Thermal–oxidative Stability of Commercial Rice Bran Oil

Abstract

The thermal–oxidative stability of commercially available rice bran oil (RBO) was evaluated at 180 °C. Commercial RBO exhibited much higher thermal-oxidative stability than commercial high oleic canola oil in terms of polar components and polymer contents. Moreover, significant prolongation of antioxidants such as tocopherol (Toc) and γ-oryzanol (Ory) was observed for the commercial system during heating compared with the modeled RBO systems, in which Toc and/or Ory were added to physically refined RBO. Excellent thermal–oxidative stability of the commercial RBO system was proved and validated by the combined effect of Toc and Ory and the obvious prolongation of antioxidant activities.

Introduction

Rice bran oil (RBO) is a by-product of rice milling and is extracted and refined using rice bran (Ali and Devarajan, 2017; Bakota et al., 2014). RBO has been extensively applied in cooking owing to its excellent anti-oxidative stability at elevated temperatures, and its balanced content of monosaturated, polysaturated, and saturated fats (Ali and Devarajan, 2017; Bakota et al., 2014). Blending with RBO is reported to improve the oxidative stabilities of other oil substrates (Fan et al., 2013; Farhoosh and Kenari, 2009; Nakajima et al., 2017). The high thermal–oxidative stability of RBO is attributed to the high level of contained natural unsaponifiables such as vitamin E and phytonutrients, enhancing its anti-oxidative stability (Fan et al., 2013). Natural antioxidants such as tocopherol (Toc) and γ-oryzanol (Ory) can decrease the oxidation rate of oil during heating (Ali and Devarajan, 2017; Bakota et al., 2014; Fan et al., 2013; Wang et al., 2002). However, the impact of minor components in RBO on the anti-oxidative behavior during heating has received little attention. Since a RBO substrate includes various exogenous natural antioxidants as minor components, the anti-oxidative behavior should be highly complex; hence, it is necessary to assess the individual and combined anti-oxidative activity in the RBO medium. For instance, comparative analysis between a different antioxidant-containing substrate and RBO can be employed. Alternatively, information on the anti-oxidative effects and respective contributions of various minor components can be evaluated using a modeled RBO, prepared by adding well-defined minor components to physically refined RBO. If minor components such as Toc and Ory are the main antioxidant additives under heating, it is theoretically reasonable that their addition to physically refined RBO would simulate the anti-oxidative behavior of commercial RBO.

In this study, the anti-oxidative effects of Toc and Ory on the thermal–oxidative stability of RBO were investigated. First, the commercial RBO system was compared with high oleic canola oil (HOCO). Since HOCO does not contain Ory, the effect of Ory on the thermal stability was investigated by comparison. Then, we compared the commercial RBO substrate with the modeled RBO systems, in which arbitrary amounts of several minor components, such as Toc and Ory, were added. As test parameters, polar component (PC) and polymer content (PLC) were compared, since these values are considered to be good indicators of the progress of lipid oxidation (Yoon et al., 1985; Aibidi and Rennick, 2003).

Experimental procedures Materials

Commercial oils    Commercial RBO and HOCO were supplied by TSUNO Co., Ltd. (Wakayama, Japan) and Showa Sangyo Co., Ltd. (Tokyo, Japan), respectively. Fatty acid (FA) composition (%), average degree of unsaturation per triacylglycerol molecule, antioxidant contents in mg/kg,% PC and% PLC in these oils are shown in Table 1.

Table 1. FA composition in %, average degree of unsaturation, antioxidant contents of Toc and Ory in mg/kg, and PC and PLC in % in the commercial RBO and HOCO that are used.

