RICE QUALITY AND PROCESSING Evaluation of Soxtec System Operating Conditions for Surface Lipid Extraction from Rice A.L. Matsler and T.J. Siebenmorgen ABSTRACT The degree of milling (DOM) of rice is a measure of how well the germ and bran layers were removed from the surface of rice kernels during milling. Since the majority of rice kernel lipids are found on the surface, measuring the surface lipid content (SLC) of rice after milling is one way to quantify the DOM of rice. While there are several methods to measure the lipid content (LC) of rice, there is not an established standard method for determining the SLC of milled rice. The objective of this study was to evaluate the primary operating variables of a Soxtec apparatus in measuring the SLC of milled rice. This was accomplished by varying the pre-extraction drying, boiling, rinsing and post-extraction drying durations, in order to achieve the maximum extraction of lipids from rice. Experiments were performed on stored Oryza sativa L. Cypress and Bengal rice. Results showed that a 1 h pre-extraction drying duration, 20 min boiling duration, 30 min rinsing duration, and 30 min postextraction drying duration provided the maximum lipid extraction from milled head rice with petroleum ether. 1 This is a completed study. INTRODUCTION Degree of milling (DOM) of rice refers to the extent to which the germ and bran layers of brown rice kernels are removed during the milling process. Thorough milling produces well-milled rice that has more bran removed and has higher starch 378
B.R. Wells Rice Research Studies 2003 content with lower lipid and protein levels than under-milled rice (Chen et al., 1998; Perdon et al., 2001). Since bran is located on the surface of the rice kernel and is chiefly made up of lipids (15 to 20%) (Juliano, 1985), the surface lipid content (SLC) of rice is directly related to how well the rice was milled, and thus gives an indication as to the DOM of the rice (Hogan and Deobald, 1961; Miller et al., 1979; Pomeranz et al., 1975). The traditional chemical analysis methods for lipid content (LC) in grain and animal feeds are the Soxhlet and Goldfisch procedures (AACC methods 30-20 and 30-25). The traditional Soxhlet and Goldfisch procedures, however, are time-consuming, requiring up to six hours per extraction, with high solvent consumption. The Soxtec is a system that has recently become widely utilized for lipid extraction. While the Soxtec operation consists of basically the same extraction procedure as the Soxhlet or Goldfisch procedures, the Soxtec provides a faster, more automated, less solventconsumptive method for performing solvent extractions. Currently there is no standard or officially accepted method for analyzing milled rice SLCs with the Soxtec. As the use of the Soxtec becomes more widespread, the importance of developing a standard method for its operation becomes more vital. The objective of this study was to evaluate the effects of Soxtec apparatus operating conditions by varying the pre-extraction drying, boiling, rinsing, and postextraction drying durations on the extraction of lipids from milled rice. MATERIALS AND METHODS Sample Preparation Cypress and Bengal rice, harvested in the fall of 2001 from the Rice Research and Extension Center at Stuttgart, Ark., at moisture contents (MCs) of 19.2% and 16.7%, respectively, were used for this experiment. After drying to approximately 12.5% MC, 150-g samples of rough rice were dehulled (Rice Machine, type THU, Satake, Tokyo, Japan) and the resulting brown rice was milled in a laboratory mill (McGill #2, RAPSCO, Brookshire, Texas). Samples were milled for 10, 30 or 60 s to obtain samples with a wide range of SLCs. Head rice (milled kernels >/= 75% of the original kernel length) was separated from brokens using a sizing device (Seedburo Equipment Co., Chicago, Ill). Experimental Design In order to determine the effect of extraction conditions on the lipid amounts removed from the rice, various durations of the four operations comprising a Soxtec system extraction (pre-extraction drying, boiling, rinsing, and post-extraction drying) were tested. Each Soxtec operation variable was tested separately. Pre-extraction drying durations tested were 0, 1, 2, 4, or 24 h. Boiling and rinsing durations tested were 0, 10, 20, 30, or 40 min. Post-extraction drying durations of 0, 0.5, 1, 2, 379
