WIWIIIUHHHIIIHlllllHlIHIHIHIHHIHIIHIIWI __THS THESE Date ' ’2 A‘o-I-pr This is to certify that the thesis entitled INFLUENCE OF ANTIOXIDANTS AND PACKAGE ENVIRONMENT ON STABILITY OF TURKEY PRODUCTS presented by NANCY ANN KING has been accepted towards fulfillment of the requirements for Masters degree inFood Science}. Humafi'NUfFTt1on X; 8W Major professor February 16, 1982 0-7639 MSU RETURNING MATERIALS: Piace in book drop to “3anng remove this checkout from 1—3:... your record. FINES wiH be charged if book is returned after the date stamped below. INFLUENCE OF ANTIOXIDANTS AND PACKAGE ENVIRONMENT ON STABILITY OF TURKEY ROASTS by Nancy King A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Food Science and Human Nutrition 1981 ABSTRACT INFLUENCE OF ANTIOXIDANTS AND PACKAGE ENVIRONMENT 0N STABILITY OF TURKEY ROASTS by Nancy King Commercially prepared turkey breast meat roasts were packaged using SaranexTM , polyethylene (PE) and a butylated hydroxyanisole (BHA) impregnated coextruded polyethylene film (PE + BHA). Package environments included vacuum, nitrogen, and air environments. All products were stored at -lB°C for six months. Lipid oxidation (warmed over flavor) was monitored by thiobarbituric acid (TBA) values, and sensory panel scores. TBA values of the turkey roasts were initially high (approximately 3) and there was considerable variation in this initial measurement. At the end of the storage period, mean TBA values of the product were lowest for turkey meat in PE + BHA film, followed by SaranexTM and PE film in that order. Packaging environments made no significant difference in lipid oxidation of turkey roasts as indicated by TBA values. Storage time had a greater influence on deterioration of flavor and color in turkey roasts than did packaging films or environments. BHA from the PE-BHA packaging film migrated into the turkey fat and may have controlled the extent of lipid oxidation. To my parents with my love, respect and admiration ii ACKNOWLEDGEMENTS My sincere gratitude is extended to Dr. L. E. Dawson for his professional guidance, inspiration, and patience as well as for his personal interest and concern. A special thanks to Violet Dawson for her friendship. The members of my guidance committee: Dr. I. J. Gray, Dr. B. R. Harte and Dr. M. R. Bennink also deserve my appreciation for their generous assistance and encouragement in completing this project. Special mention is made to Dr. H. J. Stadelman for his enthusiastic introduction to the field of Food Science and for his continued confidence in my professional abilities. Further appreciation to the Institute of Food TechnolOgists and the Department of Food Science and Human Nutrition of Michigan State University for their financial assistance throughout the course of this study. I am especially grateful to my wonderful fiance, Chuck, for his unending support, understanding and love. Finally, it is difficult to express apprOpriately my heartfelt appreciation to my outstanding parents who have provided me with abundant love, confidence and friendship in all my endeavors. TABLE OF CONTENTS Page LIST OF TABLES ....................... v LIST OF FIGURES ...................... vi INTRODUCTION ........................ l REVIEW OF LITERATURE .................... 3 Poultry Meat . . . . . . ............ . . 3 Development of Warmed Over Flavor ........ g. . 4 Lipid Oxidation ................... 5 Products of Lipid Oxidation ............. 7 Measurements of Rancidity . ............. 8 Effects of Cooking .................. lO Effects of Frozen Storage .............. ll Effects of Packaging ................. l3 Antioxidants . . . . . . . . . . . . ........ . l7 MATERIALS AND METHODS ................... 20 Source of Meat . . . ............ . . . . . 20 Sample Preparation .................. 20 Proximate Composition ................ 2l Moisture .................... 21 Fat ........ . . . . . . . . . . . . . . . 23 Protein ..................... 23 Ash ....................... 24 Chemical-Physical Analyses .............. 24 Lipid Oxidation . . . . ............ . 24 Extraction of Lipids .............. 25 BHA Analysis .................. 26 Surface Color of Thawed Turkey Roast ...... 26 Sensory Evaluation . . . . . . . . . . . . . ..... 27 Statistical Analyses ................. 27 RESULTS AND DISCUSSION ................... 28 Initial Product ................... 28 Lipid Oxidation, TBA Test .............. 3l BHA Analyses ..................... 39 Color Evaluation . . . . . . . . . . . . . . . . . . . 44 Sensory Evaluation .................. 52 SUMMARY AND CONCLUSIONS . . ................ 57 RECOMMENDATIONS FOR FURTHER RESEARCH ............ 61 APPENDIX . . . ....................... 62 LIST OF REFERENCES ..................... 63 iv TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE TABLE l0. ll. LIST OF TABLES Proximate Composition of Turkey Roasts. Initial TBA Values for Two Precooked Turkey Roasts. TBA Values for Turkey Roasts in Different Package Films During Six Months of Frozen Storage. TBA Values for Turkey Roasts in Different Package Films and Environments During Six Months of Frozen Storage. Analysis of Variance of TBA Values for Turkey Roasts. BHA Content in Turkey Meat During Frozen Storage. Color Measurements (L-Value) of Turkey Roasts During Frozen Storage. Color Measurements (a-value) of Turkey Roasts During Frozen Storage. Color Measurements (b-value) of Turkey Roasts During Frozen Storage. Flavor Scores of Turkey Roasts (Mean of All Package Films and Environments) During Frozen Storage. Analysis of Variance of Sensory Scores for Turkey Roasts. Flavor Scores for Turkey Roasts Package Treatments (Mean of Environments for Each Film) During Frozen Storage. FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE FIGURE LISTS OF FIGURES Sample Preparation and Experimental Design for Turkey Roasts. TBA Values in Turkey Roasts (mean of all environments for each film) during Six Months Frozen Storage. TBA Values in Turkey Roasts in Saranex Film during Frozen Storage. TBA Values in Turkey Roasts in PE Film during Frozen Storage. TBA Value in Turkey Roasts in PE-BHA Film during Frozen Storage. BHA Transfer from PE-BHA Film to Turkey