(“vqh wan. «mot-“Mw‘ ~~ qnfi‘ “my... ~wmm‘9!‘ THE L033 05" VOIMILE CONSTIWENTS AND STORAGE STABILITY OF- FREEZE - DRIED SOUR CREAM Thesis for the Degree of ‘Ph. D. MICHIGAN STATE UNIVERSITY CHARLES RADANOVICS 1959 15;;th Wig yfilx/‘i/[nglffli’t’flt////IWI 533;: 7 0978 W This is to certify that the thesis entitled THE LOSS OF VOLATILE CONSTITUENTS AND STORAGE STABILITY OF FREEZE-DRIED SOUR CREAM presented by CHARLES RADANOVICS has been accepted towards fulfillment of the requirements for Ph.D. degree in Food Science MA MSW Major professor Date 13 November 1969 0-169 .2“ 4 s r: "'9 ¢ "I. G 7 l . {l ‘Jd'db {1 Q “”5 ABSTRACT THE LOSS OF VOLATILE CONSTITUENTS AND STORAGE STABILITY OF FREEZE-DRIED SOUR CREAM BY Charles Radanovics The research reported in this thesis involves a study of constituents in a freeze—dried model system and the storage stability of freeze-dried sour cream. Flavor Studies.--Acidified cream with added di- acetyl or acetoin served as a model system to determine the loss of flavor volatiles. The samples were freeze- dried under conditions wherein the absolute pressure, platen temperature, tray loading and the level of vola- tile additives were varied. In addition, the samples were exposed to different levels of relative humidity to ob- serve desorption of adsorbed volatiles. ' The absolute pressure between 20-350u in the vacuum chamber did not effect the relative loss of di— acetyl or acetoin; however the retention of these com- pounds increased when freeze-dried at a pressure of 4p. The increase in platen temperature from 75-225F resulted in a linear increase in the loss of both diacetyl and Charles Radanovics acetoin and a linear decrease in the time required for freeze-drying. The relative loss of diacetyl was inde- pendent of the initial absolute concentration between the range of 5-100 ppm when freeze-dried at a platen temper- ature of lOOF. Holding the powders at 50% relative humidity for 24 hours resulted in a rapid decrease of the adsorbed di- acetyl. Samples stored at 0% humidity showed no appreci- able loss of diacetyl. These results suggest that flavor volatiles are retained in freeze-dried foods by an inter- action resulting in adsorption rather than selective permeability. Storage Studies.--Freeze-dried sour cream is known to be sensitive to oxidation. The addition of various antioxidants, use of low temperature storage, and releasing the vacuum in the dehydrator with an inert gas were used to retard lipid oxidation and thereby increase the storage life. In addition samples were reconstituted and irradiated with ultra violet light (UV) in an effort to predict the usefulness of the added antioxidants. Organoleptic evaluation and Peroxide Values (PV) were used as criteria of quality. Samples stored at 99F developed off-flavors in 4 weeks and after 8 weeks of storage, the samples were com- pletely unacceptable. The off-flavors were caused by Maillard browning and lipid oxidation. The antioxidant Charles Radanovics combinations which were found to be the most effective at both 72 and 40F were 0.01% ERA + 0.01% BHT; 0.04% PG; and 0.04% ERA. The measured hydrOperoxide levels correlated well with the organoleptic evaluation when compared to the individual control. Samples stored at 40F had much higher Peroxide Values and this was attributed to the accumula- tion of hydroperoxides at low temperatures. Releasing the vacuum in the freeze dryer with nitrogen gas (N2) resulted in improved shelf life at both 72 and 40F storage. The N2 treatment appeared to be as effective in retarding lipid oxidation as the use of an antioxidant. The peroxide values and the results of the organoleptic evaluation correlated well in these experiments. Exposing the samples to UV light resulted in increased hydroperoxide formation. The level of such hydroperoxides was signifi- cantly lower in the samples containing effective concen- trations of antioxidants. THE LOSS OF VOLATILE CONSTITUENTS AND STORAGE STABILITY OF FREEZE-DRIED SOUR CREAM BY Charles Radanovics A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Food Science 1969 DEDICATION This work is dedicated to my beloved family: wife Margaret and to my children Anthony, Maria and Kathrina. my ACKNOWLEDGEMENTS I wish to express my most sincere appreciation to Dr. C. M. Stine for his advice and guidance throughout my graduate work. I also wish to thank Dr. E. J. Benne, Dr. C. L. Bedford, Dr. H. A. Lillevik and Dr. P. Markakis for their time and effort in serving on my committee and critically reviewing this manuscript. I am grateful to Dr. B. S. Schweigert, Chairman, and to the members of the Department of Food Science for the opportunity given to me for graduate study and reé. search in the department. The author acknowledges with gratitude the financial support of the American Dairy Association and the Department of Food Science which made possible this study. iii TABLE OF CONTENTS . Page DEDICATION O O O 0 O O O O O O O O O O O O O O O O 0 ii ACKNOWLE DGEMENTS O O O O O O O O O O O O I. O O O O O i i i LIST OF TABLES O O O O O O O O O O O O O O O I O O 0 Vi LIST OF FIGURES O O O O O O O O O O O O O O O O O . Viii Chapter INTRODUCTION . . . . . . . . . . . . . . . . . . . ‘ 1 LITERATURE REVIEW . . . . . . . . . . . . . . . . 3 Freeze-Drying of Foods . . . . . . . . . . . . . 4 Freeze-Drying of Dairy Products . . . . . . . 8 Storage Stability of Dehydrated Dairy Products . . . . . . . . . . . . . . . . . . 12 The Mechanism of Oxidation . . . . . . . . . . l3 Oxidized Flavor in Dairy Products . . . . . . 15 Prevention of Lipid Oxidation in Dairy Products . . . . . . . . . . . . . . 19 The Mechanism of Antioxidation . . . . . . . . 21 The Use of Antioxidants in Dairy Products . . . . . . . . . . . . . . . . . 22 The Chemical Detection of Oxidative Rancidity . . . . . . . . . . . . . . . . 25 Peroxide Determination . . . . . . . . . . . . 26 Thiobarbituric Acid (TBA) Test . . . . . . . . 27 Miscellaneous Tests . . . . . . . . . . . . . 28 Methods to Evaluate the Effectiveness of Antioxidants . . . . . . . . . . . . . 29 The Flavor of Cultured Sour Cream . . . . 30 The Loss of Flavor Constituents in Dehydrated FOOdS o o o o o o o o o o o o o o o o o o 3 3 METHODS AND METHOWLOGY O O 0 o o o o o o 0 Op 0 o 39 Preparation of the Samples . . . . . . . . . 39 Preparation of Samples for Flavor Studies . . . 39 iv Chapter Preparation of Samples for Storage Studies . . . . . . . . . . . . . . . . . Antioxidant Studies . . . . . . . The Use of Selected Gases to Break the Vacuum in the Freeze Dryer. Analytical Techniques . . Moisture . . . . . . Fat . . . . . . . . . . Titratable Acidity . Extraction of Flavor Volatiles . Determination of Diacetyl and Acetoin . . . Peroxide Value (P. V. ) . . . . . . . . . . . Flavor Evaluation . . . . . . . . . . . . . Ultra Violet (U.V.) Exposure . . . . . . . RESULTS AND DISCUSSION 0 C. O O O O O O O O O O The Effect of Processing Variables on the Loss of Diacetyl and Acetoin in Acidified, Freeze Dried Cream . . . . . . The Effect of Absolute Pressure in the Freeze Drying Process on Losses of Volatiles . . . . . . . . . . . . . . . The Effect of Platen Temperature on the Loss of Diacetyl and Acetoin . . . Storage Stability of Freeze Dried Sour Cream. The Effect of Antioxidants on the Shelf Life of Freeze Dried Sour Cream . . . . The Effect of Inert Gas in Breaking the Vacuum of the Freeze Dryer on the Shelf Life of Freeze Dried Sour cream . U.V. Experiments - - - . . . . . - - - . - SUMMARY AND CONCLUSIONS . - . . . . . . - . . . Flavor Studies Storage Studies THESIS REFERENCES 0 o o o o o o o o o o o o o o 0 APPENDIX 0 O O O O O O O O O O O 0 0.0.0.... 0 O Page 42 42 44 44 44 45 45 45 45 47 47 47 49 49 50 54 74 74 94 99 103 103 104 106 126 LIST OF TABLES Table Page 1. Antioxidants and synergists used in freeze-dried sour cream studies . . . . . . 43 2. The effect of vacuum on the loss of diacetyl in freeze-dried acidified cream at lOOF platen temperature . . . . . 52 3. The effect of vacuum on the loss of acetoin in freeze-dried acidified cream at lOOF platen temperature . . . . . . . . S3 4. The effect of a condenser temperature on the loss of diacetyl at lOOF platen temperature . . . . . . . . . . . . . . . . 54 5. The effect of platen temperature on the loss of diacetyl in freeze-dried acidified cream . . . . . . . . . . . . . . 58 6. The effect of platen temperature on the loss of acetoin in freeze-dried acidified cream . . . . . . . . . . . . . . 6l 7. The effect of layer thickness on the loss of diacetyl and acetoin at lOOF platen temperature . . . . . . . . . . . . . . . . 65 8. The effect of different levels of diacetyl on the loss of diacetyl in freeze-dried acidified cream at lOOF platen temperature . . . . . . . . ,,, . . 69 9. The effect of 24 hours storage at various temperatures and humidities on the loss of diacetyl . . . . . . . . . . . . . 71 10. Peroxide values (PV) and intensity of oxidized flavor (0F) of freeze-dried sour cream in storage . . . . . . . . . . . 126 vi Table Page 11. Peroxide values (PV) and intensity of oxidized flavor (OF) of freeze-dried sour cream in storage . . . . . . . . . . . 127 12. Peroxide values (PV) and intensity of oxidized flavor (OF) of freeze-dried sour cream in storage . . . . . . . . . . . 129 13. Peroxide values of fat extracted from reconstituted sour cream exposed for variable periods to ultraviolet light (2537A) . . . . . . . . . . . . . . . 100 vii Figure 1. 10. LIST OF FIGURES RePP freeze dryer equipped with tempera- ture and pressure recorders . . . . . Equipment used for irradiating samples with Ultra Violet light . . . . . . Temperature profile of freeze-dried acidified cream at various platen temperatures . . . . . . . . . . . . Freeze dried acidified cream at X130 magnification . . . . . . . . . . . . The effect of extended freeze-drying on the loss of diacetyl at different platen temperatures . . . . . . . . . Drying curve and temperature profile of freeze-dried sour cream freeze dried at lOOF O O O O O O O O O I O O O 0 O Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 99F . . . . . . . . . . . . Peroxide values and detected oxidized flavor (+) of freeze—dried sour cream containing various antioxidants, stored at 99F . . . . . . . . . . . . Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 99F . . . . . . . . . . . . Peroxide values and detected oxidized flavor (t) of freeze-dried sour cream containing various antioxidants, stored at 72F . . . . . . . . . . . . viii Page 41 48 55 64 67 77 80 81 82 85 Figure Page 11. Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 72F . . . . . . . . . . . . . . . 86 12. Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 72F . . . . . . . . . . . . . . . 87 13. Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 40F . . . . . . . . . . . . . . . 89 14. Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 40F . . . . . . . . . . . . . . . 90 15. Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 40F . . . . . . . . . . . . . . . 91 16. Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream exposed to various gases at conclusion of drying cycle, stored at 72F . . . . . . 95 17. Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream exposed to various gases at conclusion of drying cycle, stored at 40F . . . . . . 97 ix INTRODUCTION During the past decade the production of conveni- ence foods has increased tremendously. In order to find consumer acceptance such foods must be capable of being readily prepared by the housewife. They must also possess excellent and uniform flavor attributes and have reasonably good shelf life. The consumption of fresh sour cream has also in- creased during this period of time. However, the perish— ability of the fresh product limits the utility and sales of sour cream, particularly to the institutional outlets. The development of a good quality dehydrated sour cream could greatly increase acceptability and use of sour cream. Freeze-drying, which generally produces a dehy- drated product of superior flavor attributes would be the method of choice for dehydration. The purpose of this research was to undertake an investigation which would contribute to the deve10pment of a freeze-dried sour cream. Such an investigation also would hopefully serve as a model for other researchers for better ultimate understanding of some of the problems en- countered in the freeze-drying of a food. The research reported in this thesis was aimed at elucidating the following matters pertinent to development of a freeze dried sour cream: 1. Better knowledge of the physical forces re- sponsible in retaining flavor volatiles during freeze- drying. 2. Evaluation of the effect of processing condi- tions on the loss of diacetyl and acetoin in a model system. 3. Study of the storage stability of freeze-dried sour cream at various storage temperatures, and the effect of addition of various antioxidants and processing tech- niques. 4. The development of a method to permit the qual- itative evaluation of the potential effectiveness of anti- oxidants without the necessity of long term storage studies. LITERATURE REVIEW Dehydration as a method of food preservation was practiced for centuries. It served the purpose of preserv- ing foods during the seasons of plenty to be consumed in the seasons of shortage. In the present economic system other factors such as convenience, ease of distribution, and extended shelf life play equally important parts. The many methods of dehydration, such as sun dry- ing, air drying, spray drying, drum drying, etc., all have certain inherent disadvantages, which have been summarized by Harper and Tappel (1957): l. Pronounced shrinkage of the particles 2. Migration of dissolved constituents to the surface 3. Varying degrees of denaturation of proteins 4. Case hardening A 5. Undesirable chemical reactions in the heat sen- sitive materials 6. Excessive loss of desirable volatile constitu- ents 7. Difficulty of rehydration. Freeze-drying, a relatively recent development in the dehydration of foods, overcomes, in whole or in part, many of these undesirable characteristics. Freeze-Drying of Foods Freeze-drying can be defined as a method of dehy- dration wherein moisture is removed from a frozen product by the process of sublimation, changing ice to vapor with- out going through the liquid state. This phenomenon was known in the 17th century; however, the first recorded usage of sublimation for the preservation of biological substances was that by Shackell (1909). Lyophilization was initially used to preserve bacterial cultures. Later Flosdorf and Mudd (1935) demon- strated the feasibility of freeze drying blood serum. The freeze drying of blood serum and other labile biological substances such as penicillin was done on a large scale during the Second WOrld War. The techniques used and the problems encountered in freeze drying these materials were discussed in detail by Flosdorf (1949). The utilization of freeze drying for food usage began after the Second World War. The demand by the mili- tary for high quality dehydrated combat rations, as well as the utilization of freeze dried foods in the various Space flight programs initiated research in the development of freeze dried foods (Graf, 1961; Nanz and Lachance, 1967; Tuomy, 1965). I The early development of freeze drying of foods, including equipment used, fundamentals of the drying proc- ess, and research findings on the chemical and physical properties of such foods have been discussed by Harper and Tappel (1957). Stein (1966) described the advantages as well as the cost involved in producing freeze dried foods as compared to foods dehydrated by conventional means. The increased usage of freeze dried foods during the next dec- ade was projected by Bird (1964). Freeze dried food pro- duction is expected to reach 250 million pounds by 1970, up from 6.5 million pounds in 1964. This expected increase in production has stimulated research into the development of better freeze drying equipment, reduction of costs and the investigation of the physical and chemical changes which develop in freeze dried foods. The increase in heat energy input without melting the ice crystals and the efficient utilization of such heat energy will result in increased capacity and utilization of freeze drying equipment. The application of direct heating, ultra high frequency heating and radiant heating, in freeze drying was investigated by Hammond (1967). The use of infra-red radiation as a heat source was described by Burghimer (1966). Circulating helium or hydrogen above the food during the freeze drying process would result in an increased drying rate by a factor of 2 or 3 (Kan and Winter, 1968). A similar principle is utilized by the "Wurcal Dryer" developed by King and Clark (1968). Lewin and Mateles (1962) described a method where diced vegetables were freeze dried under atmospheric pres- sure. The moisture was removed by circulating desiccated air abOve the food particles. Flushing the surface of the freeze dried food with heated carbon dioxide increased the rate of sublimation (Larson gt 31., 1967). The work of Sinnamon gt a1. (1968) demonstrated that a dew point of -80F can be reached by adjusting the velocity and temperature of the circulating air. The elimination of the condenser can be achieved through the installation of a molecular sieve before the vacuum pump. Such a molecular sieve would adsorb the vapors liberated during the freeze drying process (Saravacos, 1967a). A patent was issued for the freeze drying of foods by forming an azeotrOpe mixture (Bohrer, 1967). The food is first frozen and is then placed in a vessel containing ethyl acetate, forming an azeotrope with water. The boil- ing point of the azeotrope is lowered and hence the rate of sublimation is increased. Fluidized bed freeze drying was suggested by Scott gt_al, (1967) to increase the efficiency of the system. A method of bulk freeze drying was described by Hansen (1967L In the process liquid is converted into a frozen powder by atomization of the liquid into cold air. The powder is fluidized in a rotating drum under vacuum and is quickly freeze dried. This process can be adapted to continuous operation. Freeze drying usually produces a superior product when compared to products dehydrated by other means. This does not mean, however, that freeze dried foods are with- out defects. Such defects may develop during processing or during the storage of such foods. These defects normally involve chemical, physical or physicochemical changes in the constituents of the dehydrated food. The denaturation of protein is far less in freeze dried foods than in foods dehydrated by other means (de Grout, 1963). Zabik (1968) and Zabik and Figa (1968) observed that the gel strength of reconstituted freeze dried eggs was similar to fresh frozen eggs and superior to spray dried or foam spray dried eggs. Color loss and the breakdown of certain pigments in freeze dried foods were observed by Lusk et_al., 1965. The texture of freeze dried poultry (Hamre, 1966) and shrimp (Moorjani and Dani, 1968) was described as tough and rubbery. Processing variables effected the organolep- tic properties of freeze-dried beef (Tuomy et_al., 1968). Long term storage of freeze dried foods may result in deterioration to varying degrees (Goldblith et 31., 1963 a,b). Non-enzymatic browning results in darkening, deterioration of color and the development of undesirable flavors, depending upon moisture content and storage tem- perature. Decrease in nutritive quality from loss of vitamin C, and the decreased availability of lysine may oCcur, concurrent with browning. Oxidative deterioration involving the lipids, pigments, vitamins, and certain fla- vor constituents may also occur. Changes in solubility, dispersibility and water holding capacity could also develop as a result of prolonged storage of the dehydrated food. The quality of the raw material used will of course influence the final quality. Particular care must be taken to freeze dry a product without the presence of any patho- genic organism. The survival of such organisms in freeze dried food was investigated by several researchers. The findings indicate that food contaminated prior to freeze drying will have a large number of viable organisms after freeze drying (Goldblith 31 31., 1963 a,b; Sinskey 31 31., 1964; Saleh, 31 31., 1966; Pablo 33 31., 1967). Freeze Drying of Dairy Products Comparatively little work has been published on freeze dried dairy products; however, a few reviews are available (Juriet, 1964; Anastos, 1964; Mann, 1966, 1967). The quality improvement, which is obvious in other food products, is not always apparent in dairy products. Generally speaking, while there is improvement in flavor quality, the textural attributes are lacking. The cost, in general, is also excessively high. Nickerson e£_31, (1952) observed that the fat emulsion of fluid milk is destabilized due to freeze dry- ing, resulting in a great increase in "free fat" and rather poor dispersibility properties. The flavor of the freeze dried milk was initially better; however, no improvement in keeping quality was observed over spray dried whole milk during storage. The investigations carried out by Rutz 31 31. (1953) on the emulsion stability of freeze dried milk and cream showed that the addition of sodium citrate or ‘dipotassium phosphate, (compounds utilized in the dairy industry in the stabilization of cheese slurries) were in- effective in promoting emulsion stability. This research also demonstrated that destabilization occurred during the drying process, and not, as was generally believed, during the freezing cycle. The addition of certain emulsifiers and surfactants improved the dispersibility of whole milk powders, but adversely affected the emulsion stability (Hibbs and Ashworth, 1951; Mather and Hollender, 1955; Nelson and Winder, 1963). The proper selection of emulsi- fiers in hydrophile-lipophile balance (ULB) ranges of 11-14 increased the emulsion stability in freeze dried model milk systems. The stability of the milk fat emulsion also in- creased with the increased concentration of emulsifier (Mickle, 1965, 1966). 