Two samples of DODs were obtained from a rice bran oil (RBO) industry which uses the physical refining method (M/s A P Solvex Private Limited, Dhuri, Punjab) and termed as DOD-1 and DOD-2. Standard chemicals, mixed fatty acid methyl esters, phytosterols (β-sitosterol, campesterol, stigmasterol) and α-tocopherol were purchased from Sigma Chemical Co. (St Louis, USA). Solvents and other chemicals used were of analytical grade.

The two DODs were analyzed for characteristics such as free fatty acid value (FFA), peroxide value (PV), color, saponification value (SV), and unsaponifiable matter (USM) according to AOCS official methods (Firestone, 1998). The total tocopherol content of the DOD samples was determined according to the IUPAC Method No. 2.301 (Paquot and Havtfenne, 1987).

The phytosterol contents were determined according to the procedure of Holen (1985). The phytosterol composition was determined using HPLC (model LC-10AVP, Shimadzu corporation, Kyoto, Japan) with a C18 column (ODS column 150 mm × 4.6 mm id) using a UV detector set at 206nm. The mobile phase consisted of methanol/water (99:1.0; v/v), the samples were dissolved in chloroform and 10µL were injected into the HPLC. The phytosterols were identified and quantified using standard β-sitosterol, campesterol, and stigmasterol.

Since the DOD samples had high FFA contents (59.2% and 86.0%), the removal of the FFA was necessary for the estimation of the acylglycerol composition of the DOD samples. Sodium hydroxide (1N) was used to neutralize the FFA and the neutral oil from the soap was extracted using hexane (50 mL × 3) extractions. Ethyl alcohol (10%) was used to wash the combined hexane extracts until free of alkali. The neutral oil (NO) was obtained after desolventizing the hexane fraction in a flash evaporator. The acylglycerol compositions of the NO were determined according to the AOAC official method (Firestone, 1998) using silica gel column chromatography. The column was packed with silica gel (30 g) and the sample (1 g) was dissolved in chloroform (15 mL) and transferred into the column. Solvent elutions were made in the following sequence: benzene (200 mL), 10% diethyl ether in benzene (20 mL of diethyl ether + 180 mL of benzene), and diethyl ether (200 mL) to elute triacylglycerol, diacylglycerol + FFA and monoacylglycerol fraction, respectively. The solvents were removed in a flash evaporator under vacuum at 40 °C and the individual fraction of acylglycerol was determined gravimetrically.

A simple silica gel percolation method was developed to purify DOD through decolorization and deodorization. A silica gel (100–200 mesh size) was heated at 105 °C for 6 hours to remove moisture and later hydrated by the addition of water (5%). The column (60 cm length and 4 cm internal diameter) was packed with silica gel (100 g) using petroleum ether. DOD (100 g) was mixed with hexane (500 mL) (1:5, w/v) and transferred to the column and allowed to percolate at a flow rate of 2–3 mL / min and the fraction was collected and labeled as phytosterol concentrate fraction (PCF) which was devoid of color and bad odor. Later the second solvent system consisting of a mixture of two solvents, (3:1 of ethanol in hexane) was poured onto the column and allowed to pass through the column to obtain the second fraction, termed as brown color and bad odor fraction (BCBOF). The two fractions were desolventized and kept separately until analysis.

A high purity phytosterol mixture from the PCF of DOD-1 was prepared according to the procedure of Sakina Khatoon et al. (2010). Briefly, the PCF was refluxed on a boiling water bath containing potassium hydroxide (50% in water) for 2hr. After refluxing, it was cooled to room temperature, extracted 6 times to get the unsaponifiable matter fraction. Ethyl alcohol (10%) was used to remove the alkali present in the petroleum ether fraction. Later the solvent was removed in a flash evaporator, and the obtained residue was weighed to get the percentage of unsaponifiable matter. The unsaponifiable matter was diluted with hexane (1:10, w/v) and water (10%) was added and mixed well using a mechanical stirrer for 10 min, and cooled at -20 °C for 72h to obtain phytosterol crystals, which were then filtered under vacuum to get a crude product. Then, the crude product was mixed with hexane (1:10, w/v) and water (10%) once again and cooled at 5 °C for 2–5h. Afterwards, the contents were filtered under vacuum to get phytosterol crystals of high purity.