Roasts during Frozen Storage. BHA Transfer from PE-BHA Film to Turkey Roasts in Package Environments to Turkey Roasts during Frozen Storage. Meat Color (L-Value) Measurements of Turkey Roasts in different Package Films during Frozen Storage. Meat Color (L-Value) Measurements of Turkey Roasts in different Package Environments during Frozen Storage. vi INTRODUCTION Poultry products have received increasing popularity and demand in recent years. Turkey products, in particular, have contributed substantially to improving the market potential of poultry products, as well as to the growth of the industry. Trends for further processed turkey meat show that in l976, 608 million pounds of turkey meat were further processed and this increased to 909 million pounds in l980 (Brown, l98l). It has been estimated that 46 to 48% of the turkey crop in l98l was cut up or further processed (Brown, 1981). Dawson (1975) reported an increase in the use of turkey in forms other than whole birds from 250 million pounds in T965 to 900 million pounds in l974. This in an average increase of 70 million pounds yearly. During product distribution and subsequent entrance into the supermarket and home, meat products have many opportunities to become intentionally or unintentionally frozen. The handling and freezing of the product during this distribution process can be very critical to the lipid stability and overall acceptance of the product. This project was initiated based on a need for commercial processors to know more about stability of precooked turkey products during frozen storage. This study was designed to evaluate several procedures or methods for minimizing oxidation of lipids in a typical precooked turkey product. A commercial turkey breast meat product was selected to eliminate one of the variables--meat source. This product is also one of the turkey products on the market which has caused some problems in marketing due to develOpment of rancidity or off flavors during storage. Therefore, the objectives of this project were (I) to determine the effect of packaging 2 materials on oxidation in a turkey product during frozen storage; (2) to evaluate the effects of packaging environments on oxidation in a turkey product during frozen storage; and (3) to evaluate migration of butylated hydroxyanisole (BHA) from the packaging material into the turkey products. REVIEW OF LITERATURE Poultry Meat Poultry meat is nutritionally excellent, economically priced and readily available for convenient use by today's consumer. The proteins of the meat are similiar to those of other livestock (Millares and Fellers, 1948; Scott, l959). Scott (T956, 1958) determined the nutrient composition of the edible proteins of the raw and roasted turkeys of various types and ages and compared them to chicken, duck, beef, lamb, and pork. The protein content is relatively high for both chicken and turkey meat at approximately 20-24% and the fat content ranges from l-2% in the breast meat to 4-5% in the leg meat (Watt and Merrill, l963; Palmer and Bowers, l972). It has been generally accepted that poultry fat contains a high percentage of the unsaturated fatty acids and a low level of cholesterol (Pearson gt al. l977). Of special concern to food processors is the retention of freshly cooked flavor and prevention of stale and rancid flavor development in precooked frozen turkey products. The stability of flavor in turkey products is related to the inherent composition of meats, to cooking methods, to package atmosphere and to the storage conditions. According to Palmer and Bowers (1972), the qualities desired in cooked poultry meat are tenderness, juiciness, presence of typical poultry flavor, and the absence of off-flavor, microbiological spoilage and other chemical hazards. This rancid or off-flavor problem has been associated with the lipid fraction of the poultry meat and in particular the tissue or intramuscular lipids (Younathan and Watts, l960; Love and Pearson, 1971). Development of Warmed Over Flavor Tims and watts (1958) first suggested the role of tissue lipids in rancidity after noting a rapid flavor deterioration in cooked meats during refrigerated storage. They proposed that this change in flavor was caused by the oxidation of highly unsaturated protein bound phOSpholipids. Because of a high content of polyunsaturated fatty acids, the phOSpholipids are very labile to oxidation (Love and Pearson, 1971; Lea, 1957). This factor gives phospholipids their importance in the determination of meat quality in spite of their relatively small content in meats. El-Gharbawi and Dugan (1965) reported that phospholipids are oxidized first followed by the autoxidation of the neutral lipids. In examining the effects of triglycerides and phOSpholipids in the development of warmed over flavor (WOF) or off-flavors, Igene and Pearson (1979) found that total phOSpholipids contributed to the development of warmed over flavors in both beef and poultry. Poultry meat and fish (Watts, 1954, 1962; Acosta gt 31. 1966; Younathan and Watts, 1960) are higher is phOSpholipids than the red meats such as beef (Pearson et 31. 1977). Researchers have shown that poultry dark meat contains more phOSpholipids than white meat (Peng and Dugan,1965; Acosta gt 31. 1966). Contrasting results were obtained by Katz et_al. (1966) indicating that chicken dark meat has only about half as much phOSpholipid as white meat. Cooked meat has an apparent higher phOSpholipid content than raw meat whether it is expressed as a percentage of fat or as a percentage of total tissue (Campbell and Turkki, 1967; Fooladi, 1977). 5 Poultry meat is more unsaturated than beef, lamb or pork (Scott, 1958; Acosta gt a1. 