_’ ‘ u-"- l‘llllllllllll V 8 LleeARY I Michigan State University This is to certify that the thesis entitled INTERACTION 0F POLYMERIC PACKAGING MATERIALS WITH FLAVOR COMPONENTS FROM AN ONION/GARLIC FLAVORED SOUR CREAM presented by JANET MAR I E TOEBE has been accepted towards fulfillment of the requirements for M . S . degree in PAC KAG I NC W A). ,L/M Bruce R. Harte, Ph.D. Major professor Date December 3, I987 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution IVIESI.I RETURNING MATERIALS: Place in book drop to LIBRARJES remove this checkout from AnunzjnzlL. your record. FINES will be charged if book is returned after the date stamped below. EBB [)7 I334 I. v ' , . . {xiii}- ’l ‘ JAN 2 4 ZUIO 052410 INTERACTION OF POLYMERIC PACKAGING MATERIALS WITH FLAVOR COMPONENTS FROM AN ONION/GARLIC FLAVORED SOUR CREAM BY Janet Marie Toebe A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Packaging 1987 Copyright by Janet Marie Toebe 1987 ABSTRACT INTERACTION OF POLYMERIC PACKAGING MATERIALS WITH FLAVOR COMPONENTS FROM AN ONION/GARLIC FLAVORED SOUR CREAM BY Janet Marie Toebe Permeation and sorption studies were performed on a high impact polystyrene (HIPS) package containing onion/garlic flavored sour cream. For comparison. sorption studies were also conducted using high density polyethylene (HDPE) and polypropylene (PP). Dimethyl disulfide and dipropyl disulfide were chosen as probe compounds. Permeation studies were conducted using a quasi-isostatic technique. while a gravimetric method was used for sorption studies. No detectable permeation of probe compounds through the HIPS container was observed. Sorption studies show the solubility of probe compounds in HIPS to be substantially higher than in HDPE or PP. A model was developed to predict sorption of probes in the actual product. Based on this model. none of the three test materials would reach saturation levels during storage of the product. Sensory analysis showed that panelists were unable to detect off odors or flavors in unflavored products stored adjacent to the onion/garlic flavored sour cream. DEDICATION This thesis is dedicated to my parents in appreciation of their love and acceptance. Also. to Beth Waggoner whose friendship was a source of support and inspiration throughout this work. iv ACKNOWLEDGMENTS I would like to thank Dr. Bruce Harte for his support and guidance while acting as my advisor. Appreciation is expressed to Dr. Jack Giacin. Dr. Mary Ann Filadelfi. and Dr. Ian Gray for serving on the guidance committee. A very special thank you is extended to Heidi Hoojjat and Ruben Hernandez. Their patience and technical support proved to be invaluable throughout this research study. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES INTRODUCTION LITERATURE REVIEW Sorption and Permeation Behavior of Food Contacting Polymers High Impact Polystyrene: Characteristics Onion Flavor Components Methods of Analysis Sensory Analysis MATERIALS AND METHODS Materials: Product/Package Probe Compounds Experimental Methods: Extraction Methods Sorption Measurements Equilibrium Vapor Pressure Permeation Measurements Analytical: Gas Chromatography Gas Chromatography/Mass Spectrometry Sensory Analysis RESULTS AND DISCUSSION Product Characterization Permeation Sorption Equilibrium Vapor Pressure Absorption of Probe Compounds as a Function of Time Sensory Analysis CONCLUSION vi Page vi viii 45 45 46 49 60 61 75 77 APPENDICES Appendix I: Appendix II: Appendix III: Appendix IV: Appendix V: Appendix VI: Appendix VII: Appendix VIII: Appendix IX: BIBLIOGRAPHY Properties of High Impact Polystyrene Approximate Composition of Fresh and Dehydrated Onion Physical Characteristics of Toasted. Chopped. Dehydrated Onions Specifications for 8 02 High Impact Polystyrene and Polypropylene Tubs Polypropylene and High Density Polyethylene Pr0perties GC/MS Total Ion Scan of Dehydrated Onion Powder Extract Probe Compound Properties Standard Curve for Dimethyl disulfide Standard Curve for Dipropyl disulfide Consent Form for Taste Panel Members Questionnaire Questionnaire GC Analysis of Onion Flavored Sour Cream Extract Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide -- Total loss Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide -— Loss due to extraction Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide -- Loss due to concentration Concentration of Probe Compounds in Onion Flavored Sour Cream Determination of T2 vii Page 81 83 84 85 86 87 88 89 90 91 92 93 94 95 95 96 96 97 98 Table 10 11 12 13 LIST OF TABLES Recoveries of Dimethyl disulfide and Dipropyl disulfide from Onion Flavored Sour Cream Time Required for HIPS Samples to Reach Maximum Absorption Before Dissolving in Penetrant Solutions Equilibrium Sorption Rates for PP Samples Exposed to Penetrant Vapor Absorption Capacity of HIPS. PP. and HDPE and Available Probe Compound in Onion Flavored Sour Cream Sensory Evaluation Contingency Table Properties of High Impact Polystyrene Approximate Composition of Fresh and Dehydrated Onion Physical Characteristics of Toasted. Chopped. Dehydrated Onions Specifications for 8 oz High Impact Polystyrene and Polypropylene Tubs Polypropylene and High Density Polyethylene Properties Probe Compound Properties Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide -— Total loss Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide -- Loss due to extraction viii Page 47 50 59 74 75 81 83 84 85 86 88 95 95 Table Page 14 Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide -— Loss due to concentration 96 15 Concentration of Probe Compounds in Onion Flavored Sour Cream 96 16 Determination of T2 97 ix LIST OF FIGURES Figure 1 Tekmar Gas Flow System 2 Quartz Spring Sorption Apparatus 3 Electrobalance Sorption/Desorption Apparatus 4 Modified Likens - Nickerson Distillation Apparatus S Permeation Cell 6 Permeation Cell Integrity Test 7 Absorption/Desorption of Dipropyl disulfide by HIPS and HDPE 8 Absorption of Dipropyl disulfide by HIPS 9 Absorption/Desorption of Dimethyl disulfide by HIPS and HDPE 10 Absorption of Dimethyl disulfide by HIPS 11 Absorption/Desorption of Dipropyl disulfide by PP 12 Absorption/Desorption of Dimethyl disulfide by PP 13 Equilibrium Vapor Pressure of Dimethyl disulfide as a Function of Temperature 14 Equilibrium Vapor Pressure of Dipropyl disulfide as a Function of Temperature 15 Equilibrium Vapor Pressure of Dimethyl disulfide as a Function of Simulant Probe Concentration Page 16 19 20 32 38 39 51 53 54 56 57 58 62 63 64 Figure . 