PROCESSING AND PRODUCTS

Transcription

1 PROCESSING AND PRODUCTS Moisture Content and Particle Size of Dehydrated Egg Yolk Affect Lipid and Cholesterol Extraction Using Supercritical Carbon Dioxide 1,2 G. W. FRONING,3 R. L. WEHLING, S. CUPPETT, and L. NIEMANN Department of Food Science and Technology, University of Nebraska-Lincoln, Lincoln, Nebraska ABSTRACT Egg yolk was spray-dried under conditions to produce a small particle size powder and a large particle size powder. Particle size was determined using a Nikon Optiophot microscope. Spray-dried egg yolk was also adjusted to various moisture levels as follows: control (2 to 4% moisture), 7% moisture, and 12% moisture. Supercritical carbon dioxide extraction (SCE) of each of these moisture treatments at 45 C/306 atm using 30 g CO 2 /g of sample was completed. For the particle size study, 45 g CO 2 /g of sample at 45 C/306 atm was utilized. Particle size exhibited a significant effect on cholesterol and lipids extracted using SCE. As moisture content of dried egg yolk increased to 7%, there was a significant increase in lipids extracted using supercritical carbon dioxide. Moisture content had no significant effect on cholesterol extraction. After extracting SCE higher moisture spray-dried egg yolk, sponge cake volume was significantly reduced compared to that of the control. The reduced sponge cake volume may be due to protein denaturation. (Key words: supercritical fluid extraction, egg yolk, moisture content, particle size) 1998 Poultry Science 77: INTRODUCTION Fat and cholesterol reduction of processed egg yolk has been a primary concern of the egg industry. Consumers continue to emphasize low cholesterol and low lipid food products. Supercritical carbon dioxide extraction (SCE) has been considered as a feasible processing technology for reducing the lipid content of animal products because carbon dioxide is nontoxic and has a low supercritical temperature. Previous research has indicated that supercritical fluid technology can effectively reduce cholesterol and lipid levels, particularly in dehydrated animal products, including egg yolk, beef, and chicken (Froning et al., 1990, 1994; Wehling et al., 1992); SCE of liquid animal products or intact muscle has been found to be much less efficient (King et al., 1989; Froning et al., 1990). The effects of moisture content and particle size on efficiency of extraction of lipid components have received little emphasis. Eldridge et al. (1986) reported that the best SCE conditions for defatting soybean flakes were 12,000 psi, 85 C, and moisture levels of 10.5 to Received for publication February 9, Accepted for publication June 24, Published as paper Number 12118, Journal Series, Nebraska Agricultural Research Division, Lincoln, NE This research was funded by the Southeastern Egg and Poultry Association. 3To whom correspondence should be addressed: gfroning@ foodsci.unl.edu 11.5%. Recently, Zeidler et al. (1994) observed that use of a co-solvent such as ethanol or methanol with reduction of moisture content from 50 to 35% substantially increased the solubility of cholesterol. These same researchers added safflower oil to egg yolk to reduce the moisture content prior to SCE. Addition of safflower oil reduced the moisture content to 17%, thereby improving conditions for cholesterol extraction. Previous research has indicated that increased particle size may improve lipid and cholesterol extractability using supercritical carbon dioxide. Wehling et al. (1992) observed that lipids were more easily extracted from dehydrated beef chunks than dehydrated beef powder when utilizing SCE. Froning et al. (1994) noted similar results when supercritical carbon dioxide was used to extract cholesterol and lipids from dehydrated chicken. Wehling et al. (1992) indicated that irregularly shaped chunks do not pack as tightly in the extraction vessel as the powder particles, thereby providing less resistance to flow. The objective of this study was to investigate the effects of particle size and moisture content of dehydrated egg yolk on the efficiency of extraction of cholesterol and lipids using SCE. If relationships exist, perhaps guidelines can be established for optimum conditions for extraction. Abbreviation Key: SCE = supercritical carbon dioxide extraction. 1718