1966). Pearson et_al. (1977) stated that chicken fat is particularly high in oleic and linoleic fatty acids and because of this unsaturated nature, poultry meat is generally thought to be more susceptible to becoming rancid more rapidly than red meats. Turkey meat is most susceptible to the development of rancid flavors or oxidative rancidity closely followed by chicken, pork, beef and mutton (Wilson gt 31, 1976). In lipid oxidation, the decomposition of the hydroperoxides influence overall flavor as well as warmed over flavor. One of the principal products associated with lipid oxidation is hexanal, which has been implicated as a component of WOF (Love and Pearson, 1976). Ruenger .EE.21- (1978) in studying WOF in turkey meat found that heptanal and n-nona-3-6-dienal were correlated with WOF. By identifying these products of lipid oxidation, evidence supports that WOF is caused by lipid oxidation (Sato and Herring, 1973). Lipid Oxidation 0f the chemical constituents of foods, the lipids are the most susceptible to atmospheric oxidation or autoxidation. This oxidative deterioration of food lipids has been shown to be responsible for oxidative rancidity in foods and is characterized by objectionable odors and flavors, deterioration of nutritional value and changing physical pr0perties (Dugan, 1961). Deleterious autoxidative reactions are accompanied by various secondary reactions having oxidative and nonoxidative character (Gray, 1978). The generally accepted mechanism of 6 lipid oxidation proceeds through a free radical chain reaction mechanism which occurs in three stages: initiation, propagation and termination. Initation takes place when a labile hydrogen is abstracted from a site on the lipid (RH), with the formation of free radicals (R'). Initiators in this reaction include light, heat, heavy metals and oxygen. R1H initiators » R1 + H° (free radical) Propagation involves the combination of fatty free radicals with molecular oxygen to form peroxide free radicals. These compounds are then able to abstract another methylene hydrogen to form further free radicals and hydroperoxides which are capable of perpetuation of the chain reaction. Ri + 02 “*1D1 R100° (free radical) (peroxide free radical) R100° + RZH i:1P ROOH + Ré (hydroperoxide) Termination of the free radical chain mechanism results when free radicals are deactivated and stable with the creation of non-radical end products. Free radical inhibitors (RI) include antioxidants that may react with the chain continuing free radicals to form inert end products 7 as a termination step in the chain reaction mechanism. R' + R°--dDRR R' + ROO'——)ROOR R00“ + ROO'——)ROOR + 02 R' + RI—-)RRI Additional meat components have been implicated in catalytic roles in lipid oxidation. Some metals occurring in meat in trace quantities, especially ferrous iron, are efficient lipid oxidation catalysts. Research now indicates that ferrous iron is the active catalyst of lipid oxidation in meats (Waters, 1971; Love and Pearson, 1974). The role of the common meat additive, sodium chloride, in initiating color and flavor changes in cured meat is poorly understood. Some data indicate that trace metal impurities present in salt account for its effects on lipid oxidation (Lea, 1939). However, there is additional evidence for a direct role of sodium chloride in initiation of fat oxidation (Ellis et 31, 1968). According to Watts (1962), salt is believed to catalyze the oxidation of stored triglycerides. Products of Lipid Oxidation Hydroperoxides are widely considered to be the primary products of the reaction of oxygen with unsaturated lipids. They are colorless and odorless compounds that do not account for the off-flavor associated with lipid autoxidation (Sato and Herring 1973; Watts, 1954). Degradation of hydroperoxides through a series of scission and dismutation reactions 8 yield the secondary products of lipid oxidation. These secondary degradation products include aldehydes, ketones, acids, lactones, alcohols and unsaturated hydrocarbons and they are thought to be primarily responsible for the off odors and off flavors associated with oxidative rancidity (Lea, 1962; Sato and Herring, 1973). The following scheme as shown by Gray (1978) illustrates some of these routes of decomposition of fat hydroperoxides. ROUTES 0F DECOMPOSITION DIMERS. HIfiFER POLYMERS POLYMERIZATION FAT HYD A PEROXIDE FURTH DEHYDRATION OXIDATION CH=CH IN OTHER MOLECULES DIPEROXIDES ALDEHYDES KETO-GLYCERIDES EPOXIDES SEMI-ALDEHYDES ' OH-GLYCERIDES ALDEHYDO-GLYCERIDES DI OH-GLYCERIDES OH-COMPOUNDS ACIDS 4' POLYMERS Measurements of Rancidity Methods of rancidity measurement have been reviewed by Sherwin (1968). Objective methods available for the evaluation of both storage and stability were reviewed by Erickson and Bowers 9 (1976). According to these researchers, the methods available for determination of lipid stability and oxidation can be classified as methods of oxygen uptake, peroxide formation, and peroxide decomposition. The TBA test is a measurement of final reaction products and is based on the development and quantitation of a red pigment formed by the condensation of one molecule of malonaldehyde and two molecules of thiobarbituric acid. Malonaldehyde, a three carbon compound resulting from a breakdown of unsaturated fatty acids in food products, reacts with TBA to give a red pigment with a maximum absorption at 530 to 538 nm (Sinnhuber gt_al, 1958). Oxidation of muscle lipids has been related to flavor deterioration in cooked meats (Tims and Watts, 1958; Turner et al. 1954) and the thiobarbituric acid test (TBA) has been used extensively to determine the amount of oxidation (Younathan and Watts 1960). The TBA test is useful and convenient because it can be done on the entire food sample and it can be correlated with sensory tests (Younathan and Watts, 1959; Kesinel gt a}, 1964). For most tissues, off odors and flavors can be detected when the TBA number is 0.5-1.0 (Tarladgis st 21. 1960; Watts, 1962). Inconsistant correlations between sensory scores and TBA values doexist. Therefore, this range is not firmly established (Cash and Carlin, 1968). The consumer of a food product ultimately uses several senses as the primary means of judging the quality of a product. Hence, it is logical that chemical measurement of rancidity in fats and oils should correlate with sensory measurements of rancidity (Sherwin, 1968). 