		RBO	HOCO
FA composition (%)	Palmitic acid	22.4	5.5
	Stearic acid	1.5	1.8
	Oleic acid	40.5	76.5
	Linoleic acid	34.5	14.3
	Linolenic acid	1.1	1.9
Average degree of unsaturation		3.4	3.3
Antioxidant (mg/kg)	Toc	388.6	298.0
	Ory	2,216.4	0.0
PC (%)		3.1	0.9
PLC (%)		0.0	0.0

Refined oil    Commercial RBO was physically refined using activated carbon (Powder, Product code 037-02115; FUJIFILM Wako Pure Chemical Industries, Ltd., Osaka, Japan). The oil (50 g) was dissolved in hexane (200 mL), and then activated carbon (25 g) was added to the solution. After magnetic stirring of the mixture at 400 rpm for 5 min, the dispersed activated carbon was filtrated twice through a filter (diameter, 240 mm) with a particle retention capacity of 5.0 µm, and the solvent was evaporated under reduced pressure to afford the refined RBO. The physical refining process did not affect the FA and lipid compositions (data not shown).

Antioxidants    DL-α-Toc and Ory (each > 98% purity) were purchased from FUJIFILM Wako Pure Chemical Industries, Ltd.

Oxidized oils    Each oil was added to a 100 mL test tube and placed in an oil bath pre-heated to 180 °C, and the heating measurement was conducted with stirring at 400 rpm for several fixed periods (e.g., 0, 8, 16, 24 and 48 h) with overnight cooling to room temperature. The degree of lipid oxidation for each oil was evaluated by determining the amounts of PC, PLC and remnant ratios for antioxidants in the heat-oxidized oils. Additionally, the induction periods were evaluated for each oil substrate using conductometric determination with the Rancimat method.

Methods

Analysis of lipid composition and evaluation of PC content    A thin layer chromatography–flame ionization detector (TLC-FID) was used to determine the lipid composition and PC content. An Iatroscan MK-6s with silica gel rods S-IV (LSI Medience Corp., Tokyo, Japan) and a mixture of hexane/diethyl ether (87: 13, v/v) was used to quantify PC as an eluent. The composition was calculated using the peak areas in the TLC-FID analysis (Hara et al., 2006).

Analysis of FA composition    A GC-18A gas chromatograph (Shimadzu Corp., Kyoto, Japan) equipped with a flame ionization detector and a fused silica capillary column HR-SS-10 (0.25 mm × 25 m; Shinwa Chemical Industries, Ltd., Kyoto, Japan) was used to determine the FA composition of each oil substrate. Prior to GC analysis, FAs of oil substrates were methyl-esterified using hydrochloric acid/methanol as described by Jham et al. (1982).

Evaluation of PLC content    (Yoon et al., 1985; Aibidi and Rennick, 2003) A gel permeation chromatograph (Pump and RI-2031 Plus detector (JASCO Corp., Tokyo, Japan) equipped with a Shodex GPC KF-802.5 column (8.0 mm I.D. × 300 mm; Showa Denko K.K., Tokyo, Japan) was used to quantify the PLC content generated in oil substrates. For GPC, tetrahydrofuran was used as an eluent.

Estimation of antioxidant content    Quantification of Toc in oil substrates was carried out using a high performance liquid chromatography (HPLC) system, consisting of PU 880 Pump and FU-2020 Plus detector equipped with a Finepak SIL-5 column (4.6 mm I.D. × 250 mm) (JASCO Corp.). An excitation wavelength of 295 nm and an emission wavelength of 325 nm were used for detection. A mixture of hexane/2-propanol (124: 1, v/v) was used as an eluent. Quantification of Ory in oil substrates was performed using a HPLC system consisting of a LC-10AT Pump (Shimadzu Corp.), 875-UV detector equipped with a CrestPak C18S column (4.6 mm I.D. × 150 mm) (JASCO Corp.). Here, 315 nm was used as the detection wavelength and non-chlorinated solvents, a mixture of methanol/acetonitrile/acetic acid (52: 45: 3, v/v) (Yoshie et al., 2009), were used as an eluent.

