LWT 41 (2008) 1417–1424 Extraction of a-tocopherol and g-oryzanol from rice bran Photchanathip Imsanguan a , Amorn Roaysubtawee a , Ratsuda Borirak a , Suwassa Pongamphai a,à , Supaporn Douglas a , Peter L. Douglas b a Department of Chemical Engineering, King Mongkut’s University of Technology, Thonburi, Bangkok 10140, Thailand b Department of Chemical Engineering, University of Waterloo, Waterloo, Ont., Canada N2L 3G1 Received 15 May 2007; received in revised form 31 August 2007; accepted 31 August 2007 Abstract The aim of this research was to study the effect of operating mode (continuous versus batch+continuous), temperature, pressure and solvent on a-tocopherol and g-oryzanol extraction from rice bran (Oryza sativa Linn.) and compare the efficiency of three extraction methods: SC-CO 2 extraction, solvent extraction and soxhlet extraction. Three sets of experiments were performed. First, extraction using SC-CO 2 was performed over a range of temperatures and pressures (45–65 1C and at 38 and 48 MPa), and at a CO 2 flow rate of 0.45 mL/min. The results showed that the best conditions for a-tocopherol extraction were 55 1C, 48 MPa in the batch+continuous mode. For g-oryzanol, the best conditions were 65 1C, 48 MPa and in the continuous mode. In the second set of experiments, solvent extraction using hexane and ethanol at 32 and 55–60 1C was studied. The results showed that none of the solvents could extract a-tocopherol; however, ethanol at 55–60 1C was suitable for g-oryzanol extraction. Finally, soxhlet extraction experiments using hexane for a-tocopherol and ethanol for g-oryzanol were also performed. In summary, SC-CO 2 was found to be the best solvent for extracting both a-tocopherol and g-oryzanol from rice bran, because of its higher yields and extraction rate. r 2007 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. Keywords: Rice bran; a-Tocopherol; g-Oryzanol; Supercritical carbon dioxide; Solvent extraction 1. Introduction Rice (Oryza sativa Linn.) production is a significant crop in Thailand resulting in significant amounts of rice bran waste. Rice bran contains many valuable substances such as vitamin E (a-tocopherol and tocotrienol) and g-oryzanol. The major component of vitamin E in rice bran is a-tocopherol which is an antioxidant and can lower the risk of cancer formation and coronary heart diseases (Zhimin, Na, & Samuel, 2001), it is also reported to prevent Alzheimer’s disease and allergies (Mervyn, 1994). g-Oryzanol is a mixture of 10 ferulate esters of triterpene alcohol (Zhimin et al., 2001) and can be used to reduce blood cholesterol levels, treat nerve imbalance, as well as an antiox idant and preservative (Murase & Iishima, 1963; Rong, Ausman, & Nicholosi, 1994; Sasaki et al., 1990). The major components of g-oryzanol in rice bran are cycloar- tenyl ferulate, 24-methylene cycloartanyl ferulate and campestanyl ferulate (Zhimin et al., 2001). Since they are so abundant in rice bran, this research focused on the extraction of a-tocopherol and g-oryzanol (cycloartenyl ferulate, 24-methylene cycloartanyl ferulate and campesta- nyl ferulate) from rice bran. The chemical structures of a-tocopherol and g-oryzanol are shown in Figs. 1 and 2, respectively. Currently, a-tocopherol and g-oryzanol are extracted from various sources such as wheat germ (Ge, Ni, Yan, Chen, & Cai, 2002), rice by-product (Perretti, Miniati, Montanari, & Fantozzi, 2003) using supercritical carbon dioxide (SC-CO 2 ), and rice bran oil (Noppamas, 2002) using solvent extraction. The authors can find no reports in the literature describing the direct extraction of a-tocopherol or g-oryzanol from rice bran using supercritical carb on dioxide. The literature does however contain studies on the extraction of these substances from rice bran oil (Noppamas, 2002). This then leads to the motivation for this research. Carbon dioxide is the most commonly used supercritical solvent, because it ARTICLE IN P RESS www.elsevier.com/locate/lwt 0023-6438/$34.00 r 2007 Swiss Society of Food Science and Technology. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.lwt.2007.08.028 à Corresponding author. Tel.: +66 24709221x208; fax: +66 24283534. E-mail address: suwassa.pon@kmutt.ac.th (S. Pongamphai). has a relatively low critical temperature and pressure (31.1 1C and 7.39 MPa), non-toxic, non-flammable, non- corrosive, inert and inexpensive. The most suitable extraction methods for the extrac- tion of vitamin E (a-tocopherol) and g-oryzanol (cycloartenyl ferulate, 24-methylene cycloartanyl ferulate and campestanyl ferulate) from rice bran were studied and the extraction efficiency and operating conditions were determined. 2. Materials and methods 2.1. Materials Bran from Suphanburi rice, which was grown at Chacherngsow, Thailand, was used in this work. Carbon dioxide (99.95% purity) was supplied by Thai Industrial Gases (TIG), Thailand. Methanol, acetonitrile, hexane and ethanol used in this work are of HPLC grade and were ARTICLE IN P RESS Fig. 1. The chemical structure of a-tocopherol (Noppamas, 2002). Fig. 2. The chemical structures of g-oryzanol (Zhimin & Samuel, 1999). P. Imsanguan et al. / LWT - Food Science and Technology 41 (2008) 1417–14241418 procured from Merck Ltd., Thailand. a-Tocopherol and g- oryzanol standards were obtained from Sigma Chemical Co., USA and Tsuno, Japan, respectively. 2.2. Rice bran mean composition Rice bran was analyzed for particle size, moisture and oil content. The particle size was measured using a Taylor sieve. The moisture content was determined using oven drying at 110 1C for 2 h and the oil content was measured using soxhlet extraction with hexane for 24 h (AOAC, 1995). 2.3. Extraction of a-tocopherol and g-oryzanol from rice bran using SC-CO 2 SC-CO 2 extraction experiments were carried out in an ISCO Model SFX TM 3560 laboratory extractor shown in Fig. 3. SC-CO 2 extraction was to study the effect of operation mode (continuous versus batch+continuous), pressure (38 and 48 MPa) and temperature (45–65 1C). 2.3.1. Effect of operating mode 2.3.1.1. Continuous mode. A 0.5 g of rice bran was placed in the extraction vessel. The flow rate of CO 2 was set at 0.45 mL/min and allowed to flow continuously through the extraction vessel for 7 h at a temperature of 55 1C and a pressure of 48 MPa. The extract was analyzed using a Waters Corporation, Model #600-486-717 high- performance liquid chromatograph (HPLC). HPLC ana- lysis was performed using symmetry C-18 column (150 mm  3.9 mm i.d.) with mobile phase acetonitrile:- methanol (90:10, v/v), flow rate of 1.5 mL/min and UV detector of 295 nm. 2.3.1.2. Batch+continuous mode. Rice bran was placed in the extraction vessel and exposed to SC-CO 2 at 55 1C and 48 MPa for 1 h and then CO 2 was allowed to flow through the sample at a flow rate of 0.45 mL/min for 6 h. 2.4. Extraction of a-tocopherol and g-oryzanol from rice bran using solvent extraction A 1 g sample of rice bran was extracted using 100 mL of hexane at a rotating speed of 200 rpm and a temperature varying from 32 1C to about 55–60 1C for 24 h. The extract was analyzed using an HPLC. The hexane solvent was replaced with ethanol and the remainder of procedure and conditions are all the same as before. 2.5. Extraction of a-tocopherol and g-oryzanol from rice bran using vacuum soxhlet extraction A 3 g sample of rice bran was extracted in a soxhlet apparatus with 300 mL of hexane at temperatures varying from 65 to 70 1C and pressures varying from 500 to 550 mmHg for 24 h. The extract was analyzed using an HPLC. The hexane solvent was then replaced with ethanol and the remainder of the procedure and conditions are all the same as before. 