Edible fat (I.V.=54 g I₂/100 g) was purchased from Jiuyuan Oils & Fats Co., Ltd. (Tianjin, China). Olive oil (I.V.=80 g I₂/100 g) was purchased from Sinograin Oils Co., Ltd. (Tianjin, China). Sunflower oil 1# (I.V.=131 g I₂/100 g) was purchased from Sinograin Oils Co., Ltd. (Tianjin, China). Sunflower oil 2# (I.V.=132 g I₂/100 g) was purchased from Standard Food Co., Ltd. (Shanghai, China). Camellia seed oil (I.V.=85 g I₂/100 g) was purchased from COFCO Co., Ltd. (Xinjiang, China). Corn oil (I.V.=120 g I₂/100 g) was purchased from Xiwang Food Co., Ltd. (Shandong, China). Soybean oil 1# (I.V.=127 g I₂/100 g) and soybean oil 2# (I.V.=128 g I₂/100 g) were purchased from Jiusan group soybean Technology Co., Ltd. (Tianjin, China) and Wilmar International Limited, respectively. Sesame oil (I.V.=107 g I₂/100 g) was obtained from a local market. Petroleum ether (99%, 60–90 °C) was obtained from Precision Chemical Reagent Factory (Tianjin, China). Sodium sulfate anhydrous (> 99%), methanol (HPLC grade) and n-hexane (HPLC grade) were purchased from Kemiou Chemical Reagent Co., Ltd. (Tianjin, China). Sodium (> 99.5%) was obtained from Kelong Chemical Reagent Factory (Chengdu, China). 1,4-dioxane (99%) was acquired from Fuyu Fine Chemical Co., Ltd. (Tianjin, China). Hydrochloric acid (99%) was obtained from Chemical Reagent Factory (Luoyang, China). Sodium hydroxide (99%) was obtained from Yongda Chemical Reagent Co., Ltd. (Tianjin, China). Butyl-hydroxyanisol (BHA, 98%), butyl-hydroxytoluene (BHT, > 99%), t-butyl-hydroquinone (TBHQ, 98%) and sodium p-toluenesulfinate (anhydrous, 98%) were obtained from Aladdin Bio-Chem Tenchnology Co., Ltd. (Shanghai, China). All materials were used as received without further treatment.

p-Toluenesulfinic acid was prepared by the acidification of sodium p-toluenesulfinate (Snyder et al., 1982). Sodium p-toluenesulfinate was dissolved in a small amount of water and p-toluenesulfinic acid was precipitated by the addition of the calculated amount of 6 mol/L hydrochloric acid (HCl) to this solution. The precipitated acid was filtered off and washed rapidly several times with ice-cold distilled water. Finally, p-toluenesulfinic acid was thoroughly dried for 24 h at 40 °C under vacuum.

In the cis-trans isomerization experiments, 250 mg edible fat, 10 mL 1,4-dioxane and a certain amount of p-toluenesulfinic acid were added to the reactor and the resulting mixture was heated for a certain period of time under a nitrogen atmosphere. After cooling to room temperature, the solution was diluted with 25 mL of a 1 mol/L NaOH solution and then extracted with petroleum ether. The obtained organic layer was washed three times with a saturated sodium chloride solution and then the solvent was removed by a rotary evaporator under vacuum, which afforded the desired trans-lipid product.

The isomerized fat was treated by methyl esterification before the gas chromatography analysis. In detail, 2–3 drops of isomerized fat and 2 mL of n-hexane were mixed in a glass tube, and then 3 mL of 0.5 mol/L sodium methylate were added into the tube. The mixture was rigorously vibrated for 5–7 min and the mixture was then centrifuged to obtain the supernatant. The supernatant was dried by sodium sulfate anhydrous prior to the gas chromatography analysis. Analysis of the isomerized fat was performed on a Agilent Technologies 6890N gas chromatography (GC) using a 30.0 m×250 μm×0.25 μm BPX-70 capillary column and detected using a flame ionization detector (FID). The carrier was nitrogen with a flow of 1.0 mL/min. The oven was programmed from the set temperature of 170 °C to 210 °C at 2 °C/min. Injection volumes were 1 μL with the inlet set to split mode and the split ratio was 50:1. The GC injection temperature was 250 °C and the detector temperature was 300 °C. The yield of trans double bonds in the isomerized oil samples was calculated from the capillary GC data. Furthermore, the experimental results were mainly determined by the yield of total trans double bonds. The yield of total trans double bonds was calculated using Eq. (1):

Where total trans double bonds are the double bonds of the elaidic acid, trans linoleic acid and trans linolenic acid; total double bonds are the double bonds of the oleic acid, linoleic acid and linolenic acid.