Auto-oxidation test    Rancimat 678 (Metrohm Co., Ltd., Herisau, Switzerland) was used to estimate the heat-oxidation stability of oil substrates (Läubli and Bruttel, 1986). Each sample (1 g) in a 10 mL test tube was placed in the heating block at 100 °C with an air supply (20 L/h). The volatile decomposition products were corrected in the absorption vessels and the conductivity was recorded over time. The induction time was evaluated as the inflection time in the conductivity curve.

Results and Discussion

Comparison of the thermal stability of commercial RBO and HOCO    FA composition and contents of antioxidants, PC, and PLC in the commercial RBO and HOCO are shown in Table 1. RBO and HOCO showed similar average degree of unsaturation and fairly low content of linolenic acid (18:3). In contrast, RBO contained 2.5 times the linoleic acid content (18:2) and HOCO contained 1.9 times the oleic acid content (18:1). These characteristics were in good agreement with a previous report (Hosseini et al., 2016). For the antioxidant contents, both substrates contained a similar amount of Toc (300∼400 mg/kg); however, Ory was found in RBO at ca. 2 200 mg/kg but was not present in HOCO. PC of RBO was 3.1 wt%, while that of HOCO was 0.9 wt%.

Fig. 1 shows the effect of heating time on the contents of PC and PLC, and the remnant ratios (C_t / C₀) of Toc and Ory. PC and PLC increased in a heating time-dependent manner as the thermally oxidative and/or degradable products in the corresponding oils (Figs. 1a and 1b). PC and PLC in HOCO showed an abrupt increase at around 16 h. Thus, levels in HOCO were significantly higher than those in RBO after 16 h, indicating that after an extended heating time, the thermal–oxidative stability of RBO was greater than that of HOCO. Notably, Toc in the heated HOCO was completely consumed within the first 16 h (Fig. 1c). Thus, HOCO showed a rapid increase in the content of oxidative products and a complete loss of antioxidant components after 16 h. The time correspondence of both of these aspects clearly indicates that Toc was an effective antioxidant, preventing the thermal oxidation and/or decomposition of HOCO. In contrast, Toc and Ory were not completely consumed in the heated RBO at 48 h (Figs. 1c and 1d), indicating that RBO was effectively stabilized for a long period. An apparent difference in the heating effect on oxidative stability was observed between RBO and HOCO in terms of the induction period in the Rancimat analysis (Fig. 1e). The induction period for HOCO greatly decreased after heating for more than 16 h, while that for RBO decreased very slowly even after 48 h, clearly showing the high stability of RBO under the heating condition. Since the RBO system still contains “prolonged” antioxidants such as Toc and Ory even after 48 h (Figs. 1c and 1d), these remaining antioxidants effectively inhibit PC and PLC. The correlation between the loss of antioxidants and the very short induction period for HOCO at 16 h is also clearly evident.

Fig. 1.

Effect of heating time on the contents of remainin. (a) PC, and (b) PLC, and C_t / C₀ for (c) Toc, and (d) Ory for commercial RBO and HOCO, respectively, and (e) on their induction periods. C₀ and C_t were the antioxidants contents at 0 and t h, respectively.

Thus, from the results of PC, PLC, and the induction period, we concluded that RBO exhibits higher thermal stability than HOCO. Owing to the higher amount of linoleic acid (18:2) in the RBO substrate, the thermal–oxidative stability of RBO could be lower than that of HOCO. Typically, lipid oxidation has a tendency to occur more rapidly for lipids containing a higher degree of FA unsaturation. However, in this study, greater thermal stability was observed for RBO, where distinct prolongation was recognized for antioxidants, which directly enhanced the thermal–oxidative stability for an extended period.

Comparison of commercial RBO with modeled RBO    Effects of antioxidants, Toc and Ory, on the thermal stability of RBO as well as their prolongation were investigated. It can be assumed that Toc and Ory protect or spare each other (regenerate) at frying temperatures as suggested in previous reports (Chotimarkorn et al., 2008; Kochhar, 2000; Hamid et al., 2014). The initial contents of minor components in each model system are summarized in Table 2.

