Stability of Tocopherols and Retinyl Palmitate in Snack Extrudates K

JFS: Sensory and Nutritive Qualities of Food

Stability of Tocopherols and Retinyl Palmitate in Snack Extrudates K. SUKNARK, J. LEE, R.R. EITENMILLER, AND R.D. PHILLIPS

ABSTRACT: Fish- and peanut-containing half-products were obtained by extruding and drying tocopherol- and retinyl palmitate-fortified mixtures of tapioca starch and minced fish or partially defatted peanut flour (PDPF 60:40, wet basis). Half-products were puffed by deep-fat frying. Vitamins were determined simultaneously at each step of snack production using a direct solvent extraction method. Extrusion significantly reduced the content of tocopherols and retinyl palmitate in both products. Reduction of retinyl palmitate in fish and peanut extrudates during snack production was 48% and 27%, respectively. Final products contained more tocopherol than interme- diates because of the high tocopherol content in the frying oil and its uptake. Keywords: extrusion, fish, peanut, retinol, tocopherol

Introduction (1990). Reviews on the effect of extru- which is functional as an antioxidant. In XTRUSION HAS BEEN USED TO PRO- sion-cooking, especially on nutritional foods, tocopherols are used as antioxi- Educe a wide variety of foods such as quality of plant proteins, were also re- dants because the hydroxyl group of snacks, ready-to-eat cereals, textured ported by Phillips (1989) and on vitamin the chromanol ring in tocopherols acts vegetable protein, confectioneries, and retention, as reported by Killeit (1994). as an electron donor or a free-radical pet foods. To increase protein content Vitamins have different stability prop- acceptor. This reaction can delay the and nutritive value, various protein erties because of their different chemi- rate of oxidation of autoxidative species sources such as fish or peanut flour cal structures (Killeit 1994). Degradation (Nawar 1985; Counsell 1993) by sup- can be included in snack formulations of vitamins is dependent on factors en- pressing the formation of lipid peroxy prior to extrusion. Fish protein has an countered during food processing and radicals, and preventing propagation of excellent nutritive value because it con- storage, such as temperature, oxygen, the chain reaction (Sherwin 1978; Bur- tains essential amino acids and is highly light, moisture, pH, and exposure time. ton and Ingold 1986). The antioxidant digestible (Haard 1995). Moreover, fish Killeit (1994) reported that the retention activity of ␥- and ␦-tocopherols is high- is a good source of fat-soluble vitamins of vitamins decreased during extrusion, er than that of ␣- and ␤-tocopherols in such as vitamin E, found in flesh, and with an increase in barrel temperature most matrices (Nawar 1985; Rankin and vitamin A and D, found in liver (Exler and screw speed and a decrease in Pike 1993; Giese 1996). and Weihrauch 1976). Fish also contains moisture content, throughput, and die Both saponification and direct sol- several polyunsaturated fatty acids diameter. The present study seeks to vent extraction can be used to extract such as linoleic acid, eicosapentaenoic extend these observations to a previ- vitamins A and E from food materials acid, and docosahexaenoic acid (Exler ously uninvestigated system: tapioca prior to quantification (Ball 1988; Eiten- and Weihrauch 1976; Erickson 1992; starch mixed with either peanut flour miller and Landen 1995, 1999). Because Ackman 1995). Additionally, fish flavor or fish tissue that is processed in a twin- of simplicity, cost and time-saving abili- is desirable in snacks produced for the screw machine to produce half-prod- ties, direct solvent extraction is often international market. Peanut is a good ucts, then expanded by deep-fat frying. chosen over saponification as the ex- source of protein and essential fatty ac- Vitamin A can be characterized as a traction method of choice. A solvent or ids, such as oleic acid and linoleic acid long chain alcohol (Ottaway 1993). It is solvent mixtures appropriate for vita- (Adsule and others 1989). Peanut oil sensitive to oxidation in the presence min A and E extraction must effectively contains 76 to 82% oleic acid, and 30 to of oxygen, light, heat, and catalysts of penetrate the tissue and break lipopro- 35% linoleic acid (Woodroof 1983). It is oxidation (Olson 1990; Ottaway 1993). tein bonds, while minimizing oxidative also an excellent source of certain min- The ester forms, such as retinyl palmi- destruction, in order to efficiently re- erals (for example, phosphorus, calci- tate (RP) and retinyl acetate, are more move vitamin components from the um, magnesium, and potassium) and stable than retinol; therefore, they are food matrix (Ball 1988; Eitenmiller and vitamins (such as thiamine, riboflavin, used for food fortification (Olson 1990; Landen 1995, 1999). Landen (1982) suc- biotin, and folic acid) (Adsule and oth- Eitenmiller and Landen 1999). The bea- cessfully developed a method for the ers 1989). dlet forms of vitamin A are produced determination of RP and ␣-tocopherol Extrusion has both advantages and with gelatin, starch, gums, and other acetate in infant formulas, using the detrimental effects on the nutritional components or a combination of in- solvent mixture of isopropanol and me- quality of food. The changes in the gredients as encapsulating agents to thylene chloride. Vitamins were sepa- chemical and nutritional qualities dur- add oxidation stability and modify rated by high- pressure gel permeation ing extrusion-cooking have been re- functional properties (O’Brien and chromatography prior to quantification viewed by Bjorck and Asp (1983), Chef- Roberton 1993). on reversed-phase liquid chromatogra- tel (1986), and Camire and others Vitamin E is an essential nutrient phy (LC). Recently, Lee and others