10 The freeze drying of milk on a commercial scale, using a semi-continuous production method, was described by Duiven (1965). Ogden (1967) patented a method to pro- duce freeze dried liquid milk products. The milk is first concentrated under reduced pressure to one third of its original volume. The milk is then frozen in sheets, broken up, pressed into blocks and freeze dried to less than 5 per cent moisture. Jureit (1964) discussed the economic considerations of producing freeze dried milk, skim milk, youghurt and quarg. According to Bernhard and Nelson (1967) the freeze drying of a great variety of dairy products can be success- fully accomplished. Freeze drying as a preservation of cheese has been suggested by several workers. EvsrateVa 33 31. (1959) suggested freeze drying as a means of preserving two vari— eties of lactic-coagulated cheese in Russia. The structure and physical properties of various freeze dried cheeses was studied by Schultz (1966). The experiments included the following types of cheeses: quarg, double cream cheese, Edam, Camemberg, Gouda, Emmental, Romadur, Tilsit, Danish blue and processed cheese. The body of most of these cheeses was unsatisfactory upon rehydration. The so-called "high moisture" cheeses, such as quarg and double cream cheese could be more readily reconstituted. 11 The feasibility of freeze drying quarg on a commer- cial scale was successfully demonstrated by a German manu- facturer. The freeze dried product, packaged into plastic bags, possessed good rehydration properties and was readily soluble in cold water (Anon, 1964). Meyer and Jokay (1959) and Jokay and Meyer (1959) found that the body of freeze dried creamed cottage cheese, Cheddar, Brick, Muenster, blue and cream cheeses, after re- constitution had soft but natural consistency. A nutritionally balanced dehydrated food bar, for survival type rations, using Swiss or Cheddar cheese as a base was developed by Jokay (1964). Slurried cheese is mixed with cooked potatoes, fat and starch, formed into bars and freeze dried. The freeze drying of ice cream or ice milk was suggested by several workers (Anon, 1966; Grace, 1966; Bernhard and Nelson, 1967). Cultured dairy products such as butter milk, sour cream and youghurt, with their pleasant acid flavor and distinctive aroma, are prone to flavor deterioration when dehydrated. The improvement of these flavor attributes is considerable when such products are freeze dried. Dasai (1966) found that when sour cream is freeze dried, higher free fat values are obtained in comparison to spray dried powders. The product was of pleasant flavor, but the body did not resemble the body of a typical sour cream. The improvement in body characteristics can be 12 achieved through the addition of starches and thickening agents (Stine 31; 31. 1968, 1969). Jekabs and Mezciems (1966) patented a process which claims to improve body characteristics of dehydrated fermented milk products. The product containing 25 per cent total solids is cultured by conventional means, lactic acid is added to raise the acid level to 1.6 per cent, and the acidified mixture is then spray dried. The powder is reconstituted with fluid whole milk to a final acidity of about 0.7 per cent. A recent patent issued to Hackenberg (1967) suggested the removal of serum before freeze drying to improve on the cost factor. The process calls for coag— ulation of the milk or milk product by acid or rennet prior to freeze drying. The stabilization of the fat in a spray dried sour cream was achieved by culturing skim milk, and then homog— enizing butteroil (milkfat) into the cultured product (Noznick 33 31,, 1963). Storage Stability of Dehydrated Dairy Products The successful storage of foods implies that the food is acceptable organoleptically at the time of con- sumption. Other attributes such as nutritional factors, while important, are of secondary consideration to the consumer . 13 The stored food is exposed to a variety of hazards (Hearne, 1964): (1) Infestation of rodents, insects and mites, (2) contamination by dust and foreign odors, (3) microbiological attack, (4) oxidation, hydrolysis and re- version in fats, (5) oxidation of pigments, (6) browning reactions, (7) changes in structure of proteins, (8) staling, (9) crystallization and colloidal modifications, (10) physical changes including caking, (ll) enzymatic activity, (12) changes in nutritional value, (13) inter- action with the container, (14) container damage or de- terioration. No effort has been made in this thesis to review the vast literature available on the deterioration of foods in storage, unless it pertained specifically to the storage stability of dried milk systems. The Mechanism of Oxidation Autoxidation involving the lipid portion of a dairy product in the presence of atmOSpheric oxygen is in- fluenced by a variety of factors: (1) saturation of the fat, (2) storage temperature, (3) effect of light, (4) ionizing radiation, (5) enzymes, (6) catalysts (organic and inorganic), (7) presence of oxygen, (8) use of anti- oxidants (Lea, 1962). The development of oxidized flavor in dairy prod- ucts would follow the well accepted hydroperoxide mechanism l4 established for lipid systems. The spontaneous reaction between atmospheric oxygen and an unsaturated fat will re- sult in the formation of hydroperoxides: RH + 02 -------- > ROOH The consequent steps in the oxidation of fat be- come more complex due to breakdown involving the initially formed hydroperoxides (Lea, 1962). DIMERS, HIGHER POLYMERS polymerization .'\ HYDROPEROXIDE‘ I L//’ \‘ ‘\\\“N”“ -t~ ‘ further fission dehydration oxidation of oxidation 9 CH=CH in other molecules V V’ l . DIPEROXIDES ALDEHYDES KETO- EPOXIDES 'I SEMI- GLYCERIDES OH-GLYCERIDES \, ALDEHYDES DI OH-GLYCERIDES POLYMERS ALDEHYDO- GLYCERIDES OH-COMPOUNDS l ACIDS The formed hydroperoxides are tasteless and the detected off flavors are caused by hydroperoxide decompo- sition products in the form of saturated and unsaturated monocarbonyl compounds. 15 The volatile carbonyl compounds formed are organ— oleptically detectable at the concentrations as low as one part per billion (Patton 33 31. 1959, and Lillard and Day, 1961). Oxidized Flavor in Dairy Products Many reviews dealing with the oxidized flavor of dairy products have been published, the most prominent be- ing those of Brown (1940), Greenbank (1948), Riel and Som- mer (1955), Jenness and Patton (1959), Shipe (1964), Wilkinson (1964), and Kinsella 33 31. (1967). These re- views show the extent of intense research efforts done by literally hundreds of independent investigators. The liberation of butyric acid from the hydrolysis of milk fat is due to lipase activity. The flavor of such products is often.described as rancid, but it is distinc- tively different from oxidized flavor caused by the pres- ence of atmospheric oxygen (Herrington, 1956; Jensen, 1964). Pasteurization inactivates the enzyme and therefore pas- teurized dairy products are free from this defect (Frankel and Tarrasuk, 1959; Scwartz 33.31, 1956). Presently, vari- ous regulations require the pasteurization of dairy prod- ucts, and therefore rancidity caused by lipase activity is almost non-existent. According to Patton (1962) the particular type of off flavor developed in oxidized dairy products would ’16 depend on the relative presence or absence of water as a solvent for the reaction. The typical off flavor observed in an aqueous sys- tem can be described as "cardboard" or "cappy", while those flavors developed in the absence of water are usually described as "oily" or "tallowy." Several of the compounds responsible for the oxidized off flavors have been identi- fied. Kinsella 33 31. (1967) compiled the data in a recent review as follows: Some Descriptive Flavors and Associated Compounds Identified in Oxidized Milk Fat Flavor Compounds Oxidized oct-Z-ene-3-one, octanal, hept-2-enal, 2,4-heptadienal, n-alkanals (C2-C9) Cardboard, tallowy n-octanal, n-alkanals (C9, C11), alk-Z-enals (C8, C9), 2,4-dienals (C7, C10), 2,6-dienal (C9) Oily n-alkanals (C5, C6, C7), hex-Z-enal, 2,4-dienals (C5, C10) Painty n-alkanals (Cs-Clo), alk-2-enals (C5- C10): 2,4-dienal (C7),2-alkanone (C7) Fishy n-alkanals (Cs-Clo), alk-Z-enals (C5- C10), 2,4-dienal (C7), 2-a1kanones (C3-C11), oct-l—ene-3-one Grassy alk—2-ena1 (C6), 2,6-dienal (C9) Metallic oct-l-en-3-one Beany alkanals, non-2-enal .Mushroom oct-l-en-3-ol Cucumber 2,6-dienal (C9) Nutmeg octadienal; 2,4-dienals Creamy 4-cis-heptena1 Fruity n-alkanals (C5, C6, C8, C10) 17 The development of oxidized flavor is greatly en- hanced in the presence of iron and copper. The addition of 0.5 ppm copper to heated milk accelerated the develop- ment of oxidized flavor (Hollender and Tracy, 1942). Supplee (1923) showed that the addition of copper to milk before drying resulted in the development of tallowiness in air packed powder. King 33 31: (1959) observed that added iron does not become associated with the fat globule, whereas some of the added copper does. Aulakh (1967) investigated the binding of c0pper to the various fractions of milk pro- tein, inCluding the fat globule membrane protein, showing that the latter has the highest affinity for copper ions. The sulfhydryl groups liberated during the heat- ing of milk are in some way capable of inducing protection against the catalytic effect of copper (Corbett and Tracy, 1941) and ferrous iron (Gould, 1939). Gould and Sommer (1939) reported that the free sulfhydryl groups in high heat treated milk will significantly retard lipid oxida- tion. Similar results were obtained by Stine (1957) in his investigation of the effect of pre-heat treatment on the storage of dry whole milk, with or without the pres- ence of sulfhydryl group blocking agents. The level of moisture in dehydrated foods has a definite influence on the development of lipid oxidation (Tamsma §_t_:_ §__1_., 1961, 1964; Aceto 3E 31., 1965, 1966). 18 Tamsma 33 31, (1961, 1964) showed that the low moisture levels will adversely affect the storage stability of foam spray dried milk. Similarly Aceto 33 31. (1965, 1966) also concluded that low moisture levels will ac- celerate~lipid oxidation in the storage of foam dried milk. Holm 3£_31, (1927) reported that roller dried milk powders with a 2-3% moisture content kept better than those of lower moisture. -The oxidative effect of metal catalysts such as iron and COpper is decreased in the presence of water (Uri 1956). Labuza 33 31. (1966) has observed that in a model system water depressed cobalt catalyzed oxidation. The oxidation of linoleate in a freeze dried model system de- creased with the increase in water activity values, and this effect was most pronounced in the initial stages of oxidation (Maloney 33 31., 1966). The possible attachment of the water molecules to functional groups involved in oxidation of proteins and fats may prevent the direct contact of such functional groups with atmospheric oxygen. The water necessary for this effect could.be calculated from the BET1 equation and would represent the monomolecular layer of water (Lea, 1958; Salwin, 1959). The role of light in catalyzing the development of lipid oxidation in dairy products is well known. The flavor thus developed has been described as "solar 1Brunnauer, Emmett and Teller. 19 activated," "burnt feathers," "sunlight" or "tallowy" (Weinstein 33 31., 1951; Stull, 1953; Dunkley 33 31., 1962). .A recent review by Wishner (1966) discusses the photo- oxidation of lipids. Due to the effect of light, there is photolysis of trace quantities of hydroperoxides. Wishner concluded that there is no clear-cut distinction between the effect of light on lipid material and the classical c0pper induced oxidation. Light will also affect other components of the milk, in particular the protein fraction, developing additional off flavors (Greenbank, 1948; Stull, 1953; Patton, 1954; Dunkley, 33 31., 1962). Stull (1953) pointed out that the development of lipid oxidized off flavor and the activated or burned flavor would develop concurrently in dairy products exposed to light. This would account for the unusual composite of off flavors which characterize the "solar activated" defect. Prevention of L1pid Oxidation in Dairy Products The prevention or retardation of oxidized off flavor development in dairy products has been intensively investigated. The various federal, state and local regu- lations limiting the use of additives in dairy products have channeled research into the development of various measures to inhibit oxidation in milk. These measures have encompassed special feeding regimens for the dairy cow, elimination of all copper or copper alloy equipment, the 20 'use of stainless steel processing equipment in the dairy lplant, activation of sulfhydryl groups of the protein, lnomogenization, deodorization, use of appropriate natural (or synthetic antioxidants, and the reduction of oxygen in 'the package for dehydrated products. High temperature heat treatment to cream and fluid Inilk prior to drying enhances resistance of oxidative de- ‘terioration. Such heat treatment produces active sulf- hydryl compounds which may possibly function as primary antioxidants (Gould and Sommer, 1939; Josephson and Doan, 1939; Jack and Henderson, 1942; Harland 33 31., 1949). Gould and Keeny (1957) demonstrated that the processing of cream at 190 F for 5 minutes instead of 170 for 20 minutes definitely improved the stability of such cream to oxidative changes. The findings of Findlay 33_31, (1946) indicate that increasing the forewarming temperature from 190 F to 200 F in the production of whole milk powders did 'not improve the storage stability of such powders. The packaging of dehydrated dairy products in an inert atmosphere is a very useful method of preventing lipid oxidation. Lea 33 31. (1943) showed that reduction of oxygen in the head space gas to 2 per cent or less im- proved the storage stability of spray dried whole milk powder. Coulter and Jenness (1945) and Coulter 33.31. (1948) concluded that to prevent lipid oxidation the level of oxygen in the head space gas should be reduced to less ‘21 than 1 per cent. They also demonstrated that the reduction of oxygen to such a low level is possible by the method of double gassing. The reduction of oxygen content to 0.01- 0.016 cc per gram of powder enabled spray dried milk to be stored for a year in satisfactory condition (Hetrick and Tracy, 1945). The reduction of the oxygen level in the head space to less than 0.001 per cent to prevent lipid oxidation was suggested by Tamsma 33 31. (1967). This low level of oxygen was achieved through the spontaneous reaction of in a 5% H 0 atmosphere to form water in the presence of 2 2 a platinum catalyst. The use of palladium as a catalyst in a similar system was suggested by Abbot and Waite (1955). In—package deoxygenation as a means of reducing lipid oxidation was described by Scott and Hammer (1961). Into the sealed container a packet, permeable to air but not to moisture containing glucose oxidase, catalase and substrate was placed. The enzyme system reduced the oxygen in the container to a minimal level. The Mechanism of Antioxidation Antioxidants in general function by inhibiting the formation of free radicals or by favoring the decomposi- tion of formed hydrOperoxides. The decomposition of these hydroperoxides results in the formation of primary stable products (Dugan, 1961). Bollard and Ten Have (1947) studied 22 the mechanism and kinetics of the reaction involving the mode of action of antioxidants which led to the proposal that the antioxidants (AHZ) acted as hydrogen donors: ROO- + AH2 ------ > ROOH + AH' AH“ + AH; ------ > A + AH2 Matill (1931) noted the importance of configuration of active hydroxyl groups in phenolic types of antioxi- dants. When the two hydroxyl groups were in the ortho or para position good antioxidant efficiency was observed, while none existed at meta configuration. The phenomenontofsynergism can be observed when two or more antioxidants are used, resulting in a more pro- nounced antioxidant effect than the sum of their individual effects. Synergism occurs when two phenolic antioxidants are combined (Mahon and Chapman, 1953; Dugan 33 31., 1954) or with the use of an acid with a phenolic antioxidant (Calkins, 1947; Dutton 33 31., 1949). The Use of Antioxidants in Dairy Products The addition of various antioxidants as a means of preventing lipid oxidation in dairy products was suggested by several researchers (Garrett, 1940; Stull 33 31., 1948, I, 1948, II; Coulter, 1944; Tamsma 33 31,, 1963; Abbot and Waite, 1962; Pete and Smith, 1964). The first compounds tried in dairy products to retard lipid oxidation were of plant 23 origin. Garrett (1940) reported that cereal grains and soybean exhibited antioxidant properties in milk. Gum guaiac and hydroquinone were found to be effective in dry milk powder (Hollender, 1942). According to Coulter (1944) 2% oat flour or 0.1—0.2% wheat germ oil was as effective in retarding lipid oxidation as 0.01% hydroquinone. The addi- tion of citric acid enhanced the effectiveness of both. Nordihydroguaiaretic acid (NDGA), a naturally oc- curring phenolic compound, crystallized from the plant Larrea divericata was investigatdd by several workers for its antioxidative properties (Coulter, 1944; Stull 33 31., 1948, 1949, 1951; Bush 33 31., 1952). Stull 33 31. (1948) reported that the addition of 0.0125% NDGA retarded lipid oxidation for 5 days in the presence of .3 ppm copper. The exposure of fluid milk to light with or without the presence of NDGA was investigated by Weinstein and Trout (1951). While the control was highly oxidized, milk containing 25 mg/liter of NDGA remained acceptable. NDGA was found to be ineffective in retarding light induced oxi- dation in fluid milk when used at 1-2 mg/liter (Finley 33 31. 1967). NDGA was shown to be effective in retarding lipid autooxidation in frozen sweetened or unsweetened cream (Stull 33 31., 1948, 1949). . The storage stability of spray dried milk and ice cream mix held at room temperature was increased to 12 months by the addition of 0.1 and 0.04% NDGA respectively. 