16 17 18 19 20 21 22 23 24 25 26 27 Equilibrium Vapor Pressure of Dipropyl disulfide as a Function of Simulant Probe Concentration Equilibrium Vapor Pressure of Dimethyl disulfide in Onion Flavored Sour Cream as a Function of Temperature Equilibrium Vapor Pressure of Dipropyl disulfide in Onion Flavored Sour Cream as a Function of Temperature Absorption of Dimethyl disulfide by HIPS Function of Temperature Absorption of Dipropyl disulfide by HIPS Function of Temperature Absorption of Dimethyl disulfide by HIPS Function of Time Absorption of Dipropyl disulfide by HIPS Function of Time GC/MS Total Ion Scan of Dehydrated Onion Powder Extract Standard Curve for Dimethyl disulfide Standard Curve for Dipropyl disulfide Consent Form for Taste Panel Members GC Analysis of Onion Flavored Sour Cream xi 888 888 888 888 Extract Page 65 67 68 69 7O 72 73 87 89 9O 91 94 INTRODUCTION Plastics are enjoying widespread popularity in food packaging. The value of all rigid and semi-rigid plastic containers manufactured in the U.S. was 3.77 billion dollars in 1985 (Friedman. 1985). The properties which make plastics distinct from other materials also make them attractive alternatives for use with many food products. Plastics offer several advantages to the consumer and manufacturer. Improved processing methods continue to decrease costs. while an increasing variety of plastics is available to suit specific needs. Other important advantages of plastics include resistance to breakage and its lightweight (as compared to glass and metal) (Hanlon. 1984). With the increased use of plastics. a need to characterize interactions between polymeric packages and the contents of these packages has arisen. Possible alteration of product quality and polymer properties may result from these interactions. Gaining a better understanding of the phenomena involved will aid in the appropriate selection of packaging systems to optimize product quality. The aroma barrier properties of a package system are important because retention of aroma and flavor constituents and exclusion of off odors or flavors is directly related to product quality and shelf life. Loss of volatile aroma or flavor constituents may occur by two packaging related mechanisms: (1) The mass transfer of vapor into or out of the package. (2) Sorption of organic volatiles by the packaging material. This study was designed to determine the specific interactions between a package (polystyrene container) and product whose quality is associated with retention of volatile flavor and aromas. The product studied was an onion/garlic flavored sour cream dip susceptible to the following package interactions: (1) Loss of flavor or aroma due to permeation through the package or sorption by the package wall. (2) Absorption of off odors or flavors from products stored in the same area. To monitor permeation and sorption of volatile constituents. probe compounds were selected based on their contribution to the aroma and flavor profile of the product. Dipropyl disulfide and dimethyl disulfide were used as probes because of their prominence in the aroma/flavor profile of the product. their ease of analysis. and availability. A quasi—isostatic method of analysis was developed to determine the transmission rate of probe compounds through the test material (high impact polystyrene). To quantify permeating volatile constituents. gas chromatographic analysis was performed on samples taken from the free volume of the permeation cell. Sorption of probe compounds by the test materials (high impact polystyrene. high density polyethylene. and polypropylene) was determined using a gravimetric technique. Studies were conducted over a range of temperatures for each test material and probe compound. Desorption of probe compounds by each test material was also determined. Sensory analysis was conducted to evaluate the extent of off flavor absorption by unflavored sour cream samples which had been stored under controlled conditions in close contact with onion/garlic flavored sour cream. LITERATURE REVIEW SORPTION AND PERMEATION BEHAVIOR OF FOOD CONTACTING POLYMERS The ability of a packaging system to prevent loss of volatile. low molecular weight organic compounds from foodstuffs directly influences product quality and shelf life. Organic vapors such as aroma compounds often exist in food products at low concentrations. yet contribute extensively to the product flavor profile (Parliment. 1987). Loss of these constituents may adversely affect flavor intensity. Knowing the solubility of flavor and aroma compounds in polymer structures is essential to avoid "flavor scalping" or loss due to sorption (Hernandez et al.. 1986). Since volatile aroma compounds may normally exist at very low concentrations in foodstuffs. there is the potential for loss of flavor and aroma constituents due to absorption by the contacting packaging material (i.e.. solubilization). Further. knowledge of both permeation and sorption more completely describes the transport properties of volatile aroma constituents to polymeric packaging materials. The importance of both permeability and sorption to product quality is emphasized in recent studies describing penetrant/ polymer interactions (Berens. 1978: DeLassus. 1985: Mohney. 1986). It is possible for sorption interaction to be important and sometimes even dominant in food/polymer interactions (DeLassus. 1985). Limonene has often been used as a probe compound in sorption/permeation studies. Limonene is a common flavor component present in citrus products and has a relatively high solubility in polyolefins (Mohney. 