© 2001 Institute of Food Technologists Vol. 66, No. 6, 2001—JOURNAL OF FOOD SCIENCE 897 Sensory and Nutritive Qualities of Food C by cooling water. C by Њ C for 12 h. Each fish blend C for 12 h. Њ A co-rotating twin-screw extruder Approximately 8-kg batches of the 8-kg batches Approximately (Model MPF 1700-30, APV Baker Ltd., with a England) Newcastle-U-Lyne, length-to-diameter ratio of 25:1 was used. The length and diameter of the extruder barrel were 750 and 30 mm, The slit die had dimensions respectively. a of 1 mm × 20 mm. The extruder had clam-shell barrel consisting of 4 inde- pendent temperature zones controlled by electrical heating, and 1 controlled zone at the die. Cooling water was cir- culated through the extruder barrel jacket and the die holder to maintain temperature. The barrel at the feed hopper was not heated and was main- tained at 25 to 30 Dough temperature was measured by thermocouples which extended through the barrel and contacted the moving material at 3 locations. Pressure at the slit die and torque were record- ed. Material was fed into the extruder volumetric inlet port with a K-Tron™ Corp., K2VT20, K-Tron feeder (Model Pitman, N.J., U.S.A.). The feeder was calibrated to determine the set point to give a feed rate of 27 g/min total materials for all experimental runs. A screw configuration with feed screws and for- ward paddles was set up to use for both formulations (as described in the previ- Extrusion processing Before processing, the mixtures were mixtures the processing, Before 7 thawed at was prepared in duplicate batches. was prepared by mixture were prepared peanut flour and 40% tapioca starch mixing 60% kg tapioca water (3.31 PDPF without Tocopherols kg PDPF). and 2.20 starch (as described above and RP were added Five-hundred for the fish product). were slowly grams of vitamin pre-mix and blended with and gradually added starch and PDPF the rest of the tapioca N-50G, Hobart in a mixer (Model U.S.A.) which was Ohio, Troy, Mfg.Co., and operat- beater, equipped with a flat 1) without water ed at low speed (Speed The moisture con- addition for 30 min. adjusted to 40% tent of this blend was directly in the extruder during the extrusion process with a calibrated, pro- & N-P31, Bran portioning pump (Type Lubbe, Buffalo Grove, Ill., U.S.A.). This approach was used because the tapioca starch/PDPF blend became sticky and difficult to feed, if adjusted to 40% water prior to extrusion. After blending, the mixture was packaged, stored, and thawed as described for the fish-based product. Each peanut blend was also prepared in duplicate. ® C. Њ -tocopherol ␤ -tocopherol on -tocopherol ␦ -tocopherol stan- ␦ -, and -, and ␥ -, -, and ␤ ␥ -, -, ␣ ␣ Approximately 8-kg batches of tapi- The dards, all-trans RP, and BHT were ob- and BHT were all-trans RP, dards, tained from Sigma Chemical Co. (St. Louis, Mo., U.S.A.). The oca and minced fish were prepared by blending 60% tapioca starch and 40% minced fish (4.8 kg tapioca starch and 3.2 kg minced fish). The tocopherols and RP were manually and thoroughly mixed with 20 g of tapioca starch. Then, the vitamin pre-mix was blended with additional tapioca starch until a final vitamin pre-mix of 500 g was prepared. The 500 g of vitamin pre-mix were slowly and gradually added and blended with the rest of the tapioca starch in a mixer (Model N-50G, Hobart This mixer U.S.A.). Ohio, Mfg.Co.,Troy, was equipped with a flat beater and op- erated at low speed (Speed 1) for 15 min. Then, minced fish was added and mixed for 15 min. At this time, the mixture contained approximately 36% wa- content was adjusted The moisture ter. Af- adding water. to 40% (wet basis) by ter blending, the ingredient mixture was packed in 3.25-mil plastic bags (Curlon was obtained from Henkel Corporation (La Grange, Ill., U.S.A.). All the chemi- cals and solvents were of high-performance liquid chromatography (HPLC) grade. Hexane, isopropanol, ethyl acetate, and anhydrous magnesium sulfate (Phil- Co. Baker J.T. from obtained were lipsburg, N.J., U.S.A.). natural natural a gum acacia carrier. Dry A vitamin carrier. a gum acacia form 250-SD in beadlet Type palmitate, by Hoff- IU/g) was provided (250,000 N.J., (Nutley, Inc. Roche mann-La palmitate dry vitamin A U.S.A.). The sucrose, modified starch, RP, contained oil, butylated hy- fractionated coconut (BHT),droxytoluene sodium benzoate, add- were Tocopherols and sorbic acid. of the fat content ed at 0.02% by weight (Nawar 1985), of the finished products experiments based on preliminary in final products to showing fat content addition as- Tocopherol be 37 to 41%. content of 41%. RP sumed a final fat mixture to was added to the ingredient 250 retinol provide approximately (28 g) in the fin- equivalents per serving ished products. This concentration of vitamin A provides approximately 25% of the Required Daily Allowance (RDA) 1996). for adults (FAO Grade 863, Curwood, Oshkosh, Wis., Oshkosh, 863, Curwood, Grade U.S.A.) and these were stored at –18 Sample preparation Reagents and standards —Vol. 66, No. 6, 2001 66, No. —Vol. C until Њ -tocopheryl ␣ C for 12 h, prior Њ bags, stored at – ® -tocopheryl acetate -tocopheryl ␣ ) JOURNAL OF FOOD SCIENCE Materials and Methods C and thawed at 7 Tocopherols were added to the ex- were Tocopherols Partially defatted peanut flour Catfish were purchased live and fil- Tapioca starch was obtained from starch Tapioca Suknark and others (1999) previously Suknark and others Њ trusion mixtures to act as antioxidants during the production of half-products. Mixed natural tocopherols in powder form were provided by Henkel Corpo- ration (Kankakee, Ill., U.S.A.). The powder contained at least 30% by weight of (PDPF), obtained from Pert Laborato- ries, Edenton, N.C., U.S.A., was further ground in a hammer mill (Model 6×14, Champion Products, Inc., Eden Prairie, Minn., U.S.A.) which was equipped with a 1.6 mm screen. The PDPF was sealed in plastic bags and stored at 7 used. leted at the DeKalb Farmer Market in Market Farmer leted at the DeKalb Atlanta, Georgia. Belly flaps and fillets a prepared from catfish were minced in belt-and-drum mechanical meat sepa- Ltd., Work Machinery (Yasagiya rator The mince Japan). PREF., Yamaguchi was sealed in Ziploc National Starch and Chemical Co., N.J., U.S.A. Bridgewater, 898 Tocopherols and RP Tocopherols Partially defatted peanut flour (PDPF Minced fish Tapioca starch Tapioca (1999) analyzed all-rac- analyzed (1999) acetate, tocopherols, and RP in extrud- and RP acetate, tocopherols, foods, using isopropanol ed weaning an extract- acetate as and hexane/ethyl followed by quantification ing solvent, recoveries LC. The with normal-phase all-rac- for RP and for extruded were 98.6% and 103.2% Like- respectively. products, soybean (1998) successfully wise, Lee and others extraction to the applied direct solvent E in peanuts and analysis of vitamin tree nuts. production of reported the successful snacks using fish- and peanut-based These foods were twin-screw extrusion. by consumers found to be acceptable 1998). Since such (Suknark and others extruded foods can be used to deliver micronutrients, the objective of this study was to determine the stability of RP and tocopherols in both fish- and peanut-based foods during extrusion and frying. 18 to use. Tocopherol and Retinyl Palmitate in Snacks . . . . Snacks in Palmitate Retinyl and Tocopherol