24 »Citric acid (CA) as a synergist was also beneficial (Stull 33 31., 1951). The effectiveness of CA as a synergist for NDGA was demonstrable only in samples held at least 9 months at 85 F (Bush 33 31., 1952). Bush et a1. (1952) also observed that the antioxidant activity of NDGA is manifested by a decrease in peroxide formation but an in- crease in ferricyanide reducing values. NDGA was used by Cox 33_31: (1957)to protect Vitamin A in fluid and airpacked dry whole milk. The addition of NDGA resulted in the develOpment of fuity flavor in dry whole milk (Coulter, 1944) and bitter flavor was observed in stored butter (Negoumy and Hammond, 1962). In stored Spray dried or foam spray dried cheddar cheese lipid oxidation was retarded by the use of NDGA (Bradley and Stine, 1964). BHA has been shown to be ineffective in retarding lipidautoxidation in dry whole milk. It was further demonstrated that dodecylgallate was very effective, while various flavone compounds also possessed good antioxidant .properties (Abbot 33 31., 1962). The addition of ethyl gallate at 0.07% level increased the storage life of whole milk powder 2 1/2 - 3 fold in the accelerated storage test. The ethyl gallate remained unchanged chemically during the storage trials (Findlay, 33 31., 1945). The long chain esters cf gallic acid exhibited better antioxygenic proper- ties in whole milk powder than NDGA, while BHA and BHT were 25 the least effective. Control samples packaged in an at- mosphere of nitrogen were as acceptable as samples comtain- ing antioxidants (Pete and Smith, 1964). Tamsma 33 31. (1963) noted that lauryl gallate, propyl gallate, NDGA and BHA, when added individually to foam spray dried whole milk, exhibited decreased effectiveness in the order listed. A combination of BHT and BHT + isopropyl citrate was inef- fective in butter stored at -18 and 38 F (Nagoumy and Ham- mond, 1962). Propyl gallate in the presence of added 0.5 ppm copper increased the storage stability of fluid whole milk to 14 days (Chilson, 33 31., 1950). The distribution of various antioxidants in milk was investigated by Martinez 33_31. (1958). The fat and serum phase contained 70 - 90% of added BHA but only 5% of the NDGA could be recovered in this study. The feeding of high level of tocopherol and ethoxy- quinin was demonstrated to be beneficial in improving the stability of fluid milk (Dunkley 33 31., 1967). The Chemical Detection of Oxidative Rancidity Rancidity, a highly objectionable off flavor in dairy products, is readily detectable organoleptically. The organoleptic determination, however, is quite arbitrary, and would vary from individual to individual, depending upon his flavor threshold. Chemical tests measuring the changes during lipid oxidation can sometimes be correlated 26 to flavor scores and hopefully would be less variable (Lillard 33 31., 1961). The chemical tests are usually divided into two categories: the measurement of hydrOper- oxide formed and the measurement of the subsequent break- down products of such hydroperoxides. Peroxide Determination The determination of peroxides, based upon their ability to quantitatively oxidize iodide to iodine, was suggested by Powick (1923). Wheeler (1932) designed an iodometric method where oil, chloroform and glacial acetic acid were heated in the presence of potassium iodide and subsequently titrated with sodium thiosulfate. This method is generally utilized when the formed hydroperoxides are in the range of 100-500 milliequivalents (MEQ) of oxygen per kg of fat. Dairy products exhibit detectable oxidized flavors when the peroxide value is in the range of 0.5-1 MEQ per kg of fat. The method developed by Chapman and McFarlane (1943) is suitable for the measurement of these low levels of hydroperoxide. The method is based on the oxidation by hydroperoxides of ferrous to ferric ion in the presence of ammonium thiocyanate, with the resultant formation of a red complex of ferric thiocyanate. The replacement of acetone as the solvent medium by a mixture of benzene-methanol mixture was recommended by Hills and Thiel (1946). The 27 liberation of butterfat from fluid as well as from dry whole milk was suggested by Stine 33 31. (1954) to facili- tate the determination. In this method de-emulsification was achieved by employing a nonionic surface-active agent in the presence of an alkaline salt. A different de-emul- sification technique was developed by Pont (1955) employ- ing normal butyl alcohol and sodium citrate. Fat obtained in this manner gave peroxide values equivalent to the values obtained in the method suggested by Chapman and McKay (1949). A colorimetric test for peroxides was described by Téufel (1957). The sample is spotted on paper and is treated with a ferrous ammonium thiocyanate reagent to develop the color. Thiobarbituric Acid (TBA) Test The TBA test has been applied in the estimation of the extent of lipid oxidation in meats, fats, oils, fish, cereals, and dairy products. The pigment produced in the reaction is the condensation product of two molecules of TBA with one of malonic dialdehyde. Patton 33 31. (1951) adapted the TBA test to the detection of lipid oxidation in milk fat. The extraction procedure suggested by Dunkley33 31. (1951) involved a mix- ture of pyridine and isoamyl alcohol to separate the color com- plex from fluid milk, thereby allowing quantitative deter- mination of the reaction products. The development of 28 yellow color due to various interferring substances such a as lactose can be removed by column chromatography (Yu and Sinnhuber, 1962). Miscellaneous Tests Manometric respirometers have been adapted to measure the oxidation of lipids. The amount of oxygen ab- sorbed is measured by a decrease in pressure, and converted to micromoles of oxygen uptake (Thomas, 1954; Lundberg, 1961). The measurement of carbonyl substances, one of secondary degradation products in fat oxidation, can be measured colorimetrically by the 2,4-dinitr0pheny1hydrazine procedure (Henick 33 31., 1954). The separation and iden- tification of carbonyl compounds in dairy products, includ- ing those naturally present and those present as a result of lipid oxidation, have been reviewed by Badings and Wassik (1963) and Day (1965). The classification of peroxides in complex lipid mixtures was accomplished by thin layer chromatography (Oette, 1965). The change in concentration of unsaturated fatty acids can be studied by gas chromatography (Roubal, 1967) or by the decrease in iodine values (Lea, 1956). 29 Methods to Evaluate the Ef- fectiveness of Antioxidants The effectiveness of an antioxidant can be measured by its ability to prolong the period of useful shelf life, i.e. before the food becomes rancid or unfit for use. At present, the best method to evaluate the ef- fectiveness of an antioxidant is long term storage studies under various packaging, temperature, and humidity condi- tions. The stored food is evaluated organoleptically and chemically at regular intervals. Various techniques have been devised to accelerate the process of evaluation. The modified Schaal oven method (Eastman Kodak Co., 1963) employs high temperature storage and daily organoleptic evaluation. The aeration of the samples at 210F and periodic measurement of the increase in peroxide values is the basis of the active oxygen method (AOM) (Reimenschenider 33_31., 1943). The uptake of oxygen as a function of oxidation is measured in a method suggested by Stucky 33 31. (1958). The method is a modification of the ASTM bomb calorimeter method. Thompson 33 31. (1950) and Bickoff 33 31. (1955) suggested the measurement of change in carotene levels as a function of lipid oxidation. The antioxidant present, if effective, will preserve the level of carotene and residual carotene in the control is then compared to the sample containing antioxidant. 30 Marco (1968) described a method where the coupled oxidation of linoleic acid with B-carotene in the presence or absence of antioxidants as a function of the B-carotene loss was measured in a model system. Radanovics and Stine (1969) noted the inhibition of peroxide formation in the presence of various antioxi- dants, in sour cream samples irradiated with short wave- length ultra violet light. The Flavor of Cultured Sour Cream The flavor of cultured sour cream is a complex or- ganoleptic sensation. Besides tactile attributes, which will influence flavor acceptance, the flavor of the sour cream is also influenced by the raw material, the proces- sing methods and the flavor developed during the fermenta- tion process. ‘ Cultured cream can be described as a product pos- sessing a clean, sharp acid flavor, a delicate diacetyl aroma, and a firm, moderately heavy body with a smooth texture. The potential flavors attributable to milk fat are among the most important in dairy products. In his review, Kinsella 33 31. (l967)discussed the flavor of milk fat and its biochemical origin. The various methyl ketones, ali— phatic lactones and flavors developed during the oxidation of milk fat are all part of the flavor spectrum. 31 While the original flavor attributes of the raw material are important in the manufacture of a high quality sour cream, the typical flavor profile would be a result of the compounds produced in the fermentation process. The development of acidity and flavor can be at- tributed to the following species of bacteria (Hammer and Babel, 1943; Babel and Hammer, 1944; Riel and Gibson, 1961): Streptococcus lactis and Streptococcus cremoris, the primary lactic acid producers, Leuconostoc citrovorum and Leuconostoc dextranicum, responsible for the develop- ment of flavor, and Streptococcus diacetilactis, producing lactic acid and various aroma compounds. The work of van Niel 33 31. (1929) demonstrated that diacetyl is the major flavor compound in cultured cream butter. Since then several workers have reconfirmed the presence of diacetyl in cultured products, and they discussed its importance. According to these workers the optimum level of diacetyl is between 2-5 ppm (Prill and Hammer, 1939; Babel and Hammer, 1944; Krishnawamy, 1951). Bennet 33_31. (1965) reported that acetaldehyde and acetic acid lowered the flavor threshold of diacetyl in cultured dairy products. The importance volatile acids play in the flavor profile of cultured products was shown by Prill and Hammer (1939), and Hammer and Babel (1943). The major acid in the volatile fraction was found to be acetic acid (Hammer 33_31,, 32 1923). Chou (1962) and Chou and Harper (1963) established the presence of butyric, valeric and propionic acids in cultured buttermilk. The carbon dioxide produced by the bacteria could contribute to the desirable flavor of cultured dairy prod- ucts (Hammer and Babel, 1943; Schock, 1947; Andersen, 1961). The loss of CO2 in cultured buttermilk and sour cream re- duced the original pleasant aroma, even when the diacetyl and acetoin levels remained the same (Anderson, 1961). Day 33 31. (1962) reported the following compounds to be present in ripened cream: ethanal, propanal, n- butanal, 2-methyl butanal, n-pentanal, acetone, 2-butanone, diacetyl, acetoin, ethanol, n-butanol, ethyl acetate, methyl sulfide and acetic acid. More recently Lindsay (1965) and Lindsay 33 31. (1965) confirmed the presence of these compounds using gas chromatography combined with a Fast Scan mass spectrometer. In addition Lindsay (1965) also identified the following compounds in cultured butter: ethyl formate, methyl acetate, methyl butyrate, methane, methyl chloride, methanol, formic acid, 2-furfura1, 2- furfurol. Day 33 31, (1964) concluded that dimethyl sulfide at the ppb level plays an important role in flavors of cultured butter. This compound also served to smooth the harsh flavor of synthetically flavored products. 33 The "green" flavor often observed in cultured products can be attributed to acetaldehyde (Harvey, 1960; Harvey 33 31., 1961; Badings and Galesloot, 1962). The Loss of Flavor Constitu— ents in Dehydrated Foods Inevitable loss of flavor volatiles occures in the process of dehydrating foods (Bradley, 1964; Desai, 1966; Boudreau 33 31., 1966; Malkki and Veldstra, 1967; Reinec- cius, 1967; Saravacos and Moyer, 1968). The retention of volatile constituents, however, is relatively higher than expected when the volatility is extrapolated relative to the loss of water. This would suggest that physical forces such as diffusion, ad- sorption and selective permeability must play a role. The adsorption of volatile organic compounds can be studied from their adsorption isotherms (Issenberg 33 31., 1968, 1969; Boskovic and Issenberg, 1968). Issenberg 33 31. (1968) using a frontal analysis gas chromatographic appara- tus, studied the vapor-solid adsorption of hexane and ace- tone on microcrystalline cellulose. The partial pressures of both compounds were maintained at the l millitorr range. The resulting isotherms indicate the interaction existing between the adsorbent and the various adsorbates. Later similar experiments were conducted by Boskovic 33_31. (1968) using aliphatic straight chain alcohols, and by Issenberg 33 31, (1969) employing the acetates of aliphatic straight 34 chain alcohols. The results obtained from these experi- ments prompted the workers to suggest the possible energy interaction existing between the food and volatile com- ponents. It was further suggested that these measurements would aid in the understanding of flavor losses occurring due to dehydration and storage of foods. The formation of a membrane selectively retaining volatiles but not water in a model system was demonstrated by Menting 33_31, (1967 a,b). The suspended drop of malto- dextrin acetone mixture was dried, and the loss of acetone was measured. During the initial drying phase the loss of acetone was high. The formation of a membrane at the out- side of the drOplet resulted in minimal losses from there on. Patents utilizing the principle of membrane formation in the dehydration of foods have been issued (Olsen and Seltzer, 1941; De Gruyter, 1965). The retention of aroma in the spray drying of con- centrated solutions can also be explained, by assuming the formation of a film, acting as a selective membrane, around the drOplet (Brooks, 1965). The retention of flavor is in direct prOportion to the concentration of solids (Rey and Bastien, 1962; Sivetz and Foote, 1963; Menting and Hoogstad, 1967 a,b). The re- tention of acetone increased with the increase of glucose concentration in the freeze drying of glucose-acetone mix- tures. At 25% glucose concentration, 45% of the acetone 35 was retained, while losses were much higher at 5% glucose concentration (Rey and Bastien, 1962). Similarly, Sivetz and Foote (1963) noted that the aroma losses are low in the spray drying of coffee concentrates at 45% total solids content, but 90% loss can occur at lower solids concentra- tion. In the vacuum drying of fruit juices Saravacos and Moyer (1968) observed that the retention of volatile flavor compounds can be improved if concentrated juices are used. The loss of ethyl acetate and ethyl butyrate at 30% water evaporation was 90% but at 100% water evaporation the fur- ther loss of these compounds was minimal. Freeze drying experiments show that the loss of flavor volatiles at the initial stages of freeze drying is also high but overall, freeze drying resulted in retention of a much higher level of volatiles than vacuum drying (Saravacos, 1967b). McBean .2E.El° (1964) observed that drying conditions which in- creased the role of water removal increased the sulfur di- oxide (802) retention of fruits. The SO2 moved freely across the fruit tissue due to gaseous diffusion or because of solute concentration gradients. The retention of SO2 was less in the smaller fruits, possibly because of the larger total absorbing area. The loss of diacetyl in the vacuum concentration of skim milk-diacetyl mixture was approximately 90% at 15% total solids concentration, but further losses were minimal 36 during the continuation of the condensing process (Reineccius, 1967). The critical moisture content of a malto—dextrin model system was found to be 9%. At moisture content low- er than 9% no adsorption of acetone occurred, while at higher moisture levels the uptake of acetone increased with the increase of moisture (Menting and Hoogstad, 1967 a,b). Similarly, the retained volatiles in a freeze dried model system were not removable, unless the level of mois- ture was increased above the critical moisture level (Flink and Karel, 1969). The rate of freeze drying is influenced by a vari- ety of factors and essentially is controlled by the speed with which the water can be removed from the system. Such rates would also influence the removal of flavor volatiles. The freeze drying of various gels varied from 4.3 hours for cellulose gum to 11.5 hours for egg albumen under Iidentical conditions (Saravacos, 1965). Saravacos further demonstrated that the drying time decreases linearly as the temperature increases, but it increases linearly with the increase in layer thickness. The rate of freeze drying seems to be governed by the permeability of the samples. The slowly frozen sam- ples are more permeable due to the formation of large ice crystals than samples frozen rapidly, where small ice crys- tals are formed. The formation of larger ice crystals 37 formed in slow freezing will accelerate the removal of water vapor by the formation of micro channels in the freeze dried food (Quast and Karel, 1968). The thermal conductivity of the sample during freeze drying would de— pend on the type and size of pores (Saravacos, 1965). The boiling point of the flavor compound would be in direct relation to the losses occurring. The lower the boiling point, the higher will be the expected loss (Boudreau st 31., 1966; Katayama e£_al., 1966; Bradley and Stine, 1967; Saravacos and Moyer, 1968). The gas chromatographic analysis of spray dried cheese slurries indicated that the majority of low boiling compounds are lost during processing (Bradley, 1964; Bradley and Stine, 1967). Similarly the cooking or freeze drying of apples resulted in the loss of low boiling flavor compounds (Katayama 33 al., 1966). The research of Sara- vacos and Moyer (1968) shows that the volatility of aroma compounds was directly related to the vapor-liquid equilib- ria in the aqueous solution during the vacuum drying of fruit juices. The loss of butyric, caproic, caprylic, capric and lauric acids during spray drying of butter was directly related to the boiling points of these compounds. The loss of butyric acid was 69.4% while only 20.4% lauric acid was lost under identical conditions (Boudreau et 31., 1966). 