3. Results and discussion 3.1. Rice bran mean composition Bran from Supha nburi rice (O. sativa Linn.) was used in this research. Moisture and fat/oil content of rice bran were found to be 14.01% and 19.08% (w/w), respectively. The ARTICLE IN P RESS Fig. 3. ISCO Model SFX TM 3560 supercritical CO 2 extractor. P. Imsanguan et al. / LWT - Food Science and Technology 41 (2008) 1417–1424 1419 majority of rice bran particles were found to be 297–595 mm in size as shown in Table 1. 3.2. SC-CO 2 extraction 3.2.1. Effect of operating mode A comparison of two methods of SC-CO 2 extraction, (i) continuous mode (ii) batch+continuous mode, is shown in Figs. 4 and 5 for a-tocopherol and g-oryzanol, respectively. The extraction was performed at 55 1C, 48 MPa and a CO 2 flow rate of 0.45 mL/min. The results show that the extraction of a-tocopherol reached a maximum of both continuous mode and batch+continuous mode operations after which no additional extraction was observed; the average extraction rate was 24.80 mg/kg (dry basis )/h and 31.83 mg/kg (dry basis)/h, respectively and the extrac- tion rate of g-oryzanol was 2322.79 and 2464.73 mg/kg/h, respectively. Fig. 4 shows that batch+continuous mode results in higher extraction rates than continuous mode. a-Tocopherol is found in a stable matrix form in rice bran, so continuous mode is inadequate for high a-tocopherol extraction. Under batch+continuous mode, CO 2 was confined inside extraction vessel for 1 h, thereby increasing the CO 2 penetration into rice bran and increas- ing extraction resulting in an increase in the a-tocopherol extraction. On the other hand, Fig. 5 shows a lower g-oryzanol extraction rate when batch+continuous mode was used. 3.2.2. Effect of pressure SC-CO 2 extraction was carried out at a constant temperature of 65 1C, a carbon dioxide flow rate of 0.45 mL/min and a pressure varying from 38 to 48 MPa. The extraction results for a-tocopherol and g-oryzanol are shown in Figs. 6 and 7, respectively. Both figures indicate that when the pressure was increased, the extraction rate and yields increased. This effect is expected since an increase in the pressure increases the density and solvent ability of supercritical fluid thereby allowing it to dissolve more solute. ARTICLE IN P RESS Table 1 Size distribution of rice bran particles Particle size Rice bran (%) Mesh mm 418 41000 8.7 18–30 595–1000 25.4 30–50 297–595 36.2 50–80 177–297 26.7 80–100 149–177 2.8 o100 o149 0.2 012 6 time (h) 0 20 40 60 80 100 120 140 alpha-tocopherol (mg/kg dry basis) 754 3 Fig. 4. Effect of operating mode on a-tocopherol extraction using SC- CO 2 (T ¼ 55 1C, P ¼ 48 MPa, CO 2 flow rate ¼ 0.45 mL/min): (’) continuous and (m) batch+continuous. 047 time (h) 0 2000 4000 6000 8000 10000 12000 gamma-oryzanol (mg/kg dry basis) 65321 Fig. 5. Effect of operating mode on g-oryzanol extraction using SC-CO 2 (T ¼ 55 1C, P ¼ 48 MPa, CO 2 flow rate ¼ 0.45 mL/min): (’) continuous and (m) batch+continuous. 01 5 time (h) 0 20 40 60 80 100 120 140 alpha-tocopherol (mg/kg dry basis) 76432 Fig. 6. Effect of pressure on a-tocopherol extraction using SC-CO 2 (T ¼ 65 1C, CO 2 flow rate ¼ 0.45 mL/min): (’) 38 MPa and (m) 48 MPa. P. Imsanguan et al. / LWT - Food Science and Technology 41 (2008) 1417–14241420 3.2.3. Effect of temperature The effect of temperature on the extraction was studied at a constant pressure of 48 MPa, using a CO 2 flow rate of 0.45 mL/min and at temperatures from 45 to 65 1C. The results of a-tocopherol and g-oryzanol extraction are shown in Figs. 8 and 9, respectively. These figures indicate that increasing extraction temperature results in increase extraction rates. This effect is also expected since increasing the temperature decreases the viscosity and increases the diffusivity resulting in an increased extraction rate. Based on the range of experimental conditions, the best conditions