Sensory and Nutritive Qualities of Food Tocopherol and Retinyl Palmitate in Snacks . . . ous study by Suknark and others, 1999). screw-cap test tubes, flushed with ni- and mixed with a spatula before soni- The water pump was calibrated to de- trogen gas, then stored at –18 ЊC. All cating with an FS-30 Fisher Scientific™ termine the set point to give a 40% analyses were conducted in duplicate for 5 min. Approximately 9 g of magne- moisture content for the PDPF mixture. on samples from each extrusion run. sium sulfate was added to absorb water, Each duplicated batch was extruded followed by 30 mL of extraction solu- twice in separate runs. Proximate analysis of extruded tion (hexane:ethyl acetate in a ratio of Upon exiting the die, extrudates were products 85:15, v/v with 0.1% BHT), then the cut into 15-cm lengths and dried in a The half-products and fried prod- sample was mixed by spatula. An Lincoln Impinger® oven (Model 1450, ucts were analyzed for proximate com- amount of anhydrous magnesium sul- Lincoln Foodservice Products, Inc., position. Moisture content was deter- fate (MgSO4) equivalent to the weight of Fort Wayne, Ind., U.S.A.) at 90 ЊC for 10 mined according to Procedure 925.09 water in the mixture, plus 1 gram, was min to remove surface moisture and of the AOAC (1995). Fat content was ex- then added. The mixture was blended prevent sticking. The extrudates were tracted with petroleum ether for 18 h, with a Polytron® homogenizer at medi- sealed in Ziploc® freezer bags and using a Goldfisch extractor (Model um speed for 1 min and the homoge- stored at –18 ЊC prior to drying. The ex- 35001, Laboratory Construction Co., nizer was rinsed with 10 mL extraction trudates were cut into approximately Kansas City, Mo., U.S.A.), according to solution. The mixture was filtered 2.0 ϫ 0.1 cm2 pieces before being dried Procedure 30.25 of the AACC (1976). through a medium-porosity glass filter in a forced air oven (Blue M Electric and a vacuum bell jar filtration appara- Co., Blue Island, Ill., U.S.A.) at 50 ЊC for Water activity tus (Kontes, Vineland, N.J., U.S.A.) to a 5 h to obtain half-products with 10 to Water activity in half-products and 125-mL Erlenmeyer flask. The vacuum 12% moisture content. fried products was determined using an was released and the filter cake was AquaLab Water Saturated magnesium broken with a spatula, and washed with Deep-fat frying chloride and sodium chloride solution 10 mL of extraction solution. The filter One liter of soybean oil was heated was used to calibrate the equipment. cake was then transferred to the same in a Kitchen Kettle™ fryer (Model 125 mL-round glass bottle to process 06000, National Presto Industries, Inc., Tocopherols and RP analysis the second extraction. Five milliliters of Eau Claire, Wis., U.S.A.) for 30 min to In the tocopherols’ and RP’s stan- isopropanol (heated to near-boiling obtain equilibrium temperature of dards preparation, purity and stability point) and 20 mL of extracting solution 200 Ϯ 5 ЊC. Half-products weighing 20 g of the standards were measured by a were added to the mixture, followed by each (26 to 30 pieces) were deep-fat Beckman DU-64 Spectrophotometer® homogenization and filtration, using fried for 1.5 min. After frying, the fried (Fullerton, Calif., U.S.A.). The specific the original 125-mL Erlenmeyer flask in 1% products were drained on paper towels extinction coefficient (E 1cm) of the which to collect the combined filtrate. and cooled at ambient temperature. all-trans RP standard in isopropanol, The filtrate was then transferred to a which exhibits the maximum ultraviolet 100-mL volumetric flask and diluted to Experimental design (UV) absorption, is 960 at 326 nm (Ol- volume with extraction solution. The 1% ␣ ␤ ␥ Two formulations, a fish mixture and son, 1990). The E 1cm of -, -, -, and accurately determined volume of com- a PDPF mixture, were extruded, using ␦-tocopherols’ standard in ethanol are bined filtrate was evaporated with ni- the optimum condition of temperature 71 at 294 nm, 86.4 at 297 nm, 92.8 at 298 trogen gas, and then made to the ap- and screw speed (which were selected nm, and 91.2 at 298 nm, respectively propriate concentration of the sample from the consumer acceptance test in a (Scott 1978). with hexane-BHT solution before injec- previous study) (Suknark and others Stock standard solutions were pre- tion. The extraction of each sample was 1998). Tocopherols and RP were added pared by an accurately weighed stan- performed in triplicate. A spiked sample to duplicate formulations. The fish for- dard and dissolved in hexane with 0.01% was assayed at each step of the extrac- mulation was extruded at temperature BHT. Each stock standard was prepared tion process to obtain the recovery (Lee in the last two zones, near the die and separately. The working standard solu- and others 1999). at the die, of 97 ЊC at screw speed of 285 tion of tocopherols was prepared from a rpm. The PDPF formulation was ex- stock standard solution diluted with hex- HPLC quantification truded at temperature in the last two ane-BHT solution containing 4 com- The normal-phase HPLC system zones, near the die and at the die, of bined tocopherols. The RP working stan- consisted of an LC-6A pump equipped 100 ЊC at screw speed of 250 rpm. The dard solution was prepared separately with an RF-10A spectrofluorometric temperatures of the first (near the feed by diluting stock solution with 0.01% detector (Shimadzu Corp., Columbia, hopper), second, and third zones were BHT in hexane. Md., U.S.A.), and a SpectraSeries AS100 set at 60, 90, and 120 ЊC, respectively. autosampler (Thermo Separation Prod- Moisture content of the mixtures was Extraction ucts Inc., Calif., U.S.A.), and a 25 cm ϫ 4 adjusted to 40% (wet basis) and the feed Samples of 2 to 2.5 g of ground ex- mm, 5 µm Lichrosorb Si60 column rate was 27 g/min. trudate were accurately weighed in a (Hibar Fertigsaule RT., Darmstadt, F.R. The mixtures of raw materials, the 125 mL-round glass bottle. Hot deion- Germany) which was equipped with a extrudates immediately after extrusion, ized water (80 ЊC, about 8 mL) was add- pre-column packed with Perisorb A 30- the half-products (extrudates dried at ed to the sample with dry sample-to- 40 µm (Darmstadt, F.R. Germany). The 50 ЊC), and the fried products were water ratio of 1:4 and then it was mixed isocratic mobile phase contained 0.9% sampled and sealed in Ziploc® freezer with a spatula to dissolve the coating isopropanol in n-hexane, obtained from bags, flushed with nitrogen gas, then materials on the tocopherols and RP. J.T. Baker Chemical Co., Phillipsburg, placed in air-tight glass containers, and Isopropanol with 0.01% BHT was heat- N.J., U.S.A.. The flow rate was 1.0 mL/ stored at –18 ЊC. Samples of soybean oil ed to near-boiling point, then added in min. The mobile phase was filtered us- before and after frying were placed in the amount of 15 mL to the sample, ing a 0.22 µm nylon membrane filter