2 SUPERCRITICAL CARBON DIOXIDE EXTRACTION OF DRIED EGG YOLK 1719 TABLE 1. Effect of supercritical carbon dioxide extraction (SCE) of higher moisture content dried egg yolk on proximate composition, cholesterol, and phosphatidyl choline content Phosphatidyl Sample Moisture Protein 1 Fat 1 Cholesterol 1 choline (%) (mg/g) Initial Control 3.13 ± 0.44 C ± 0.92 C ± 0.72 A ± 0.81 A ± 3.7 C SCE 2.06 ± 0.31 C ± 0.65 B ± 0.51 C ± 0.58 B ± 2.6 B 7% Control 7.04 ± 0.44 B ± 0.92 C ± 0.72 AB ± 0.81 A ± 3.7 C SCE 3.33 ± 0.31 C ± 0.65 A ± 0.51 D ± 0.58 B ± 2.6 A 12% Control ± 0.44 A ± 0.92 C ± 0.72 B ± 0.81 A ± 3.7 C SCE 6.80 ± 0.31 B ± 0.65 A ± 0.51 D ± 0.58 B ± 2.6 A A-FMeans ± SEM in the same column with no common subscript differ significantly (P < 0.01). 1Values are reported on a dry basis. MATERIALS AND METHODS Sample Preparation for Moisture Study Dehydrated egg yolk, including type Y-1 (standard yolk) and Y-2 (glucose removed), for the moisture study was obtained from Henningsen Foods Inc.4 Egg yolk samples for each of the two replicates were obtained from different lots. Initial moisture content of each sample was determined. The water required to adjust spray-dried yolk to 7 and 12% moisture was added dropwise over a 5-min time period while mixing at speed 1 with a wire whip attachment in a Hobart Model N-50 mixer.5 Samples were mixed an additional 15 min and allowed to equilibrate for 2 wk in air tight plastic bags at 3 C, then stored at 31.8 C before extraction. Sample Preparation for Particle Size Study Liquid egg yolk from two different lots was obtained from Henningsen Foods Inc.6 for each of two replicates. Samples were dried using a Henningsen Foods Model 20-5 spray dryer4 custom designed for the University of Nebraska-Lincoln Food Processing Center Pilot Plant. The inlet temperature was 167 to 173 C, the outlet temperature was 69 to 72 C, and the atomizing air pressure was 2.0 atm. A medium internal mix nozzle7 was used to spray-dry the samples with small particle size. For the samples with large particles, a large external mix nozzle7 was used. Additionally, samples with large particles were collected from the main chamber, whereas samples with small particles were taken from the cyclone portion of the spray drier, where the finest particles collect. 4Henningsen Foods Inc., Omaha, NE Hobart Corp., Troy, OH Henningsen Foods Inc., David City, NE Spray Systems Co., Wheaton, IL Nikon Inc., Melville, NY Extraction Conditions Dried egg yolk was extracted with supercritical carbon dioxide using the SCE apparatus described by Froning et al. (1990). Approximately 100 g of dehydrated egg yolk was placed in the extraction vessel and extracted with commercial grade CO 2. Samples were extracted at 45 C/ 306 atm, which provided a fluid density of 0.90 g/cm3. Extraction vessel pressure was maintained at ± 3.5 atm and temperature was maintained at ± 1 C. Pressure in the separation vessel was regulated at 34 ± 3.5 atm for all extractions. The flow rate of carbon dioxide through the system was 5 to 10 L/min. For the moisture study, 30 g of CO 2 /g sample was used, whereas 45 g of CO 2 /g sample was utilized for the particle size study. The lower quantity of CO 2 was used for the moisture study to insure that exhaustive extraction did not occur, allowing for a true measure of the effect of the moisture content. Duplicate extractions were performed for each replicate. Analytical tests were performed in duplicate on each extracted or control sample. Analytical Techniques Total lipids of control and extracted samples were determined by solvent extraction with chloroform/ ethanol ( Association of Official Analytical Chemists, 1990). Protein was measured by the Kjeldahl method (925.31, Association of Official Analytical Chemists, 1990) using a conversion factor of 6.25, and moisture content was determined by the vacuum oven drying method (925.30, Association of Official Analytical Chemists, 1990). Cholesterol was extracted and analyzed using HPLC as reported by Froning et al. (1990). Phosphatidyl choline was also determined by HPLC using the method reported by Froning et al. (1990). Particle size distribution was measured using a Nikon Optiphot microscope8 as described by Froning (1979).