1986). DeLassus (1985) determined the effect of barrier location on sorption and permeation using limonene as the permeant compound. He found that the sorption interaction could be minimized by placing the barrier layer on the inside surface of the container. Hirose (1987) studied the influence of limonene absorption on the mechanical and barrier properties of polyethylene and two types of SurlynO. Absorption of limonene was found to affect modulus of elasticity. tensile strength. percent elongation. seal strength. impact resistance. and oxygen permeability of the film. Mohney (1986) performed permeability and solubility tests on cereal package liners using limonene as the penetrant vapor. Both permeability and sorption were important in loss of limonene vapor from the intact package. Berens (1978) found that absorption of limonene by polyolefin structures caused swelling of the polymer matrix. This swelling altered diffusion. permeability. and solubility properties of the material. Unlike inert gases such as oxygen and carbon dioxide. organic vapors exhibit non-ideal diffusion and solubility due to interaction with packaging materials. Swelling of the polymer matrix by sorbed organic vapors alters the configuration of polymer chains. resulting in concentration dependent diffusion (Bagley and Long. 1958: Fujita. 1961: Crank. 1975: and Berens. 1977). Permeability of the polymer must be described as a function of penetrant diffusion (D) and penetrant solubility (S). The diffusion coefficient (D) represents a measure of the ease with which a penetrant molecule moves through a membrane. while (8) describes the number of penetrant molecules permeating the barrier (Crank and Park. 1968). For fixed or noninteractive gases. the permeability coefficient (P) is related to two fundamental mass transfer parameters by: F=st This relationship is not applicable to organic permeants or multilayer laminant structures. Concentration dependency caused by swelling of the polymer matrix results in non-ideal diffusion and solubility properties (Bagley and Long. 1958: Fujita. 1961: Crank. 1975: and Berens. 1977). For such systems it is necessary to characterize diffusion. solubility. and permeability behavior to completely describe mass transfer between organic vapors (such as aroma compounds) and polymeric barrier structures. HIGH IMPACT POLYSTYRENE: CHARACTERISTICS High impact polystyrene is used extensively for protecting. storing. and serving many different food products (Monte and Landau—West. 1982). The wide range of physical properties and available formulations make polystyrene an attractive alternative to more expensive resins (Swett. 1986). In Appendix I. Table 6. the physical properties of HIPS are shown. Polystyrene is the most widely used resin in packaging with an estimated 838MM 1b consumed or 46% of the 1985 total resin consumption for miscellaneous plastic containers. A major portion of this resin is converted into tubs used to package such foods as cottage cheese. sour cream. margarine. and delicatessen foods (Rauch. 1985). Impact polystyrene has most of the advantages of polystyrene such as rigidity. ease of fabrication. and color and granule size availability. In addition. impact polystyrene has proven to be both tough and resistant to abuse (Herman et al.. 1964). To enhance its toughness. high impact polystyrene usually contains 5-10% polybutadiene or styrene—butadiene copolymer. Bergen (1968) investigated the effect of solvent contact on stress cracking in styrene polymers. Stress cracking can be defined as external or internal cracks in a plastic caused by tensile stresses less than that of its short-time mechanical strength. The development of such cracks is frequently accelerated by the environment to which the plastic is exposed (Beckman et al.. 1979). To reduce stress cracking and improve overall strength. Bergen suggested use of a rubber copolymer with a particle size of 2-5 microns. Some contacting substances may cause substantial change in the structure and properties of HIPS. vom Bruck et al. (1981) found that interactions between HIPS and fat containing foods are so intense that under stress the containers often suffer from stress cracking. In vom Bruck's et al. research the presence of stress cracks indicated a definite interaction with a food component (i.e.. fat). The reverse situation. no stress cracking. no interaction. cannot be assumed. Monte and Landau-West (1982) have shown certain foods to be incompatible with polystyrene. This incompatibility was due to the dissolution of polystyrene by certain essential oils present in the foods. Phillips (1979) inferred that polystyrene was dissolved by the essential oils in a lemon flavored tea to such an extent that those who drank this mixture from a polystyrene container “will also consume an appreciable amount of the container itself in solubilized form." Monte and Landau—West (1982) found that citronella. terpinene. and limonene (constituents of many flavor oils) were excellent solvents for polystyrene. At room temperature these compounds solubilized almost half a gram of polystyrene per gram of solvent. D-limonene was one of the most active of the solvents tested. Lemon oil in low concentrations such as in lemon tea. was absorbed almost completely by the contacting polystyrene. Rapid loss of certain flavor components from orange juice packaged in polystyrene was shown by Durr et al. (1981). Orange juice packaged in glass experiences a much lower rate of flavor loss. ONION FLAVOR COMPONENTS The flavor compounds found in onion are extremely complex. Abraham et a1. (1976) described a total of 74 compounds identified in fresh and processed onion. Members of the genus Allium (onions. garlic. leeks. chives. etc.) possess strong characteristic aromas and flavors not found in other vegetables. Stoll et al. (1951) found that the characteristic odor of onion is produced by volatiles enzymatically produced when injury occurs. Most members of this genus have no odor unless the plant tissue is cut or damaged in some way. The most important volatile flavor and aroma constituents of onion have been identified as sulfur containing compounds (Wahlroos and Virtanen. 1965; Bernhard. 1966). 10 Whitaker (1976) noted that 26—34 sulfur compounds have been isolated from intact onion. Of these. 8 account for 90% of the total sulfur content of onion. Saghir et a1. (1964) and Bernhard (1968). found that the di— and tri-sulfides are the primary flavor components of the Allium family. Bernhard (1968). was able to quantify the disulfide volatiles in fresh onion and reported the results in descending order of concentrations: di—n-propyl. n-propyl allyl. methyl—n-propyl. methyl allyl. dimethyl. and diallyl. Bernhard (1968) found this order to be markedly altered in dehydrated onion. Methyl-n-propyl was the principal disulfide. followed by dimethyl. methyl allyl. di-n—propyl. n-propyl allyl. and diallyl. Not only was the prominence of each sulfur compound altered. but the total quantity of volatiles was significantly decreased. Total disulfide loss ranged from 89.0% to 99.70%. Loss of all volatiles (including disulfide) averaged 98% (Bernhard. 