AAES Research Series 517 or 4 h were evaluated (Table 1). Each operation duration was tested in triplicate with six samples comprising each extraction replication for a total of 18 samples tested per experimental treatment combination for both Bengal and Cypress rice. Once an operation variable was tested, statistical analysis was performed using JMP software (version 5.0.1.2, SAS Institute, Cary, N.C.) to compare the mean values of the 18 samples tested for each experimental treatment duration. Based on the outcomes of the statistical analysis testing for an operation, the experimental treatment level that resulted in the greatest SLC for that operation was held constant for subsequent testing of the other variables. Lipid Extraction Surface lipid extraction was performed using a Soxtec system (Avanti 2055, FOSS North America Inc., Eden Prairie, Minn.). To evaluate the effect of the preextraction drying duration, thimbles containing rice samples were dried in a convection oven maintained at 100 C. After pre-extraction drying, samples were placed in a desiccator to cool for approximately 10 min. After cooling in the desiccator, the thimbles containing the rice were attached to the magnets below the condensers in the Soxtec unit (Fig. 1). When all six samples were in place, extraction cups were placed on the hot plate below the thimbles and 70 ml of petroleum ether (boiling point 35-60 C) was measured into each cup. When the extraction cups were secured into place and the hot plate below the extraction cups was heated to 135 C, the extraction cycle was started by immersing the thimbles into the solvent (Fig. 1). After the specified boiling duration, the thimbles were raised out of the solvent and the rinsing operation was initiated. During rinsing, the evaporated solvent from the extraction cups condensed when contacting the condensers, which had approximately 20 C cooling water running through them (Fig. 1). The condensed solvent dripped down through the sample to rinse remaining lipids from the rice into the extraction cups. After the rinsing duration, the lever on the Soxtec was moved into the solvent collection position to stop the solvent flow from the condensers back into the cups. The remaining solvent in the extraction cups was evaporated out of the cups and collected in the solvent reservoir of the Soxtec for later disposal. The cups containing the extracted lipids were removed from the Soxtec and placed into an oven maintained at 100 C. After the post-extraction drying operation, the extraction cups containing the extracted lipids were placed in a desiccator to cool to room temperature for approximately 30 min. After cooling, the extraction cups were removed from the desiccator and weighed. The difference in weight between the cups containing the extracted lipids and boiling beads and the original weight of the cups and beads was then calculated to obtain the weight of the extracted lipids. The SLC was calculated as the weight of extracted lipids expressed as a percentage of the weight of the original milled rice sample. 380
B.R. Wells Rice Research Studies 2003 RESULTS AND DISCUSSION The pre-extraction drying duration providing the greatest SLCs for both Bengal and Cypress across all milling durations was 1 h (Table 2A). The 20 min boiling duration provided the greatest extraction for both Bengal and Cypress rice milled for all durations (Table 2B). Both the 30 min rinsing and 30 min post-extraction drying durations showed a slightly greater level of extracted lipids (Tables 2C and 2D, respectively) for Bengal and Cypress rice milled for most durations, except for Bengal milled for 30 s. This rice had the greatest extraction level with zero rinsing and zero post-extraction drying durations. The reason for this finding is unknown and was not observed on the Cypress rice. Since the SLCs of all other samples were greatest with a 30 min rinsing and 30 min post-extraction drying duration, these durations were selected as the recommended rinsing and post-extraction drying. SIGNIFICANCE OF FINDINGS In evaluating the primary operating variables of a Soxtec extraction system for use in determining the SLC of milled rice, the extraction durations providing the greatest SLC values across both Cypress and Bengal samples milled for 10, 30, and 60 s durations were: 1 h pre-extraction drying, 20 min boiling, 30 min rinsing, and 30 min post-extraction drying. The next step in this research is to determine if this Soxtec procedure is extracting only surface lipids from the rice kernels, or is also extracting lipids from within the kernel. This information is necessary to ensure that the proposed method can be used as a reference method in comparing more rapid methods of determining the DOM of rice. Developing a rapid method of determining the DOM of rice will benefit the rice milling industry by providing a tool that will help maintain uniformity in DOM measurement. LITERATURE CITED American Association of Cereal Chemists. 