3 1720 FRONING ET AL. TABLE 2. Effect of supercritical carbon dioxide extraction (SCE) of higher moisture content dried egg yolk on functional properties and Hunterlab color values Sponge Emulsion Hunterlab color values Sample cake volume stability L a L b L (cm 2 ) (g oil/5 g mayonnaise) Initial Control 277 ± 8.4 A ± A ± 0.61 A 0.77 ± 0.58 B ± 0.39 B SCE 270 ± 5.9 A ± A ± 0.43 A 0.92 ± 0.41 B ± 0.27 B 7% Control 267 ± 8.4 A ± A ± 0.61 B 4.45 ± 0.58 A ± 0.39 A SCE 235 ± 5.7 B ± A ± 0.41 A 0.68 ± 0.40 B ± 0.26 B 12% Control 252 ± 8.4 A ± A ± 0.61 B 5.82 ± 0.58 A ± 0.39 A SCE 203 ± 5.9 C ± A ± 0.43 A 1.04 ± 0.41 B ± 0.27 B A-CMeans ± SEM in the same column with no common subscripts differ significantly (P < 0.01). Functional and Color Properties Emulsion stability (mayonnaise test) was determined using the method reported by Lieu et al. (1978) and sponge cake volume as described by Froning et al. (1986). Color values (L, a L, and b L values) were measured using a Hunterlab Tristimulus Colorimeter Model D25 M-9.9 Samples for color measurement were packed into 50 mm (diameter) 9 mm Falcon Type 1006 plastic Petri dishes10 and measured in triplicate. A white color plate (Number C2-6070) having an L value of 95.47, a L value of 0.61, and b L value of 0.40 was used to standardize the instrument. Statistical Analysis Data were subjected to analysis of variance using SAS software (SAS Institute, 1985). A linear model used type by moisture level by replicate by treatment mean square as the error term for the moisture study, whereas replicate by size by treatment mean square was utilized for the particle size study. Differences between means were determined using the SAS least significant mean method. Significance levels were set at 1 or 5%, and levels used are indicated in the tables. Moisture content of dried egg yolk prior to SCE exhibited a significant effect on efficiency of fat extraction (Table 1). Because the type of yolk (Y-1 or Y-2) did not significantly affect any results, both types were combined for presentation of results. Fat extraction was significantly improved as the moisture content of the dried yolk was increased to 7%; however, an additional increase in moisture content to 12% did not give further improvement in the fat extraction. Changes in moisture content did not significantly affect extraction of cholesterol. As found in previous research (Froning et al., 1990), protein and phosphatidyl choline were concentrated with the removal of fat and cholesterol, with the greatest concentration occurring at the higher moisture levels. Although the higher moisture levels did not significantly improve extraction of cholesterol, extraction at or above 7% moisture would be recommended to achieve increased lipid removal. Table 2 presents data on how SCE of dried egg yolk solids at different moisture contents affects functional properties. Sponge cake volume was significantly lowered as moisture content of dried egg yolk was increased prior to SCE. Although emulsion stability appeared to be adversely affected at the higher moisture levels, the results were not significant. The effect on cake volume may be related to possible protein denaturation when using SCE on yolk samples at the higher moisture levels. Color values are presented in Table 2. Moisture content did not significantly affect color of SCE samples as compared to the initial samples with no moisture RESULTS AND DISCUSSION 9Hunter Associates Inc., Reston, VA Falcon Co., Oxford, CA TABLE 3. Particle size distribution from spray dried egg yolk dried under conditions to produce small or large particle sizes Small particle Large particle Particle size size group size group (mm) (%) 0 to to to to to to to to to

4 SUPERCRITICAL CARBON DIOXIDE EXTRACTION OF DRIED EGG YOLK 1721 TABLE 4. Effect of spray dried egg yolk particle size on efficiency of supercritical carbon dioxide extraction (SCE) 1 Phosphatidyl Sample Moisture Protein 1 Fat 1 Cholesterol 1 choline (%) (mg/g) Small Control 1.52 ± 0.66 a ± 1.79 c ± 2.07 a ± 1.36 a 141 ± 14.9 b SCE 1.43 ± 0.43 a ± 1.15 b ± 1.34 a ± 0.87 a 194 ± 9.6 a Large Control 3.73 ± 0.66 a ± 1.79 c ± 2.07 a ± 1.36 a 160 ± 14.9 b SCE 1.42 ± 0.43 a ± 1.15 a ± 1.34 b ± 0.87 b 220 ± 19.6 a 1Values are reported on dry basis. a-cmeans ± SEM in the same column with no common subscript differ significantly (P < 0.05). adjustment. However, when comparing the unextracted moisture adjusted treatments, L, a L, and b L values were significantly different than that of the SCE moistureadjusted groups. Although the 7% and 12% moistureadjusted sample color values were not significantly different, SCE did significantly increase L values and