Vol. 66, No. 6, 2001—JOURNAL OF FOOD SCIENCE 899 Sensory and Nutritive Qualities of Food - C ␤ Њ -toco- ␣ (0.08) - and Retinyl ␣ -tocopherol ␣ 0.05). -tocotrienol were C and 14.6% mois- Ͻ (0.28) Њ

␤ p tocopherol palmitate* -tocotrienol in the mild ␣ (0.07) -tocotrienol. The loss of ␣ Erickson (1992) reported that the Hakansson and others (1987) studied -tocopherol and pherol and and severe conditions were about 85 to The losses of 87% and 94%, respectively. ␤ ture content). These two extrusions were conducted at a screw speed of 200 rpm and feed rate of 318 g/min. They found that the losses of both about 63 to 67% in the mild condition and 78% in the severe condition, which were lower than in the and tocopherols in that study was higher than those in the fish and peanut extrudates of this study. and 24.6% moisture content), and severe condition (197 of tocopherol in peanut extrudate com- of tocopherol in peanut during extrusion pared to fish extrudate was probably due to the difference in fatty acid composition in the different raw materials. The most predominant fatty acids occurring in fin-fish were reported to be palmitic acid (C16:0), oleic acid (C18:1), and the highly polyunsatu- such fatty acid in the n-3 family, rated as eicosapentaenoic acid (C20:5) and docosahexaenoic acid (C22:6) (Exler 1976). and Weihrauch predominant fatty acid in the total lipid fraction for three catfish strains was oleic acid, and the predominant polyunsaturated fatty acids were linoleic acid (C18:2) and docosahexaenoic acid. Fish oil contains more highly polyunsaturated fatty acids that are more susceptible to oxidation, which results in reduction of tocopherol (Machlin 1990; Li and others 1996). The susceptibility to rancidity and the rate of autoxidation increase as the number of double bonds increases in the fatty acid (Mach- lin 1990; Erickson 1992). the effect of extrusion-cooking on vitamin E in white flour from wheat with two conditions: mild condition (148 (0.15) (0.31) (0.12) (0.46) (0.11) -, -, ␥ ␣ -, ␣ -toco- Tocopherol Total ␤ ␣␤␥␦ - and ␣ (0.04) (0.01)(0.08) (0.24) (0.07) (0.13) (0.14) (0.26) (0.42) (0.43) (0.08) (0.04) C and 250 rpm. 1.22 c 0.15 c 4.12 c 2.17 c 7.66 cb 1.34 Њ -tocopherols de- -tocopherols -tocopherol was ␦ ␥ C. A lower reduction Њ - and ␥ 0.05) during extrusion. -tocopherol in the lowest -tocopherols in peanut ex- ␤ ␦ Ͻ