10 According to Dugan (1976) all objective methods available for determining lipid stability have their own limitations; therefore, sensory methods are necessary for confirmation. Sensory tests can be as simple as having an individual taste or smell a food sample to determine any obvious off-flavors or odors. These tests can be as complex as a multi-membered panel of individuals who have been previously trained for odor and taste evaluations. This is then followed by conduction of a statistically designed sensory evaluation on a series of samples under conditions designed to minimize the bias and human error inherent in such a test. Effects of Cooking Both the chemical and physical changes that take place in poultry meat during cooking have been extensiVely reviewed (Bratzler, 1971; Palmer and Bowers, 1972; Paul, 1972; Meyer, 1975). MacNeil and Dimick (1970) studied the cooking losses of turkey roasts andifound that when they were cooked to an internal temperature of 170°F, cooking losses were higher in breast meat than in thigh meat. The effects of endpoint cooking temperature and rate of cooking on sheer values of turkey breast were determined by Goodwin gt 31. (1962). They stated that breast meat cooked to 88°C and 94°C appeared drier and tended to crumble more than turkey breast cooked to a lower endpoint temperature. H0ke.§£.21~ (1967) studied the effect of selected internal and oven temperatures for roasting or braising on the eating quality of the cooked meat. They reported that as internal temperature increased, yields and juiciness of the cooked meat decreased, but the palatability factors increased. It is generally accepted that develOpment of oxidized flavor or ll randicity in meats is accelerated by heating (Younathan and Watts, 1959; Chang et_al, 1961; Sato and Herring, 1971; Keller and Kinsella, 1973). Labuza (1971) suggested that the rapid rate of oxidation in cooked meat may be due to the denaturation of the myoglobin during the cooking process. He concluded that the unfolding of the protein allows greater exposure access of the iron to the previously formed peroxide. Younathan and Watts (1959) and Sato and Herring (1971) suggested that the rapid oxidation of lipids in cooked meat is initiated by the conversion of the iron in the porphyrin ring of the myoglobin pigments to the ferric form. Thus, during heating the pigment ferric hemochromogen is an active catalyst for unsaturated fats. The severity of heat treatment seems to be related to the extent of lipid oxidation as reviewed by Yamauchi (1972a). Meat subjected to long periods of heating and/or high temperatures had lower TBA values than did meat cooked at lower temperatures for a shorter period of time (Huang and Green, 1978; Dawson and Schierholz, 1976). Cooked meat has higher TBA values than fresh meat (Keller and Kinsella, 1973). Consumer acceptance of poultry products is strongly influenced by the color of the product (Mugler gt al. 1970). Two heme pigments, myoglobin and hemoglobin, make uncooked poultry flesh pink to red. According to Mugler gt_al, (1970) consumers often object to variation from the normal appearance of uncooked meat or from the normal white to golden brown of cooked turkey. Pool (1956) reported that turkey meat acquired a pink color when oven roasted in an uncovered container. Effects of Frozen Storage Frozen poultry meat and the importance of low temperature storage 12 in retarding rancidity has been the topic of much research (Cook and White, 1939, 1940; Ramsbottom, 1947; Koonz, 1947; Klose 25.21- 1957). It is generally accepted that lipid oxidation proceeds more rapidly at high temperatures, therefore, the use of low temperature storage is recommended. In frozen meats, the degree of unsaturation as well as the composition of the lipid determines the storage stability of the product (Watts, 1954; Greene, 1969; and Igene, 1976). The lipids in poultry meat, having high relative degree of unsaturation are influenced by storage conditions which makes frozen storage of products of primary importance. Stadelman (1974) reviewed the storage stability of turkey meat. However, he did not estimate the shelf life of turkey rolls because they are closely linked to product formulation. According to Essary and Rogers (1968), fresh turkey rolls had higher organoleptic values than frozen turkey rolls held at -29°C up to eight months and at fluctuating temperatures. Small flavor losses during low temperature storage of chicken meat have been reported (Jacobson and Koehler, 1970). Steinberg et 31. (1949) indicated that there was a significant decrease in palatability of beef with an increase in the oxygen content of the atmOSphere surrounding the meat during frozen storage. According to Hiner et al. (1951), deterioration in the palatability of beef, pork and lamb occurred due to oxidation of fat. They stated that during freezer storage, vacuum packaging produced the most desirable product with the least decline in quality. Hanson et 31. (1950) reported that the type of package is of greater importance than freezer storage 13 temperature for retention of flavor of precooked frozen creamed turkey and chicken. Rancidity in turkey meat, as indicated by the TBA test, was lower in freshly frozen turkey than in precooked frozen turkey roasts before storage at 00F (Cash and Carlin, 1968). Johnson and Bowers (1974b) reported that freshly cooked turkey breast meat had lower TBA scores than precooked and freshly cooked meat stored at -13°C for five weeks. They also reported that phOSphate treated precooked and freshly cooked meat had lower TBA values than the control without phosphates. In precooked frozen turkey roast, Cash and Carlin (1968) found that TBA values varied among duplicate samples from the same slice of meat. They also reported that taste panelists noted that the rancid flavor was present around the edges or perhaps in one specific area of the piece of meat. These results suggest that in initial stages of rancidity, oxidative deterioration may develOp unevenly throughout the meat as indicated by off-flavor. Effects of Packaging Between the time a food product is processed and packaged and eaten by the consumer, the food travels through a complex distribution and storage system. During this time it is essential that the food remain free from deleterious chemical reactions and physical deteriorations and remains safe and palatable. Food