2000. Approved methods of the AACC, 10th ed. Methods 30-20 and 30-25. The Association: St. Paul, Minn. AOCS. 1993. Official methods and recommended practices of the American Oil Chemists Society. 4th ed. Method Ac 3-44. The Society: Champaign, Ill. Chen, H., T.J. Siebenmorgen, and K. Griffin. 1998. Quality characteristics of longgrain rice milled in two commercial systems. Cereal Chem. 75:560-565. Hogan, J.T. and H.J. Deobald. 1961. Note on a method of determining the degree of milling of whole milled rice. Cereal Chem. 38:291-293. Juliano, B.O. 1985. Rice Bran. Pages 647-688 in: Rice Chemistry and Technology. Bienvenido O. Juliano, ed. The Association: St. Paul, Minn.. Miller, B.S., M.S. Lee, Y. Pomeranz, and R. Rousser. 1979. A rapid, objective method to measure the degree of milling in rice. Cereal Chem. 56:172-180. 381
AAES Research Series 517 Perdon, A.A., T.J. Siebenmorgen, A. Mauromoustokos, V.K. Griffin, and E.R. Johnson. 2001. Degree of milling effects on rice pasting properties. Cereal Chem. 78:205-209. Pomeranz, Y., R.A. Stermer, and E. Dikeman. 1975. NMR-oil content as an index of degree of rice milling. Cereal Chem. 52:849-853. Condenser through which cold water runs. Arm that raises and lowers thimble into the extraction cup Magnet to attach thimble Extraction cup - clamps to condenser and contains solvent during extraction. Thimble containing rice sample Hot plate on which extraction cup sits Fig. 1. Schematic and explanation of Soxtec system components. 382
B.R. Wells Rice Research Studies 2003 Table 1. Experimental design for evaluating the four operations of the Soxtec Avanti 2055. Each row represents the Soxtec operation being evaluated in a particular experiment and the variable durations used while testing each specific operation. Operation Pre-extraction Boiling Rinsing Post-extraction being evaluated drying duration duration drying (h) -------------- (min) ------------- (h) Pre-extraction drying 0, 1, 2, 4, 24 z 20 30 0.5 Boiling duration 1 0, 10, 20, 30, 40 30 0.5 Rinsing duration 1 20 0, 10, 20, 30, 40 0.5 Post-extraction drying 1 20 30 0, 0.5, 1, 2, 4 z Eighteen extractions were performed for each experimental treatment combination. 383
AAES Research Series 517 Table 2. Mean surface lipid contents of Bengal and Cypress rice milled for 10, 30,or 60 s attained with the indicated levels of different Soxtec operations; pre-extraction drying (A), boiling (B), rinsing (C), and post-extraction drying (D). Numbers in bold represent the treatment duration resulting in the greatest lipid extraction concentration across the two varieties and three milling durations. Each value is the average of extractions from 18 samples. Durations of the other operating variables used while testing a specific operating variable are shown in Table 1. A Bengal Cypress Pre-extraction drying duration 10 30 60 10 30 60 (h) 0 1.17 bc 0.42 c 0.25 a 0.96 b 0.41 b 0.21 a 1 1.23 a 0.51 a 0.25 a 0.99 a 0.43 a 0.22 a 2 1.21 a 0.49 a 0.25 a 0.94 a 0.41 b 0.16 b 4 1.17 c 0.48 b 0.25 a 0.91 c 0.40 b 0.17 b 24 1.20 ab 0.37 d 0.16 b 0.76 d 0.39 b 0.12 c B Bengal Cypress Boiling duration 10 30 60 10 30 60 (min) 0 1.14 b 0.40 b 0.23 b 1.0 a 0.33 b 0.19 b 10 1.21 a 0.39 c 0.24 ab 0.96 bc 0.32 c 0.18 b 20 1.24 a 0.43 a 0.25 a 0.99 ab 0.35 a 0.22 a 30 1.23 a 0.37 d 0.24 ab 0.93 c 0.32 c 0.19 b 40 1.22 a 0.40 bc 0.25 a 0.98 ab 0.34 ab 0.19 b C Bengal Cypress Rinsing duration 10 30 60 10 30 60 (min) 0 1.06 c 0.41 a 0.22 c 0.86 c 0.29 bc 0.14 c 10 1.17 b 0.39 b 0.22 c 0.93 ab 0.30 ab 0.20 ab 20 1.18 b 0.37 c 0.23 bc 0.92 b 0.31 a 0.18 b 30 1.23 a 0.38 bc 0.25 ab 0.95 ab 0.32 a 0.22 a 40 1.23 a 0.37 bc 0.26 a 0.95 a 0.28 c 0.19 b D Bengal Cypress Post-extraction drying duration 10 30 60 10 30 60 (h) 0 1.23 a 0.56 a 0.26 a 0.95 bc 0.43 a 0.20 b 0.5 1.23 a 0.48 c 0.20 ab 0.98 a 0.44 a 0.22 a 1 1.17 b 0.52 b 0.18 b 0.97 ab 0.42 a 0.17 c 2 1.17 b 0.50 c 0.22 ab 0.93 c 0.44 a 0.19 b 4 1.16 b 0.49 c 0.19 ab 0.94 bc 0.43 a 0.18 bc z Mean values in the same column with different letters represent operation durations that were significantly different. 384