Before oxidation assay (day 0), only 10 types of carbonyl compounds in camellia oil and 9 types in palm oil were found, and sunflower and perilla oils showed a lower number of carbonyls, 6 and 5 types, respectively (Figure 2). It is well-known that the thermoxidative degradation of unsaturated fatty acids may result in the presence of many other secondary products, and their concentrations increased with time. As shown in Figure 3, the highest increment of total carbonyls (the sum of all detected aldehydes and ketones) measured as area count during 35 days of oxidation was detected in sunflower oil (2.57 × 104) from 3.33 × 102 to 2.60 × 104, followed by camellia oil (1.92 × 104). By contrast, there were lower increments in perilla oil (5.54 × 103) and palm oil (5.21 × 103). That means the order of oxidation stability was sunflower oil < camellia oil < perilla oil < palm oil. Interestingly, the order was the same as that in comparison of TOTOX of samples. Figure 5 shows oxidative volatile profiles obtained for four oils by plotting five major carbonyl compounds’ content (chosen from Figure 2) versus time of thermal treatment at 62 12550
dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552
Journal of Agricultural and Food Chemistry
Article
(1.04 × 103 or 1.05 × 103). The order of oxidation stability was sunflower oil < camellia oil < perilla oil < palm oil. Hence, the yield of indicator could be used to demonstrate the oxidation state of oil. Meanwhile, the correct choice of oxidation indicator was necessary. Regression Analysis. As shown in Figure 6, good linear correlations were observed between the total nonpolar carbonyl amount (y) and TOTOX (x) during 35 days of accelerated oxidation: palm oil (y = 34.4x + 312, R2 = 0.97), camellia oil (y = 75.5x − 783, R2 = 0.95), sunflower oil (y = 30.8x + 1680, R2 = 0.96), and perilla oil (y = 26.4x + 715, R2 = 0.89). These linear trends suggested that the amount of nonpolar carbonyls could be used to estimate oxidative deterioration of oils as TOTOX. To our knowledge, this is the first report on this linear relationship between nonpolar carbonyl amount and TOTOX during lipid oxidation. TOTOX was calculated by POV and p-AV, and also linear correlations were found between nonpolar carbonyl amount and POV in the four oils (R2 = 0.92−0.97; Supporting Information, Figure S1). In a sense, generation of carbonyl compounds may be closely related to the production of hydroperoxides. Richards et al.37 established a model relating ln[POV] to the volatile oxidation products: ln[POV] = −2.50 + 0.308 × ln[hexanal] + 0.229 × ln[trans,trans-2,4-heptadienal], but R2 was much lower (only 0.26). Hence, the total amount of nonpolar carbonyls was a good index for oxidation assessment, despite their difficult and time-consuming determination. Generally, POV, p-AV, and TBARS are the common and foundational oxidation indices in lipid analysis. Actually, hexanal has become a well-known indicator owing to its being a major secondary oxidation product in lipid.38−40 However, the profile of carbonyl compounds may be closely linked to the fatty acid composition of lipid. In our previous study, we found that oleic, linoleic, and linolenic acids had their characteristic carbonyl compounds during accelerated oxidation.41 Thus, it is necessary and significant to choose corresponding oxidation indicators for oil samples with different fatty acids. It was found that both hexanal and 2-nonenal were good oxidation indicators for sunflower and palm oils, as both had good linear relationship with TOTOX (Figure 6): hexanal = 7.10 TOTOX + 342, R2 = 0.78, and nonenal = 6.65 TOTOX + 276, R2 = 0.84 for palm oil; hexanal =9.56 TOTOX + 913, R2 = 0.90, and nonenal = 10.5 TOTOX + 691, R2 = 0.95 for sunflower oil. For camellia oil, nonanal was a good indicator because of the good linear correlation (nonanal = 21.6 TOTOX − 595, R2 = 0.92). Propanal was a good indicator for perilla oil, and also a good linear relationship was found between the amount of propanal and TOTOX: propanal = 8.10 TOTOX + 75.0, R2 = 0.90. Hence, the yield of the indicator was a key index for evaluating the oxidation state of vegetable oil. In summary, in this work, palm, camellia, sunflower, and perilla oils were selected in oxidation assays in consideration of their typical fatty acid compositions. Camellia (20 types), sunflower (19 types), and palm (19 types) oils were found to have more types of nonpolar carbonyls than perilla oil (14 types) on the 35th day. Besides, alkanal contained a higher proportion of nonpolar carbonyl than alkenal and alkadienal in the four oils. Interestingly, the four oils had their own characteristic carbonyl products, which could be used as markers to monitor the oxidation state of oils. Nonanal was a good oxidation indicator for camellia oil (oleic acid mainly); hexanal and nonenal were good oxidation indicators for
sunflower oil (linoleic acid mainly) and palm oil (palmitic and oleic acids mainly); propanal was a good oxidation indicator for perilla oil (linolenic acid mainly). In addition, these indicators had good linear correlations with their corresponding TOTOX and the amounts of total nonpolar carbonyls in oils (R2 > 0.8) during 35 days of accelerated oxidation. According to the change patterns of POV, TOTOX, total nonpolar carbonyls, and oxidation indicator, the corresponding order of oxidation stability was uniformly sunflower oil < camellia oil < perilla oil < palm oil. Hence, the yield of indicator was a key index for evaluating the oxidation state of vegetable oil. In addition, the determination of indicator was a novel approach to monitor oil oxidation. They played an important role as TOTOX or POV in the oxidation assessment of lipid.