The fatty acid composition of the samples was determined by GC. For this purpose, the sample was converted into fatty acid methyl esters according to the procedure of Van-Wijngaarden (1967). DOD (200 mg) was taken in a test tube and methanolic sodium hydroxide (2 mL) (0.5N) was added and saponified by heating at 50 °C for 20min. After saponification, the unsaponifiable matter was extracted by adding hexane (2 mL) and the hexane extract rejected. The aqueous layer was acidified by adding concentrated HCl (0.2 mL) to break up the soap and to facilitate the release of fatty acids. Then hexane (2 mL) was added to recover the fatty acids. The obtained fatty acids were converted into fatty acid methyl esters by treating with boron trifluoride (1mL) in methanol. The identification of fatty acids was carried out using standard fatty acid methyl esters. The prepared fatty acid methyl esters were analyzed using a gas chromatograph (model GC-2010, Shimadzu Corporation, Kyoto, Japan) equipped with a data processor and fitted with an FID detector and RTx-1 fused silica column (30 m × 0.32 mm id). Injector temperature: 230 °C, detector temperature: 250 °C and column oven initial temperature: 120 °C with 5 °C increase per minute and finally held at 220 °C for 10 minutes. The total run-time of the program was 30 minutes.

All experiments and analyses were carried out in triplicate, and the obtained results were analyzed for mean and standard deviations. Data were analyzed using the statistical program-GraphPad InStat Demo [DATASET1.ISD]. A two-tailed p value was determined to show the significant differences and a significant difference value was considered only when the p value ≤0.05.

The three main objectives of the current study were to analyze the physico-chemical characteristics of two different DODs, the purification of DODs and the preparation of high purity phytosterols. The two DOD samples (DOD-1 and DOD-2) were examined for their physico-chemical characteristics and are presented in Table 1. Both the DODs were dark brown in color, of soft solid consistency at room temperature. The free fatty acid value, peroxide value and color values of DOD-1 and DOD-2 were 59.2%, 86.0%; 2.93 meq O₂· kg⁻¹, 2.87 meq O₂·kg⁻¹, and 55.0, 52.0 Lovibond units, respectively. The neutral oil (NO) contents were 22.1% and 6.3% for DOD-1 and DOD-2 with the saponification values of 153.6 mg KOH·g⁻¹ and 176.7 mg KOH·g⁻¹ indicating varied glyceride composition and unsaponifiable matter. The acylglycerol composition of the DOD-1 was 18.7%, 2.9% and 0.2% for triacylglycerol, diacylglycerol and monoacylglycerol, whereas for DOD-2 it was 5.4%, 0.7% and 0.05%, respectively, indicating variations in the compositions of the two DODs studied. Also, the unsaponifiable matter and phytosterol contents were 18.7%, 7.7% and 8.7%, 4.2% for DOD-1 and DOD-2, respectively, indicating variations in their compositions of unsaponifiable matter. The PCF of DOD-1 was chosen for the preparation of high purity phytosterols as it had higher unsaponifiable matter and phytosterol content when compared with DOD-2. The total tocopherol contents (which includes all tocopherols and tocotrienols) of DOD-1 and DOD-2 were 1238.7 and 1095.1 mg·kg⁻¹. The PCF fraction of the DODs may also be considered as raw materials for the preparation of a nutraceutical products having the original nutraceuticals present in the starting DODs. The deacidified PCF from DOD-1 and DOD-2 had enriched unsaponifiable matter (70.3%, 54.2%), phytosterols (34.4%, 31.1%) and total tocopherol (4800 mg·kg⁻¹, 6900 mg·kg⁻¹) contents making it a product suitable for further purification and utilization as a nutraceutical product for use in foods.

The presence of nutraceuticals such as phytosterols, tocopherols and tocotrienols in DOD make it a valuable raw material for their isolation. DOD contains phytosterols in the range of 4.2–8.7%, which makes DOD the obvious starting material for the isolation and preparation of phytosterols (Figure 1) (vegetable oils contain 0.1–1.0% of phytosterols). The major disadvantage of the DOD is the presence of bad odor, brown color and toxic oxidized substances. DOD was declared as having a brown color and bad odor after comparing it with the physically refined RBO available commercially in the Indian markets. The results discussed henceforth are presented in Table 2. The physical appearance of the PCF of DOD-1 and DOD-2 had a light yellow color, and soft solid consistency at room temperature. The free fatty acid contents of the PCF of DOD-1 and DOD-2 were 64.0% and 80.8%, whereas the peroxide values of DOD-1 and DOD-2 were 5.83 meq O₂·kg⁻¹ and 14.0 meq O₂·kg⁻¹. The unsaponifiable matter and neutral oil contents of DOD-1 and DOD-2 were 24.6%, 7.9% and 12.0%, 9.8% in the PCF. The color of the PCF of DOD-1 and DOD-2 was reduced by 72.8% (color units reduced to 15 from 55 units) and 73.0% (color units reduced to 14 from 52 units). The color values of BCBOF of DOD-1 and DOD-2 were increased by approximately 100% (color units increased to 109 from 55 units) and 111.5% (color units increased to 110 from 52 units). The phytosterol contents of DOD-1 and DOD-2 were enriched by 42.5% (Increased to 12.4% from 8.7%) and 40.2% (Increased to 5.9% from 4.2%) in the PCF. As the PCF was free of any bad odor with 90% of phytosterols in it and the BCBOF fraction had a bad odor and brown color, the phytosterol content of BCBOF was back-calculated to have 5–10% of the starting DOD. This is probably the first report on the decolorization and deodorization of DOD from rice bran oil or any other DOD studies reported so far.