p -tocopherols were reduced signifi- -, and ␦ C at screw speed of 285 rpm, where- ␥ The fish mixture was extruded at a During extrusion, the fish and pea- Њ -, as the peanut mixture was extruded under conditions of 100 These conditions were chosen based on observations made during preliminary the exiting product However, studies. temperature from the extruder of both formulations was observed to be in the range of 91 to 93 temperature in the last two zones of 97 nut mixtures were extruded under slightly different conditions. creased more than the pherols (which was similar to the find- the ings in the fish extrudate). However, reduction of tocopherols in the peanut extrudate was lower than the reduction of tocopherols in the fish extrudate during extrusion. The reduction of ␤ trudate during extrusion was 18, 11, 28, of The reduction and 30%, respectively. total tocopherol in the peanut product during extrusion was 27%. The varying stability of the various forms of tocopherol is a complex function of the matrix and processing conditions. An insig- nificant loss of all tocopherol forms was found in the peanut half-products during drying (similar to the results noted for the fish half-products). Fried peanut snacks contained the highest amount of tocopherols within the series, as was true with the fish snacks. * Mean values in fat-free dry basis * Mean values in fat-free Sample materialRaw mixture 1.67 b 0.25 b 7.02 b represent the standard deviations. Numbers in parentheses (0.23) ( with different letters are significantly different Means within the same column 3.75 b (0.02) b 12.79 2.65 a Table 1—Tocopherols (mg/100 g) and retinyl palmitate (mg/100 g) of fish snack g) of fish (mg/100 palmitate retinyl g) and (mg/100 1—Tocopherols Table of extrusion process in different steps products Half-product Fried product 2.70 a 0.57 a 25.61 a 13.54 a 42.30 a 1.38 b Extrudate immediatelyExtrudate after extrusion 1.28 c 0.18 c 4.17 c (0.03) 2.24 c (0.03) 7.75 cb 1.44 found in the highest amount (8.6 mg/ 100 g) and and the Moreover, amount (0.27 mg/100 g) in the PDPF- and-tapioca starch mixture. The cantly ( tocopherol and total tocopherol con- tocopherol and total peanut snack are tents (dry basis) of the the significant shows 2. It Table in given difference of the vitamin content during snack production. - -, ␦ OF —Vol. 66, No. 6, 2001 66, No. —Vol. ␤

-, - and ␣ ␣ - and ␥ -tocopherol, ␤ EXTRACTION

(DJMRT). Signifi- © 8.02%. Recovery of Ϯ C for 1.5 min, the fish 4.28%. SOLVENT Њ

Ϯ -tocopherol was found in ␤ -tocopherol) and total toco- ␦ DIRECT

JOURNAL OF FOOD SCIENCE Results and Discussion 0.05) from the extrudates. After 0.05. HE tocopherols and RP in extruded

In the fish snack, the content (dry The data were subjected to analysis The data were subjected Recovery of tocopherols in the sam- In the peanut snack, the individual With the exception of Ͼ Յ -tocopherols, respectively. The reduc- respectively. -tocopherols, -, and p basis) of individual tocopherols ( 900

Effect of extrusion and frying on vitamin E T cant difference was established at cant difference was p Tocopherol and Retinyl Palmitate in Snacks . . . Snacks in Palmitate Retinyl and Tocopherol U.S.A.) and Mass., Westboro, Inc., (MSI The by stirring under vacuum. degassed in the of tocopherols concentrations the av- calculated, using samples were between areas compared erage peak and 330 and samples at 290 standards wave- and emission nm for excitation injections. The lengths after duplicate changed to 325 and wavelengths were and emission for 470 nm for excitation RP. the determination of differ- The mean of variance (ANOVA). using Duncan’s ence was determined test multiple range samples, using hot isopropanol and hexane/ethyl acetate at the ratio of 85:15, proved to be reliable method of sample extraction. This method elimi- nated saponification which hydrolyzes the ester linkage and produces some in- terfering compounds in the extract and others 1989), and emulsion. (Tuan ples was 100.43 RP was 95.85 ␥ the lowest amount (0.25 mg/100 g). A significant decrease in tocopherols oc- 1). De- during extrusion (Table curred creases were about 40% for pherol are shown in Table 1. All toco- Table in shown pherol are pherol forms were found in the mixture of fish-and-tapioca starch raw material, which was fortified with mixed natural tocopherols. Gamma-tocopherol was found in the highest amount (7.02 mg/ 100 g) and snacks contained all tocopherols at a significantly higher level than samples at other points of the process, because samples absorbed the soybean oil. the loss of tocopherols during drying was in the range of 1 to 5%. The contents of all tocopherols in half-products were not significantly different ( frying at 200 tocopherols and 23 and 28 % for ␤ tion of total tocopherol in fish extrudate during extrusion was 39%.