38 The method of dehydration employed would be a major factor in the extent to which flavor volatiles are lost (Bradley, 1964; Desai, 1966; Reineccius, 1967). Reineccius (1967) compared the various levels of retention of diacetyl and acetoin in skim milk systems, employing different methods of dehydration. Retention was negligible in roller drying or tray drying. Freeze drying resulted in retention of 63% of the volatile constituents, while the spray dried powder retained 70-56% of the ini- tially added volatiles, depending on the drying conditions. The loss of volatile acidity, diacetyl, and di- acetyl + acetoin was higher in foam spray dried sour cream than its freeze dried counterpart (Desai, 1966). The ob- servations of Bradley and Stine (1967) indicate that foam spray drying is more suitable in retaining the naturally present flavor volatiles, especially the low boiling flavor compounds, when compared to conventional spray drying of cheddar cheese slurries. This increased retention was at— tributed to the layer particle size. Similarly Boudreau 35 El' (1966) concluded that larger particle sizes result in better flavor retention in the spray drying of butter. The increase of wall thickness can be directly correlated in improved flavor retention in the manufacture of instant coffee (Sivetz and Foote, 1963). METHODS AND METHODOLOGY Preparation of the Samples Fresh raw milk was heated to 100F, separated and the cream standardized with skim milk to 20 i 0.3% milkfat. The standardized cream was pasteurized at 165F for 30 minutes, double homogenized at 1500 psi and cooled to 40F. Preparation of Samples for Flavor Studies Six hundred gram portions of cream were weighed into stainless steel beakers. ,Under constant agitation 50% lactic acid was added dropwise to the cold cream until a titratable acidity of 0.75% (as lactic acid) was reached. Without interrupting the stirring the preselected and cal- culated flavor component dissolved in water was added to the acidified cream. In order to achieve constant condi- tions of volume and weight, the total volume of all addi- tives was maintained at 27 ml, using distilled water to make to mark. Five hundred grams of the prepared acidified cream was poured into alumiAum pans measuring 11 1/4 x 7 1/2 x 1 1/2 inches. Thermocouples (iron-constantan) were placed on the bottom of the tray before the addition of the 39 40 samples, and at the surface of the liquid acidified cream. The trays were covered with a plastic film and placed in a freezer maintained at 0F. Appropriate quantities of liquid samples were taken and stored at 40F to be analyzed later. After completion of freezing the samples were placed in the freeze dryer on platen number 4 and cooled to -50F. The samples were freeze dried in a Virtis RePP #FFD 42 WS freeze dryer (Figure 1), equipped with automat- ic controls for condenser and platen temperatures, vacuum adjustment, weight system with a recorder for determination of drying curves, constant recording of absolute pressure and thermocouple connections. The capacity of the freeze dryer is 50 lbs of water removal per run. The adjustable temperature range for the platen was ambient to 250F. Eight thermocouple connections for temperature measurements were installed inside the vacuum chamber. The thermocouples were connected to a Honeywell Electronic 15 Multipoint Strip Chart Recorder (Figure l). The temperature of the platen and the temperature of the product at the bottom and at the surface were con- stantly measured. .The process of freeze drying was ter- minated when the temperature measured at the inside bottom of the tray was within 10F of the platen temperature. The 42 freeze dried powder was screened, sized and stored at 40F for analysis. Preparation of Samples for Storage Studies Antioxidant Studies Various antioxidants, dissolved in propylene glycol, were obtained from Eastman Kodak Company, Kingsport, Ten- nessee. They were proprietary blends of antioxidants and synergists approved for use in foods. The level and type of antioxidants used throughout the experiments are shown in Table l. The premeasured antioxidants were dispersed in the fresh pasteurized cream by recirculating in the holding tank of the homogenizer. The cream was then homogenized at pasteurization temperature and cooled to 72F. A commercial frozen starter culture obtained from Dairytechnics, Sarasota, Florida which had been selected for optimum acid production and flavor characteristics was added at a concentration of 0.5%(v/v). The samples were incubated at 72F until a titratable acidity of approximately 0.7% expressed as lactic acid was attained. The sour cream was then held at 40F for 48 hours to develop the flavor and body of a typical sour cream. The cold sour cream was thoroughly mixed and spread in a 3/4 inch thick layer onto stainless steel trays meas- . uring 18 l/4 x 23 3/4 inches. Thermocouples measuring the 43 Table 1.--Antioxidants and synergists used in freeze-dried sour cream studies Sample Number Level of Antioxidants and Synergists Used*(w/w) 10 ll 12 13 None BHA (0.01%) PG (0.004%) CA (0.003%) BHA (0.01%) BHT (0.01%) PG (0.008%) CA (0.008%) BHA (0.02%) CA (0.02%) None PG (0.01%) CA (0.005%) BHA (0.04%) CA (0.04%) BHA (0.02%) CA (0.004%) BHA (0.01%) BHT (0.01%) None PG (0.02%) CA (0.01%) BHA (0.01%) BHT (0.01%) PG (0.04%) CA (0.02%) *BHA Butylated Hydroxyanisole BHT Butylated Hydroxytoluene PG Propyl Gallate CA Citric Acid temperature of the product were inserted at the inside bottom of the trays. to -50F, The trays were placed into the freeze dryer, frozen and freeze dried at lOOF platen temperature and at 44 an absolute pressure of 5 microns of mercury. The freeze drying was discontinued when the product temperature was within 10F of the platen temperature. The freeze dried sour cream was removed from the trays, sized to pass through a 0.8. Standard #8 sieve (openings 2.38 mm) and placed in storage in brown glass jars at 40, 72, and 99F. The Use of Selected Gases to Break the Vacuum in the Freeze Dryer The above procedure was used for preparing samples, omitting the addition of antioxidants. After completion of the freeze drying cycle a pure gas rather than air was introduced into the vacuum chamber until the inside pres- sure reached equilibrium with the outside atmosphere. The samples were kept in the chamber for 30 minutes before sizing, packaging and storing. Analytical Techniques Moisture The percentage moisture in the samples used in storage studies was determined with a Cenco Infra-red Moisture Balance. The percentage moisture in the samples used in flavor studies was determined by the method of the Association of Official Agricultural Chemists (AOAC, 1965), established for dairy products. I 45 Fat The official Babcock procedure for liquid cream was used (AOAC, 1965). Titratable Acidity Titratable acidity, expressed as lactic acid, was determined on a 9.0 gram sample by titrating with 0.1 N sodium hydroxide to the phenolphthalein endpoint. Extraction of Flavor Volatiles Appropriate quantities of liquid acidified cream were accurately weighed to contain a final concentration of approximately 2 to 10 ug flavor constituent after dilu- tion. Sufficient tricholoacetic acid (TCA, 25%) added to the sample to bring the final solution to a concentration of 12.5%(w/w) TCA. The precipitated samples were filtered through a Whatman #1 filter paper and the filtrate was stored in glass vials in an ice bath to await analysis. The freeze dried samples were homogenized with water in a Potter-Elvehjam type homogenizer for one minute. The resulting homogenate was handled as described for the liquid samples. Determination of Diacetyl and Acetoin The method of Westerfield (1943) was used with the following modifications: 46 l. The wave-length was chosen to be 505 mu, rather than the recommended 530 mu. This was necessary to minimize the interference by a slight blue color which also developed. 2. Holding time for the samples used in the deter- mination for diacetyl was 10 minutes in a 90F water bath, and 5 minutes at room temperature. For acetoin, a 35 minute hold in a 90F water bath, and 5 minutes at room temperature were employed. 3. Due to the presence of a slight amount of pre- cipitate at the end of incubation the sample was centri- fuged in an Adams angle head centrifuge for 3 minutes at speed 5 to remove the sediment. Recovery tests conducted indicated that this sediment did not interfere with the determination. A standard curve prepared at the wave length of 505 mu was linear within the range of the levels determined. The effect of homogenization as a possible source of flavor loss was experimentally determined through re- covery studies and this effect was shown to be negligible and within the experimental error of the method. The ef- fect of TCA on the development of the color was determined. Since the increase in color was demonstrated also to be a linear function of concentration, the standard curve was established using TCA. 47 Peroxide Value (P.V.) The method of Stine gt a1, (1954) was used with the following modifications: The surface active agent was re- placed with the de-emulsification reagent suggested by Pont (1955). The use of the ice bath was omitted and the samples were placed into a cold water bath for 5 minutes. Flavor Evaluation Each batch of freeze dried sour cream was evaluated concommitant with the chemical analysis, initially and thereafter at monthly intervals. The samples were recon- stituted to contain 27% solids in a Waring blendor, homog- enized in a hand homogenizer by a single pass, cooled to 40F, and organoleptically evaluated. Ultra Violet (U.V.) Exposure A Model R81 Minearalight which generates light at 2537 A was used as the source of UV light. Twenty ml of reconstituted sour cream was poured onto a petri dish. Four samples were placed randomly into an area covered by the ultraviolet light at a distance of 10 inches from the light source. The samples were exposed for a predeter— mined time and appropriate samples removed for determina- tion of peroxide values. The equipment used for the irradiation studies is shown in Figure 2. 48 Fig. 2.--Equipment used for irradiating samples with Ultra Violet light RESULTS AND DISCUSSION The development of a completely acceptable dehy- drated sour cream is a complex task. Desai (1966) showed that freeze drying as a dehydration process for sour cream was superior to spray drying. The Effect of Processing Variables on the Loss of Diacetyl and Acetoin in Acidifiéd, Freeze Dried Cream A model system was chosen in preference to natural, cultured sour cream because such a system is uniform for all experiments, the concentrated fladbr compounds can be more precisely controlled, the presence of interfering metabolic byproducts is eliminated and handling and pre- paration procedures are more readily standardized. The samples were analyzed before and after freeze drying for moisture and levels of flavor volatiles and all results are expressed on a dry weight basis. Sufficient numbers of experiments were conducted to be able to make a conclusive evaluation of the condition tested. The standard error of the mean indicates that a close correlation exists between the various runs. 49 50 The Effect of Absolute Pres- sure in the Freeze Drying Process on Losses of Vola- tiles The pressure in the vacuum chamber of the freeze dryer was maintained :5% of the predetermined level. Platen temperature was maintained at 100 F (:3F) in all experiments unless indicated otherwise. The temperature profile of the product showed no significant deviation from results obtained under optimum Operating conditions at an absolute pressure of 4p. The results shown in Tables 2 and 3 indicate that absolute pressure variations in the range 20-350u had no effect on the loss of diacetyl or acetoin. The loss of di- acetyl and acetoin shows no conclusive trend and the mean value for the experiments are 68.7 i 2'2 and 73.2 i 5.1% respectively. The loss of acetoin seems to be somewhat higher than diacetyl, but due to the high mean values no significant difference could be established. The retention of diacetyl and acetoin is significantly higher then the minimum attainable pressure of 40 was achieved in the vacuum chamber. The losses of diacetyl and acetoin under these conditions were 43.5% and 56.1% respectively (data pre- sented in Table 2 and 3 are transposed from Table 5 and 6). Such results would tend to indicate that there is a change in the vapor-pressure equilibrium as the absolute 51 pressure is raised from 4“ to 20“ but that further increase of absolute pressure to 3500 would not effect further such vapor-pressure equilibrium. Similarly the results of other workers indicate that at relatively high vacuum the varia- tion in the chamber pressure did not affect significantly either the freeze drying time or the final product quality (Saravacos, 1965; Lusk, gt_§1,, 1965). It is interesting to note that the losses observed for both diacetyl and acetoin at absolute pressures of 20-3500 are approximately the same as losses observed at 200 F platen temperature (Tables 4, 5) and the time of freeze drying was also reduced. The condenser temperature is normally adjusted to -62F to minimize the vapor pressure of water in the cham- ber.’ The readjustment of temperature to -40F resulted in increased flavor losses at 100F platen temperature (Table 4). The higher flavor loss was assumed to be a func- tion of the change of water vapor pressure in the vacuum chamber. The absolute pressure was approximately 150 at -40F condenser temperature compared to 4H at -62F. The effect of increased absolute pressure was shown to result in a higher level of flavor losses (Tables 2, 3) and these results correlate well with the previous findings. It can Table 2.-—The effect of vacuum on the loss of diacetyl 52 freeze—dried acidified cream at 100F platen ' r‘, temperature Vacuum Sample Diacetyl Level* (ppm) Diacetyl (u) Number Before / After Loss Freeze Drying ppm % 4** l --- --~ --- 48.5 20' 2 41.0 12.3 28.7 70.0 3 41.0 12.3 28.7 70.0 4 38.0 12.8 25.2 66.3 5 38.0 13.3 24.7 65.0 50 6 42.5 12.5 30.0 70.6 7 42.5 12.5 30.0 70.6 350 8 36.2 12.0 24.2 66.9 9 36.2 10.5 25.7 71.0 10 36.5 12.0 24.5 67.1 11 36.5 11.0 25.5 69.6 Mean: 68.7 O: 2.2 *Expressed on dry weight basis (g/100g). **From Table 5. 53 Table 3.--The effect of vacuum on the loss of acetoin in freeze-dried acidified cream at lOOF platen temperature Vacuum Acetoin Leve1* (ppm) Acetoin (U) Before / After Loss Freeze-Drying ppm % 4** ~—— -—- --- 56.1 20 44.0 9.2 34.8 81.5 44.0 8.1 35.9 79.1 41.6 12.5 29.1 69.9 41.6 12.5 29.1 69.9 50 50.1 15.0 35.1 69.8 50.1 11.5 38.6 77.0 350 43.0 12.4 30.6 71.2 43.0 14.6 28.4 66.0 36.0 8.1 27.9 77.5 36.0 10.8 25.2 70.0 Mean: 73.2 i 5.1 **From Table 6. *Expressed on dry weight basis (g/lOOg). 54 Table 4.-—The effect of condenser temperature on the loss of diacetyl at 100F platen temperature Condenser Vacuum Loss of Temperature (u) Diacetyl (F) (%) -62* 4 i 1 48.5 i 3.6 -40 ' 15 i 5 64.4 i 1.7 *From Table 3. be assumed that the capacity of the pump is the limiting factor in handling the excess vapors at higher condenser temperatures, resulting in an effective increase of chamber pressure. The Effect of Platen Tempera- ture on the Loss of Diacetyl and Acetoin The temperature profile of product and the tempera- ture of the platens during freeze drying were measured with the aid of thermocouples (Fig. 3.). These temperature pro- files show close similarity with freeze drying curves ob- served when dehydrating shrimp and salmon steaks (Gold- blith g5 31., 1963 and Lusk g3 31., 1965). The temperature of the platen on which the samples were placed was 20-30F lower than the control platens and mmusbmwmmEmu ceumae msoHum> um Emcwo pmwwapflom pmfiupnmmomuw mo maflmoum musumummfiwhuu.m .mflm 3.85 ms: e223 55 “dual ..- P .‘ 8 '8‘“: g e!!! 3 g. g fiiig 34.2% .._o muflmam E a F misam ...o 20:8 2.. m 5.2.: 3.3% I m zmmmna somMZOU..u f: wmnbnmwazwh w. v. _ . qqiaN-thqWfi E-quumd Own 00.. V 4\ V I(|\ U m c 38 \J l\ 8 8. (:1) BUOLVUBdINBL 56 the desired "set" temperature was reached only by the last half hour of the drying cycle. This lower temperature, measured at the interface of the surface of the platen and sample trays, was attributed to several factors: 1. The constant BTU requirement for the latent heat of sublimation, 2. The cooling effect of the sample, which had a significantly lower temperature averaging about -20F, 3. The inadequate heat transfer from the heating fluid, circulating at the bottom of the platen, to the surface of the platen. The product temperatures, measured at the bottom and surface during the freeze drying process, are shown in Fig. 3. The temperature was constantly monitored and the freeze drying was terminated when the temperature of the sample at the bottom was within 10F of the sample platen temperature. This "endpoint" was used after conducting pre-trial experiments, which established that under these conditions all the ice in the product was sublimed and the moisture content of the food had equilibrated at l-l.5%. (A similar endpoint determination was described by Kan (1962). The surface temperature of the product increased above the mean product temperature at the beginning of the freeze drying and rapidly increased as the freeze drying progressed. The difference is observable at all platen 57 temperatures, but it becomes more significant as higher platen temperatures are used. At 200F and 225F platen temperatures, the mean surface temperature was above 100F. This high temperature could be detrimental to the final product quality (Harper and Tappel, 1957; Lusk g£_gl., 1965; Tuomy E£.E£-v 1968). While the surface temperature was above the mean product temperature, it did not approach the temperature of the platen except during the last part of the freezing cycle. It is assumed that the subliming water vapors dif- fusing through the already freeze dried portion of the food exerted a cooling effect on the dried food. The mean product temperature increased as the platen temperature was elevated; however, it never reached the eutectic temperature of the product. The time of freeze drying appears to be a loga- rithmic function of platen temperature (Table 5). While an average of 20 hours was required to freeze dry at 75F the drying time was reduced to 6.3 hours at 225E platen temperature. The reduction of freeze-drying time as the platen temperature is increased is well substantiated in the literature (Zamzow 22.2l3' 1952; Harper and Tappel, 1956; Goldblith gg,gl., 1963; Lusk g£_gl,, 1965: Saravacos and Meyer, 1968). Based on such findings, the rate of freeze drying can be predicted and controlled, depending on the platen temperatures used. The freeze drying 58 Table 5.--The effect of platen temperature on the loss of diacetyl in freeze-dried acidified cream Platen Drying Diacetyl Level* (ppm) Diacetyl Tempera- Sample Time Before / After Loss ture Number (hours) Freeze—Drying ppm % (F) 75 1 20.25 41.9 24.2 17.7 42.2 2 20.25 41.9 25.7 16.2 38.7 3 22.0 42.6 23.8 19.1 44.8 4 22.0 39.6 21.5 18.4 46.1 5 19.0 40.7 23.8 16.9 41.5 6 19.0 40.7 21.4 19.3 47.4 Mean: 43.5 i 3.2 100 1 15.5 41.4 22.1 19.3 46.6 2 15.5 42.4 23.1 19.3 45.5 3 16.5 41.8 20.1 21.7 51.9 4 16.5 41.8 20.5 21.3 51.0 5 15.5 44.0 22.7 21.7 49.3 6 15.5 44.0 23.4 20.6 46.8 Mean: 48.5 i 3.6 150 1 10.25 40.9 16.9 24.0 58.7 2 10.25 40.9 15.4 25.5 62.3 3 10.50 42.6 17.3 25.3 59.4 4 10.50 42.6 18.4 24.2 56.8 5 10.50 43.1 16.2 26.9 62.4 6 43.1 18.7 24.4 56.6 Mean: 59.4 i 2.5 Table 5.--Continued 59 Platen Drying Diacetyl Leve1* (ppm) Diacetyl Tempera- Sample Time Before / After Loss ture Number (hours) Freeze—Drying ppm % (F) 200 l 8.25 43.6 14.4 29.2 67.0 2 8.25 43.6 13.6 30.0 68.8 3 7.75 42.2 13.0 29.2 69.2 4 7.75 42.2 11.3 30.9 73.2 5 8.00 42.6 14.4 28.2 66.2 6 8.00 42.6 15.0 27.6 64.8 Mean: 68.2 i 2.9 225 1 6.50 43.3 9.2 34.1 78.8 2 6.50 43.3 8.5 34.8 80.4 3 6.25 43.3 10.2 33.1 76.4 4 6.25 43.3 8.4 34.9 80.6 Mean: 79.1 i 1.9 *Expressed on dry weight basis (g/lOOg). 60 conditions however should be chosen to achieve the most ef- ficient utilization of the freeze dryer without adversely affecting the product quality. The change in the loss of diacetyl and acetoin as a function of platen temperature is shown in Tables 5 and 6. The loss of diacetyl increases as higher platen tem- peratures are employed (Table 5). While only 43.5% loss is observed at 75F platen temperature, the loss is almost twofold (79.1%) at 225F. The loss of diacetyl appears to be a linear function of platen temperature between the temperature range of 75-225F. This straight line rela- tionship would allow extrapolation of losses at tempera- tures not tested. Elevated platen temperatures also resulted in in- creased loss of acetoin (Table 6). The losses after freeze drying were between the range of 32.6-80.8% for platen temperatures of 75-225F respectively. The loss is a linear function of temperature between the temperature range of 100-225F but it is significantly less at 75F. The overall loss of acetoin appears to be higher than diacetyl. This is contrary to what would be expected if the losses were a function of the boiling point alone. The loss of fatty acids belonging to a homologous series was shown to be a function of boiling point in the Spray drying of butter (Boudreau E£.El°v 1966). The ad- sorption of acetone was 7.5 times that of hexane in 61 Table 6.