1968). Loss in flavor intensity of dehydrated onion has been directly related to the decrease in concentration of the highly odorous dipropyl disulfide (Mazza et al.. 1979: Yagami et al.. 1980). Even with such significant losses. there is still a sufficient concentration of volatile components remaining in the dehydrated product to be readily detected by the human nose. 11 In Appendix II the approximate composition of fresh and dehydrated onion (Table 7) and physical characteristics of toasted chopped dehydrated onion (Table 8) are shown. The major loss of volatiles from dehydrated onion occurs during the slicing and/or chopping operations and is not associated with drying (Stephenson. 1949). Mazza and LeMaguer (1978) found that dipropyl disulfide was lost almost exclusively after chopping and before the onion reached a critical moisture content during dehydration. Many factors influence the retention and behavior of onion volatiles. Processing and initial concentration of solids and volatiles are among these factors (Mazza and LeMaguer. 1978). Several studies have investigated the effect of processing methods on dehydrated onion and garlic products (Pruthi et al.. 1959: Bhatti and Asaghar. 1965: Mazza and LeMaguer. 1978). Accelerating the drying rate and increasing air temperature were found to enhance percent volatile retention. Saimbhi et al. (1970) studied the effect of onion variety on dehydration traits and concluded that a fresh onion with a high degree of pungency. good flavor. pure white flesh. and a low drying ratio (pounds of fresh produce to 1 lb dry) was the best candidate for dehydration. Several methods have been developed to identify volatile components in onion. Mazza et a1. (1980) used headspace sampling of dehydrated onion and gas chromatographic 12 analysis. All major volatile constituents found in fresh onion were retained in the dehydrated sample. Constituents with low retention times (12-15 minutes) were difficult to detect. Dipropyl disulfide was not detectable unless sufficiently rehydrated prior to analysis. Loss of dimethyl disulfide was not as great as loss of other volatiles. It remains in amounts adequate to become the dominant disulfide component of dehydrated onion (Bernhard. 1968). The measured total disulfide content of fresh and dehydrated onion differs by 89%. However. this has not adversely affected the use of dehydrated onion. They are sufficiently flavored to be used in place of raw onions by most food processors (Bernhard. 1968). METHODS OF ANALYSIS The fraction responsible for the aroma and flavor of foods is often composed of very low levels of compounds with widely different polarities. solubilities. and volatilities. Difficulties in analysis may result due to volatile levels which may be in the subparts per million (ppm) range. Presence of soluble and insoluble solids. as well as lipid materials. and large amounts of water may add to the problems of sample collection and preparation (Parliment. 1987). Early studies used direct 13 gravimetric analysis to measure loss of odor and flavor compounds from products packaged in plastic lined containers (Becker et al.. 1983). In the last two decades much more effective means of flavor and aroma analysis have been developed. Sample preparation techniques for flavor analysis can be classified as follows (Parliment. 1987): (1) Headspace sampling (2) Headspace concentrating (3) Distillation/extraction (4) Direct analysis of aqueous samples (5) Direct adsorption of aqueous samples (6) Direct extraction of aqueous samples A method commonly used to separate volatile compounds from foods and beverages is steam distillation followed by extraction wih an organic solvent (Parliment. 1987). The Likens-Nickerson extraction apparatus can be used to continuously and concurrently steam distill and solvent extract organic volatiles. A volatile aroma concentrate is produced which can be further concentrated for gas chromatographic analysis. Great care must be taken when concentrating the sample since losses of low—boiling volatiles and concentration of solvent impurities may result (Parliment. 1987). Mazza et al. (1980) found that headspace sampling. when compared to solvent extraction procedures. does not reliably quantify the higher boiling 14 point constituents of onion. The level of aroma compounds in the gaseous environment over foods is often very low. Preconcentration is often required to achieve suitable analytical sensitivities (Olafsdottir et al.. 1985). This is particularily necessary if onions have been pickled. canned. boiled. fried. dehydrated. or freeze—dried. since such processing is accompanied by considerable flavor loss (Freeman and Whenham. 1974). Adsorption of volatiles onto a porous solid is a commonly used procedure to concentrate samples for analysis by gas chromatography. Tenax—GC (2-6 diphenyl-paraphenylene oxide polymer) has emerged as a widely used porous polymer for these applications (Olafsdottir et al.. 1985). A headspace concentration technique was developed by Olafsdottir et a1. (1985) which employed Tenax-GC adsorption of volatile compounds followed by ether extraction. Wyatt (1986) used this same technique to study off odors in foods which were present due to migration from contacting polymeric packages. Using new instrumentation. concentration of samples can be fully automated. A dynamic headspace analysis technique has been developed which significantly enhances recovery of volatiles. The Tekmar Model 4000 concentrator system (Tekmar Co.. Cincinnati. OH) is designed for this application (Westendorf. 1984). Figure 1 presents a general schematic of the Tekmar gas flow system. 