for a-tocopherol extraction using supercritical carbon dioxide were found to be 48 MPa, 55 1C, CO 2 flow rate of 0.45 mL/min with batch+c ontinuous mode and for g-oryzanol extraction were found to be 48 MPa, 65 1C, CO 2 flow rate of 0.45 mL/min with continuous mode. The extraction yields of a-tocopherol and g-oryzanol at these conditions were 127.33 and 11,371.79 mg/kg (dry basis), respectively. Fig. 10 shows an HPLC chromatogram of the extract; retention times of a-tocopherol and 3 of the 10 components of g-oryzanol (cycloartenyl ferulate, 24-methylene cycloar- tanyl ferulate and campestanyl ferulate) were identified and found to be 10.52, 21.58, 24.61 and 30.18 min, respectively. The area under a particular peak in the chromatogram indicates the amount of that particular component in the rice bran extract. 3.3. Solvent extraction The solvent extraction experiments were carried out at atmospheric pressure and at various temperatures. Solvent extraction experiments showed neither ethanol nor hexane could extract a-tocopherol. The results showed that this is because a-tocopherol is in a matrix form in rice bran, so extraction at low pressure (i.e. atmospheric pressure) is inadequate for a-tocopherol extraction, because this pressure cannot eliminate the interferences from other molecules such as proteins and carbohydrates, that are non-soluble in organic solvents (Ruperez, Martın, Herrera, & Barbas, 2001). Perhap s at elevated pressures and longer extraction times a-tocopherol might be extracted from rice bran. The results of solvent extraction of g-oryzanol using hexane are shown in Fig. 11. This figure shows that increasing temperatur e increases the quantity of g-oryzanol extracted. Increasing temperature decreases the viscosity and increases the diffusivity of the solvent resulting in increasing extraction rate and yield. In this work, the highest extraction yield of g-oryzanol was found to be 7349.25 mg/kg (dry basis) at temperatures between 55 and 60 1C. The effect of solvent type on the extracti on was ARTICLE IN P RESS 0 time (h) 0 2000 4000 6000 8000 10000 12000 gamma-oryzanol (mg/kg dry basis) 7654321 Fig. 7. Effect of pressure on g-oryzanol extraction using SC-CO 2 (T ¼ 65 1C, CO 2 flow rate ¼ 0.45 mL/min): (’) 38 MPa and (m) 48 MPa. 012 6 time (h) 0 20 40 60 80 100 120 140 alpha-tocopherol (mg/kg dry basis) 5743 Fig. 8. Effect of temperature on a-tocopherol extraction using SC-CO 2 (P ¼ 48 MPa, CO 2 flow rate ¼ 0.45 mL/min): (K)451C, (m)551C and (’)651C. 01 time (h) 0 2000 4000 6000 8000 10000 12000 gamma-oryzanol (mg/kg dry basis) 765432 Fig. 9. Effect of temperature on g-oryzanol extraction using SC-CO 2 (P ¼ 48 MPa, CO 2 flow rate ¼ 0.45 mL/min): (K)451C, (m)551C and (’)651C. P. Imsanguan et al. / LWT - Food Science and Technology 41 (2008) 1417–1424 1421 carried out at atmospheric pressure and temperature of 55–60 1C. Fig. 12 shows the solvent extraction of g-oryzanol using bot h ethanol and hexane. Ethanol was found to be a better solvent than hexane. This is because g-oryzanol, which consists of triterpene alcohol or sterols and fururic acid, is a polar molecule (Zhimin & Samuel, 1999), so it is easily dissol ved in ethanol, which is also a polar molecule; extraction yield was 9414.02 mg/kg (dry basis). 3.4. Vacuum soxhlet extraction The effect of solvent type on the vacuum soxhlet extraction was carried out at pressures of 500–550 mmHg and temperatures of 65–70 1C. The results showed that hexane and ethanol were the best solvents for a-tocopherol and g-oryzanol extraction, respectively. This is because hexane and a-tocopherol are non-polar molecules, wher eas ethanol and g-oryzanol are polar molecules. Extraction yields of a-tocopherol and g-oryzanol were found to be 172.23 and 9808.79 mg/kg (dry basis), respectively. More- over, the quantity of both substances using the vacuum soxhlet extraction was higher than that achieved using the solvent extraction. The reasons being that the extraction temperature of the soxhlet extraction was higher than the solvent extraction, and the solvent used in soxhlet extraction can be evaporated and condensed in the same way as the steam distillation, so efficiency of the soxhlet extraction is higher than the solvent extraction (Pattira, Krisda, & Kittipong, 2003). 