--The effect of platen temperature on the loss of acetoin in freeze-dried acidified cream Platen Drying Acetoin Leve1* (ppm) Acetoin Tempera- Sample Time Before / After Loss ture Number (hours) Freeze-Drying ppm % (F) 75 1 20.25 45.0 29.8 15.2 33.8 2 20.25 45.0 33.0 12.0 26.3 3 22.0 46.5 28.9 17.6 37.8 4 19.0 46.5 31.4 15.1 32.5 32.6 i 4.76 100 1 15.5 39.7 14.9 24.8 58.5 2 15.5 39.7 17.4 22.3 56.2 3 16.5 50.6 21.9 28.7 56.3 4 16.5 50.6 23.7 26.9 53.2 56.1 i 2.17 150 l 10.25 49.5 16.1 33.4 67.5 ' 2 10.25 49.5 19.7 29.8 70.2 3 10.50 49.1 18.7 30.4 61.9 4 10.50 49.1 16.2 32.9 67.0 66.7 i 3.46 225 1 6.50 48.9 10.0 38.9 79.6 2 6.50 48.9 11.0 37.9 77.6 3 6.25 38.8 6.0 32.8 84.5 4 6.25 38.8 7.2 31.6 81.5 80.8 i 2.94 *Expressed on dry weight basis (g/lOOg). 62 isotherm studies conducted with micro crystalline ce11u~ lose. This difference in adsorption could be attributed to higher affinity of the acetone to micro crystalline cellulose (Issenberg gg g1,, 1968). The volatiles lost during the vacuum drying of fruit juices were not related to their vapor pressures 'or relative volatility (Saravacos and Moyer, 1968). Losses of flavor volatiles of various fruit juices in- creased as the processing temperature was raised (Malkki and Veldstra, 1967; Saravacos and Meyer, 1968). The losses appeared to be the highest at the initial stage of evapora- tion. Malkki and Veldstra (1967) observed that this is utilized in the flavor stripping of the most volatile fla- vor components of fruit juices under either high tempera- ture or vacuum before initiating the main concentration step. Several workers have demonstrated that the loss of flavor volatiles is a function of solids concentration (Rey, 1962; Sivetz g 511., 1963; Menting gt 33., l967a,b; Reineccius, 1967). The retention of flavor volatiles during the de- hydrating process at present is attributed to two forces: selective permeability and adsorption. Menting g£_g1. (1967) demonstrated that the reten- tion of acetoin increased after the formation of a carbo- hydrate membrane when air drying various carbohydrate solutions. Bradley and Stine (1968) and Sivetz and Foote 63 (1963) observed higher levels of volatile retention as particle size increased in the spray drying of cheese and coffee. This higher retention was attributed to the vola- tiles trapped in the particles. In recent experiments Issenberg et a1. (1968) demonstrated that adsorption might also play an important role. It was also demonstrated during these experiments that the affinity of the volatile compound for the system in which it is present would deter- mine its ability to adsorb. The results obtained on the losses of diacetyl and acetoin in the freeze dried acidified cream as a function of platen temperature and absolute pressure indicated that the retention is probably due to adsorption. During the freeze drying process there is no experimental evidence to support the theory of membrane formation. The water and other volatile substances will pass through the vari- ous layers by hydrodynamic flow through the porous media and the driving force is a total pressure gradient between the ice interface and the surface (Harper and Tappel, 1957; Kramers, 1958). Rapid freezing has been shown to result in increased drying rates (Mackenzie gE_g1:, 1963; Quast, 1968). The increase in rate was attributed to the smaller ice crystals formed, permiting the formation of "micro channels" for the rapid diffusion of water vapor (Mackenzie g5 g1., 1963). The freeze dried acidified cream appeared to be of 64 porous structure. Microscopic examination of the sub- cellular particles showed, however, that they are amor— phous crystalline in appearance, without any observable distinctive particles (Fig. 4). Fig. 4.--Freeze-dried acidified cream at X130 magnifi— cation In view of such findings, and the apparent dis— agreement in the literature on the retention of volatile substances in dehydrated foods, further research was ini- tiated to establish if adsorption is responsible for the retention of flavor volatiles during the freeze drying process. The influence of layer thickness on the relative losses of diacetyl and acetoin was investigated at lOOF 65 platen temperature and 4“ absolute pressure. The results of the experiments are as follows: Table 7.—-The effect of layer thickness on the loss of diacetyl and acetoin at lOOF platen tempera- ture Sample Number Drying Loss of Loss of Weight of time Diacetyl Acetoin (9) runs (hours) (%) (%) 500* 6 16.0 48.5 56.1 250 2 8.25 46.5 88.1 *From Tables 5 and 6 The reduction of volume from 500g to 2509 did not result in any significant decrease or increase in the loss of diacetyl or acetoin. This result indicates that the volatiles are probably adsorbed as the dehydration progresses, and the exposure of the already freeze dried layer to high vacuum and temperature would result in only minimal additional losses. Reducing the volume of the material to be dried by one half resulted in a 50% decrease of drying time. 66 The direct relationship of layer thickness to drying time is well known. The effect of reduced layer thickness was found to be a linear function of time in the freeze drying of gelatin and starch gels (Saravacos, 1965), and beef slices (Harper and Tappel, 1957). However, the same workers also showed that the drying rate would be different depending on the type of product freeze dried. Under identical conditions the time required to freeze dry various gels varied from 4.3 hours for cellulose gum to 11.5 hours for egg albumen (Saravacos, 1965). The effect of extended freeze-drying on the loss of diacetyl is shown in Fig. 5. The loss of flavor is greatest during the normal drying cycle before a moisture content of 1-l.5% is reached. This part of the curve is extrapolated on the basis of analysis on the initial and freeze dried samples. Before the completion of freeze- drying, the sample consists of a freeze—dried layer where flavor losses are measurable and a layer of ice where no flavor loss is expected. The precise analysis of such a complex system therefore would be meaningless. After completion of the freeze drying cycle further losses are observed at both 100 and 200F platen tempera- tures. The slope of the curve indicates however that such losses are of lesser magnitude than in the initial stage of freeze drying. It is interesting to note that high 67 mmusumnmmsou cmumHm ucmuwMMHp um HmumomHU.wo mmoH we» so mcflmuoumummum pmpcmuxm mo pomwmm m£Bn|.m .mflm $.35 m2... 02:8 ¢m11 _ mm mm 1 NW _ MWi _ m? _ w _ W. _ ADO. .. 1.. moo. 083.128 1 ezsmo mNmmE / I 8 853.128 / J .. 0255 HUME / / L // / 1 / / 49 - // / L // / .. / / is / / T / /1 / Pl _ _ e (%) $501 1113:»: 10 68 vacuum (4U) and 200F platen temperature for 12 hours failed to remove all the diacetyl adsorbed onto the acidified cream. Such results tend to indicate that certain bond- ings, such as adsorption, must play an important role in retention of these volatiles. Issenberg gE_g1, (1968) also suggested that there might only be a small number of active sites available for the adsorption of low levels of volatile compounds usually found in foods and that at high concentrations the relative level of adsorption of these volatiles could be at a dif- ferent rate. To determine if the diacetyl level would have any influence on the retention of flavor volatiles, 5-100 ppm diacetyl was added to model systems prior to freeze drying. The results of these experiments are shown in Table 8. The loss of diacetyl appears to be a function of the concentration added to the acidified cream. In two experiments where 5, 10, 25, 50 and 100 ppm of diacetyl were added to individual samples the absolute loss of di- acetyl increased, with increase in concentration but the relative loss remained the same. The loss of diacetyl in these ten experiments was 48.0 :_2.0%. The results ob- tained in a separate series of experiments (Table 5) where 10 ppm of diacetyl was added the relative loss was 48.5 i 3.6%. This excellent correlation further substantiates that between the range of 5-100 ppm the relative flavor 69 Table 8.-—The effect of different levels of diacetyl on the loss of diacetyl in freeze-dried acidified cream at 100F platen temperature Diacetyl Added Sample Diacetyl Leve1* (ppm) Diacetyl to Liquid Number Before / After Loss Cream (ppm) Freeze-Drying (%) 5 1 21.1 10.3 48.9 10 2 41.4 22.1 46.6 25 3 87.4 42.4 51.5 50 4 178.6 96.3 46.1 100 5 368.7 190.8 48.3 5 6 21.3 9.8 50.7 10 7 42.4 23.1 45.5 25 8 107.0 54.3 49.3 50 9 182.7 97.3 46.7 100 10 363.9 194.0 46.7 Mean: -28T0~’ : 2.0 *Expressed on dry weight basis (g/lOOg). 70 losses are independent of the concentration of volatile substances. The existing active sites therefore either were not saturated or such active sites are not the sole basis of adsorption. Based on the results obtained, the desired level of volatile flavors can be adjusted in the final product by appropriately increasing the initial concentration of such materials. This increase in the original concentra- tion can be achieved by the proper selection of raw in- gredients, or by the direct addition of natural or chem- ical flavor components. The various flavor components would readily adsorb or desorb as the moisture of the freeze dried product is raised (Menting and Hoogstad, 1967a; Flink and Karel, 1969). Menting and Hoogstad showed that ethyl acetate and acetone were readily adsorbed by maltodextrin powders at' moisture levels of 9% or above but no adsorption was ob- served below 9%. Flink and Karel (1969) observed that as the moisture of the freeze dried model system was raised to the critical moisture range the retained acetone, methyl acetate and a series of alcohols, readily desorbed. These two experiments indicate that flavor vola- tiles could have adsorption-desorption isotherms similar to those of water vapors, which are well established for dehydrated foods. In the experiments conducted in this research, a powder containing 15.6 ppm diacetyl was placed under 71 different storage and humidity conditions for a period of 24 hours. The results observed are as follows: Table 9.—-The effect of 24 hours storage at various tem- peratures and humidities on the loss of diacetyl Storage Conditions Moisture of the Diacety1* Loss powder (%) (ppm) (%) Initial level 1.9 15.5 -- Storage at 40F 1.9 15.6 0 Storage at 75F 1.8 15.6 0 0% humidity, 75F 0.9 15.8 0 52% humidity, 75F 4.6 2.7 82.6 100% humidity, 75F 11.0 2.8 81.3 *Expressed on dry weight basis (g/100g). Control (low moisture) powders stored at room temperature or in refrigerated storage showed no loss of diacetyl. These findings were substantiated in prelimi- nary experiments where ten various batches of freeze dried acidified cream containing various levels of acetoin and diacetyl were stored in the refrigerator for 4 months and periodically analyzed. During this period of storage no substantial flavor loss was observed. An increase of moisture to 4.6% resulted in a di- acetyl loss of 82.6%. Further increases in relative humidity, resulting in a final moisture level of 11%, 72 resulted in similar losses. This indicates that there is a critical moisture level somewhere above 1.9% where de- sorption occurs. It is interesting to note that placing the powder into a 0% humidity chamber resulted in no change in the level of diacetyl; however the moisture content of the powder was further reduced to an equilib- rium moisture content of 0.9% from the initial 1.9%. This indicates that the diacetyl is adsorbed more strongly to the freeze-dried model system than water. The loss of diacetyl due to the increase of mois- ture content, could be somewhat analogous to observations in the gas packaging of milk powders. Gas adsorbed and entrapped inside the particles could not be removed by a single vacuum treatment, and equilibration in the inert atmosphere was required to reduce the 02 level (Lea g£_g1. 1943; Coulter and Jenness, 1945). Similarly, adsorbed volatile compounds appear to reach a vapor pressure equilibrium with the water vapors under constant humidity conditions. In the process of reaching an equilibrium in a constant humidity chamber, the flavor volatile would constantly desorb and adsorb competitively with the water vapors. It seems feasible to calculate or experimentally determine the existence of such equilibrium by establishing various sorption iso- therms for water vapor on volatile constituents under study. 73 Based on the findings of these experiments, and evaluating the data presented by others (Issenberg g3 31., 1968, 1969; Boskovic gE.g1,, 1968), it can be concluded that adsorption rather than the formation of a membrane (Olsen, SE.E£°' 1941; Sivetz and Foote, 1963; Brooks, 1965; De Gruyter, 1965; Menting g2 g1., 1967a,b) is re- sponsible for the retention of flavor volatiles. The findings of these workers, interpreted appropriately, would also support the theory of adsorption. During the initial stage of drying, volatile fla- vor losses are much higher (Rey g3 g1., 1962; Sivetz, gg gl., 1963; Menting gg_gl,, 1967a; Reineccius, 1967; Saravacos, SEHEL°v 1968). At this stage of drying the volatile flavor constituents are concentrated in the water phase, and the molecules of flavor constituents and the solid particles are relatively far apart. The water vapor therefore can literally strip off flavor volatiles, without any possibility of prolonged adsorption (Malkki and Veldstra, 1967). As the drying progresses the liquid be- comes more concentrated, and at this stage greater adsorp- tion of the flavor volatiles can take place. In certain drying methods, such as spray drying, the formation of a membrane is soon observable after ini— tiation of drying (Olsen EE.E£" 1941; Rey, 1962; Menting gg_gl., 1967a,b). Such a membrane would then act as a concentrated layer of solids, where adsorption can take place for both water and flavor volatiles. While the 74 water molecules would desorb as the drying continues, the forces of bonding between the solids and flavor volatiles are much stronger, and therefore losses of flavor vola- tiles would diminish. Once all water is removed from the system, the further loss of flavor volatiles would be governed by the force of bonding, the relative moisture content of the food and the conditions which such food would be exposed to. Storage Stability of Freeze Dried Sour Cream The Effect of Antioxidants on the Shelf Life of Freeze Dried Sour Cream Good quality sour cream was freeze dried, sized, placed in brown glass jars for storage, and evaluated at monthly intervals. Organoleptic evaluations and peroxide values were used as criteria of quality. In the preparation of the samples of freeze dried sour cream all processing conditions were carefully con- trolled to minimize variation in the finished product. The fresh cream was standardized to 20% milkfat, and following processing the fat content was again determined. In this way the fat content of the samples could be main- tained at 20% i 0.3%. Lactic acid, responsible for the pleasant acid taste of sour cream (Hammer and Babel, 1953) was determined, 75 and under the conditions used for culturing varied from 0.68 - 0.73% with a mean value of 0.70%. This level of lactic acid is considered to be normal in fresh sour cream. Drying conditions were established to yield a good quality sour cream with optimum flavor retention and mini- mal drying time. After establishing the optimum condi- tions in pretrial experiments, the frozen sour cream was freeze dried at lOOF platen temperature and an absolute pressure of Sn. The temperature of the sour cream and the tempera- ture of the platens was measured during freeze drying and the drying was terminated when the temperature differen- tial between the sample and platens was 10F. Under such processing conditions, the moisture content of the freeze- dried sour cream was between 1.5 - 2.0%. It has been established that moisture level plays an important part in the storage of dehydrated dairy products. The reduction of moisture below the level necessary to form a monolayer accelerates the oxidation of fats by permitting direct contact with the oxygenof the air, and too high a moisture content initiates other chem- ical reactions such as browning and enzyme activity (Holm, 1927; Aceto SE.E£°' 1965, 1966: Tamsma gg_gl., 1961, 1964). 76 During the desorption experiments conducted in this research (Table 9) it was noted that increasing the moisture in the powder to 11% resulted in the development of strong off flavors, typical of Maillard browning, with- in 24 hours at 72F. According to Patton (1955) the upper limit of moisture for good storage stability for dehydrated nonfat milk, whole milk and whey is 4.0, 2.5, and 5.0% respec- tively. A typical drying curve, and the temperature pro- file of the freeze dried sour cream are presented in Fig. 6). ‘ Since the temperature profile of freeze dried products was shown and discussed previously (Fig. 3), discussion here is limited to other conditions of these experiments (Figs. 3 and 6). As can be seen, the time of drying is considerably longer in the freeze drying of sour cream than in the model system. This longer drying time is attributed to the larger samples freeze dried. The typical load in the freeze dryer was 5 X 8 lbs of sour cream while only 4 X 1 lbs were dried in the model system. The loss of weight during freeze drying appears to be a curvilinear plot up to the point at which drying be- comes essentially completed. A similar drying curve was observed by Goldblith EE,El° (1965) and Harper and Tappel (1957) in the freeze drying of shrimp, salmon and beef. 77 TEMPERATURE (FL 8 )2 8 as o 9* 8 4141.... % mooa um Umflup mummum Ame—SI V m2; 0.. 8 la fimmuo HSOm pmauc neummuw mo 0Hflmoum musumummEmu pew m>uso wcflxueut.o wmnkdmwQZMk zwkjn. wJafidm $323.23 5.2.4.1 .65.on /‘ b??? % (%) SSO'I .LHOIBM ./ .mE 1.1% wdidm “.0 SOS—.00 Pd wmuhdmmaiwk 78 The rate of moisture loss appears to be constant and no sudden changes in the curve are observable. The falling rate of the drying cycle, which is typical of spray dried and air dried products (Van Arsdel and Copley, 1963) is completely absent in freeze drying. The dehydrated sour cream was air packed in brown glass jars, sealed and placed in storage at 40, 72, and 99F. At intervals portions were removed for organoleptic evaluation and for determination of peroxide values. The body of the reconstituted sour cream was weak and grainy. This grainy texture tended to adversely influence flavor scores when the samples were judged. The homogenization of small quantities of samples with a hand homogenizer re- sulted in a smooth body similar to liquid buttermilk and such samples were scored higher for flavor. All samples consequently were homogenized prior to evaluation. The flavor of the freshly freeze dried and reconstituted sour cream was clean and no off flavor was detectable. How- ever, compared to a fresh sample, the reconstituted freeze dried samples were somewhat "flat" indicating that some loss of volatile flavor constituents had occurred. The data'obtained as a result of storage studies are presented in Figs. 7-15 for ease of comparative eval- uation, and in Tables 10, 11 and 12,* (see Appendix) which *Since the data are presented graphically, further reference to these tables will be omitted. The reader may refer to them if verification is required. 