15 In dynamic headspace analysis the sample is purged with an inert gas (i.e.. bubbling through a liquid sample or sweeping the headspace of a solid sample) driving volatiles into the gas phase. The purge gas is then passed through a short column containing a porous polymer adsorbent (typically Tenax-GC). The adsorbent selectively retains the sample compounds while allowing the purge gas and any water vapor to pass through. By purging in this manner. the entire organic contents of the vapor phase can be analyzed by gas chromatography. Further partitioning of the volatiles into the vapor state is accomplished by continual removal of the purge gas from above the sample. After the purge step is completed the adsorbent column is heated to desorb the organic compounds. The sample is then backflushed to the GC via a 6-port valve. A cryogenic trap is used to focus the sample prior to injection into the capillary column. Utilizing this method. detection limits of low part-per—billion levels are obtainable with a good reproducibility. The use of this technique has experienced rapid growth in recent years and is now in routine use in a number of laboratories (Westendorf. 1984). Quantification of aroma and flavor compounds is most commonly achieved by means of flame ionization detector (FID) systems. such as that used in gas chromatographic analysis. Ionization detectors operate on the principle 16 80umwo sue H.om wee awe am.m am.me menueemfia H>nuoewo amouo uaom a“ >uo>ooom a mmoq annoy :ofiumuucoocoo cowuomuuxm canomEou onoum newuwuucoocoo onoum ca 099 moon on can mmoa Emouo uzom couo>mam newco Scum mofiuanmfio axmoumwa new wowuasmwa H>suoefia no mowuo>ooom a oHnt 48 testing period. percent vapor loss continued to increase for all closures. Equilibrium vapor loss was not achieved. therefore it was not possible to determine a reproducible error that would account for leakage. Following permeation cell integrity testing. permeation studies were conducted using the Teflono gasket closure system over a time period of 6 days. The HIPS tub containing the onion/garlic flavored sour cream was placed in the cell at room (24°C/75.2°F) and refrigeration temperatures (SOC/41°F). After 6 days. no measurable permeation of probe compounds at either temperature could be detected by GC analysis. By extending the test period to 8 weeks and preconcentrating the samples using the Tenax procedure. detection of permeant compounds was possible. The peak responses could not be identified as either of the probe compounds; therefore. quantification of permeants was not possible. It was not apparent after the 8 week test as to whether permeation occurred through the container wall or whether there was leakage through the seal area. Distribution time for the sour cream normally requires less than 4 weeks (Campbell. 1987). Retention of volatiles during this time period is most crucial to maintain product quality. Since permeation (from the seal and/or through the package wall) could not be detected in this time period. it was not considered to be a primary mechanism of volatile flavor loss. Subsequent studies were focused on package/volatile sorption interactions. 49 SORPTION The results of absorption and desorption studies for the 2 probe compounds (dimethyl disulfide and dipropyl disulfide) and the 3 test materials (HIPS. PP. and HDPE) are summarized in Figures 7-12. Studies were conducted at several temperatures using saturated vapor pressures for each of the probe compounds. Percent absorption and desorption were determined by quantifying the gain or loss of probe by the polymer per unit of initial polymer weight. Absorption and desorption were monitored as a function of polymer weight change. The probe compounds were highly soluble in HIPS over the range of test temperatures (#17°c. 5°C. 24°C. and 35°C). At 35°C. HIPS became pliable and difficult to handle after 4 days of exposure to dipropyl disulfide and after only 7 hours of exposure to dimethyl disulfide. After this time period these samples were completely dissolved in the penetrant solutions. All HIPS samples eventually dissolved in the dipropyl disulfide and dimethyl disulfide solutions: therefore. equilibrium sorption of the two probe compounds in HIPS was not obtained. This effect appeared to be independent of temperature. Probe compounds were not as soluble in the PP and HDPE polymer samples. In Table 2 times required for HIPS samples to reach maximum absorption before dissolving in dimethyl disulfide and dipropyl disulfide 50 penetrant solutions are given. These materials did not soften and remained intact throughout the sorption studies. In Figure 7 is shown the absorption of dipropyl disulfide by HIPS at 4 temperatures (-17°C. 5°C. 24°C. and 35°C). Desorption of dipropyl disulfide by HIPS is shown at 24°C. Absorption and desorption of dipropyl disulfide by HDPE is also shown at 24°C. Maximum absorption of dipropyl disulfide by HIPS was 43.5% (wt/wt) after 13 days at 5°C. Further measurements were not possible since polymer samples dissolved in the penetrant solutions. At 24°C the HIPS sample desorbed 69.6% of the sorbed dipropyl disulfide. 30.4% of the probe compound initially sorbed by HIPS was retained. Table 2 Time Required for HIPS Samples to Reach Maximum Absorption Before Dissolving in Penetrant Solutions Dimethyl Dipropyl Temperature disulfide disulfide 35°C 8 hr 120 hr 24°C 8 hr 240 hr 5°C 8 hr 312 hr —17°C 28 hr 1.020 hr (Z) HBSORPTIUN / DESURPTIUN 51 45 _ 0 -17c nee-HIPS 0 24C Fibs-HDPE +\ - 24c Des-HDPE .A t A 40 - \R+ A 5 c ebe- HIPS +\ A 35* n D 240 ebe- HIPS +.+ {I 30 _ *6 35C Hoe-HIPS “+5 x I + 24c Des-HIPS + 1 IE \+ g 25 . I +/ .,. .\ qbd? 20 - 3‘7 " 01M [3.I0 I .00 “1.00 IOOJIJ HIILOO TIME (hours) FIGURE 7. HWT/UV / DESUPPUU/V 6F 01/0900”. BISZZFJUE 8)” HIPS 6WD HOPE 6;»..1 .