Sensory and Nutritive Qualities of Food Tocopherol and Retinyl Palmitate in Snacks . . .

Table 2—Tocopherols (mg/100 g) and retinyl palmitate (mg/100 g) of peanut peanut extrudates during extrusion and snack products in different steps of extrusion process drying was 20 and 12%, respectively. Tocopherol Total Retinyl The fat-free fried product contained Sample ␣␤␥␦tocopherol palmitate* 2.00 mg/100 g RP (dry basis) which was Raw material mixture 1.60 b 0.27 b 8.58 b 4.11 b 14.44 b 2.74 a about a 27% reduction from the initial (0.11) (0.04) (0.47) (0.54) (1.02) (0.12) amount in the raw material. Retinol was Extrudate immediately 1.32 c 0.24 bc 6.15 c 2.86 10.55 c 2.19 b also not found in any samples of peanut after extrusion (0.08) (0.02) (0.21) (0.15) (0.41) (0.05) production. Half-product 1.26 c 0.20 c 6.03 c 2.84 c 10.36 c 1.92 c Vitamin A is susceptible to oxidation (0.02) (0.01) (0.33) (0.13) (0.48) (0.03) in the presence of oxygen, and the de- Fried product 3.07 a 0.65 a 26.26 a 13.20 a 43.18 a 2.00 c (0.21) (0.07) (1.40) (0.67) (2.16) (0.10) velopment of off-flavors accompanies its degradation (Arya and Thakur 1990). Numbers in parentheses represent the standard deviations. Means within the same column with different letters are significantly different (p Յ 0.05). Lee and others (1978) studied the stabil- * Mean values in fat-free dry basis ity of vitamin A and provitamin A (caro- tenoid) during extrusion at a cooking Њ Table 3—Tocopherols (mg/100 g) of soybean oil before and after frying fish temperature of 130 C and a screw and peanut snack products speed of 700 to 1,000 rpm. They reported that ␤-carotene destruction was Tocopherol Total about 75% after extrusion, which made ␣␤␥ ␦ Sample tocopherol it uneconomical to add a provitamin A Oil before frying 6.16 a 1.29 a 60.75 a 31.38 a 99.58 a source before extrusion. The retention (0.03) (0.04) (0.36) (0.45) (0.66) of vitamin A acetate and vitamin A

Oil after frying 5.77 b 1.18 b 57.09 c 29.81 b 94.09 c palmitate were 90 to 100 and 52 to 91%, fish snack (0.07) (0.03) (0.70) (0.37) (0.91) Oil after frying 5.83 b 1.25 a 59.04 b 30.74 a 96.75 b respectively. Low screw speeds provid- peanut snack (0.12) (0.03) (0.94) (0.70) (1.66) ed long extrusion residence time and Numbers in parentheses represent the standard deviations. increased vitamin A destruction at Means within the same column with different letters are significantly different (p Յ 0.05). 130 ЊC. Cheftel (1986) noted that an increase in screw speed at low moisture content probably increased vitamin de- Table 4—Moisture content, fat content, and water activity (a ) of half-products w struction, because shear force in- and fried products creased temperature. However, an in-

Sample Moisture (wb) Fat (db) aw crease in screw speed at high moisture Fish half-products 11.15 (0.09) 3.841 0.66 (0.01) content might decrease the vitamin de- Peanut half-products 10.40 (0.08) 4.691 0.60 (0.01) struction due to the significant effect Fish fried products 1.17 (0.12) 37.41 (2.28) 0.16 (0.02) from short residence time. Peanut fried products 1.03 (0.07) 37.35 (0.33) 0.15 (0.01) Numbers in parentheses represent the standard deviations. Chemical and nutritional qualities 1Values calculated from composition of ingredients of extruded products Fat content, moisture content, and water activity of half-products and fried products are presented in Table 4. Both Effect of frying on vitamin E in min A palmitate, decreased significantly fish and peanut half-products con- soybean oil during extrusion. The reduction of RP tained low fat content of 3.84 and The tocopherol content of soybean was about 46% (a reduction of 2.65 to 4.69%, respectively. Water activity of oil (before and after frying) of the 1.44 mg/100 g, as shown in Table 1). half-products (0.60 to 0.66) were higher sample is presented in Table 3. The Drying at 50 ЊC for 5 h and frying at than fried products (0.15 to 0.16). Fish predominant tocopherol homologue 200 ЊC for 1.5 min did not significantly and peanut fried products had high fat detected in soybean oil was ␥-toco- affect the contents of the RP. Extrudate content (37%). Tocopherol and vitamin pherol which was followed by ␦-, ␣- immediately after extrusion, half-prod- A in fish snacks per serving (25 to 28 g) and ␤-tocopherols, respectively. After uct, and fried product contained a sig- were about 15, and 14 to 15% of the soybean oil was heated at 200 ЊC for 30 nificantly lower RP content than the RDA, respectively. Tocopherol and vitamin to obtain equilibrium tempera- raw material mixture. Measured on a min A in peanut snacks per serving (25 ture, the half-products were fried for fat-free basis, the products contained to 28 g) were about 16, and 20 to 22% of 1.5 min. A significant reduction of total 1.38 mg/100 g RP, which was a 48% re- RDA, respectively. tocopherol content in soybean oil oc- duction from the initial amount in the curred during use (Table 3). The re- raw material. Retinol was not detected Conclusions duction of tocopherols ranged from 2 in the samples at any point in the fish T IS POSSIBLE TO PRODUCE NUTRITIOUS to 9%. snack production, as indicated by its Ifish and peanut snacks fortified with absence in chromatograms. Therefore, tocopherols and RP using extrusion. Ex- Effect of extrusion and frying on extrusion did not hydrolyze the ester trusion significantly affected the degra- vitamin A linkage of RP. dation of both tocopherols and RP. The In the fish snack, RP contents in the In the peanut snack, extrusion and reduction of total tocopherol and RP in samples were calculated based on a fat- drying did significantly affect the RP fish extrudates was higher than in pea- free dry basis. RP in the raw fish mix- content (fat-free dry basis) in the sam- nut extrudates, probably because of ture, which was fortified with dry vita- ples (Table 2). The reduction of RP in higher concentrations of polyunsaturat-