79 contain all of the data shown in the graphs. Arrows, placed at appropriate points on the graphs indicated the time of storage at which the particular sample was rated organoleptically "objectionably oxidized." Samples stored at 99F showed detectable off fla- vors in 4-6 weeks (Figures 7, 8, 9) representing three different experiments). Further storage at this tempera- ture resulted in an unacceptable product in 8 weeks. The off flavors were characterized as a combination of lipid oxidation and Maillard browning. These off flavors were detectable in the control as well as the samples contain- ing antioxidants regardless of type and level used. The degree of intensity of oxidized flavor was difficult to assess in the samples stored at 99F due to the pronounced level of other coincidental off flavors. The organoleptic evaluation could not show significant differences between browning and lipid oxidized off flavors. The powder ap- peared to be slightly darker than the 40F samples as the time of stoarge progressed. The change in hydroperoxide formation as a function of storage time is shown in Figures 7, 8 and 9.. The overall obServation of peroxide values indi- cates that at 99F the formation of hydroperoxides is either inhibited or the formed hydroperoxides break down to subse- quent oxidation products at this elevated temperature. The control samples show a significant increase in P.V. as 80 '5') I r 1 T r l I r I l I CONTROL D L4 -- BHA .Oll°/o PG .OO4°/o CA .003 % O _. BHALCHVKpBHTKNIVB PG.OOEW% (3 CA .008°/o "' BHA .02°/o CA .0279 O -1 (.2- .1 F- E Loi— 4 s _ 4 8 0.8— .1 ES 8 i '- lA405 ..J .J =1 T Z _ i cud _( 1 1L 1 1L1 1 ° 1 g (o (a 4 8 MONTHS OF STORAGE,99F Fig. 7.--Peroxide values and detected oxidized flavor (l) of freeze—dried sour cream containing various antioxidants, stored at 99F 81 L5 1 I l f I l I I 1 l 1 L _ l4... CONTROL D - PG .OF%)CA..00596 I) BHA .04°/o CA .O4°/o 0 - BHA .02°/o CA .004°/o . -( BHATINVCIBHTINQC ‘3 LZ- _. ”' ‘1 '5. 1.0L — 3’ \ - ‘l «I O O 0.8 _ q 2 :3 _ a O Q _JIQ§.. -( .-'.. 2 E T OMH- “ 1 -7 %% MONTHS OF STORAGE, 99F Fig. 8.--Peroxide values and detected oxidized flavor (+) of freeze-dried sour cream containing various antioxidants, stored at 99F 82 LG I | I l ' I ' I ' I ' CONTROL U _ PG 02% CA .0l°/o . .1 ‘4 BHA .0I°/. BHT .0I°/. a PG 04% CA.02°/o 0 I2 - "' g... _. .. 6 ad Lo .— 3." \ .4 N I— O . dei- " _>. D _ .4 O E - :1 0.6 - =E _ _ 04- ° 7 0. ,f < 7 l l J___ L . 1 + 1 l l _%_ l _—.8 D '2 4 MONTHS OF STORAGE , 99F Fig. 9.--Peroxide values and detected oxidized flavor (I) of freeze-dried sour cream containing various antioxidants, stored at 99F 83 storage time increases. It is interesting to note that the addition of 0.01% and 0.02% propyl gallate (PG) re- sulted in accelerated hydroperoxide formation, at a much faster rate than the control sample containing no antiox- idant (Figs. 8, 9). Further increases of PG to 0.04% (Fig. 9) resulted in lower peroxide values. The peroxide value of the samples containing antioxidants remains es- sentially the same during the whole duration of storage trials except in the case of samples containing only 0.01% PG, as previously noted. The effect of elevated temperatures on the storage stability of dehydrated dairy products is well documented in the literature (Krienke and Tracy, 1946: Coulter gg.gl., 1951; Patton, 1955). Dry whey containing high levels of lactic acid is subject to browning and the development of ~off flavors much more rapidly when exposed to high tem- peratures or high humidity than is dry whole milk (Doob gg gl., 1942). It was pointed out by Patton (1955) that the Maillard-type browning requires a relatively low order of energy for its initiation, and that it exhibits auto- catalytic qualities after the initiation of the reaction. An excellent review on the browning of foods, including the various chemical reactions involved, is presented by Reynolds (1963, 1965). Samples stored at 72F showed good storage stabili- ty for 3-4 months without the addition of any antioxidants 84 (Figures 10, 11 and 12). The samples containing 0.01% BHA + 0.01% BHT (Figures 11, 12) were of good quality after six months of storage and slightly oxidized but still acceptable after eight months. The addition of 0.04% PG was found to be effective in retarding lipid oxidation (Figure 12), but 0.02% PG was ineffective (Fig. 12) and 0.01% PG showed accelerated autoxidation (Fig. 11). The addition of 0.04% BHA extended storage life to 8 months, while the addition of 0.02% BHT resulted in samples with good stability for periods up to six months (Fig. 11). The trend of the peroxide values correlated well with the organoleptic observations. Generally the control samples had significantly higher peroxide values than samples con— taining the antioxidants. The decrease of P.V. as compared to the control resulted in increased storage stability. Occasionally, at low antioxidant levels, benefi- cial effects were noted in one sample but were lacking in a duplicate experiment. The addition of 0.02% BHA with 0.02% CA resulted in good storage stability in one experi- ment (Figure 10) but in the duplicate run (Figure 11) 0.04% BHA was required to achieve the same storage sta- bility. These variations are not unusual in a complex biological system such as this which contains natural an- tioxidants like tocopherol and carotene (Dunkley 2E.El°' 1967), trace metal contaminants (Greenbank, 1948) and vary- ing levels of "activated" sulfhydral groups, due to the I I 1 I l I l I ] I T I | OONTROL A C] _‘ ' BHA .Ol'o/o PG .OO4°/o CA .00370 O BHA .OII°/o BHTDII ‘79 PG .OOB°/o A CA .ooeolo —( BHA .0270 CA 02%: . l2 - .. P 1‘5 Io)— .. a; .. _I ca<18L- . _. 2 D - _ 8 j 06$- —I ..-.-.' 2 _. .. Q4... - .. 02 "' ' ' 1’? '7 - .Ao’Or ‘ . o J o 1 i 1 ‘i 1 1% AL i? I ~—i5__J———1Q MONTHS OF) STORAGE . 72F Fig. 10.--Peroxide values and detected oxidized flavor (I) of freeze-dried sour cream containing various antioxidants, stored at 72F 86 |.6 fl I T l 1 I I I I I I (4.. CONTROL 0 _ PG 01% CA 005% t BHA 04% CA 04% 0 t BHA 02% CA 0047.0 _ BHA .0I% BHT .0I% A I.2 b 4 '— 6 l.0)- ._ =3 .1 O O 0.8 _ fl 5 t _ g .JIOSr- ‘- =4 2 _ -1 0.4 — ' "I °2 1‘ ‘4' . . .. l i l 41 1 _6L 1 g [(+1 1 MONTHS OF STORAGE, 72F Fig. ll.--Peroxide values and detected oxidized flavor (I) of freeze-dried sour cream containing various antioxidants, stored at 72F 87 LEV 1* Al I I I I I 1 T l I A—T A — CONTROL [1 - PG 02% CA 0| % 0 L4 BHA .0I% BHT. .0I% A _ 7 P6 04% CA .02 % o - '- .J '02 P d l.— E r- .. 0 d 4‘ L . \ °F N -1 o I- ); (18" ‘.A D O - _. E! j 0.6 - —I 2 _ . .. 04 . ' ~ — ‘I 02 Ar ‘ ’ .1 «'11 A .r ._ A I I I ' ' 12L l 4 I 7)? I a I I0 12 MONTHS OF STORAGE . 72F Fig. 12.-~Peroxide values and detected oxidized flavor (I) of freeze-dried sour cream containing various antioxidants, stored at 72F 88 heat treatment involved in processing (Gould g: 31., 1939, 1957; Findlay S£,31f' 1946). In commercial operations, the addition of the higher level would be recommended to assure good storage stability. Samples stored at 40F exhibited the best storage stability. The results of the experiments are shown in Figures 13, 14 and 15. After six months of storage, sam- ples were judged to be slightly stale and in some cases very slightly oxidized, but still acceptable. The effec- tiveness of the antioxidants in samples stored at 40F cor- related well with the samples held at 72F. The best com- bination to improve shelf life was found to be 0.01% BHA + 0.01% BHT (Figures 14, 15) or 0.04% PG (Figure 16). The addition of 0.04% BHA (Figure 13) was also satisfactory in retarding lipid oxidation. These samples were still satis- factory after 12 months of storage at 40F. The observed increase in peroxide value correlated well with the detected off flavors. While values cannot be transposed from experiment to experiment, the trend of hydroperoxide formation and the effectiveness of certain antioxidants in retarding lipid oxidation becomes obvious, when observing the trends developed during storage. The observed peroxide values were the highest in samples stored at 40F. As the storage temperature in- creased, the peroxide values for the corresponding samples I' 89 '5 I I I I I r I I I I I F A CONTROL U BHA.(NF%IPGJOO496(HKJJO396 C) L4 I" BHA .0II% BHT.OII°/o PG 008% A - CA JOOGOC I. _ BHA 02% CA 02% .. “2" H '- .J I— EEI‘*_ _ 3' \ L _. 3 , 98)" -I 2 CD3 ‘- —I w j 064— - . . a 2 E ' , agn- t. . . § 0.2» , . y . . 00/ L ./A A :/‘ . I L l l L l | J 3 ‘4 A” a . MONTHS OF STORAGE ,4OF Fig. l3.--Peroxide(va1ues and detected oxidized flavor (I) of freeze-dried sour cream containing various antioxidants, stored at 40F 90 IST’ ' l I ‘I I 1 I r, 1’ I r - --( CONTROL ' D ‘ I-4~ BHA 04% CA 04% O - BHA .0279 CA .OO4o/o . I )- BHA ' .O|°/o BHT .O|°/o A _1 PG .Ol’lo CA .00579 ‘ Ia ”T ' I U. 3’ \ Loh— . _. N O _ . _ >' ES CENT -( O . 9.4 I- .1 _l :4 . J 2 96 ” ' 1., r- —( 04 - /' I ' . ‘ A -L% Li we 5 L I'% MONTHS OF STORAGE . 40F Fig. 14.--Peroxide values and detected oxidized flavor (I) of freeze-dried sour cream containing various antioxidants, stored at 40F 91 LG 1 I I I l ' ' ' (gm. L89) I __ (IOMO. 2.08) _ I.4- ' - I2 I- 1 LOL' .4 r CONTROL (3 a PG 02% CA .0I% . BHA .0I% BHTOI’Io A PG 104% CA 02% O MILLIEQUIV. Oz/kg FAT 8 I 0.6. h- _ . .. o ‘ ‘ 4 .J 0.2! "‘ OJ _ 1 l l l ll_g 1 LI 1 l 1 ’ E 4 6 8 I0 1 MONTHS OF STORAGE , 40F Fig. 15.-~Peroxide values and detected oxidized flavor (I) of freeze-dried sour cream containing various antioxidants, stored at 40F 92 were lower. The relationship of peroxide value versus temperature was established by Greenbank (1048). In view of the findings in the presented research, one may conclude that P.V., when compared to a suitable control, is useful for measuring the extent of lipid oxida- tion. The comparison of the data at three different tem- peratures indicates that lower temperatures tend to result in higher hydroperoxide levels but improved flavor attri— butes. The obvious conclusion is that for each storage temperature, a different peroxide value would be repre- sentative for an objectionable off-flavor. Lillard and Day (1961) reported that the correla- tion coefficient between peroxide values and TBA values and the reciprocal of the absolute flavor threshold of autooxidized milk fat were significant at the 1% level. However, Negoumy and Hammond (1962) found that P.V. and TBA tests in cold stored butter did not always correspond to the results of organoleptic evaluation, and Kliman gg g1. (1960) stated that in the evaluation of dried whole milk, the P.V. showed poor correlation to flavor. The ability of various antioxidants to retard the formation of hydroperoxides and to prolong the storage life of the freeze dried sour cream is well substantiated. The addition of BHA at 0.04% level was found to be effective in retarding lipid oxidation, while Others have found that BHA was ineffective in dry whole milk (Tamsma 93 EE.E£°' 1963; Pete g£.g1:, 1964). Pete gE'gl. (1964) showed that the long chain esters of gallic acid were more effective than NDGA, while the addition of BHA and BHT showed only limited improvement in the storage of dry whole milk. Butter stored at ~18 and +38F showed no im— provement in flavor when BHT or BHT + isopropyl citrate were added (Nagoumy and Hamond, 1962). In this research, however, excellent results were obtained by adding .01% BHA + .01% BHT. The results are probably due to synergistic action of the two phenolic antioxidants (Mahon gg'g1., 1953; Dugan g£_gl., 1954). Propyl gallate in the presence of added copper has been shown to increase the storage stability of fluid whole milk (Chilson, 1950). This antioxidant also successfully retarded lipid oxidation in freeze dried sour cream in this research. The effect of propyl gallate was considered to be superior to NDGA in milk fat held at 40C (Hill gg_gl., 1969). Other researchers have found that NDGA has excel- lent antioxigenic properties in dairy products (Stull, 1948, 1949; Bush gE_g1., 1952; Cox g£_g1., 1957). How- ever, food regulations at present do not allow the addi- tion of NDGA. Cornell 25.2lf (1969) suggested that the oil/water distribution coefficients of the various antioxidants would influence the pattern of distribution, and therefore the 94 efficient protection of dried milk products might require special processing conditions. Martinez g£_gl. (1958) demonstrated that different antioxidants would be dis— tributed in various fractions of milk. The required level of antioxidant in the case of PG and BHA was about double that used in normal practice. Since these antioxidants are aromatic hydrocarbons, it is possible that at the very low absolute pressures employed in freeze drying, there would be substantial losses of these compounds. Other workers have demonstrated such losses during dehydration of foods. The Effect of Inert Gas in Breaking the Vacuum of the Fieeze Dryer on the Shelf Life of Freeze Dried Sour Cream In an effort to improve the storage stability of freeze dried sour cream without the use of antioxidants, inert gas was used to break the vacuum in the freeze dryer. Results indicate that highly significant reduction in the formation of hydroperoxides can be achieved if the vacuum is broken with N2 or CO2 rather than air. The flavor of the freeze dried sour cream was also improved correspondingly. The air exposed samples showed oxidized flavor in four months at 72F storage while the N2 and C02 treated samples were still satisfactory after five months (Figure 16). The organoleptic evaluation corresponded well with 95 l ] I l | I 1 ' I I l I l. VACUUM BROKEN _. [3 Air 0 _. €02 A - 02 . LZI- " E _ LOH" C» .1: ... \ F" 8 , 03i- " 2 D — _ O 9.1 _| 06;- " =3 2 _ _. ‘14h- " r _ (32- -' l I l l l .fir__;___1£ ' 7i” 4 Tel? + MONTHS OF STORAGE ,TZF Fig. 16.--Peroxide values and detected oxidized flavor (I) ' of freeze-dried sour cream exposed to various gases at conclusion of drying cycle, stored at 72F 96 the P.V. Air treatment resulted in three times higher P.V. than when the samples were treated with N2 and C02. Samples stored at 40F showed improved storage stability over samples stored at 72F. Based on organo- leptic evaluation, N treatment extended the storage life 2 over eight months, while the storage life of air, 02 and CO2 treated samples was 5, 5, and 8 months respectively (Fig. 17). The level of hydroperoxides formed showed trends similar to those in the antioxidant studies. The effect of N2 (Fig. 17) in retarding hydroperoxide forma- tion was of similar magnitude, as the use of an antioxi- dant such as 0.04% P.G. (Fig. 15), or 0.01% BHA + 0.01% BHT (Figs. 14, 15). Such results would indicate that breaking the vacuum with N2 could result in improvement similar to that obtained with antioxidants, in the event that antioxidants were too costly or not permitted. Carbon dioxide was not as effective as N2 in re- tarding lipid oxidation. Since CO2 is known to absorb into food particles this decrease in effectiveness is not unusual. Following absorption of the CO2 the surface of the food may no longer be protected by a layer of inert gas and lipid oxidation will proceed. The use of pure oxygen instead of air resulted in no further acceleration of lipid oxidation. This tends to indicate that the oxygen content in the air is suffi- ciently high to initiate maximum lipid oxidation. 97 L6 I l I I I l I 1' I T 37 r. VACUUM BROKEN N2 '3 ~ -I L4. Air 0 002 A )— Q . -( I2 I- . — t I '2 UL K)F' ‘ - ‘3 ~ ‘ . I . 08+. 0 1 2 :3 _ 4 O LLJ _| 0.6 -- "i d 2 L.- ‘ . .4 (14- -a 02 - ' '4 ; /- a 1 L l L l I MONTHS OF STORAGE,40F Fig. 17.--Peroxide values and detected oxidized flavor (I) of freeze-dried sour cream exposed to various gases at conclusion of drying cycle, stored at 40F 98 Reports on the introduction of inert gas into the vacuum chamber are limited; however, it was suggested by Goldblith (1963a) and Landman (1960). Coulter (1947) demonstrated that the flavor of milk powders would be im- proved if they are spray dried into an inert atmosphere. The improvement in the quality of freeze dried sour cream by the introduction of an inert gas can be ex- plained as follows: during the process of freeze-drying intercellular spaces created by removal of sublimed water are void of any gas. At the time of introduction of an inert gas, these intercellular spaces will be filled with inert gas. The possible adsorption of such inert gas on the surface of the dry food particles may also serve to en- hance stability by limiting the accessibility of atmospheric oxygen to labile constituents. The necessity of double gassing to reach 1% oxygen level was demonstrated by Coulter gg g1. (1945). This double gassing was necessary because the adsorbed oxygen or the oxygen trapped in the amorphous lactose glass was not removed by a simple vacuum treatment and required a 24 hour equilibration with an inert gas. The reduction of oxygen to 0.01% required the use of oxygen scavengers or enzyme treatments (Abbot g£_gl,, 1955; Scott g£_g1., 1961; Tamsma g£.gl., 1967). The oxygen removal in such a system would be continuous and would only be limited by the avail- ability of oxygen and the capacity of the system employed. Apparently the reduction to such low oxygen tension requires 99 forces capable of removing the adsorbed oxygen molecules. Similarly, therefore, in the absence of any oxygen, the introduction of an inert gas would result in adsorption at the surface of the particle. U.V. Experiments In an effort to predict the usefulness of an anti- oxidant, reconstituted freeze dried sour cream with or without added antioxidants was exposed to U.V. light at 2537 A wave length for various lengths of time. The effect of light on the formation of hydro- peroxides is well known (Holman, 1954: Wishner, 1966) and the utilization of such energy to accelerate lipid oxida- tion was the basis of this research. The results of the experiments are presented in Table 13. The reader is re- ferred to Table 1 for the various antioxidants used in this experiment. The results presented are the average of duplicate determinations. The samples containing no antioxidant or synergist (designated control) showed significantly higher peroxide values in all three experiments. The combination of anti- oxidants which retarded lipid oxidation in storage also resulted in lower P.V. after 50 minutes of exposure to U.V. light. The U.V. experiments show that 0.02% PG acceler- ated the formation of hydroperoxides (Table 13, Sample 11) 100 Table 13.—-Peroxide values of fat extracted from reconsti- tuted sour cream exposed for variable periods to ultraviolet light (2537A) Sample Time of Exposure (min.) Number* 0 10 20 50 Experiment 1 1(control) 0.135 0.195 0.210 0.507 2 0.115 0.144 0.146 0.278 3 0.105 0.161 0.165 0.238 Experiment 2 5(control) 0.090 0.140 0.210 0.342 6 0.080 0.205 0.220 0.270 7 0.108 0.148 0.170 0.222 9 0.090 0.122 0.198 0.242 Experiment 3 10(control) 0.168 0.191 0.285 0.450 11 0.140 0.262 0.232 0.460 12 0.142 0.245 0.280 0.362 13 0.120 0.175 0.212 0.295 ' *Level of antioxidants in samples are shown in Table 1, page 43. 101 while the addition of 0.04% PG to the sample resulted in a decrease in P.V. Similarly, the addition of 0.04% was required in the long term storage studies (Table 13, Sample 13) to extend storage life. The addition of 0.01% BHT + 0.01% BHT resulted in excellent storage stability, and also resulted in a decrease in P.V. in the samples exposed to U.V. light (Table 13, Sample 9, 12). It is interesting to note that an exposure of 50 minutes was required before the effectiveness of the vari- ous antioxidants became evident. The action of antioxi- dants involves initially the reaction between the unsat- urated fat and the -OH group of the phenolic antioxidant, resulting in hydroperoxide formation. In the propagation and termination stages the antioxidant will inhibit the formation of hydroperoxides and therefore retards lipid oxidation (Bollard gg_g1.. 1947; Dugan, 1961). The correlation of flavor to peroxide values in this experiment was unsuccessful. The exposure to 10 minutes of U.V. light resulted in a flavor which can be described as "solar" or "activated." The effect of ex- posure to light could result in the initiation of chemical reactions with other components of the liquid sour cream. Several reports are available which indicate that other components of a milk system, particularly the proteins, contribute to the development of off flavors due to light exposure (Greenbank, 1948; Stull, 1953: Dunkley g£.g1., 102 1962). The development of such off flavors would be con- current with lipid oxidation (Stull, 1953). While the full utilization of such U.V. exposure to predict long term storage stability is not fully elucidated in these experiments, it is felt that further investigation is warranted. Other methods are either too cumbersome and require expensive equipment (Reimenschenider, 1943; Stucky g£.g1,, 1958), or are only applicable in a purified system (Thompson, 1950; Marco, 1968). SUMMARY AND CONCLUSIONS Flavor Studies l. Acidified cream with added diacetyl and acetoin was freeze-dried under conditions in which absolute pres- sure, platen temperature, tray loading and volatile addi- tives were varied. 