packaging serves four primary functions according to Ball (1967), (l) to protect the food product against contamination by microorganisms and filth; (2) to retard or to prohibit the gain or loss of moisture; (3) to facilitate handling; and (4) to shield the product from light and oxygen. Flexible packaging 14 is very prevalent among poultry products. There are numerous advantages in the use of flexible packaging material. One important property is the low permeability to moisture, oxygen, nitrOgen, carbon dioxide, and desirable or undesirable volatiles (Ball, 1967). Polyethylene is a common packaging material used in the food industry because it is low in cost, strong, tough, pliable, therm0plastic and high in resistance. Disadvantages of polyethylene include its relatively high permeability to gas, low resistance to breakdown by heat and limited transparency. Gray and Giacin (1980) suggested that the extent to which a package renders protection to a food product is partly determined according to its ability to act as a barrier between the external environment and the internal package environment in which the food is in contact. They further pointed out that by using a high oxygen barrier packaging material, lipid oxidation in a food system can be inhibited or retarded by restricting the diffusion of oxygen into the internal package environment. Various packaging techniques have been used by the food industry to retard quality deterioration of food products. Packaging materials and methods may be used to reduce the partial pressure of oxygen within the packaged product (Ball, 1967; Labuza, 1971). Vacuum packaging is one such technique in which all available oxygen in removed from the package creating a vacuum around the food product (Ramsbottom, 1971). Modified gas atmosphere packaging is a procedure that utilizes the gas barrier properties of flexible packaging materials to retard oxidative spoilage in meat and poultry products (Pinto, 1979 and Sander, 15 1978). In this operation the product is placed in a plastic container with an inert gas such as nitrogen and then the package is sealed. This inert atmosphere along with a slightly permeable packaging material provides protection of the product by two mechanisms. The inert atmosphere inhibits growth of aerobic bacteria which thrive in direct proportion to the amount of oxygen available. In addition, the permeation of oxygen through the packaging material restricts the growth of anaerobic bacteria (Gray and Giacin, 1980). Rey and Kraft (1971) reported that freezing poultry meat prior to refrigerated storage enhances the occurance of microorganisms if packaging films highly permeable to oxygen are used. They also found that a packaging film impermeable to oxygen reduces the growth and proteolytic and 1ypolytic activities of psychrophilic bacteria on poultry. This effect is increased by vacuum packaging. Seideman gt_gl. (1976) found that the higher the degree of vacuum the more effective it was in producing a higher desirability rating in beef cuts. In another study comparing packaging methods for precooked chicken, the meat in polyester/polyethelene laminate pouches packaged under vacuum conditions showed lower TBA values than that packaged without vacuum (Arafa and Chen, 1976). Seideman gt 31. (1979) in investigating the physical and sensory characteristics of beef packaged in modified atmOSphere, suggested that the use of a gas mixture of 20% C02/80% N was at least equal to if not superior to vacuum packaging. Mechanically deboned chicken meat and turkey meat had significantly lower TBA numbers when packaged under vacuum or N2 conditions rather than 16 02 altered and vacuum induced cling packaging techniques. These packaging techniques are far superior to unwrapped samples at 3, 6 and 9 months of storage (Jantawat and Dawson, 1979). Other research indicates that there is no significant difference in flavor scores and rancidity which could be attributed to the effect of packaging materials and conditions during frozen storage (Jeremiah, 1980). Another technique of packaging involves the treatment of a packaging material with an antioxidant providing protection to a food product from oxidative reactions (Gray and Giacin, 1980). They hypothesized the mechanism of antioxidant activity as being dependant upon the volatility of the antioxidant and the migration or transfer of the BHA from the waxed liner to the cereal product. After incorporation into the product the antioxidant functions in a termination of the free-radical chain oxidation reaction. Gray and Giacin (1980) concluded that using a packaging material as a antioxidant carrier could result in an overall reduction of costs, even though this packaging material which has been impregnated must be kept tightly wrapped and stored under controlled conditions. Various environmental conditions, including temperature and relative humidity, can result in the oxidation of the antioxidant during storage (Daun g; gfl, 1974), as well as migration of the antioxidant out of the packaging material under time and temperature conditions of storage. Because of these limitations long term storage of these packaging materials may not be desirable. In researching the migration of plasticizers in polyvinylchloride 17 films, it was reported that the migration of the plasticizer, di(2-ethylhexyl) adipate increased in the samples of meat with higher fat content. Furthermore, after 48 hours, the migration was almost complete (Daun and Gilbert, 1977). Antioxidants Antioxidants are compounds that slow down or prevent the oxidation of autoxidizable materials such as food lipids. The characteristics of the antioxidants, as well as the requirements of the food system,’ determine the choice of antioxidant (Dugan, 1960). .He also indicated that the qualities of an antioxidant are effectiveness at low concentrations, ability to impart desirable characteristics in the food system, low in cost and can be conveniently handled. The generally accepted mechanism of antioxidants and their commercial use have been reviewed by Stukey (1962; 1968) and Uri (1961). Labuza (1971) classified antioxidants into three types which were