Vol. 66, No. 6, 2001—JOURNAL OF FOOD SCIENCE 901 Sensory and Nutritive Qualities of Food ed. Westport, ed. rd inquiries to author inquiries to ed. New York: Marcel York: ed. New nd Conn.: AVI. p 165-179. AVI. Conn.: Handbook of Vitamins. 2 Vitamins. of Handbook Dekker, Inc. p 1-57. Inc. Dekker, in Vitamins of Technology The editor. PB, Ottaway & Professional Blackie Academic York: New Foods. Hall. p 90-113. Chapman & Phillips RD, of plant proteins. In: tional quality of and the Effect Quality Protein editors. JW, Finley p 219- Inc. Dekker, Marcel York: New Processing. 246. several natural antioxi- inhibition varies among system. J Food Sci dants in an aqueous model 58(3):653-655, 687. Plenum York: 2. New Vol book of Lipid Research. Press. p 133-210. Chem Soc 55: 809-814. and oil processing. J Am Oil consumers of extrud- tance by American and Asian products. J Food Sci ed fish and peanut snack 63(4):721-725. snacks by twin-screw ex- fish and tapioca-peanut trusion and deep-fat frying. J Food Sci 64(2):303- 308. tion of vitamin E homologues in infant formulas by HPLC using fluorometric determination. J Mi- cronutr Anal 6(1):35-45. Pro- Peanuts: JG, editor. Woodroof In: of peanuts. 3 Products. duction, Processing, MS 20000601 This research was partially supported by a grant from the is Georgia Commodity Commission for Peanuts. Gratitude also expressed to the Anantamahidol Foundation under the Thailand, for The King of Majesty, of His Patronage Royal support of Dr. Jangchud (nee Suknark) during her doctoral studies. Dept. University’s is with Kasetsart Suknark Author of Product Development and is on the Agro-Indus- try faculty in Bangkok, Thailand. Author Lee is with at Science and Technology of Food the Dept. Au- of Korea. Republic Univ., Chingbuk National of thors Eitenmiller and Phillips are with the Dept. of Georgia, at Univ. Science and Technology Food Athens, GA 30602, and Griffin, GA 30223-1797, Aaddress U.S.A., respectively. Phillips ([email protected]). Olson JA. 1990. Vitamin A. In: Machlin LJ, editor. Machlin A. In: Vitamin JA. 1990. Olson Ottaway PB. 1993. Stability of vitamins in food. In: 1993. Stability of vitamins Ottaway PB. of extrusion 1989. Effect on the nutri- RD. Phillips Cholesterol autoxidation Rankin SA, Pike OA. 1993. Hand- editor. HF, DeLuca Vitamin E. In: Scott ML. 1978. and antioxidants in fat Sherwin ER. 1978. Oxidation KH. 1998. Accep- McWatters RD, K, Phillips Suknark Tapioca- 1999. Y-W. Huang K, Phillips RD, Suknark QK. 1989. Determina- Wei Chou CC, TF, S, Lee Tuan value and nutritive JG. 1983. Composition Woodroof Presented as: FAO Presented . ed. New York: Marcel York: ed. New nd ). J Sci Food Agric 59(4):529-536. Food Agric ). J Sci ed. New York: Marcel Dekker, Inc. p Inc. Dekker, Marcel York: ed. New nd -tocopheryl acetate in infant formulas. J ␣ talurus punctatus talurus Diet Assoc acids in foods. J Amer tion of fatty 69(4):243-248. paper] [technical ity Control 139-244. editor. Ottaway PB, foods (specific application). In: York: New Vitamins in Foods. of Technology The Blackie Academic & Professional Chapman & Hall. p 114-142. AOAC 65:810-816. 1998. Application of direct solvent extraction to the LC quantification of vitamin E in peanuts, peanut Sci 25(2):123-128. Peanut and selected nuts. butter, RR. 1999. Rapid liquid chromatographic assay of vitamin E and retinyl palmitate in extruded weaning foods. J Food Sci 64(6):968-972. ity of vitamin A and provitamin A (carotenoids) in extrusion-cooking processing. AICHE Symp Ser 74:192-195. Storage, heating, and tocopherols affect cholesterol oxide formation in food oils. J Agric Food Chem 44(12):3830-3840. 2 Vitamins. of Handbook Fisheries Food and Nutrition Paper No. 60. 