2. The loss of flavor volatiles during freeze- drying was dependent upon the processing conditions used and the affinity of such flavor compounds to adsorb onto the solid components of the system employed. 3. The selection of a suitable freeze-drying con— dition has to be economically favorable to reduce the unit cost, and at the same time it should retain the maximum level of desirable flavor components. 4. The increase in platen temperature resulted in increased losses of both diacetyl and acetoin and a de- crease in the freeze-drying time. Both the losses and the time of freeze drying appeared to be linear functions of the change in platen temperature. 5. Varying the absolute pressure between 20-3SOu in the vacuum chamber during freeze~drying resulted in no Ichange in the relative loss of diacetyl and acetoin; how- ever, losses were significantly lower at 40. 103 104 6. The reduction of the layer thickness in the trays by one half during freeze drying reduced the drying time by 50% but did not affect the loss of diacetyl or acetoin. 7. Diacetyl adsorbed onto the powder when the acidified cream was freeze dried readily desorbed when these powders were placed into a constant humidity chamber and allowed to equilibrate for 24 hours to 4.5% moisture. 8. The relative loss of diacetyl was independent of the initial absolute concentration in the freeze drying of acidified cream. Storage Studies 9. Samples stored at 99F developed off flavors in 4 weeks and after 8 weeks of storage, the samples were com- pletely unacceptable. The off flavors were caused by Maillard browning and lipid oxidation. The effectiveness of the various antioxidants at this elevated storage tem- perature could not be established because of the browning flavors which developed concomitant with oxidized flavors. 10. The control samples stored at 72F were accept- able after 3 months storage, and the addition of 0.01% BHA + 0.01% BHT, 0.04% PG or 0.04% BHA effectively retarded lipid oxidation and extended the storage life to >8, >8 and 8 months respectively. The trend of peroxide values correlated well with the organoleptic observations. 105 11. The samples containing no antioxidants and stored at 40 F were of acceptable quality after 6 months. The addition of 0.01% BHA + 0.01% BHT, 0.04% PG or 0.04% BHA extended the shelf life to over one year. Peroxide values showed good correlation to flavor scores; however, they were much higher than peroxide values obtained from samples stored at 72 or 99F. The accumulation of hydro- peroxides was attributed to the slower rate of breakdown to subsequent secondary products. 12. Breaking the vacuum in the freeze dryer with nitrogen (N2) resulted in improved shelf life at both 72 and 40F storage. The NZ treatment appeared to be as ef- fective in retarding lipid oxidation as the use of an antioxidant. Breaking the vacuum with CO2 was only par- tially effective, due to the slow absorption of CO2 into the particles. 13. Exposing the samples to UV light resulted in hydroperoxide formation. The level of such hydroperoxides was significantly lower in the samples containing antioxi- dants found to function effectively to retard lipid oxi- dation and extend the storage life of the freeze dried sour cream. THES IS REFERENCES 10. THESIS REFERENCES Abbot, J. and Waite, R., 1955, Gas packing milk powder with mixture of nitrogen and hydrogen in the presence of palladium catalyst. J. Dairy Res. fig: 285. Abbot, J. and Waite, R., 1962, The effect of anti- oxidants on the keeping quality of whole milk pow- der. I. Flavones, gallates, butyl hydroxyanisole and nordihydroquaiaretic acid. J. Dairy Res. 29:55. Aceto, N.C., Craig, J. C. Jr., Eskew, R. K. and Talley, F. B., 1966, Storage aspects of continuous vacuum foam dried whole milk. International Dairy Congress, 17th Munich Proceedings E/F 189-196. Aceto, N. C., Sinnamon, H. I., Schoppet, E. F., Craig, J. C. and Eskew, R. K., 1965, Moisture con- tent and flavor stability of batch vacuum foam dried milk. J. Dairy Sci. 18:544. Aulakh, J. S., 1967, A study of lipid oxidation in model systems of, Copper-Proteins, milk lipids and milk dialyzate. Ph.D. Thesis. Michigan State Uni- versity. Anastos, C., 1964, Freeze-drying of dairy products. Milk Dealer 23:40. Andersen, V. B., 1961, Cultured buttermilk and sour cream from single strain cultures of lectic strep- tococci. Milchwissenschaft 1§:(3) 128. Anonymous, 1964, Freeze dried quarg from Dahlenburg, Molkerei- u Kasereiztg. Anonymous, 1966, Freeze drying of dairy products. Milk Ind. 58:22. Arsdel Van W. B., Copley M. J., 1963, Food dehydra— tion Vol. I. The AVI Publishing Co., Inc., West- port, Conn. 106 ll. 12. l3. 14. 15. 16. 17. 18. 19. 20. 21. 107 Association of Official Agricultural Chemists (AOAC), 1965, Official methods of analysis, Ninth Edition. Published by the AOAC, Washington, D.C. Babel, F. J. and Hammer, B. W., 1944, Action of but- ter cultures in butter. A Review. J. Dairy Sci. .21:141. Badings, H. T. and Galesloot, T. E., 1962, Studies on the flavour of different types of butter start- ers with reference to the defect "youghurt flavour". Butter, Proceedings 16th Int. Dairy Cong. K¢benhaven. Vol. B:199. Badings, H. T. and Wassik, J. C., 1963, Separation and identification of aliphatic aldehydes and ketones by thin layer chromatography of the 2,4-dinitro— phenylhydrazones. Neth. Milk and Dairy J. 11:132. Barnhart, J. L., Nelson, K. E., 1967, Freeze drying dairy products. Idaho Agricultural Research Progress Rep. No. 120. Bennett, Grace, Liska, B. J., Hempenius, W. L., 1965, Effect of other flavor components on the pre- ception of diacetyl in fermented dairy products. J. Food Sci. 39:35. Bickoff, E. M., Livingston, A. L., and Thompson, C. R., 1955, Bisphenol derivatives as antioxidants for carotene. J. Am. Oil Chem. Soc. 22:64. Bird, K., 1964, The freeze-drying industry: pro— jections of capital and labor requirements. A bul- letin: United States Department of Agriculture, Economics Research Service, Washington, D. C. Bohrer, B. B., 1967, Azeotropic drying process. U. S. Patent 3,298,109. Bollard, J. L., and Ten Have, P., 1947, Kinetic studies in the chemistry of rubber and related ma- terials. IV. The inhibitory effect of hydroquinone on the thermal oxidation of ethyl linoleate. Trans. Faraday Soc. 13:201. Boskovic, M. and Issenberg, P., 1968, Adsorption of volatile organic compounds in dehydrated food sys- tems. II. Sorption of some alcohols on cellulose. 28th Ann. Meeting, Inst. Food Technol. Philadelphia, Penn. Paper No. 185. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 108 Boudreau, A., Richardson, T. and Amundsen, C. H., 1966, Spray-dried butter and loss of volatile fatty acids during spray drying. Food Technol. 22:100. Bradley, R. L., 1964, Spray-drying of natural cheese, Ph.D. Thesis, Michigan State University. East Lansing. Bradley, R. L. Jr., Stine, C. M., 1964, Foam spray drying of natural cheese. Manif. Milk Prod. J. 55:8. Brooks, R., 1965, Spray-drying of flavoring materi— als. Birmingham Univ. Chem. Eng. 22:11. Brown, W. C. and Thurston, L. M., 1940, A review of oxidation in milk and milk products as related to flavor. J. Dairy Sci. 22:629. Burghimer, F. T., 1966, A study of high intensity near infrared radiation as energy source for freeze drying. Ph.D. thesis, University of Illinois. Busch, A. A., Decker, C. W., and Asworth, U. 8., 1952, The keeping quality of whole milk powder. IV. The antioxidant properties of NDGA and certain emul- sifying agents. J. Dairy Sci. 22:525. Calkins, V. P., 1947, The mechanism of the antioxy— genic synergism of quinones and quinols with phos- phoric acid and other acids in fat systems. J. Am. Oil Chem. Soc. 225384. Chapman, R. A. and MacFarlene, W. D., 1943, A color- imetric method for the determination of fat-per- oxides and its application in the study of the keep- ing quality of milk powders. Canad. J. Res. 218:133. Chapman, R. A. and MacKay, K., 1949, The estimation of peroxides in fats and oils by the Ferric Thio- cyanate Method. J. Am. Oil Chem. Soc. 22:360. Chilson, W. H., Martin, W. H. and Whitman, C. H., 1950, Use of propyl gallate to defer development of oxidized flavor in market milk. J. Dairy Sci. 22: 925. I Chou, T. C., 1962, The chemical nature of the char- acteristic flavor of cultured buttermilk. Ph.D. thesis, Ohio State University, Columbus, Ohio. 34. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 109 Chou, T. C. and Harper, W. J., 1963, Chemical nature of the characteristic flavor of cultured buttermilk. J. Dairy Sci. 22:614. Corbett, W. J., and Tracy, P. H., 1941, Effect of condensing on the dev010pment of oxidized flavor. J. Dairy Sci. 22:109. Cornell, D. G., DeVilbiss, E. D., and Pallansch, M. J. Distribution of antioxidants in heterogeneous systems. Paper No. M.47, 64th Annual Meeting. Am. Dairy Sci. Assoc., Minnesota. Coulter, S. T., 1944, Dehydrated milk and milk products, use of antioxidants. J. Dairy Sci. 22: 607. Coulter, S. T, 1947, The keeping quality of dry whole milk spray dried in an atmosphere of an inert gas. J. Dairy Sci. 22:995. ‘ Coulter, S. T., Jenness, R., 1945, Packaging dry whole milk in inert gas. University of Minn. Agr. Expt. Sta. Tech. Bull. 167. Coulter, S. T., Jenness, R., and Geddes, W. F., 1951, Physical and chemical aspects of the production storage and utility of dry milk products. Adv. in Food Res. 2:1951. Coulter, S. T., Jenness, R. and Crowe, L. K., 1948, Some changes in dry whole milk during storage. J. Dairy Sci. 22:986. Cox, D. H., Coulter, S. T., and Lundberg, W. O» 1957, Effect of Nordihydroguiaretic acid (NDGA) and other factors on the stability of added Vitamin A in dry and fluid milks. J. Dairy Sci. 22:464. Day, E. A., 1965, Some properties of carbonyl com- pounds encountered in flavor isolets of dairy prod- ucts. A review. Food Technol. 22:129. Day, E. A., Larsen, P. B., Lindsay, R. C. and Elliker, P. R., 1962, Some observations on the vola- tile flavor compounds of ripened cream butter. J. Dairy Sci. 22:660. Day, E. A., Lindsay, R. C. and Forss, D. A., 1964, Dimethyl sulfide and the flavor of butter. J. Dairy Sci. 22:197. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 110 Desai, J. Rohini, 1966, A study of some physical and chemical properties of dehydrated sour cream. M.S. thesis, Michigan State University, East Lansing, Michigan. Doob, H. Jr., Willman, A. and Sharp, P. F., 1942, Influence of moisture and browning of dried whey and skim milk. Ind. Eng. Chem. 22:1460. Dugan, L. R. Jr., 1961, Development and inhibition of oxidative rancidity of foods. Food Technol. 22: 10. Dugan, L. R., Jr., Marx, L., Weir, E. E. and Kray- bill, H. R., 1954, Butylated hydroxytoluene. A new antioxidant for food use. Am. Meat Inst. Foundation Bull. No. 18. Duiven, J. E., 1965, Freeze drying of milk. Con- serva. 22:251. Dunkley, W. L., Franklin, J. D., and Pangborn, R. M., 1962, Influence of homogenization, copper, and ascorbic acid on light activated flavor in milk. J. Dairy Sci. 22:1040. Dunkley, W. L. and Jennings, W. C., 1951, A procedure for application of the thiobarbituric acid test to milk. J. Dairy Sci. 22:1064. Dunkley, W. L., Ronning, M., Franke, A. A. and Robb, J., 1967, Supplementing rations with tocopherol and ethoxyquin to increase oxidative stability of milk. J. Dairy Sci. 22:492. Dutton, H. J., Schwab, A., Moser, H. A. and Cowan, J. C., 1949, The flavor problem of soybean oil. V. Some consideration in the use of metal scavengers in commercial operations. J. Am. Oil. Chem. Soc. 22: 441. Eastman Kodak Company, 1963, Tenox, Eastman food grade antioxidants, a Bulletin, Kingsport, Tenn. Evsrateva, E., Kuriachev, and Sinicyn, 1959, Pre- servation of lactic cheese by a method of freeze- drying. (In Russian, English translation) Mol. Prom. 22;7, 10. Findlay, J. D., Higginbottom, C., Smith, J. A. B., Lea, C. H., 1946, The effect of the preheating 58. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 111 temperature on the bacterial count and the storage life of whole milk powder spray dried by the Krauss Process. J. Dairy Res. 22:378. Findlay, J. D., Smith, J. A. B., and Lea, C. H., 1945, Experiments on the use of antioxidants in spray dried whole milk powder. J. Dairy Res. 22:165. Finley, J. W. and Shipe, W. F., 1967, Observation relative to light induced off flavors in milk. J. Dairy Sci. 22:943. Flake, J. C., Weckel, K. C., Jackson, H. C., 1939, Studies of the activated flavor in milk. J. Dairy Sci. 22:153. Flink, J. and Karel, M., 1969, Studies on mechanisms of retention of organic volatiles in freeze dried systems. 29th Ann. Meeting, Inst. Food Technol. Chicago, Ill. Paper No. 85. Flosdorf, E. W., 1949, Freeze—drying (drying by sub- limation). Reinhold Publishing Corp., New York, N. Y. Flosdorf, E. W., and Mudd, S., 1935, Procedure and apparatus for preservation in 1yphi1e form of serum and other biological substances. J. Immunol. 22: 389. Frankel, E. N. and Tarrassuk, N. P., 1959, Inhibi- tion of lipase and lipolysis in milk. J. Dairy Sci. 42:409. Garrett, O. F., 1940, The antioxidant action of milled oat flour on milk. Milk Plant. Monthly 22: 40. Goldblith, S. A., Karel, M. and Lusk, G., 1963, The role of food science and technology in the freeze dehydration of foods. a) First of two parts, Food Technol. 22:139. b) Second of two parts, Food Technol. 22;258. Gould, T. A., 1939, Cooked and oxidized flavors of milk as effected by ferrous iron, J. Dairy Sci., 22:1017. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 112 Gould, T. A. and Keeney, P. G., 1957, Certain factors effecting the concentration of active sulfhydryl compounds in heated cream. J. Dairy Sci. 22:297. Gould, T. A. and Sommer, H. H., 1939, Effect of heat on milk with especial reference to the cooked flavor. Mich. Agr. Expt. Sta. Tech. Bull. 164. Grace, W. R. and Co., 1966, Dehydration of frozen ice milk confections. Br. patent 1,042,530. Graf, R. L., 1961, Present status of freeze dried Fa foods for the military, in Activities Report, Quartermaster Corps, Vol. 13. Greenbank, G. R., 1948, The oXidized flavor in milk and dairy products. A review. J. Dairy Sci. 22: 913. “is. ‘ E Greenbank, G. R., Wright, P. A., Deysher, E. F. and Holm, G. E., 1946, The keeping quality of samples of commercially dried milk packed in air and in inert gas. J. Dairy Sci. 22:55. Grout, A. P. de, 1963, The influence of dehydration of foods on the digestibility and the biological value of the protein. Food Technol. 22:339. Gruyter en Zoon N. V. P. de, 1965, Process for the removal of water by selective permeation and evapor- ation. Dutch patent application No. 6400502. Hammer, B. W. and Babel, F. J., 1943, Bacteriology of butter cultures. A review. J. Dairy Sci. 22: 168. Hammer, B. W. and Sherwood, F. F., 1923, The volatile acid produced by starters and by organism isolated from them. Iowa Agr. Expt. Sta. Res. Bull. 80. Hammond, L. M., 1967, Economic evaluation of UHF dielectric v.5. radiant heating for freeze drying. Food Technol. 22:735. Hamre, M. L., 1966, The effect of freezing method on the quality of frozen and freeze dried cooked diced chicken meat. Ph.D. thesis, Purdue University, Indiana. 81. 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 113 Harper, J. C. and Tappel, A. L., 1957, Freeze drying of food products, p. 171, in Advances of Food Re- search, Vol. 7, Academic Press Inc., New York, N. Y. Harland, H. A., Coulter, S. T. and Jenness, R., 1949, Some factors influencing the_reducing systems in dry whole milk. J. Dairy Sci. 22:334. Harvey, R. J., 1960, Production of acetone and acetaldehyde by lactic streptococci. J. of Dairy Res. 22:41. Harvey, R. J. and Collins, E. B., 1961, The role of citratase in acetoin formation by Streptococcus 22- acetylactis and leuconostoc citrovorum. J. of Bac- teriol. 82:954. Hansen, 0., 1967, Fluidized freeze-drying. Food Engineering 22(6)73. Hearne, J. F., 1964, Long term storage of foods. Food Technol. 22:318. , Henick, A. S., Benca, M. F. and Mitchell, 1954, Estimating carbonyl compounds in rancid fats and foods, J. Am. Oil Chem. Soc. 22:88. Herrington, B. L., 1954, Lipase: a review. J. Dairy Sci. 22:775. Hetrick, J. M. and Tracy, P. H., 1945, Some observa- tions on the keeping quality of spray dried whole milk stored at room temperature. J. Dairy Sci. 22: 687. Hibbs, R. A. and Asworth, U. S., 1951, The solubility of whole milk powder as affected by protein stabil- izers and by emulsifiers. J. Dairy Sci. 22:1084. Hills, G. Loftus, and Thiel, C. C., 1946, The Ferric Thiocyanate Method of estimating peroxide in fat of butter, milk and dried milk. J. Dairy Res. 22g340. Hollender, H. A., Tracy, P. H., 1942, The relation of the use of certain anti-oxidants and methods of processing to the keeping quality of powdered whole milk. J. Dairy Sci. 22:249. Holm, G. E., Wright, P. A. and Greenbank, G. R., 1927, Variations in the susceptibility of the fat in 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. 114 dry whole milk to oxidation when stored at various temperatures and in various atmospheres. J. Dairy Sci. 22:33. Holman, R. T., 1954, Autoxidation of fats and re- lated substances. In: The Chemistry of Fats and Other Lipids 2:51, Pargamon Press, London. Issenberg, P., Boskovic, M. and Hwang, J. S., 1969, Adsorption of volatile organic compounds in dehy- drated foods and model systems. Sorption of some alcohols and acetates on cellulose. 29th Ann. Meet— ing, Inst. Food Technol. Chicago, Ill. Paper No. 84. Issenberg, P., Greenstein, G. and Boskovic, M., 1968, Adsorption of volatile organic compounds in dehydrated food systems. I. Measurement of sorp- tion isotherms at low water activities. J. Food Sci. 22:621. Jack, E. L. and Henderson, J. L., 1942, Preventing off-flavors in dried whole milk. Food Inds. 22:50. Jekabs, D. Mezciems, I., 1966, Dehydrated fermented milk products. Br. Patent 1043, 194. Jenness, R., and Patton, S., 1959, Principles of Dairy Chemistry. John Wiley and Sons, Inc., New York. Jensen, R. G., Lipolysis. A symposium, 1964. J. Dairy Sci. 21:210. Jokay, L., 1964, Carbohydrate enriched cheese base survival-type dehydrated food bar. U.S. Patent 3,121,014. Jokay, L. and Meyer, R. T., 1959, Application of lyophilization to cheese products. II. The develop- ment of freeze dehydrated creamed cottage cheese. J. Dairy Sci. 22:908. Josophson, D. V. and Doan, F. J., 1939, Observations on cooked flavor in milk. Its source and signifi- cance. Milk Dealer 22:54. Jureit, S., 1964, Technological and economic prob- lems of freeze-drying milk and dairy products. Dtsch. Molkereiztg 22:867. 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 115 Kan, B. and Winter, F. de, 1968, Accelerated freeze drying, through improved heat transfer. Food Tech- nol. 22:1269. Katayama, 0., Watanabe, A and Yamato, I., 1966, Armoa components of fruits and vegetables. I. Volatile components of apple. Nippon Shokuhin Kogyo Gakkaishi 22:(10) 415 Chem Abstracts:66-ll4731g. King, J. C. and Clark, J. P., 1968, Referred to in freeze dries with Helium. Food Engineering 40(6) 175. King, R. L., LuiCk, J. R., Litman, I. I., Jennings, W. G., and Dunkley, W. L., 1959, Distribution of natural and added copper and iron in milk. J. Dairy Sci. 22:780. Kinsella, J. E., Patton, S., Dimick, P. S., 1967, The flavor potential of milk fat. A review of its chemical nature and biochemical origin. J. Am. Oil Chem. Soc. 22:449. Kliman, Phillis, Tamsma, A., and Pallansch, M. J. Abstra. Pap. Amer. Chem. Soc. Sept. 1960:pp. 5A. Kramers, H., 1958, Rate controlling factors in freeze drying. In: Fundamental Aspects of Dehydration of Foodstuffs. Macmillan Co., New York. Krienke, W. A. and Tracy, P. H., 1946, Factors re- lated to the brown discoloration of powdered whole milk. J. Dairy Sci. 22:488. Krishnawamy, M. A., and Babel, F. J., 1951, Diacetyl production by hemolytic and non hemolytic strepto- cocci from pathogenic sources and milk. J. Bac- teriol. 