previously designated by Scott (1965). Type I are the free radical chain stoppers and include butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), and tocopherol. This group consists mainly of phenolic compounds capable of donating a labile hydrogen. The phenolic antioxidants are widely used in the food industry today. Labuza (1971) reported that phenolic antioxidants offer very good protection in animal fats and most vegetable fats contain high enough amounts of tocopherol to be fairly stable. The use of phenolic antioxidants in stabilizing fats has been reviewed by Pokorny (1971). He indicated that phenolic antioxidants reacted with free radicals and 18 resulted in a non—propagating end product. Thus, the ratio of hydrOperoxide formation and its breakdown products would be reduced. A combination of antioxidants are commonly used in order to obtain effectiveness and the desirable properties of the various individual antioxidants (Dugan, 1960). Many of the phenolic antioxidants are decomposed under processing conditions such as baking or frying; however, BHA and BHT were found to survive and remain effective throughout these treatments (Nickerson, 1967). According to Kraybill gt_gl. (1949), BHA has been reported to be very valuable in the stabilization of food which has been heat processed. Marion and Forsythe (1964) reported a significant delay of autoxidation of turkey lipids with the addition of BHA at .04%. Igene gt_gl, (1976) found evidence that potent natural antioxidants, the tocopherols, are present in meats. Natural toc0pherol levels in turkey meat have been shown to be lower than in any other poultry meat. Mecchi gt 31. (1953) found that the lower toc0pherol content of turkey meat compared to chicken meat may be the fat component reSponsible for superior storage stability of chicken meat. Applications of antioxidants either alone or in combination can be achieved by numerous methods and have been detailed and summarized by Lindsay gt_gl. (1975). Some of these methods include applying the antioxidant directly to the food, dipping the food into the antioxidant, applying antioxidant as a spray or adding the antioxidant to the packaging material. According to Nickerson (1967) the lowered effectiveness of the antioxidant in flesh type food is caused by the high moisture content of the food in which the antioxidant is not soluble. In 19 addition, he stated that the fat on the surface of the flesh product will absorb the antioxidant with difficulty. Lund gt 31. (1976) using various methods of application, concluded that obtaining a uniform distribution of the applied antioxidant was not easily achieved. Stukey (1968) reported some success in achieving product protection by adding a highconcentration of antioxidant to the food packaging material such as the inner waxed liner of a food package. The antioxidant can then migrate to the surface of the packaged product and provide protection to the lipid portion of the food. MATERIALS AND METHODS Source of Meat Turkey meat used in this study was obtained from Bil Mar Foods Inc., Zeeland, MI. Source or sex of the live birds was not identified. Three pound Split Breast turkey roasts were commercially manufactured with 1.5% salt and .5% phosphates at the plant. These roasts were precooked and vacuum packaged in polyethylene bags. Sample Preparation Upon receiving the product in the research laboratory, they were cut into one and two pound roasts and then repackaged. One pound roasts were used for chemical analyses and two pound roasts for sensory analyses. Three packaging materials were used in the repackaging: polyethylene (control), polyethylene impregnated with the antioxidant BHA and SaranexTM (Dow Chemical, Midland, MI) package material The polyethylene used was a 2 mil thick, low density film and served as the control. SaranexTM is a coextruded multilayered thermoplastic film which is basically a layer of Saran resin between outside layers of polyethylene. This film is recommended as being durable for freezer storage. BHA impregnated film (2 mil) is coextruded polyethylene film with the antioxidant BHA impregnated into the film. The inside layer is low density for heat sealing purposes. Pouches were made from these packaging materials by cutting the packaging film to size, folding the sides together and sealing with a temperature pressure controlled heat sealing instrument. When turkey meat was placed in the package the entrance of the pouch was sealed using the Kenfield Vacuum 20 21 Sealer (Model C-14 International Kenfield Distributing Co.). Turkey roasts in one third of the pouches were vacuum packaged with the Kenfield vacuum sealer. Another third of the pouches were given a vacuum treatment, followed by backflushing with nitrogen and sealing with the same Kenfield instrument. The remaining third of the samples were sealed in an air environment. The vacuum operated at 25 in. Hg and the nitrogen was flushed into the pouch at 20 pounds per square inch. After sealing, all of the pouches were stored in a walk in type freezer at -18°C. The design of this experiment is shown in Figure l. Replicates of each sample were produced in order that each analysis could be performed on 2 individually packaged samples. Samples were taken during frozen storage at 2, 4 and 6 months. All analyses were run in duplicate after samples had been thawed for 24 hours at 4°C. Proximate Composition Moisture. The A.0.A.C. (1975, 25.003b) procedure was used to determine the moisture of the initial Turkey product prior to storage. Triplicate 5 9 samples were weighed into tared aluminum pans and dried to a constant weight at 100°C (18 hours) in a forced air oven. Moisture was expressed as percent weight lost during drying. The following equation was used: % Moisture = weight of moisture lost (9) x 100 weight of initial sample (9) 22 .pmmom xmxesh com cmmmmo ngcmewgmaxm new cowumgcqwca mpqemm ._ «gamed LE .0 comoEz .m E:=om> .< .5: Mn. .m 8.: a “one. a. .. . m_m>_mc< Eomcow w_m>_mc< .mo_Eo;0 /\ #mmocm >35... :75. 