£°'°' 0 O: on 00:: ow HIKII.00 52 In comparison. the HDPE sample absorbed a maximum of 5.2% (wt/wt) at 24°C and desorbed an equivalent amount. None of the probe compound was retained by the polymer. In Figure 8 is shown the absorption of dipropyl disulfide by HIPS at -17°C. 5°C. 24°C. and 35°C. The only storage temperature which had a significant effect on the absorption rate was -17°C. The samples at 5°C. 24°C. and 35°C experienced similar absorption rates. Several investigators have established a relationship between solubility and temperature. Monte and Landau—West (1982) found a direct correlation between temperature and solubility of polystyrene in several solvents. As temperature was decreased. the solubility of polystyrene also decreased. Penetrant vapor pressures decrease with decreased temperatures. More of the penetrant will remain in the liquid phase. This results in fewer vapor molecules for the polymer to interact with: therefore. absorption rates are lowered. The HIPS sample at -17°C absorbed only 12% of the probe compound after 69 days of exposure (Figure 7). The maximum absorption at 35°C was 30% achieved in only 5 days of exposure. This comparison illustrates the dynamic effect temperature can have on absorption. In Figure 9 is shown the absorption of dimethyl o disulfide by HIPS at 4 temperatures (—17°C. 5°C. 24 C. and 35°C). Desorption of dimethyl disulfide by HIPS 53 45 /A 0 - A El 4 A 40' 17C 6 C 2 C/ X 35 c A 35* 30- X / n N (A3 o ”13/ 25' DU A’ E r/ A/ E / g 207 [/A U) / CD A CE 15* IO“ 5' [A 0 o__o_——0-—-0——O—"'O”'—’— 0 Gig—“OTO'E’: l l 1 I J L 1 0255075100125150175200226250275300325 TIME (hours) FIGURE 8. HWUW 05 DIP/mall DIEM/EMF BY HIPS 54 HBSDRPTION /DESORPTION (Z) 0 -17C abs-HIPS A 5 c nee-HIPS D 24c Hos-HIPS 35 1. at 35c Hos-HIPS I] + 24c Des-HIPS 30- I l 241: Des-HDPE 0 24C Rbs-I-DPE 25 ~ 20L 15 -\ \‘ \+ \+ 10 ~ ‘ + . 00005-0») ..A' “ ' ++1~ #0" 4| . . 1 4.3;“...H oI’J 1.00 10.00 100.00 1000.00 FIGURE 9. TIME (hours) HWUQV / HEW/071W 0F flfMEff/IZ UfflFflf BY HIPS fl/VU H095 55 is shown at 24°C. Absorption and desorption of dimethyl disulfide by HDPE is shown at 24°C. Maximum absorption of dimethyl disulfide by HIPS was 33.2% (wt/wt) after 7 hours at 24°C. Samples held at 35°C. 24°C. and 5°C dissolved in the penetrant solutions after 8 hours of exposure. The sample evaluated at -17°C dissolved after 28 hours of exposure to the dimethyl disulfide vapor. At 24°C the HIPS sample desorbed 99.4% of the sorbed probe compound. Only .6% of the dimethyl disulfide was retained by the polymer. In comparison. the HDPE sample absorbed a maximum of 5.4% (wt/wt) at 24°C and desorbed an equivalent amount. None of the probe compound was retained by the polymer. In Figure 10 is shown the absorption of dimethyl disulfide by HIPS at —17°c. 5°C. 24°C. and 35°C. Polymer samples at 35°C and 24°C experienced absorption rates significantly higher than samples at 5°C and -17°C. This effect can be attributed to the increased penetrant vapor pressure at 35°C and 24°C. The polymer is exposed to more vapor molecules than those samples at 5°C and—17°C. In Figures 11 and 12 the absorption and desorption of dipropyl disulfide and dimethyl disulfide by PP are shown. For both probe compounds. equilibrium absorption was approximately 9% (wt/wt) at all test temperatures. HBSURPTIUN (Z) 56 35 1:1 3/ 0-17c *6 35:: 30' J A 5 5 1:1 240 25' éf 20.. J 0 15* /A I0?! A’A U ’5' 1 KA 0/ 5.- Q)’bqy00’ Au 0 l 1 l I l 0 5 10 I5 20 25 TIME (hours) FIGURE 10. HWIIW 0F DIM-’IHIZ DISZ/IFIDE BY HIPS RBSURPT I 0N / DESDRPT I [IN ( Z ) 57 —‘ N (J -5 (II I r 05 CFIbs _ 3K240068 USBCFIbs A 24 c Hbs I0” 9’I 8- 7 U 0 IA 5 A 0 4::" 9 Q 3 2 I. IV 06 l l l I 1 1 l l l w 0 I00 200 300 400 500 600 700 800 900 I000 I100 FIGURE 11. TIME (hours) HHSUPPIIU/V / DESUPPIICW CF DIP/300% HISIIFIUE HY PP HBSDRPTI 0N / DESORPTI ON C 7.’ ) 58 01 p O -‘ N (a) A I I T r “)0thme 05 CFle A24ce5e Us5caee 3‘ 24 C Use fi==fi=I—_I+I.*_.I__1 0 25 50 75I0012515OI75200225250275300325350 TIME (hours) FIGURE 12. HHSU‘BPIIW / DESOPPIIU/V 05 DIMEIHYI HISIIFIHE HY PP 59 As shown in the following table. time required for PP samples exposed to dimethyl disulfide to reach equilibrium was much shorter than that for samples exposed to dipropyl disulfide. Table 3 Time Required to Reach Equilibrium Sorption for PP Samples Exposed to Penetrant Vapor* Dimethyl Dipropyl Temperature disulfide disulfide 35°C 22 hr 130 hr 24°C ' 214 hr 370 hr 5°C 310 hr 465 hr *All samples achieved equilibrium at approximately 9% absorption Dimethyl disulfide and dipropyl disulfide were not retained by the polymer. At 24°C the PP completely desorbed both probe compounds. The solubility of dimethyl disulfide and diprOpyl disulfide in HDPE and PP is substantially lower than in HIPS. HIPS retains a significant portion of the sorbed dipropyl disulfide. while HDPE and PP desorb 100% of both probes. Equilibrium distribution of diprOpyl disulfide and dimethyl disulfide between the product and container would result in much lower probe compound concentrations in HDPE and PP structures than in the HIPS. This is significant in avoiding the effects of "flavor scalping" 60 or loss due to sorption for a product whose quality is associated with the retention of volatile aroma constituents. EQUILIBRIUM VAPOR PRESSURE The equilibrium vapor pressures of dimethyl disulfide and dipropyl disulfide were determined for the pure probe and simulant systems over a range of concentrations and temperatures. Simulant systems with probe compound concentrations of 20 ppm. 100 ppm. and 1000 ppm (wt/wt) were analyzed at temperatures of 5°C. 26°C. and 35°C. These concentration levels were chosen based on estimates of the actual amount of dimethyl disulfide and diprOpyl disulfide in the onion flavored sour cream (Appendix VIII. Table 15). At these concentrations equilibrium vapor pressures were used to extrapolate equilibrium vapor pressure for 80 ppm dimethyl disulfide and 10 ppm dipropyl disulfide by substitution of these values into equations 5—9. In Appendix IX. Table 16. are listed all equations used in this section. Sensitivity of GC analysis did not allow quantification at all concentration levels and temperatures. Equilibrium vapor pressures were not obtained for dipropyl disulfide 20 ppm simulant samples. Dipropyl disulfide samples at 5°C (278°K) and 100 ppm were also below detectable limits. In Figures 13 and 14 are the probe equilibrium vapor 61 pressures (expressed as ppm in the vapor phase) as a function of temperature for the simulant and pure probe systems. In Figures 15 and 16 are the probe equilibrium vapor pressures as a function of simulant probe concentration from Figures 13 and 14. The experimental data shown was derived by linear regression analysis using the least square method (Gacula and Singh. 