1995 and Nutrition Paper Fisheries Food of an FAO report Italy; 20-23; Rome, November Ita- Rome, from FAO, Available technical meeting. ly. 50(11):73-80. Technol Food id oxidation. compounds. In: fish proteins and other nitrogen Compo- Products: Fish and Fishery A, editor. Ruiter Guild- and Stability. Properties sition, Nutritive 77-115. ford, U.K.: Biddles Ltd. p processes on pro- The effects of various thermal in vitamins, and selenium content tein quality, Sci 6: J Cere wheat and white flour. whole-grain 269-282. ing. Food Chem 49(2):149-155. reverse phase chromatography and non-aqueous liquid chro- chromatography to high-performance matographic determination of retinyl palmitate and Dekker, Inc. p 99-144. Inc. Dekker, 2 Chemistry. Exler J, Weihrauch JL. 1976. Comprehensive evalua- Comprehensive JL. 1976. Weihrauch J, Exler and Qual- Technology Fortification: 1996. Food FAO. O’Brien A, Roberton D. 1993. Vitamin of fortification 1993. D. A, Roberton O’Brien Lee J, Landen WO Jr., Phillips RD, Eitenmiller RR. Eitenmiller RD, Phillips Jr., WO Lee J, Landen Eitenmiller RD, Phillips Y, K, Kluvitse Lee J, Suknark 1978. Stabil- Alid G, Chichester CO. T, Chen T-C, Lee JJ. 1996. Sim RT, G, Ahn DU, Hardin Li SX, Cherian LJ, editor. Machlin Vitamin E. In: LJ. 1990. Machlin Giese J. 1996. Antioxidants: Tools for preventing lip- for preventing Tools J. 1996. Antioxidants: Giese 1995. Composition and nutritive value of NF. Haard Oste R, Akesson B. 1987. Hakansson B, Jagerstad M, in extrusion cook- retention Vitamin Killeit U. 1994. gel permeation of 1982. Application WO. Landen Nawar WW. 1985. Lipids. In: Fennema OR, editor. Food OR, editor. Fennema In: 1985. Lipids. WW. Nawar Ic- —Vol. 66, No. 6, 2001 66, No. —Vol. Vol. Boca Vol. nd ed. Arlington, Va: Association of Association Va: ed. Arlington, References th JOURNAL OF FOOD SCIENCE composition in three strains of channel catfish ( Vitamin Analysis for the Health and Food Scienc- es. Boca Raton, FL: CRC Press LLC. p 109-148. IJ, Ikins WG, editors. Analyzing Food for Nutrition Analyzing Food editors. WG, IJ, Ikins York: New Contaminants. Labeling and Hazardous p 195-281. Inc. Dekker, Marcel taway PB, editor. The Technology of Vitamins in Vitamins of Technology The editor. taway PB, & Professional Blackie Academic York: New Foods. Chapman & Hall. p 143-171. cooking. Food Chem 20:263-283. and nutritional change in food during extrusion. Crit Rev Food Sci Nutr 29(1):35-57. of the principles of physical organic chemistry to the exploration of its structure and function. AACC Chem Res 19:194-201. ing on nutritional value: A literature review. J Food review. A literature ing on nutritional value: Eng 2(4):281-308. Analysis: A Comprehensive Review. New York: El- York: New Review. Analysis: A Comprehensive 326 p. sevier. vitamin A degradation in wheat flour (ATTA). J Food in wheat flour (ATTA). vitamin A degradation Proc Pres 14(2):123-134. Official Analytical Chemists. of Analysis. 16 of Analysis. Raton, Fla: CRC Press, Inc. p 193-214. Raton, Fla: CRC Press, Inc. Salunkhe DK, Kadam SS, editors. RC Handbook of Salunkhe DK, Kadam SS, Pro- Chemistry, Legumes: Nutritional Food World and Utilization. 2 cessing Technology, of fish and shellfish lipids. In: Ruiter A, editor. Fish A, editor. Ruiter In: of fish and shellfish lipids. Composition, Nutritive and Fishery Products: U.K.: Biddles Guildford, and Stability. Properties Ltd. p 117-156. of the American Association of Cereal Chemists. of the American Association St. Paul, Minn.: AACC. Erickson MC. 1992. Variation of lipid and tocopherol Variation MC. 1992. Erickson Eitenmiller RR, Landen WO Jr. 1999. Vitamin E. In: 1999. Jr. WO RR, Landen Eitenmiller Eitenmiller RR, Landen WO. 1995. Vitamins. In: Jeon In: Vitamins. 1995. WO. RR, Landen Eitenmiller Counsell JN 1993. Vitamins as food additives. In: Ot- Vitamins as food additives. JN 1993. Counsell Cheftel JC. 1986. Nutritional effects of extrusion- Camire ME, Camire A, Krumhar K. 1990. Chemical Burton GW, Ingold IU. 1986. Vitamin E: Application IU. 1986. Ingold GW, Burton Bjorck I, Asp NG. 1983. The effects of extrusion-cook- Ball GFM. 1988. Fat-Soluble Vitamin in Food Assays GFM. 1988. Fat-Soluble Ball Arya SS, Thakur BR. 1990. Effect of water activity on Arya SS, Thakur BR. 1990. AOAC. 1995. Procedure 925.09. In: Official Methods AOAC. 1995. Procedure 925.09. 902 Adsule RN, Kadam SS, Salunkle DK. 1989. Peanut. In: Adsule RN, Kadam SS, Salunkle Ackman RG. 1995. Composition and nutritive value Ackman RG. 1995. Composition AACC. 1976. Procedure 30.25. In: Approved Methods AACC. 1976. Procedure 30.25. Tocopherol and Retinyl Palmitate in Snacks . . . Snacks in Palmitate Retinyl and Tocopherol direct solvent in fish. A acids ed fatty and hex- using isopropanol extraction, sol- acetate as an extracting ane/ethyl toco- extracted both vent, efficiently simultaneously. and RP pherols

Sensory and Nutritive Qualities of Food