The proposed mechanism for trans carbon-carbon double bond formation in edible oils using p-toluenesulfinic acid as catalyst is presented in Scheme 1. According to the previously proposed mechanism (Gibson et al., 1976), the p-toluenesulfinic acid catalyst would produce an arylthiyl radical when p-toluenesulfinic acid was heated in the reaction system. All of the radicals would promote cis-trans isomerization. Hence, the reaction heating temperature and antioxidant compounds in edible oils would both affect the isomerization reaction.

As we know, edible oils, mainly vegetable oils, contain natural antioxidants and synthetic antioxidants which would inhibit the lipid peroxidation and other radical reactions. Thus, edible oil without any antioxidants is an ideal candidate for the cis-trans isomerization reaction through a radical process. Therefore, edible fat without any antioxidants was selected as the model lipid for the isomerization reaction because the content of natural antioxidants in edible fat is scarce. In addition, the fatty acid composition of the edible fat used in this work showed that the content of unsaturated fatty acids in edible fat is about 50.0%, which includes 36.8% oleic acid and 13.1% linoleic acid. Refined edible fat was suitable to study this free radical-like cis-trans isomerization of unsaturated carbon-carbon double bonds of fatty acids to prepare trans-lipid.

Firstly, the reaction time was one of the key factors for isomerization because the radical intermediates generated by the p-toluenesulfinic acid are responsible for the fat isomerization. The effect of reaction time on the isomerization of edible fat is shown in Table 1. The isomerization was performed at 120 °C under a nitrogen atmosphere using p-toluenesulfinic acid as catalyst (oil/catalyst=250:20) to investigate the effect of reaction times (10 min, 20 min, 30 min, 60 min, 90 min and 120 min) on the isomerization process. Obviously, the generation of radical intermediates required a certain period of time and a much lower yield of TFAs (2.8%) was generated after heating for only 10 min. When the reaction time was increased to 20 min, the yield of TFAs in the fat was increased rapidly and a 72.5% yield of TFAs was detected. Prolonging the reaction time (30–90 min) led to a slight increment in TFA and a 79.2% yield of TFAs was produced in fat after heating for 90 min. Continuing to increase the reaction time (120 min) resulted in no significant changes in the yield of trans double bonds. Therefore, an operation reaction time of 90 min was chosen in the following experiments.

The effect of reaction temperature on the isomerization of edible fat was investigated and shown in Table 2. The experiments were performed under a nitrogen atmosphere using p-toluenesulfinic acid (250:20 edible fat to catalyst) as catalyst to investigate the effect of different reaction temperatures (60 °C, 80 °C, 100 °C, 120 °C and 140 °C) on the isomerization process. Based on the reaction mechanism, the generation of the free radical catalyst from p-toluenesulfinic acid required a definite temperature.

It could be observed in the results that no generation of trans double bonds was observed when the reaction was performed below 100 °C (60 °C and 80 °C). When the experiment was carried out at 100 °C for 90 min, the yield of trans double bonds in the fat could reach 79.6%. Further increasing the reaction temperature (120–140 °C) had almost no effect on the yield of trans double bonds in the fat. This is probably because the temperature was not the rate-determining step in this cis-trans isomerization reaction. In other words, it was impossible to increase the yield of trans double bonds through increasing the reaction temperature. Therefore, 100 °C was chosen as the optimum reaction temperature.

The catalyst amount had a certain impact on the cis-trans isomerization of the edible fat. The effect of the amount of catalyst in terms of different fat/catalyst mass ratios (250:5, 250:10, 250:20 and 250:40) was also investigated and the results are listed in Table 3. Heating the fat at 100 °C without any catalyst did not lead to any production of TFAs after heating for 90 min. The use of less catalyst (250:5) was not conducive to the production of trans-fat. The ratio of 250:20 led to a higher conversion of TFAs (79.6%) in the fat. Further increasing the catalyst amount (250:40) showed no improvement in the yield of trans double bonds in the fat product. Therefore, 250:20 was selected as the optimum mass ratio of fat and catalyst.