■
ASSOCIATED CONTENT
* Supporting Information S
Figures S1 and S2. This material is available free of charge via the Internet at http://pubs.acs.org.
■
AUTHOR INFORMATION
Corresponding Authors
*(Y.F.) Phone/fax: +86 791 88304402. E-mail: fanyawei@tom. com. Mail: State Key Laboratory of Food Science and Technology, Nanchang University, 235 Nanjing East Road, Nanchang 330047, Jiangxi, China. *(Z.-Y.D.) E-mail:
[email protected] Funding
This study was supported by the National Natural Science Foundation of China (31460427; 31071561) and Jiangxi Provincial Natural Science Foundation of China (20132BAB204001). Notes
The authors declare no competing financial interest.
■
ABBREVIATIONS USED AV, acid value; POV, peroxide value; p-AV, p-anisidine value; TOTOX, total oxidation value; DNPH, 2,4-dinitrophenylhydrazine; FAME, fatty acid methyl ester(s); DAD, diode array detector; UV, ultraviolet; ESI, electrospray ionization
■
REFERENCES
(1) Choe, E.; Min, D. B. Mechanisms and factors for edible oil oxidation. Compr. Rev. Food Sci. Food Saf. 2006, 5, 169−186. (2) Beltrán, A.; Ramos, M.; Grané, N.; Martín, M. L.; Garrigós, M. C. Monitoring the oxidation of almond oils by HS-SPME-GC-MS and ATR-FTIR: application of volatile compounds determination to cultivar authenticity. Food Chem. 2011, 126, 603−609. (3) Poyato, C.; Ansorena, D.; Navarro-Blasco, I.; Astiasarán, I. A novel approach to monitor the oxidation process of different types of heated oils by using chemometric tools. Food Res. Int. 2014, 57, 152− 161. (4) Petersen, K. D.; Kleeberg, K. K.; Jahreis, G.; Busch-Stockfisch, M.; Fritsche, J. Comparison of analytical and sensory lipid oxidation parameters in conventional and high-oleic rapeseed oil. Eur. J. Lipid Sci. Technol. 2012, 114, 1193−1203. (5) Guillén, M. D.; Uriarte, P. S. Aldehydes contained in edible oils of a very different nature after prolonged heating at frying temperature: presence of toxic oxygenated α,β unsaturated aldehydes. Food Chem. 2012, 131, 915−926. (6) Goicoechea, E.; Guillén, M. D. Volatile compounds generated in corn oil stored at room temperature. Presence of toxic compounds. Eur. J. Lipid Sci. Technol. 2014, 116, 395−406.