Table 2. Antioxidant contents of Toc and Ory in mg/kg in various RBO systems. These contents for commercial RBO were shown in Table 1.

	Refined RBO	Anti-oxidant added RBO system
		Toc,Ory	Toc	Ory
Toc (mg/kg)	0.0	368.7	393.5	0.0
Ory (mg/kg)	0.0	2,432.9	0.0	2,342.3

Fig. 2 shows the effect of heating time on the contents of PC and PLC, and C_t /C₀ of Toc and Ory of commercial, refined, and modeled RBO systems. In all systems, the contents of PC and PLC gradually increased as a function of time (Figs. 2a and 2b). Owing to the lack of antioxidants, much higher contents of PC and PLC were found in the refined RBO than in the RBO systems containing antioxidants during the experiment. The individual additions of Toc or Ory prevented the generation of PC and PLC, and the addition of both antioxidants effectively prevented their generation, demonstrating the effectiveness of the combined use of Toc and Ory.

Fig. 2.

Effect of heating time on the contents of (a) PC, and (b) PLC, and C_t / C₀ for (c) Toc, and (d) Ory for commercial, physically refined, and various modeled RBO systems where Toc and/or Ory were added. C₀ and C_t were the antioxidants contents at 0 and t h, respectively.

However, while the PC content in the modeled RBO system with Toc and Ory became similar to that in the commercial RBO at 48 h, the model system showed higher PLC content than the commercial system, though the initial amounts of Toc and Ory added were slightly higher (Tables 1 and 2). Furthermore, the half-life for the commercial system in terms of Toc and Ory, shown in Figs. 2c and 2d, were much longer among the RBO systems studied; the half-life, which corresponds to the time when C_t / C₀ becomes 0.5, of Toc was 5.6, 6.7, and 15.1 h for the Toc-added, Toc and Ory-added, and commercial RBO system, respectively; while that of Ory was 14.2, 16.4, and 35.6 h for the Ory-added, Toc and Ory-added, and commercial RBO system, respectively. These results indicated that the rate of Toc and Ory consumption became much lower for the commercial RBO system than for the model systems, demonstrating that the commercial system exhibits higher thermal stability than the modeled RBOs, although the initial amounts of Toc and Ory prior to heating were slightly greater. In contrast, while Ory was reported to exhibit a protective effect against Toc degradation during heating (Chotimarkorn et al., 2008; Kochhar, 2000; Hamid et al., 2014), in this study, a reduced prolongation effect, owing to Ory, was observed on Toc. It was reported that crude RBO contains various anti-oxidative components besides Toc and Ory (Goufo and Trindade, 2014), and these minor components can impact the anti-oxidative and/or prolongation effects in the system. For instance, the anti-polymerization effect of sterol additives (Wang et al., 2002) and the combined behavior of minor components with Ory and γ-tocotrienol in maintaining Toc activity in the RBO substrate (crude) (Hamid et al., 2014) have been suggested.

Conclusion

Commercially available RBO showed higher thermal–oxidative stability than HOCO in terms of PC and PLC contents. The products in RBO after heating at 180 °C for more than 16 h lowered the PC and PLC contents compared to commercial HOCO. Besides, the addition of Toc and Ory for the three modeled RBO systems effectively decreased the production of PC and PLC in the refined RBO under heating. However, the commercial RBO generally showed a higher thermal-oxidative stability than the modeled RBOs. In addition to the combined anti-oxidative effects owing to Toc and Ory, a distinct mechanism enhancing their prolongations must be dominant in the commercial RBO system. It is necessary to carry out further comprehensive and systematic investigations including more complex systems to clarify the mechanism.

Acknowledgements    We would like to thank TSUNO Co., Ltd. (Wakayama, Japan) and Showa Sangyo Co., Ltd. (Tokyo, Japan) for the supply of the oils, respectively.