23 .533. Solids obtained from moisture determinations were used for Goldfisch extraction (A.0.A.C., 1975, 24.005b). The samples were continuously extracted for three and one-half hours using anhydrous ethyl ether. Ether was evaporated and the lipid extract dried at 100°C for 30 minutes. Total fat was calculated on a fresh weight basis using the weight of the cooled extracted material and the following equation: % fat = weight of dried extract (9) x 100 weight of initial sample (9) Protein. Protein was determined using a modified A.0.A.C. (1975, 23.009) semi-micro Kjeldahl procedure. Triplicate 0.5 g turkey samples were digested by l g sodium sulfate, 7 ml concentrated sulfuric acid and 1 ml of 10% (w/v), copper sulfate solution, with heat to a clear pale green endpoint. The digested sample was neutralized with a 35 ml of 50% sodium hydroxide and distilled into 20 ml 2% boric acid. The distillate was back titrated to a colorless endpoint using a standardized 0.1N sulfuric acid with brom cresol green as an indicator. Percent protein was calculated on a fresh weight basis using the following equation: % Protein = (N of H2S04) (net m1 H2504) (.014) (6.25) x 100 fresh sample weight in grams 24 53h. Total ash was determined using a variation of the A.0.A.C. (1975, 29.012) method. Triplicate 5 9 samples of fresh turkey were weighed into previously ashed and tared Coors 50 ml (size 2) porcelain crucibles and then dried at 1000C for 18 hours. These dried samples were pre-ashed over a Fisher burner. Crucibles were then placed in a muffle furnace and heated at 525°C until a uniform white ash was obtained (approx. 24 hours). Ashed crucibles were held in a desiccator until cool before weighing. Percent ash was calculated as a function of the combustible material on a fresh weight basis using the following equation: % Ash = weight of ash residue (9) x 100 weight of initial sample (9) Chemical Physical Analyses Lipid Oxidation 2-Thiobarbituric acid (TBA) analyses were done according to the distillation method of Tarladgis gt 31. (1960) to test for lipid oxidation. A randomly selected 10 9 sample of turkey meat was homogenized with 50 ml distilled water in a Vir-Tis macrohomogenizer Model 23 (285 ml flask) for two minutes. Each blended sample was quantitatively transferred by washing with 47.5 ml distilled water to a 500 ml distilling flask with 2.5 ml hydrochloric acid added (4N HCL). Several glass beads were added along with Antifoam A spray (Dow Corning, Midland, MI) to prevent excessive foaming. The distillations were 25 performed on a unit using a 300 mm Vigreux column attached to a 479 mm Leibig condenser with a 750 elbow. Fifty mililiters of the distillate were collected in a graduated cylinder within 10-15 minutes. Special consideration was given to maintain uniform heating times among distillations. TBA reagent (0.02M 2-thiobarbituric acid in 90% acetic acid) was prepared by dissolving 1.4416 g thiobarbituric acid (Eastman Organic Chemicals, Rochester, New York) with 50 ml distilled water and making a 500 m1 volume with glacial acetic acid. To help in dissolving the TBA reagent, an ultrasonic cleaner (Mettler Electronics Corp., Anaheim, California) was used. The sample distillate was mixed and 5 ml were reacted with 5 ml of the TBA reagent in capped culture tubes (200 mm X 25 mm). The samples were heated for 35 minutes in a boiling water bath and then cooled in a cold water bath for ten minutes. Spectrophotometric quantitation was performed (Beckman DB Spectrophotometer, Beckman Inc., Fullerton, California) using absorbance at 532 nm against a blank consisting of 5 ml of distilled water. Duplicate reactions were run for each distillate. Distilled water was used as the blank (reagent). The TBA number expressed as mg malonaldehyde per 1000 grams of meat was calculated by multiplying the mean absorbance by the constant 7.8. Extraction of Lipids Total lipids were extracted from all the turkey samples by a modified procedure of Folch gt 31. (1957). Approximately 100 grams were homogenized at high speeds in a Vortex homogenizer with a volume of 2:1 cholorform to methanol. The mixture was filtered using a Buchner funnel 26 fitted with #1 Whatman filter paper and transferred to a separatory funnel to allow the chloroform and the aqueous layer to separate. The solvent layer was evaporated at 20°C by using a rotary vacuum evaporator (Rinco Instrument Co.). Following evaporation traces of chloroform were further evaporated under a N2 stream. The sample was stored at -18°C when not used immediately. BHA Analyses BHA analyses were performed using a Waters Associates ALC 201, 202 High Pressure Liquid Chromatography (HPLC) equipped with a Model 440 absorbance detector at 284 nm with a sensitivity of AUFS=0.02. The column, 30 cm X 3.9 mm was packed with u-Porasil and the injector system was model U6K. The carrier solvent was chloroform. The flow rate was set at 1.5 ml/min with a retention time of 7.5 minutes. A sample of 50 microliters was injected for each analysis. The emerging peaks were quantatively identified using retention times of standard mixtures of known concentrations of BHA. Peak areas were calculated quantitatively as the product of peak height and concentration injected. Results were expressed as milligran BHA per gram fat. Surface Color of Thawed Turkey Roast After thawing at 4°C for 24 hours, each turkey roast was sliced and the surface color of the slices was evaluated using a Hunter Lab Model D-25 Color and Color Difference meter (Hunter Associates Laboratory, Fairfax, Virginia). Using a white standard, Hunter L,a,b, were obtained for each slice of meat. Two slices from each roast were 27 evaluated and two readings were made at 900 angles on each slice to average any deviation due to irregular surface refraction. Sensory Evaluation Sensory evaluations were obtained by sixteen panelists who were randomly selected from students, faculty, and staff of the Department of Food Science and Human Nutrition. All turkey roasts were thawed for 24 hours at 40C and then each slice was cut into four pieces. Samples were coded with two digit random numbers and evaluated under white light in individual panel booths. Panelists were given four samples at one sitting to evaluate and they returned the following day to evaluate the remaining