Water exists widely in oil-based food and has some impact on the reactions in food. The effect of water addition on the cis-trans isomerization of edible fat is shown in Table 4. It was evident that water addition ranging from 5–20% had no significant impact on the cis-trans isomerization of the edible fat. The total trans double bonds remained at the same level with increasing the amount of water. Because the small amount of added water probably evaporated into the gaseous state under heating (100 °C) conditions, this resulted in separating it from the reaction system. These results made this method more convenient to produce trans-lipids without extra drying or dehydration of the oil sample before the isomerization reaction.

The cis-trans isomerization of carbon-carbon double bonds in edible fat was caused by free radicals. During the reaction, p-toluenesulfinic acid cleaved thermally to radicals which resulted in the production of trans fat. Hence, the antioxidant as free radical quencher would affect the radical reaction. Considering the complexity of natural antioxidants in different edible oils, various commonly used synthetic antioxidants are investigated in the cis-trans isomerization of carbon-carbon double bonds in edible fat. BHA (butyl-hydroxyanisol), BHT (butyl-hydroxytoluene) and TBHQ (t-butyl-hydroxyquinone) are the most commonly used synthetic antioxidants for delaying the lipid oxidation process. Table 5 shows the effect of different antioxidants on the isomerization of edible fat. The results indicate that introducing 50–200 mg/kg BHA in the reaction system had no significant impact on the isomerization of the edible fat. However, the addition of 200 mg/kg BHT completely inhibited the isomerization of edible fat at 100 °C and no TFA was detected. Increasing the reaction temperature from 100 °C to 120 °C, the cis-trans isomerization took place smoothly with the addition of 200 mg/kg BHT and afforded the TFAs a yield of 79.3%. Because the high radical inhibition activity of BHT was most likely obtained at 100 °C, the isomerization of edible fat with BHT could be implemented by elevating the reaction temperature to 120 °C. Moreover, introducing 200 mg/kg TBHQ into the reaction system had no significant impact on the isomerization of edible fat at 100 °C. Therefore, cis-trans isomerization at 100 °C or 120 °C could efficiently take place in edible fat containing no more than 200 mg/kg synthetic antioxidants.

The results of cis-trans isomerized unsaturated fatty acids in edible fat monitored by GC are shown in Figure 1. The A and B diagrams in Figure 1 show the fatty acid compositions by GC before and after the cis-trans isomerization, respectively. It was clear that trans double bonds were formed. The results were achieved under the optimum conditions which were conducted at 100 °C for 90 min. The fatty acid composition of isomerized edible fat (Table 6) showed that the yields of isomerized double bonds of oleic acid and linoleic acid were 79.7% and 79.3%, respectively. The total yield of trans double bonds in the final fat was 79.6% without a change in the location of the double bonds (Snyder et al., 1982).

IR spectroscopy is one of the most efficient tools to directly determine trans-double bonds. Therefore, the cis-trans isomerization of edible fat was also detected by React IR^TM spectroscopy. The React IR^TM spectra (Figure 2) clearly showed the generation of trans-double bonds. The new bonds, associated with trans fat formation, were observed at a bond of 966 cm^-1. These results indicated that the cis-trans isomerization of edible fat was realized. Moreover, React IR was also used to monitor the cis-trans isomerization of fat process (Figure 3), and the trans fat was generated rapidly.

Based on the above results in hand, we investigated the cis-trans isomerization of unsaturated fatty acids in other edible oils, mainly vegetable oils (Table 7). It could be seen that the yields of trans double bonds in the eight kinds of isomerized edible oils were all close to 80.0%. This is similar to an earlier report employing pure fatty acids as substrates (Snyder et al., 1982). In fact, vegetable oils contain various natural antioxidants with different kinds and concentrations, which are more complicated than edible fat or pure fatty acids. The yields of elaidic acid were all over 79%, and the yields of trans-linoleic acid fluctuated between 78.0% and 79.8%. Therefore, nearly 80.0% yields of TFAs generated in various vegetable oils are an acceptable result for the production of trans-lipids without changing the degree of unsaturation.