2:79. Krukovsky, V. N., 1961, The biochemical properties of milk and lipid deterioration in milk and milk prod- ucts as influenced by natural varietal factors. J. Agr. Food Chem. 2: 439. Labuza, T. P., Maloney, J. F. and Karel, M., 1966, Autoxidation of methyl linoleate in freeze-dried model systems. II. Effect of water on cobald- catalyzed oxidation. J. Food Sci. 22:885. 117. 118. 1195 121.1 122. 123. 124. 125. 126.; 127. 116 Landmann, W. A., 1960, Role of water and 02 in the deterioration of freeze dried meat. Q. M. C. Food Cont. Inst. Report #328. Larson, R. W., Steinberg, M. P. and Nelson, A. I., 1967, Effect of gas flushing at several pressures on the rate of freeze drying of beef. Food Technol. 21:401. Lea, C. H., 1956, Some observations on the prepara- tion and properties of phosphatidyl-ethanolamine. In: Biochemical Problems of Lipids, p. 81, Inter- science Publishers, Inc., New York. Lea, C. H., 1958, Chemical changes in the preparation and storage of dehydrated foods. In: Fundamental Aspects of the Dehydration of Food Stuffs, Macmillan ‘Co., New York. Lea, C. H., 1962, The oxidative deterioration of food lipids. In: Symposium on Foods: Lipids and their oxidation. The Avi Publishing Co., Westport, ‘ Conn. Lea, C. H., Moran, T., Smith, J. A. B., 1943, The gas packing and storage of milk powder. J. Dairy Res. 22:162. ~Lewin, L. M. and Mateles, R. I., 1962, Freeze drying without vacuum: a preliminary investigation. Food Technol. 22:94. Lillard, D. A. and Day, E. A., 1961, Autoxidation of milk lipids. II. The relationship of sensory to chemical methods for measuring the oxidized flavor of milk fats. J. Dairy Sci. 22:621. Lindsay, R. C., 1965, Flavor chemistry of butter culture. Ph.D. thesis, Oregon State University, Cor- vallis, Oregon. Lindsay, R. C., Day, E. A. and Sandine, W. E., 1965, Identification of volatile flavor components of but- ter culture. J. Dairy Sci. 22:1566. Lundberg, W. 0., 1961, Oxidative rancidity and its prevention. In: Autoxidation and Antioxidants, Ed. by W. O. Lundberg. Interscience Publishers, Inc., New York, Vol. I, II. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 138. 139. 117 Lusk, G., Karel, M. and Goldblith, S. A., 1965, Effect of some processing parameters on the rates of freeze drying of shrimp. Food Technol. 22:620. Mackenzie, A. P., and Luyet, B. J., 1963, Effect of recrystallization upon the velocity of freeze dry- ing. In: "Progress in Refrigeration Science and Technology," Vol. 2, pp. 1573, Pergamont Press, Oxford. Mahon, J. H. and Chapman, R. A., 1953, The relative rates of destruction of propyl gallate and buty- lated hydroxyanisole in oxidized lard. J. Am. Oil Chem. Soc. 30:34. Malkki, Y., and Veldstra, J., 1967, Flavor retention and heat transfer during concentration of liquids in centrifugal film evaporator. Food Technol. 22:1179. Maloney, J. F., Labuza, T. P., Wallace, D. M. and Karel, M., 1966, Autoxidation of methy linoleate in freeze dried model systems. I. Effect of water on the autocatalyzed oxidation. J. Food Sci. 22:878. Mann, E. J., 1966, Major development in the dairy technology. Dairy Ind. 22:457. Mann, E. J., 1967, Freeze-drying of dairy products. Dairy Ind. 22:452. Marco, G. J., 1968, A rapid method for evaluation of antioxidants. J. Am. Oil. Chem. Soc. 22:594. Mather, R. W. and Hollender, H. A., 1955, The influ- ence of some surfactants upon the self dispersion and churning of Whole milk powder. J. Dairy Sci. 38:217. Matill, H., 1931, Antioxidants and the autoxidation of fats. J. Biol. Chem. 22:141. Martinez-Mateo, J., Choate, L. and Edmondson, L. F., 1958, Distribution between fat phase and the aque— ous phase of certain antioxidants added to whole milk. J. Dairy Sci. 22:703. McBean, M. G., Pitt, J. I. and Johnson, A. A., 1964, Retention of $02 in fruit tissue during drying. Food Technol. 22:141. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 150. 151. 118 Menting, L. C. and Hoogstad, B., 1967a, Selectivity of carbohydrate films as influenced by their mois- ture content. Experientia 22:738. Menting, L. C. and Hoogstad, B., 1967b, Volatiles retention during the drying of aqueous carbohydrate solutions. J. Food Sci. 22:87. Meyer, R. I. and Jokay, L., 1959, Application of lyophilization to cheese products. I. Some observ- ations on the characteristics and stability of freeze-dehydrated cheddar, brick, Muenster, blue, cream, and cottage cheese. J. Dairy Sci. 22:908. Mickle, J. B., 1965, Influence of emulsifiers and milk solids-not-fat on fat stability in freeze-dried milk products. J. Dairy Sci. 22:779. Mickle, J. B., 1966, Influence of emulsifiers and nonfat milk solids on physical stability of recon- stituted lyophilized milk products. J. Dairy Sci. 49:1117. Moorjani, M. N. and Dani, N. P. Textural and recon- stitutional properties of freeze-dried shrimp. Food Technol. 22:886. Nanz, R. T. and Lachance, 1967, Acceptability of food items developed for space flight feeding. Food Technol. 22:1361. Negoumy-El, A. M. and Hammond, E. G., 1962, Effect of antioxidants on the flavor of cold storage butter. J. Dairy Sci. 22:848. Nelson, W. E., and Winder, W. C., 1963, Effects of some surfactants on the reconstitution characteris- tics of dried whole milk. J. Dairy Sci. 22:594. Nickerson, T. A., Coulter, S. T. and Jenness, R., 1952, Some prOperties of the freeze dried milk. J. Dairy Sci. 22:77. Niel, yan C. B., Kluyver, A. J. and Derx, H. G., 1929, Uber des Buttergroma. Biochemische Zeitschrift 210:234. Noznick, P. P., Bundus, R. H., Eggen, I. B., 1963, Methods of making dried sour cream. U.S. Patent 3,090,688. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 162. 163. 164. 119 Oette, K., 1965, Identification of some lipid per- oxides by thin-layer chromatography. J. Lipid Res. 6:449. Ogden, R. P., 1967, Method of freeze drying liquid milk products. U. S. Patent 3,297,455. Olcott, H. S. and Einset, E., 1958, A weighing method for measuring the induction period of marine and other oils. J. Am. Oil Chem. Soc. 22:161. Olsen, A. G. and Seltzer, 1941, Preparation of flavoring materials. U. S. Patent 421,988. Pablo, S. S., Sinskey, T. J. and Silverman, G. J., 1967, Selection of micro-organisms due to freeze drying. Food Technol. 22:748. Patton, S., 1954, The mechanism of sunlight in fla- vor formation in milk with special reference to Methionine and Riboflavin. J. Dairy Sci. 22:446. Patton, S., 1955, Browning and associated changes in milk and its products. J. Dairy Sci. 22:457. Patton, S., 1962, Autoxidation of Foods: Dairy Products. In: Symposium on Foods: Lipids and Their Oxidation. AVI Publishing Co., Inc., Westport, Conn., p. 190. Patton, S., Barnes, I. J., and Evans, L. E., 1959, n-Deca-2,4-dienal, its origin from linoleate and flavor significance in fats. J. Am. Chem. Soc. 36:280. Patton, S. and Kutz, G. W., 1951, 2-Thiobarbituric acid, a reagent for detecting milk fat oxidation. J. Dairy Sci. 22:664. Pete, J. W., and Smith, J. A., 1964, The use of antioxidants in dry whole milk. A review. Int. Dairy Fed. Ann. Bull. Pont, E. G., 1955, A de-emulsification technique for use in the peroxide test on the fat of milk, cream, concentrated and dried milks. Aust. J. Dairy Tech. 10:72. Powick, W. C., 1923, Compound developed in rancid fats, with observation on the mechanism of formation. J. Agr. Research 22:823. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 175. 176. 177. 120 Prill, E. A. and Hammer, B. W., 1939, Production of diacetyl from citric acid in butter cultures. J. Dairy Sci. 22:67. Quast, D. G., and Karel, M., 1968, Dry layer permea- bility and freeze drying rates in concentrated fluid systems. J. Food Sci. 22:170. Radanovics, C. and Stine, C. M., 1969, Effects of antioxidants on the storage stability of freeze dried sour cream. Paper No. 164, Institute of Food Technol. 29th Ann. Meeting, Chicago, Ill. Reimenschenider, R. W., Turrer, J. and Speck, R. M., 1943, Modifications of the Swift Stability test. Oil and Soap 22:169. Reineccius, G. A., 1967, Retention of volatile com- pounds during the spray drying of skim milk systems. M.S. thesis, University of Minnesota, Minneapolis, Minnesota. Rey, L. R. and Bastien, M. C., 1962, In: “Freeze Drying of Foods," Natl. Acad. Sci., Natl, Res. Council, Washington, D. C., p. 25. Reynolds, T. M., 1963, Chemistry of nonenzymatic browning 1. Adv. in Food Res. 22:1. Reynolds, T. M., 1965, Chemistry of nonenzymatic browning II. Adv. in Food Res. 22:168. Riel, R. R. and Gibson, C. A., starter distillate for flavoring butter. Technol. 22:137. 1961, The use of Food Riel, R. R. and Sommer, H. H., 1955, Isolation and characterization of some oxidation products asso- ciated with oxidized flavor. J. Dairy Sci. 22:1215. Roubal, W. T., 1967, Oxidative deterioration of flesh lipids of Pacific cod (Gadus macrocgphalus). J. Am. Oil Chem. Soc. 22:325. Emulsion sta- J. Dairy Sci. Rutz, W. D. and Winder, W. C., 1953, bility of cream dried by sublimation. 22:566. Saleh, B. A., Silverman, G. J., and Goldblith, S., 1966, Microbial evaluation of commercial freeze dried foods. Food Technol. 22:671. 178. 179. 180. 181. 182. 183. 184. 185. 186. 187. 188. 189. 190. 121 Salwin, H., 1959, Defining minimum moisture require- ments for dehydrated foods. Food Technol. 22:594. Saravacos, G. D., 1965, Freeze-drying rates and water sorption of model food gels. Food Technol. 19:625. Saravacos, G. D., 1967a, Freeze-drying using molecu- lar sieve adsorbants. Food Technol. 22:187. SaravaCOS, G. D., 1967b, Effect of the drying method of the water sorption of dehydrated apple and potato. J. Food Sci. 22:81. Saravacos, G. D., and Moyer, J. C., 1968, Volatility of some aroma compounds during vacuum-drying of fruit juices. Food Technol. 22:623. Schultz, M. E., varieties for freeze drying. 559. 1964, Suitability of different cheese Milchwissenshaft l9: Schwartz, D. P., Gould, J. A. and Harper, W. J., 1956, The milk lipase system. I. Effect of time, pH, and concentration of substrate activity. J. Dairy Sci. 22:1364. Scott, packet for in-package deoxigenation. 15:99. D. and Hammer, F., 1961, Oxygen scavenging Food Technol. Scott, K. R., Tape, N. W., and Aref, M. M., 1967, Pilot plant unit for fluidized-bed freezing and drying. Food Technol. 22:1090. Seitz, E. W., Sandine, W. E., Elliker, P. R., and Day, E. A., 1963, Studies on the diacetyl biosyn- thesis by Streptococcus diacetilactis. Canad. J. Microbiol. 2:431. Shackell, L. F., 1909, An improved method of de- siccation with some applications to biological prob- lems. Am. J. Physiol. 22:325. Shipe, W. F., 1964, Oxidation in the dark. Sci.‘2l:221. J. Dairy Schock, A., and Breazegle, D. F., 1948, Cultured buttermilk from non-fat dry milk solids. S. Dak. Agr. Expt. Sta. Bull. 391. 191. 192. 193. 194. 195. 196. 197. 198. 199. 200. 201. 122 Sinnamon, H. I., Komanowsky, M., and Heiland, W. K., 1968, An experimental apparatus for drying particu- late foods in air. Food Technol. 22:219. Sinskey, T. J., McIntosh, A. M., Pablo, T. S., Silverman, G. J., and Goldblith, S. A., 1964, Con- siderations in the recovery of micro-organisms from freeze dried foods. Health Lab. Sci. 2:297. Sivetz, M. and Foote, M. E., 1963, Coffee Processing Technology: Vol. 1, p. 383. AVI Publishing Co., Westport, Conn. Stein, M., 1966, Freeze drying the present position surveyed. Food Manufacture 22:51. Stine, C. M., Harland, H. A., Coulter, S. T., and Jenness, R., 1954, A modified peroxide test for de- tection of lipid oxidation in dairy products. J. Dairy Sci. 22:202. Stine, C. M., 1957, A study of lipid oxidation in dry whole milk. Ph.D. thesis, University of Min- nesota, Minneapolis. Stine, C. M., Hamilton, M. O. and Radanovics, 1968, Freeze-dried sour cream. Annual Report on Project 71-2546. American Dairy Association. Stine, C. M., Hamilton, M. 0., and Radanovics, C., 1969, Freeze-dried sour cream. Annual Report on Project 71-2546, American Dairy Association. Stuckey, B. N., Sherwin, E. R. and Hannah, F. D. Jr., 1958, Improved techniques for testing fats and oils by the oxygen bomb method. J. Am. Oil Chem. Soc. 22:581. Stull, J. W., 1953, The effect of light on acti- vated flavor development and on the constituents of milk and its products. A review. J. Dairy Sci. 22:1153. Stull, J. W., Herreid, E. O. and Tracy, P. H., A study of the use of the antioxidant Nordihydro- guiaretic acid in dairy products. 1948, I. Its antioxygenic properties in milk. J. Dairy Sci. 22:449. 1948, II. Its antioxygenic properties in un- sweetened frozen cream. J, Dairy Sci. 22: 1024. 202. 203. 204 . 205. 206. 207. 208. 209. 210. 123 1949, III. Its antioxygenic properties in sweetened frozen cream. J. Dairy Sci. 22: 301. 1951, IV. Its antioxygenic properties in spray dried whole milk and ice cream mix with and without added synergist. J. Dairy Sci. 22:80. Supplee, C. G., 1923, Keeping quality of dry milk. Proc. World's Dairy Cong. 1248. Swartling, P. and Johansson, 8., 1954, Treatment of winter cream at different temperatures with regard to butter aroma. Suenska Mejeritidin 46:309 (Chem. Abstracts 22:9574C) Tarrasuk, N. P. and Jack, E. L., 1948, A study of the browning reaction in whole milk powder and ice cream powder. J. Dairy Sci. 22:255. Taufel, K., 1957, Fette-Seifen-Anstrichmittel 59: 87. Referred in: Lundenberg, W. 0., Oxidative Rancidity in food fats and its prevention, "Autoxi- dation and Antioxidants, Vol. II, Interscience Pub., New York. Tamsma, A., Kurtz, F. E., and Pallansch, M. J., 1967, Effect of oxygen removal technique on flavor stability of low-heat foam sprayed dried whole milk. J. Dairy Sci. 22:1562. Tamsma, A., Mucha, T. J. and Pallansch, M. J., 1963, Factors related to the flavor stability during storage of foam-dried whole milk. III. Effect of antioxidants. J. Dairy Sci. 22:114. Tamsma, A. and Pallanasch, M. J., 1964, Factors re- lated to the storage stability of foam dried whole milk. IV. Effect of powder moisture content and in-pack oxygen at different storage temperatures. J. Dairy Sci. 22:970. Tamsma, A., Pallanasch, M. J., Mucha, T. J. and Patterson, W. I., 1961, Factors relating to flavor stability of foam dried whole milk. I. Effect of oxygen level. J. Dairy Sci. 22:1644. Thomas, A., 1954, A study of the mechanism of sulf- hydryl action in milk. Ph.D. thesis, University of Minnesota, Minneapolis. 211. 212. 213. 214. 215. 216. 217. 218. 219. 220. 221. 222. 124 Thompson, C. R., alfalfa meal, effect of antioxidants. Chem. 22:922. 1950, Stability of carotene in Ind. Eng. Tuomy, J. M., 1965, Development of dehydrated meat and fish salads for military use. Food Technol. 19:936. Tuomy, J. M., Stentz, R. L., and Helmer, R. L., 1968, Effect of processing conditions in the quality of cooked, sliced, freeze-dried beef. Tech. Rep. 68-53-FL. U.S. Army Natick Lab., Natick, Mass. Uri, N., 1956, Metal ion catalysis and polarity of environment in the aerobic oxidation of unsaturated fatty acids. Nature 177:1177. Weinstein, B. R., and Trout, G. M., 1951, The solar— activated flavor of homogenized milk. II. The role of oxidation and the effectiveness of certain treat- ments. J. Dairy Sci. 22:559. Westerfeld, W. W., tion of blood acetoin. 1945, A colorimetric determina- J. Biol. Chem. 161:465. Wheeler, D. H., 1932, Peroxide formation as a meas- ure of autoxidative deterioration. Oil and Soap 9:89. Wilkinson, R. A., 1964, Theories of the mechanism of oxidized flavor development in dairy products. In- ternal Report No. 4, Division of Dairy Research, Commonwealth Scientific and Industrial Research Or- ganization, Melbourne, Austr. Wishner, L. A., 1964, Light induced oxidation in milk. J. Dairy Sci. 22:216. Yu, T. C. and Sinnhuben, N. 0., 1962, Removal of in- terfering pigment in determining malonaldehyde by the thiobarbituric acid reaction. Food Technol. 2226):115. Zabik, M. E., 1968, Comparison of frozen, foam-spray- dried, freeze dried, and spray-dried eggs. Food Technol. 22:1465. Zabik, M. E., and Figa, J. E., 1968, Comparison of frozen, foam-spray dried, freeze dried and spray- dried eggs. Food Technol. 22:1169. 125 223. Zamzow, W. M., and Marshall, W. R., Jr, 1952, Freeze drying with radiant energy. Chem. Eng. Progress 48:22. APPENDIX 126 Table lO.--PerOxide values (PV) and intensity of oxidized flavor (OF) of freeze-dried sour cream in storage Storage Sample Number Time 2 3 (Months) . PV OF PV OF PV OF PV OF Air Packed, Stored at 40°F Initial 0.11 - 0.09 - 0.06 - 0.09 - 1 0.15 - 0.12 - 0.08 - 0.12 - 2 0.26 - 0.15 - 0.12 - 0.15 - 3 0.27 - 0.20 - 0.12 - 0.17 - 4 _ _ _ - 5 0.37 - 0.26 — 0.16 - 0.18 - 6 0.46 + 0.35 - 0.20 - 0.25 — 7 + 0.40 - 0.22 - — 8 0.60 + 0.48 + 0.20 + 0.30 + 9 0.60 ++ 0.50 + 0.22 - 0.29 - 10 0.59 ++ 0.45 + 0.18 - - 11 0.62 ++ ++ ++ 0.30 + 12 0.64 +++ 0.46 ++ 0.20 + 0.28 + Air Packed, Stored at 72°F Initial 0.11 - 0.09 - 0.06 - 0.09 - 1 0.15 - 0.12 - 0.08 - 0.12 - 2 0.24 - 0.15 - 0.12 - 0.15 - 3 0.27 — 0.20 - 0.12 — 0.17 - 4 _ _ _ - 5 0.26 + 0.15 + 0.10 - 0.12 - 6 0.30 + 0.17 - 0.12 - 0.16 + 7 ++ - - - 8 0.34 ++ 0.19 + 0.16 - 0.22 + 9 0.33 +++ 0.24 + 0.15 + 0.15 - 10 0.38 +++ 0.26 + 0.16 + 0.17 ++ Air Packed, Stored at 99°F Initial 0.11 0.09 0.06 0.09 1 0.18 + 0.13 + 0.12 + 0.11 + 2 0.17 ++ 0.11 ++ 0.11 ++ 0.10 + 3 0.27 +++ 0.17 ++ 0.14 ++ 0.14 +++. 4 +++ +++ +++ 5 0.16 +++ 0.09 +++ 0.08 +++ 0.09 +++ 6 0.32 +++ 0.11 +++ 0.10 +++ 0.12 +++ - no oxidized flavor + slight oxidized flavor ++ moderate oxidized flavor +++ objectionable oxidized flavor 127 Table ll.--Peroxide values (PV) and intensity of oxidized flavor (OF) of freeze-dried sour cream in storage Stor- Sample Number age Time 3 (“‘05- 1 PV OF PV OF PV OF PV OF PV OF Air Packed, Stored at 40°F Ini- tial 0.09 - 0.09 - 0.07 - 0.10 — 0.10 1/2 0.18 - 0.15 — 0.14 - 0.17 — 0.16 1 0.31 - 0.29 - 0.23 - 0.26 - 0.21 2 0.44 - 0.45 - 0.33 - 0.39 - 0.28 3 _ _ .. _ 4 0.60 - 0.63 — 0.45 — 0.40 - 0.23 5 0.74 - 1.04 - 0.43 - 0.57 - 0.31 6 0.92 + 0.87 + — - 7 1.05 + + 0.45 - 0.71 — 0.37 9 + 1.20 + 0.60 + + 10 1.28 ++ + 0.56 + 0.83 + 0.40 11 1.34 ++ 1.27 ++ 0.57 + 0.94 + 0.43 Air Packed, Stored at 72°F Ini- tial 0.09 - 0.09 - 0.07 - 0.10 — 0.10 1/2 0.20 — 0.20 - 0.09 — 0.12 - 0.12 1 0.23 - 0.28 - 0.17 - 0.21 - 0.16 2 0.28 - 0.39 — 0.26 - 0.25 - 0.20 3 - - + - 4 0.31 + 0.38 + 0.20 - 0.21 - 0.15 5 0.39 + 0.43 + 0.22 + 0.26 - 0.18 6 0.40 ++ 0.44 ++ ++ + 7 ++ ++ 0.30 + 0.22 ++ 0.18 8 0.41 +++ 0.45 +++ 0.34 +++ 0.25 +++ 0.17 128 Table 11.--Continued Stor- Sample Number age Time 1 2 3 4 5 (mos.) pv OF PV OF PV OF PV OF PV OF Air Packed, Stored at 99°F Ini- tial 0.09 - 0.09 - 0.07 0.09 - 1/2 0.11 - 0.15 - 0.08 0.10 - 0.09 - 1 0.18 - 0.22 - 0.10 0.12 - 0.12 ~ 2 0.19 + 0.26 + 0.13 + 0.14 + 0.13 + 3 ++ ++ ++ + 0.13 + 4 0.17 +++ 0.32 +++ 0.13 +++ 0.15 +++ 0.11 +++ 5 0.25 +++ +++ 0.13 +++ 0.13 +++ 0.14 +++ - no oxidized flavor + slight oxidized flavor ++ moderate oxidized flavor +++ objectionable oxidized flavor 129 Table 12.--Peroxide alues (PV) and intensity of oxidized flavor (OF) of freeze-dried sour cream in storage Stor- Sample Number age Time 1 2 3 4 5 (mos.) PV OF PV OF PV OF PV OF PV OF Air Packed, Stored at 40°F Ini- tial 0.10 - 0.12 - 0.11 - 0.09 0.14 — 1/2 0.55 - 0.34 - 0.38 - 0.27 0.27 — l 0.56 - 0.38 - 0.25 - 0.41 0.33 - 2 0.95 - 0.60 - 0.66 - 0.40 0.37 - 3 1.46 - 0.95 - 0.95 - 0.43 0.42 - 4 _ .. .. _ 5' 1.55 - 1.08 - 1.13 - 0.51 0.43 + 6 - - _ - 7 1.34 + 1.25 + 1.27 - 0.52 0.55 + 8 1.89 ++ 1.47 + 1.52 + 0.54 0.60 + 9 ++ + + + 10 2.08 ++ 1.61 ++ 1.66 ++ 0.67 0.71 + Air Packed, Stored at 72°F Ini- tial 0.10 - 0.12 - 0.11 - 0.09 0.14 - 1/2 0.45 - 0.09 - 0.29 - 0.25 0.24 - 1 0.50 - 0.14 - 0.26 - 0.22 0.20 - 2 0.48 - 0.33 - 0.25 - 0.22 0.24 - 3 0.85 - 0.46 - 0.45 - 0.25 0.23 - 4 + - - - 5 0.87 + 0.59 - 0.48 - 0.29 0.25 - 6 ++ + + - 7 1.00 ++ 0.56 + 0.57 + 0.27 0.27 - 8 1.07 +++ 0.59 + 0.60 + 0.28 0.26 + Tabl 130 e 12.--Continued Stor- Sample Number age Time 1 2 3 (mos . ) pv OF PV OF PV OF PV OF PV OF Air Packed, Stored at 99°F Ini- tial 0.10 - 0.12 — 0.11 - 0.09 - 0.14 - 1/2 0.30 - 0.18 — 0.15 — 0.18 - 0.17 - 1 0.33 + 0.21 + 0.16 + 0.13 + 0.17 + 2 0.29 ++ 0.27 ++ 0.14 + 0.15 ++ 0.19 ++ 3 ++ ++ ++ ++ 4 0.30 +++ 0.25 +++ 0.17 ++ 0.14 +++ 0.19 +++ 5 0.40 +++ 0.49 +++ 0.18 +++ 0.16 +++ 0.31 +++ + ++ +++ no oxidized flavor slight oxidized flavor moderate oxidized flavor objectionable oxidized flavor GQN STRTE UNIV LIB IIIIIIIIIIIIlsIIalIIIIIIII IIIIIIIIIIIIIILII|7II|I8IIII|