significantly decrease a L and b L values as compared to the unextracted moisture-adjusted treatments. Particle size data are shown in Tables 3, 4, and 5. Particle size distribution of dried yolk indicated that the samples designated as having large particles had substantially more large particles than that observed from the small particle size group (Table 3). The large particle size group had 32.2% of the particles that were 25 mm or greater, whereas the group designated as the small particle size group only had 0.6% particles 25 mm or greater and none above 50 mm. The SCE of dried egg yolk samples with a large particle size significantly removed more fat and cholesterol than that noted from the small particle size group (Table 4). This is probably due to easier and more uniform penetration of CO 2 throughout the sample, which minimizes the formation of channels through the particle bed. For large particle samples, equivalent extraction levels can be achieved with a smaller CO 2 volume than that required with small particles. As lipids were extracted, the protein content was again concentrated as well as the phosphatidyl choline. Moisture contents of all particle size groupings were not significantly different. Control samples with a large particle size produced sponge cake volumes significantly higher than all of the other treatments (Table 5); however, cake volumes from SCE samples with varying particle sizes were not significantly different. Emulsion stability (Table 5) was not significantly affected by the particle size and SCE. The L values of large particle size dried egg yolk samples were significantly increased when subjected to SCE (Table 5). The b L values were significantly decreased when dried egg yolk of either particle size was extracted with SCE. Hunterlab a L values were not significantly affected by any of the treatments. These studies separately investigated moisture content and particle size as they affect the efficiency of SCE. Both moisture content and particle size have been shown to influence lipid removal during SCE. Future studies should likely focus on the combined effect of moisture content and particle size to ascertain whether any synergistic influence is observed. TABLE 5. Functional properties and Hunterlab color values of small and large particle size egg yolk extracted using supercritical carbon dioxide (SCE) Sponge Emulsion Hunterlab color values Sample cake volume stability L a L b L (cm 2 ) (g oil/5 g mayonnaise) Small Control 306 ± 2.5 b ± a ± 0.48 a 1.69 ± 0.83 a ± 0.35 b SCE 294 ± 1.6 b ± a ± 0.31 a 0.64 ± 0.54 a ± 0.23 c Large Control 323 ± 2.5 a ± a ± 0.48 b 0.32 ± 0.83 a ± 0.35 a SCE 304 ± 1.6 b ± a ± 0.31 a 1.35 ± 0.54 a ± 0.23 b a-cmeans ± SEM in the same column with no common subscript differ significantly different (P < 0.05).

5 1722 REFERENCES Association of Official Analytical Chemists, Official Methods of Analysis. 15th ed. Association of Official Analytical Chemists, Arlington, VA. Eldridge, A. C., J. P. Friedrich, K. Warner, and W. F. Kwolek, Preparation and evaluation of supercritical carbon dioxide defatted soybean flakes. J. Food Sci. 51: Froning, G. W., Characteristics of bone particles from various poultry meat products. Poultry Sci. 58: Froning, G. W., F. Fieman, R. L. Wehling, S. L. Cuppett, and L. Niemann, Supercritical carbon dioxide extraction of lipids and cholesterol from dehydrated chicken. Poultry Sci. 73: Froning, G. W., R. L. Wehling, S. L. Cuppett, M. M. Pierce, L. Niemann, and D. K. Siekman, Extraction of cholesterol and other lipids from dried egg yolk using supercritical carbon dioxide. J. Food Sci. 55: Froning, G. W., J. C. Acton, H. R. Ball, C. J. Brekke, R. J. Hasiak, and N. A. Cox, Recommended Methods for Analysis FRONING ET AL. of Eggs and Poultry Meat. North Central Regional Research Publication No University of Nebraska, Lincoln, NE. King, J. W., J. H. Johnson, and J. P. Friedrich, Extraction of fat tissue from meat products with supercritical carbon dioxide. J. Agric. Food Chem. 37: Lieu, Eng-Hwa, G. W. Froning, and R. Dam, Effect of storage on lipid composition and functional properties of dried egg products. Poultry Sci. 57: SAS Institute, SAS Users Guide. SAS Institute Inc., Cary, NC. Wehling, R. L., G. W. Froning, S. L. Cuppett, and L. Niemann, Extraction of cholesterol and other lipids from dehydrated beef using supercritical carbon dioxide. J. Agric. Food Chem. 40: Zeidler, G., G. Pasin, and A. King, Removing cholesterol from liquid egg yolk by carbon dioxide supercritical carbon dioxide. Pages in: Egg Uses and Processing Technologies New Developments. J. S. Sim and S. Nakai, ed. CAB International, Wallingford, Oxon, UK.