1984). Equations describing the curves (Equations 5-9 for Figures 15 and 16) for various storage temperatures were then derived. These expressions describe the vapor phase concentration of dimethyl disulfide and dipropyl disulfide as a function of simulant probe concentration. By substituting the probe concentrations existing in the product into equations 5—9. the actual equilibrium vapor pressure of dimethyl disulfide and dipropyl disulfide in the product was determined for several temperatures (Appendix IX. Table 16). ABSORPTION OF PROBE COMPOUNDS AS A FUNCTION OF TIME In order to relate the behavior of the pure probe system to the actual product. it was first necessary to 2. T2 15 the temperature at which the equilibrium vapor pressure determine the temperature referred to as T (ppm) of the pure probe standard is equivalent to the PPM 62 I40 ———~~—~un»——-—_~JW- 0 Pure standard A 20 ppm simulant 120- I 100 ppm slmulant 01000 ppm simulant 100' 80' 50* 40* 20' o... 0 LA “AT 72‘ 1 5.1 285 28) 2&5 2A) 2E5 am fifi :am as :xm :IE 2M0 TEMPERHTURE (K) FIGURE 13. PHI/II IHPIU/I/ I/HPUH PHI-“SSW? 0F DI ME I HYI HISIIFIDE HS H FHIZ‘IISI’ 0F IEMPEPHTII‘I’E - PPM 63 0.8 0 Pure standard I 100 ppm simulant 0.7" (3 01000 ppm simulant 0.5" 0.5" 0.4" 0.3“- 0.2*' 0.1 ' na_ ‘0 0.0 l é==: . L l r L l H 270 275 280 285 290 295 300 305 310 TEMPERRTURE (K) FIGURE 14. PHI/II. [HRH/H VHPUH PPESSZ/PE 0F DIP/PUP”. UISIIFIUE HS H FUAZ‘IIW 0F IEMPEPHIIIPE PPM VFIPOR 64 4.0 0308K A299K I27s1< 3.5 " / 0 3.0 r 2.5 - A/ 2.0 - 1.5 r '/ 1.0I 0.5 P 0.05%: L L 0 I00 200 3C!) 400 500 600 700 800 900 1000 IIOO PPM SIMULHNT FIGURE 15. EZRA/II IBPIUM I/HPUP PPESSUPE UF UIMEIHYI DISUIFIDE HS H FIWCI/U/V 0F SI/WIHNI PPUHE CUI/Cfl/ IPHI I [W PPM VHPOR 65 0.025r 0.020 '- 0.015I _' 0.010 0.005 t 0308K A2991< / 0 A L 1 4‘ J 0.000 0 I00 200 300 400 500 600 700 800 900 1000 I100 PPM SIMULHNT FIGURE 16. PHI/II I 3?] UH I/HPUP PHESSUPE 0F HIPPUPYI ISUIFIUE HS H PHI-If 710V 05 SIHMHI/I PHUSE CUIZ‘H/ I PHII O/V 66 equilibrium vapor pressure in the actual product at the normal storage temperature (T1=5°C). To determine equilibrium vapor pressure as a function of temperature. linear regression analysis was performed on results derived from equations 5-9. Equations 10 and 11 and Figures 17 and 18 describe the vapor phase concentration of probe compound in the product as a function of temperature. T =5°C(278°k) can be substituted into 1 equations 10 and 11 to find a T which represents the 2 temperature of the pure probe that approximates behavior of the actual product. From substitution of T1 into equations 10 and 11. results are shown which represent the actual concentration of probe vapor in the product at 5°C. Equations 12 and 13 derived from Figures 13 and 14 describe equilibrium vapor pressure for the pure probe systems as a function of temperature. By substituting the concentration of probe vapor found at T into equations 1 12 and 13 (representing the'pure standard). T2 can be determined. Results for T2 from equations 12 and 13 are given in Appendix IX. Table 16. In order to relate the amount of sorption occurring at T2 to time. Figures 19 and 20 were constructed using data from sorption studies. From these results it was possible to determine percent absorption at T2=-26.4°C for dimethyl disulfide (results were extrapolated) and T2=—3.3°C for dipropyl disulfide at several time PPM 0.30 0.25 0.20 P 0.15" 0.10” 0.05 ' 0.00 67 L l J 275 280 2% 290 295 3CD 305 310 315 320 TEMPERFITURE C K ) FIGURE 17. EWII [SWIM I/HPUP PPESSUHE UP BIMEIHYI DISUIFIDE I/V 0/1/[0/1/ FIHI/QPED SUI/P SPEH/‘I HS H FUNCIIU/V 0F TEMPE/397%? PPM X 10 E+3 05 UAW 03“ 02" DJ“ 00 68 J g l L L i 2W5 am 255 2%) an) 35 (N0 36 TEMPERHTURE (K) FIGURE 18. Edi/II [HPIW l/HPII‘? PHESSUPE 0F DIP/WWI flIfl/IFIUE I /V PM 0/1/ HHI/UPEIY SUI/P [PE/9M HS H FU/VC I I 0/1/ OF I EMPEPH I USE mm HBSDRPTION (Z) 35 30 69 0T1me= 1 hour * AT1me=2houre ° Dles-4Ixnms 013me==8111ms 25* [Ir—’9 20- 154 A A/ 105 L L 1 l l 1_ 0 I I l I I -83 45 -40 -5 0 E5 K) 15 K) 25 a) £5 TEMPERHTURE (C) FIGURE 19. HHSUR‘WUV [F HI/IIEIHYI HISIIFIIJE HY HIPS HS H FUM‘IIIW 0F IE/VPEPHIIIPE 4 i I 1 I l i J 0 FIBSDRPTI ON C Z ) 70 40F °Tims=1dcy ITIme=4dcgs UTime=6 XTIms-Bdags OTIms-IOdags 35° o/T’\O 30" x/r * 20- TEMPERHTURE C C ) FIGURE 20. HWIIW 05 DIPPUPIZ HISZIFIIYE HY HIPS HS H FWIIU/V 0F IEIFEPHIIHE dogs [I 0 l l l l l -20 -I5 -10 -5 0 5 10 I5 20 25 30 35 40 71 intervals. Figures 21 and 22 were prepared from these data. These graphs of percent absorption vs. time for the 80 ppm dimethyl disulfide and 10 ppm dipropyl disulfide simulant systems approximate behavior of the actual product stored at 5°C. The capacity of a container to absorb dimethyl disulfide and dipropyl disulfide is a function of the equilibrium concentration of probe in the polymer (maximum absorption). The amount of probe actually available for absorption is a function of the concentration of probe compound in the product. By using equations 1 and 2. the capacity of the container to absorb the probe and the actual amount of probe available can be determined. Capacity of container to absorb probe (mg) = Cc x C 100 (l) where Cc = equilibrium concentration of probe compound in the polymer (% maximum absorption from Figures 19 and 20) C = weight of polymer container (mg) Amount of probe compound available (mg) = CP x P (2) where Cp = concentration of'probe compound in the product (ppm from Table 1) P = weight of product (kg) HBSORPTION (Z) 72 45 OEDIxMIsHmNant 40' 35 30' 25" 20' L5" L0 05” 0.0L . ' . . 0 1 2 3 4 5 6 7 TIME (hours) FIGURE 21. HEISU‘IPIIQV 05 UIIEIHIZ [YISZIFIUE HY HIPSHSH FU/IEIIO/V IF II/‘E HBSORPTIUN (Z) 73 20 24" 010 ppm simulant 18’ 10' 14' I2“ 10- ° 0 1 2 3 AI 5 6 77 8 9 10 TIME (dogs) FIGURE 22. HHSU‘PIICW IF UIPPUPIZ DISZ/IFIDE H)” HIPS HS H FWIIIJ/V 0F IIME J H 12 74 The capacity of the HIPS container used in this study to absorb dimethyl disulfide and the actual amount of. dimethyl disulfide available in the product were determined as follows: HIPS Capacity to absorb Mezsz = (4.2) (9500mg) = 399mg 100 MeZS2 Available = (80.1 ppm) (.236kg) = 18.9 mg In Table 4 are results for the three test materials and two probe compounds. In all cases the quantity of available probe is much smaller than the absorption capacity of the container. Each test container has the potential to absorb 100% of the dimethyl disulfide and dipropyl disulfide contained in the onion flavored SOUE cream. Table 4 Absorption Capacity of HIPS. PP. and HDPE and Available Probe Compound in Onion Flavored Sour Cream. Dimethyl disulfide Dipropyl disulfide