3.5. Extraction rate Figs. 4–9, 11 and 12 show that extraction rate is increasing in the beginning of extraction, since there is a lot of concentration gradient of substances in bulk fluid and interior of rice bran. The extraction rate then ARTICLE IN P RESS Fig. 10. Chromatogram of the rice bran extract using SC-CO 2 (P ¼ 48 MPa, T ¼ 65 1C). 04812162024 time (h) 0 2000 4000 6000 8000 10000 12000 gamma-oryzanol (mg/kg dry basis) Fig. 12. Effect of solvent type on g-oryzanol extraction: (m) hexane and (’) ethanol. 0481216 20 24 time (h) 0 2000 4000 6000 8000 10000 12000 gamma-oryzanol (mg/kg dry basis) Fig. 11. Effect of temperature on g-oryzanol extraction using hexane: (m) 32 1C and (’) 55–60 1C. P. Imsanguan et al. / LWT - Food Science and Technology 41 (2008) 1417–14241422 continually decreases due to the decreasing of concentra- tion gradient of the solute in the solvent. Finally, the extraction process approaches equilibrium. 3.6. Extraction efficiency The comparison of three extraction methods, SC-CO 2 , solvent and soxhlet extra ction, is shown in Figs. 13 and 14 for a-tocopherol and g-oryzanol, respectively. The extrac- tion efficiency of a-tocopherol for SC-CO 2 and solvent extraction were 73.93% and 0%, respectively, as compared to soxhlet extraction. For g-oryzanol, extraction yield of SC-CO 2 extraction is greater than the soxhlet extraction, because soxhlet extraction time and temperature were higher than that of SC-CO 2 extraction, so some g-oryzanol was degraded. Both figures indicated that SC-CO 2 extrac- tion resulted in higher extraction yields, lower extraction time and avoided degradation of substances. 4. Conclusions The results indicated that the extraction of a-tocopherol and g-oryzanol from rice bran is successfully performed using SC-CO 2 . The extraction yields of a-tocoph erol and g-o ryzanol depend on the pressure and temp erature during extraction. The best conditions for a-tocoph erol extraction were pressure of 48 MPa, temperature of 55 1C and batch +continuous mode. Extraction yield was 127.33 mg/kg (dry basis). For g-oryzanol extraction, the best condition s w ere p ressure o f 48 MPa, temperature of 65 1C in the continuous mode. Extraction yield was 11,371.79 mg/kg (dry basis). Neither ethanol nor hexane could extract a-tocopherol at atmospheric pressure but ethanol at 55–60 1C was found to be best for extracting g-oryzanol. For soxhlet extraction, hexane and ethanol were the best solvent for a-tocopherol and g-oryzanol extraction, respectively. Based on a comparison of three extraction methods (SC-CO 2 , solvent extraction and s oxhlet), it was concluded that SC-CO 2 extraction was the best extraction method for a-tocopherol and g-oryzanol extraction from rice bran. Extraction using SC-CO 2 results in h igh yields a nd the h ighest extraction rates u sing a non-toxic solvent, C O 2 . Although a greater y ield of a-tocopherol r esults from soxhlet extraction with hexane than SC-CO 2 , the extraction rate is significantly higher when using SC-CO 2 than hexane (31.83 mg/kg/h with SC-CO 2 compared to 7.18 mg/kg/h with soxhlet). Since the SC-CO 2 extraction ti me was m uch l ower than the soxhlet extraction time (4 h compared to 24 h ) at similar temperatures the possibility of thermal degradation is greatly reduced. Acknowledgment The authors gratefully acknowledge the financial sup- port from the Thailand Research Fund (TRF) through