samples. Panelists were asked to judge only flavor on a scoring system of one to nine, as follows: 9-extremely fresh flavor, 8-very fresh flavor, 7-moderately fresh flavor, 6-slightly fresh flavor, 5-f1at (neutral), 4-warmed over flavor (slightly stale), 3-sta1e, 2-rancid and l-very rancid. See appendix for score sheet. Statistical Analyses The "Statistical Package for the Social Sciences" (Nie gt_gl, 1975) program for the CDC 6500 computer operated by Michigan State University was used to assist statistical analyses. Three-way analyses of variance were determined using "Anova" program. Mean squares were reported after rounding and Tukey mean separations were determined to denote significant differences. RESULTS AND DISCUSSION The Initial Product The proximate composition of the turkey roast is outlined in Table 1. It can be seen that this turkey product has a high protein content, approximately 22%, and a very low lipid content, approximately 2%. The turkey roasts, prior to cooking, had a mean TBA value of 1.4. This value was somewhat higher than anticipated. Possible causes for this high measurement include the method by which the birds were killed and processed, and the holding time of the turkey before use in this product. Following cooking, TBA values of the precooked turkey, prior to storage were evaluated using two individual turkey roasts. TBA values from these roasts are reported in Table 2. These data indicate that TBA values were initially high, with a mean of approximately 3.00. As can be seen from Table 2 there is substantial variance among the samples of the product; varying from a mean of 2.5 to 3.4. This fluctuation and scattering at 0 time among samples could influence the variability in TBA values obtained throughout the study. Variability in TBA numbers of the meat samples as well as accuracy of the TBA test could have influenced this initial fluctuation. Cash and Carlin (1968) reported similar findings and suggested that early stages of rancidity and oxidative deterioration may occur first in one area, perhaps the outer edge of the meat and is not evenly distributed throughout. The high TBA values of the initial cooked product could be the result of the cooking process or the addition of salt. Watts (1962) also found that cooked meat had larger TBA values 28 29 Table 1. Proximate Composition of Turkey Roast. COMPOSITION Component Protein Moisture Lipid Ash Mean % 22.3 72.8 2.2 2.6 30 Table 2. Initial TBA values for two precooked Turkey Roasts. Beau. ngplg 1, TBA Values7? 1. 2.25 2.35 2. 2.59 2.59 3. 2.78 2.78 4. 2.35 2.32 Mean 2.49 2.51 W Sample TBA Values 1. 3.42 3.37 2. 3.37 3.37 3. 3.46 3.51 4. 3.37 3.35 Mean 3.40 3.40 II (I) 4 samples per roast, 2 replicates per sample of meat ... n than raw meat. In cooked meat phOSpholipids have been shown to be the lipid component most rapidly oxidized (Younathan and Watts, 1960). According to Lea (1957) the tendency of phOSpholipids to oxidize rapidly is partially due to their high content of unsaturated fatty acids. An additional possible explanation for the increase in TBA values after cooking is that the raw meat lipid is bound to protein and exists in a 31 1ip0protein complex. High cooking temperatures could result in the breakdown of this complex. The lipid fraction then could be released and become increasingly susceptible to oxidation attack. Lipid Oxidation, TBA Test TBA values of the turkey meat in all of the three package films (not considering environments) over the six months of frozen storage are reported in Table 3. These results are graphically illustrated in Figures 2, 3, 4, and 5. Table 4 reports TBA values for turkey meat in all of the package materials and environments during the six month storage period. Figure 2 shows the mean TBA values of turkey meat packaged in all films (not considering environments) vs. time. Turkey lipids increased in TBA value or rancidity over the six month storage period in meat packaged in all of the films. Turkey packaged in PE + BHA film had the lowest mean TBA value, 3.83, after six months of frozen storage. This is significantly lower than turkey meat in the control (PE) which had a TBA value of 4.69. The largest increase in TBA values in the meat occurred between zero and two months of frozen storage in all of the films. 32 Table 3. TBA values for Turkey Roasts in Different Package Films During Six Months Frozen Storage. StorageiTiméTTmonths) 6 Package Film OQO’ ‘2 4* mean TBA Values SARANEXTM 2.95 3.44a 3.92a 4.19ab 3.73 PE + BHA 2.95 4.15b 4.35a 3.83a 4.00 PE 2.95 4.50b 4.15a 4.69b 4.39 MEAN 3.99 4.13 4.24 All values within columns that have the same letter are not significantly different as determined by Tukey's test. .mmogoum co~ogu msucoz xwm m:_c=o Aep_m :oam Lee mocwscogw>cm Ppn mo covey mumoom moxeap cm mmzpm> mxm2h Z. mm34<> m§m=h Z_ wm24<> mxm2h Z. mm:._<> \ ooo .......... .\ ‘00 .v ZWGOCh—z l..............“‘..... 1 II I’ L m_< I II \ II \ Is A SE <=mlm5 $me 2. mm:._<> <2. 41 Table 6. BHA Content in Turkey Meat During Frozen Storage Storage Time (month) Package Treatments micrograms BHA/g fit BHA - vacuum 28.31 8.18 20.33 BHA - N2 15.09 9.82 7.38 BHA - Air 10.94 7.76 5.62 oxidation as indicated by TBA values in this experiment. As stated previously, BHA is present at its highest level after two months of frozen storage. TBA values also show the greatest increase between 0 and 2 months of storage. The possible correlation is that TBA values rose rapidly as the result of a short induction period for oxidation reactions due to the prior cooking; thus, oxidation took place before the BHA was present or available to slow down the oxidative reactions. Very slight changes in TBA values took place between two and six months of storage which could be attributed to the antioxidant present and available to slow down oxidation in the turkey. In summary BHA does seem to have an effect on TBA values or lipid 42 .mmmgoum :mNoeu newczo mummoa zmxezp o. cmemcmeh zm wmmxooa cw Loemcoch m¥m:._. O._. mmumzst. ugv eo—ou new: .m me=m_m 9.2.29). m V N . . . . . P am A E san1VA-1 co 1\ £01.00 ._.=u mmmxuom cw mumeom amxgzp mo mangmgsmmmz Am:_m>-gv co_ou “mm: .m mezmwm th202 m V G q £0400 ._.