12551
dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552
Journal of Agricultural and Food Chemistry
Article
(7) Guillen, M. D.; Goicoechea, E. Formation of oxygenated α,βunsaturated aldehydes and other toxic compounds in sunflower oil oxidation at room temperature in closed receptacles. Food Chem. 2008, 111, 157−164. (8) Seppanen, C. M.; Saari Csallany, A. Simultaneous determination of lipophilic aldehydes by high-performance liquid chromatography in vegetable oil. J. Am. Oil Chem. Soc. 2001, 78, 1253−1260. (9) Zhu, H.; Li, X.; Shoemaker, C. F.; Wang, S. C. Ultrahigh performance liquid chromatography analysis of volatile carbonyl compounds in virgin olive oils. J. Agric. Food Chem. 2013, 61, 12253−12259. (10) Narváez-Rivas, M.; Gallardo, E.; León-Camacho, M. Chemical changes in volatile aldehydes and ketones from subcutaneous fat during ripening of Iberian dry-cured ham. Prediction of the curing time. Food Res. Int. 2014, 55, 381−390. (11) Mathew, S.; Grey, C.; Rumpunen, K.; Adlercreutz, P. Analysis of carbonyl compounds in sea buckthorn for the evaluation of triglyceride oxidation, by enzymatic hydrolysis and derivatisation methodology. Food Chem. 2011, 126, 1399−1405. (12) Spiteller, G.; Kern, W.; Spiteller, P. Investigation of aldehydic lipid peroxidation products by gas chromatography-mass spectrometry. J. Chromatogr., A 1999, 843, 29−98. (13) Wang, S.; Hwang, H.; Yoon, S.; Choe, E. Temperature dependence of autoxidation of perilla oil and tocopherol degradation. J. Food Sci. 2010, 75, C498−C505. (14) Iqbal, S.; Bhanger, M. Stabilization of sunflower oil by garlic extract during accelerated storage. Food Chem. 2007, 100, 246−254. (15) Kawahara, F.; Dutton, H. Volatile cleavage products of autoxidized soybean oil. J. Am. Oil Chem. Soc. 1952, 29, 372−377. (16) Gaddis, A.; Ellis, R.; Currie, G. Carbonyls in oxidizing fat. V. The composition of neutral volatile monocarbonyl compounds from autoxidized oleate, linoleate, linolenate esters, and fats. J. Am. Oil Chem. Soc. 1961, 38, 371−375. (17) Shahidi, F.; Wanasundara, U. N. Chapter 14. Methods for measuring oxidative rancidity in fats and oils. In Food Lipids: Chemistry, Nutrition and Biotechnology, 2nd ed.; CRC Press: Boca Raton, FL, USA, 2002; pp 387−403. (18) Hu, J. N.; Zhang, B.; Zhu, X. M.; Li, J.; Fan, Y. W.; Liu, R.; Tang, L.; Lee, K. T.; Deng, Z. Y. Characterization of medium-chain triacylglycerol (MCT)-enriched seed oil from Cinnamomum camphora (Lauraceae) and its oxidative stability. J. Agric. Food Chem. 2011, 59, 4771−4778. (19) Cruz-Hernandez, C.; Deng, Z. Y.; Zhou, J.; Hill, A. R.; Yurawecz, M. P.; Delmonte, P.; Mossoba, M. M.; Dugan, M. E. R.; Kramer, J. K. G. Methods for analysis of conjugated linoleic acids and trans-18:1 isomers in dairy fats by using a combination of gas chromatography, silver-ion thin-layer chromatography/gas chromatography, and silver-ion liquid chromatography. J. AOAC Int. 2004, 87, 545−562. (20) Lei, L.; Li, J.; Luo, T.; Fan, Y. W.; Zhang, B.; Ye, J.; Ye, H.; Sun, Y.; Deng, Z. Y. Predictable effects of dietary lipid sources on the fatty acids compositions of four 1-year-old wild freshwater fish from Poyang Lake. J. Agric. Food Chem. 2012, 61, 210−218. (21) Cerretani, L.; Maggio, R. M.; Barnaba, C.; Toschi, T. G.; Chiavaro, E. Application of partial least square regression to differential scanning calorimetry data for fatty acid quantitation in olive oil. Food Chem. 2011, 127, 1899−1904. (22) Ramadan, M. F.; Wandan, K. M. M. Blending of corn oil with black cumin (Nigella sativa) and coriander (Coriandrum sativum) seed oils: impact on functionality, stability and radical scavenging activity. Food Chem. 2012, 132, 873−879. (23) Li, H.; Fan, Y. W.; Li, J.; Tang, L.; Hu, J. N.; Deng, Z. Y. Evaluating and predicting the oxidative stability of vegetable oils with different fatty acid compositions. J. Food Sci. 2013, 78, H633−H641. (24) Endo, Y.; Li, C. M.; Tagiri-Endo, M.; Fujimoto, K. A modified method for the estimation of total carbonyl compounds in heated and frying oils using 2-propanol as a solvent. J. Am. Oil Chem. Soc. 2001, 78, 1021−1024.