References

•  Aibidi,  S. L. and  Rennick,  K. A. (2003). Determination of nonvolatile components in polar fractions of rice bran oils. J. Am. Oil Chem. Soc., 80, 1057-1062.
•  Ali,  A. and  Devarajan,  S. (2017). Nutritional and health benefits of rice bran oil. In “Brown Rice” ed. by  A.  Manickavasagan,  C.  Santhakuma,  N.  Venkatachalapathy. Springer International Publishing, AG, pp. 135-158.
•  Bakota,  E. L.,  Winkler-Moser,  J. K., and  Hwang,  H. -S. (2014). Properties of rice bran oil−derived functional ingredients. Lipid Technol., 26, 179-182.
•  Chotimarkorn,  C.,  Benjakul,  S., and  Silalai,  N. (2008). Antioxidative effects of rice bran extracts on refined tuna oil during storage. Food Res. Int., 41, 616-622.
•  Fan,  H. Y.,  Sharifudin,  M. S.,  Hasmadi,  M., and  Chew,  H. M. (2013). Frying stability of rice bran oil and palm olein. Int. Food Res. J., 20, 403-407.
•  Farhoosh,  R. and  Kenari,  R. E. (2009). Anti−rancidity effects of sesame and rice bran oils on canola oil during deep frying. J. Am. Oil Chem. Soc., 86, 539-544.
•  Goufo,  P. and  Trindade,  H. (2014). Rice antioxidants: phenolic acids, flavonoids, anthocyanins, proanthocyanidins, tocopherols, tocotrienols, γ-oryzanol, and phytic acid. Food Sci. Nutr., 2, 75-104.
•  Hamid,  A. A.,  Dek,  M. S. P.,  Tan,  C. P.,  Zainudin,  M. A. M., and  Fang,  E.K.W. (2014). Changes of major antioxidant compounds and radical scavenging activity of palm oil and rice bran oil during deep-frying. Antioxidants, 3, 502-515.
•  Hara,  S.,  Ogawa,  E., and  Totani,  Y. (2006). Evaluation of heat-deteriorated oils. I. TLC-FID method for determining polar compounds content. J. Oleo Sci., 55, 167-172.
•  Hosseini,  H.,  Ghorbani,  M.,  Meshginfar,  N., and  Mahoonak,  A.S. (2016). A review on frying: procedure, fat, deterioration progress and health hazards. J. Am. Oil Chem. Soc., 93, 445-466.
•  Jham,  G. N.,  Teles,  F. F. F., and  Campos,  L. G. (1982). Use of aqueous HCl/MeOH as esterification reagent for analysis of fatty acids derived from soybean lipids. J. Am. Oil Chem. Soc., 59, 132-133.
•  Kochhar,  S. P. (2000). Stabilisation of frying oils with natural antioxidative components. Eur. J. Lipid Sci. Tech., 102, 552-559.
•  Läubli,  M. W. and  Bruttel,  P. A. (1986). Determination of the oxidative stability of fats and oils: Comparison between the active oxygen method (AOCS Cd 12–57) and the rancimat method. J. Am. Oil Chem. Soc., 63, 792-795.
•  Nakajima,  S.,  Takai,  M.,  Hayashi,  C.,  Tsuno,  T., and  Endo,  Y. (2017). Autoxidation of fish oil blended with rice bran oil. J. Oleo Sci., 66, 573-577.
•  Wang,  T.,  Hicks,  K. B., and  Moreau,  R. (2002). Antioxidant activity of phytosterols, oryzanol, and other phytosterol conjugates. J. Am. Oil Chem. Soc., 79, 1201-1206.
•  Yoon,  S. H.,  Kim,  S. K.,  Shin,  M. G., and  Kim,  K. H. (1985). Comparative study of physical methods for lipid oxidation measurement in oils. J. Am. Oil Chem. Soc., 62, 1487-1489.
•  Yoshie,  A.,  Kanda,  A.,  Nakamura,  T.,  Igusa,  H., and  Hara,  S. (2009). Comparison of γ-oryzanol contents in crude rice bran oils from different sources by various determination methods. J. Oleo Sci., 58, 511-518.