the Royal Golden Jubilee Ph.D. Program (1.C.KT/47/G.1). References AOAC. (1995) (16th ed.). Official methods of analysis, Vol. 41. Washington, DC: Association of Official Analytical Chemists. Ge, Y., Ni, Y., Yan, H., Chen, Y., & Cai, T. (2002). Optimization of the supercritical fluid extraction of natural vitamin E from wheat germ using response surface methodology. Journal of Food Science, 67(1), 239–243. Mervyn, L. (1994). The vitamins explained simply. Bangkok: Smith (pp. 131–139). Murase, Y., & Iishima, H. (1963). Clinical studies of oral administration of gamma-oryzanol on climacteric complaints and its syndrome. Obstetrics and Gynecological Practice, 12, 147–149. Noppamas, M. (2002). The investigation of g-oryzanol and vitamin E content in Thai rice varieties (pp. 11–67). Master of Science, Department of Bioresources and Technology, King Mongkut’s University of Technology, Thonburi. Pattira, P., Krisda, K., & Kittipong, R. (2003). Supercritical carbon dioxide extraction of vitamin E from oil palm leaves (pp. 18–20). ARTICLE IN P RESS sc-co 2 (6 h) solvent (ethanol 24 h) soxhlet (ethanol 24 h) extraction method 0 2000 4000 6000 8000 10000 12000 gamma-oryzanol (mg/kg dry basis) Fig. 14. Comparison of extraction methods (SC-CO 2 extraction, solvent extraction using ethanol and soxhlet extraction using ethanol) for g- oryzanol extraction. sc-co 2 (4 h) solvent (24 h) soxhlet (hexane 24 h) extraction method 0 20 40 60 80 100 120 140 160 180 alpha-tocopherol (mg/kg dry basis) Fig. 13. Comparison of extraction methods (SC-CO 2 extraction, solvent extraction and soxhlet extraction using hexane) for a-tocopherol extrac- tion. P. Imsanguan et al. / LWT - Food Science and Technology 41 (2008) 1417–1424 1423 Bachelor of Engineering, Department of Chemical Engineering, King Mongkut’s University of Technology, Thonburi. Perretti, G., Miniati, E., Montanari, L., & Fantozzi, P. (2003). Improving the value of rice by-products by SFE. Journal of Supercritical Fluids, 26, 63–71. Rong, N., Ausman, L. M., & Nicholosi, R. J. (1994). Rice bran oil decreases plasma LDL cholesterol by inhibitory dietary cholesterol absorption. The FASEB Journal: Official Publication of the Federation of American Societies for Experimental Biology, 8, 303–309. Ruperez, F. J., Martın, D., Herrera, E., & Barbas, C. (2001). Chromato- graphic analysis of a-tocopherol and related compounds in various matrices. Journal of Chromatography A, 935, 45–69. Sasaki, J., Takada, Y., Kusuda, M., Tanabe, Y., Matsunaga, A., & Arakawa, K. (1990). Effects of g-oryzanol on serum lipids and apolipoproteins in dyslipidemic schizophrenics receiving major tran- quilizero. Clinical Therapeutics, 12, 263–268. Zhimin, X., Na, H., & Samuel, G. J. (2001). Antioxidant activity of tocopherols, tocotrienols, and g-oryzanol components from rice bran against cholesterol oxidation accelerated by 2,2 0 -azobis(2-methylpro- pionamidine) dihydrochloride. Journal of Agriculture and Food Chemistry, 49, 2077–2081. Zhimin, X., & Samuel, G. J. (1999). Purification and identification of components of g-oryzanol in rice bran oil. Journal of Agriculture and Food Chemistry, 47, 2724–2728. ARTICLE IN P RESS P. Imsanguan et al. / LWT - Food Science and Technology 41 (2008) 1417–14241424 . at a flow rate of 0.45 mL/min for 6 h. 2.4. Extraction of a-tocopherol and g-oryzanol from rice bran using solvent extraction A 1 g sample of rice bran was extracted using 100 mL of hexane at a. ferulate and campestanyl ferulate) from rice bran were studied and the extraction efficiency and operating conditions were determined. 2. Materials and methods 2.1. Materials Bran from Suphanburi rice, . remainder of procedure and conditions are all the same as before. 2.5. Extraction of a-tocopherol and g-oryzanol from rice bran using vacuum soxhlet extraction A 3 g sample of rice bran was extracted