As the DODs were of high FFA contents, (59.2% and 87.0%) it was interesting to find out whether a single fatty acid predominates or is a mixture of a couple of fatty acids resembling that of the commercially available refined RBO. The fatty acid composition of the DOD-1 and its fractions such as neutral oil (NO) and free fatty acids (FFA) were in the range of lauric acid 0.3–1.4%, myristic acid 0.61–0.87%, palmitic acid 19.6–26.8%, stearic acid 1.0–1.6%, oleic acid 35.1–37.8%, linoleic acid 28.0–34.4% and linolenic acid 0.9–1.3%. After the purification of DOD, the fatty acid compositions of PCF and BCBOF of both DOD-1 and DOD-2 were examined and indicated similarity with the starting DODs. The fatty acid composition of the PCF and BCBOF of DOD-1 was lauric acid (0.7–1.2%), myristic acid (0.65–0.68%), palmitic acid (24.2–28.3%), stearic acid (1.4–1.5%), oleic acid (37.9–39.4%), linoleic acid (31.2–32.5%) and linolenic acid (0.5–0.6%) which was almost similar to the fatty acid composition of the refined rice bran oil (except linoleic acid content) (Table 3).

The fatty acid composition of the DOD-2, NO and free fatty acid fractions were in the range of lauric acid (0.5–1.4%), myristic acid (0.75–0.84%), palmitic acid (23.0–27.5%), stearic acid (0.6–1.8%), oleic acid (32.5–36.3%), linoleic acid (30.8–32.8%) and linolenic acid (1.2–1.7%), respectively (Table 3). The fatty acid composition of the PCF and BCBOF of DOD-2 was lauric acid (0.5–0.6%), myristic acid (0.38–0.68%), palmitic acid (16.3–19.7%), stearic acid (1.8–1.9%), oleic acid (36.1–42.0%), linoleic acid (37.8–39.3%) and linolenic acid (1.2–1.3%). Even though the fatty acid composition of DOD-1 was slightly different from rice bran oil, the fatty acid compositions of different fractions of DOD-1 showed insignificant differences. The PCF and BCBOF of DOD-1 did not show any significant difference in the fatty acid composition, and the same thing applies for DOD-2. Gopala Krishna (1993) identified the compounds responsible for the dark color in crude and dewaxed rice bran oil. The fatty acid composition of crude RBO was palmitic acid (36.5%), oleic acid (25.4%) and linoleic acid (36.6%) whereas in the case of dewaxed RBO only palmitic acid (83.1%) was present. In the current study, in both DOD-1 and DOD-2, palmitic, oleic and linoleic acids contributed to 95.1–96.1% of the total fatty acids in BCBOF. The presence of these fatty acids either in oxidized form or as oxidized monoglycerides may be responsible for the brown color of both DOD-1 and DOD-2. This is the first report regarding the compounds responsible for the brown color of the Indian DOD and fatty acid composition of different fractions such as NO, FFA, PCF and BCBOF of both DOD-1 and DOD-2.

The HPLC profile of the phytosterols present in the starting DOD, unsaponifiable matter, crude product and crystallized product were: campesterol: 1.0, 4.73, 8.02 and 9.53%; stigmasterol: 3.68, 17.37, 29.41 and 34.94%; and β-sitosterol: 4.02, 19.0, 32.19 and 38.23%, respectively (Table 4), which agrees with the results of Van Hoed et al. (2006). The new finding, which is specific to this study, was the purification of DOD to make it free of bad odor and lighter in color, which is enriched in phytosterol contents (40.2–42.5%) in both DOD-1 and DOD-2, when compared with the starting DOD. As the DOD was purified, the obtained phytosterol mixture was free of bad odor.