(25) Pang, X.; Shi, X.; Mu, Y.; He, H.; Shuai, S.; Chen, H.; Li, R. Characteristics of carbonyl compounds emission from a diesel-engine using biodiesel-ethanol-diesel as fuel. Atmos. Environ. 2006, 40, 7057− 7065. (26) Kölliker, S.; Oehme, M.; Dye, C. Structure elucidation of 2,4dinitrophenylhydrazone derivatives of carbonyl compounds in ambient air by HPLC/MS and multiple MS/MS using atmospheric chemical ionization in the negative ion mode. Anal. Chem. 1998, 70, 1979− 1985. (27) Belitz, H. D.; Grosch, W.; Schieberle, P. Chapter 14. Edible fats and oils. In Food Chemistry, 4th ed.; Springer: Berlin Heidelberg, Germany, 2009; pp 640−669. (28) Petersen, K. D.; Kleeberg, K. K.; Jahreis, G.; Fritsche, J. Assessment of the oxidative stability of conventional and high-oleic sunflower oil by means of solid-phase microextraction-gas chromatography. Int. J. Food Sci. Nutr. 2012, 63, 160−169. (29) Sjovall, O.; Kuksis, A.; Kallio, H. Tentative identification and quantification of TAG core aldehydes as dinitrophenylhydrazones in autoxidized sunflowerseed oil using reversed-phase HPLC with electrospray ionization MS. Lipids 2003, 38, 1179−1190. (30) Guillen, M. D.; Goicoechea, E. Oxidation of corn oil at room temperature: primary and secondary oxidation products and determination of their concentration in the oil liquid matrix from 1H nuclear magnetic resonance data. Food Chem. 2009, 116, 183−192. (31) Fatemi, S. H.; Hammond, E. G. Analysis of oleate, linoleate and linolenate hydroperoxides in oxidized ester mixtures. Lipids 1980, 15, 379−385. (32) Morales, M.; Rios, J.; Aparicio, R. Changes in the volatile composition of virgin olive oil during oxidation: flavors and off-flavors. J. Agric. Food Chem. 1997, 45, 2666−2673. (33) Wheatley, R. A. Some recent trends in the analytical chemistry of lipid peroxidation. TrAC−Trends Anal. Chem. 2000, 19, 617−628. (34) Guillén, M. D.; Uriarte, P. S. Simultaneous control of the evolution of the percentage in weight of polar compounds, iodine value, acyl groups proportions and aldehydes concentrations in sunflower oil submitted to frying temperature in an industrial fryer. Food Control 2012, 24, 50−56. (35) Guillén, M. D.; Ruiz, A. Oxidation process of oils with high content of linoleic acyl groups and formation of toxic hydroperoxyand hydroxyalkenals. A study by 1H nuclear magnetic resonance. J. Sci. Food Agric. 2005, 85, 2413−2420. (36) Guillén, M. D.; Goicoechea, E. Detection of primary and secondary oxidation products by fourier transform infrared spectroscopy (FTIR) and 1H nuclear magnetic resonance (NMR) in sunflower oil during storage. J. Agric. Food Chem. 2007, 55, 10729−10736. (37) Richards, A.; Wijesundera, C.; Salisbury, P. Evaluation of oxidative stability of canola oils by headspace analysis. J. Am. Oil Chem. Soc. 2005, 82, 869−874. (38) Sanches-Silva, A.; Rodriguez-Bernaldo de Quiros, A.; LópezHernández, J.; Paseiro-Losada, P. Determination of hexanal as indicator of the lipidic oxidation state in potato crisps using gas chromatography and high-performance liquid chromatography. J. Chromatogr., A 2004, 1046, 75−81. (39) Brunton, N. P.; Cronin, D. A.; Monahan, F. J.; Durcan, R. A comparison of solid-phase microextraction (SPME) fibres for measurement of hexanal and pentanal in cooked turkey. Food Chem. 2000, 68, 339−345. (40) Panseri, S.; Soncin, S.; Chiesa, L. M.; Biondi, P. A. A headspace solid-phase microextraction gas-chromatographic mass-spectrometric method (HS-SPME-GC/MS) to quantify hexanal in butter during storage as marker of lipid oxidation. Food Chem. 2011, 127, 886−889. (41) Cao, J.; Zou, X. G.; Deng, L.; Fan, Y. W.; Li, H.; Li, J.; Deng, Z. Y. Analysis of nonpolar lipophilic aldehydes/ketones in oxidized edible oils using HPLC-QqQ-MS for the evaluation of their parent fatty acids. Food Res. Int. 2014, 64, 901−907.
12552
dx.doi.org/10.1021/jf5047656 | J. Agric. Food Chem. 2014, 62, 12545−12552