est Material Capacity Available Capacity Available HIPS 399.0 mg 18.9 mg 2090.0 mg 2.1 mg PP 346.9 mg 18.9 mg 1817.2 mg 2.1 mg HDPE 558.6 mg 18.9 mg 2926.0 mg 2.1 mg 75 SENSORY ANALYSIS An untrained panel of 534 participants evaluated a control sample and a sample of unflavored sour cream which had been exposed to the onion flavored product. Samples were presented in a paired-comparison/simple difference test to determine if the unflavored product had absorbed any off odors or flavors during storage. Only two panelists detected an off flavor related to onion and/or garlic in the unflavored product. Two panelists noted these flavors in the control product not exposed to the onion/garlic flavored sour cream. In Table 5. a contingency table analysis of responses for test and control samples is shown. Table 5 Sensory Evaluation Contingency Table. Observed (0) Class 1 Class 2 Grand Total Test 2 532 534 Control g 532 534 Total 4 1.064 1.068 Expected (E) Class 1 Class 2 Grand Total Test 2 532 534 Control 2 532 534 Total 4 1.064 1.068 76 Results for Table 5 were derived from equation 3. Eij = Ricj/N (3) where R1 = Row total Cj = Column total N = Grand total Class 1 responses are defined as panelists noting onion and/or garlic off flavors or odors. Class 2 responses are defined as panelists not noting onion and/or garlic off flavors or odors. Chi-Square was calculated from equation 4. (4) observed expected 2 2 (Oij‘Eij)2 where Oij >2 .2 I / = (M)? 313 =1 3:1 Eij 2:0 The contingency table analysis of responses for test and control samples gave Chi-Square equal to zero. Thus. statistical analysis indicates that the results show no difference in panelists' responses to test and control samples. The results of the sensory studies provide additional insight into the possible mechanisms of flavor loss. Solubility of flavor components in the polymer matrix appears to be the dominant mechanism. If permeation of volatile flavor compounds does occur from flavored to unflavored product. it is not at levels which adversely affect product quality or would be detected by consumers. CONCLUSION Aroma and flavor compounds are important constituents of many foods. The ability of a package system to prevent loss of these compounds directly influences product quality. In this study the mechanisms of flavor loss for a product/ package system were determined. The major findings are summarized. (1) (2) Sorption was the major mechanism responsible for flavor loss. Absorption Studies. Dimethyl disulfide and dipropyl disulfide were very soluble in HIPS. Desorption Studies. Dimethyl disulfide and dipropyl disulfide were retained by HIPS during desorption studies. Permeation Studies. No detectable amount of permeation was quantified. Sensory Analysis. Panelists could not detect off odor or flavors which could be attributed to permeation from flavored to unflavored product. Absorption of probe compounds by the package. Based on the high capacity of the container to 77 (3) (4) (5) (6) 78 absorb and the small amount of probe compound available in the product. the system can act as an infinite sink. This accurately describes. the HIPS. PP. and HDPE test materials. Temperature effects on sorption of probe compounds were minimal except at very low temperatures. Penetrant vapor pressure was sufficiently lowered at -17°C to decrease sorption rates. Quality of the product over time will be dependent on the diffusion coefficient of flavor constituents. The rate at which these compounds come into contact with the package wall will determine the rate of flavor loss. Products stored near the onion flavored sour cream were not adversely affected by pick—up of off flavors or odors. No special storage or handling procedures are required for this product. No detectable amount of permeation was quantified. a. Sensory analysis panelists could not detect off odor or flavors which could be attributed to permeation from flavored to unflavored product. 79 Sorption may occur at a higher rate than predicted due to two factors: (1) Maximum absorption rates for HIPS were based on measurements obtained before the polymer dissolved in the penetrant solution. If the polymer remained intact. an equilibrium sorption rate could be determined. (2) The temperature T2 is lower than the actual product storage temperature. The physical state of the polymer at T2 was not taken into account when determining rate of sorption. Temperature influences sorption by affecting penetrant vapor pressures. but it can also alter the physical structure of the polymer. More sorption will occur at a higher temperature as the polymer matrix swells. For this reason a lower temperature of T2 will lead to an underestimation of sorption. Of the three materials tested. HIPS had the highest degree of interaction with the product. PP and HDPE absorbed and retained much less of the probe compounds than HIPS. HDPE and PP have the capacity to absorb 100% of the probe compounds: however. the rate of absorption is significantly lower than that of HIPS. Using HDPE or PP could reduce the effects of flavor scalping during the critical shelf life period. 80 Gaining an understanding of the mechanisms of flavor loss allows selection of packaging systems which will optimize quality and compatability. As the use of plastics in packaging increases. it becomes increasingly important to characterize interactions between products and packages. The methods used in this study can be applied to a wide variety of materials and foods. Much more accurate estimations of shelf life can be based on loss of principal flavor components. as opposed to overall quality perceptions. Accuracy of shelf life predictions could be further improved by determining the diffusion coefficients of principal flavor constituents in the produCt. The loss of flavor as a function of time could be quantified in this manner. These studies show that the quality of the product can be affected by flavor scalping. Flavor loss from a package can affect surrounding products. Reducing flavor loss can improve the quality of the product itself and others in contact with it. This could have a major impact on the food packaging industry as it continues to utilize greater amounts of plastics for an ever increasing variety of foods. APPENDICES APPENDI X I 81 APPENDIX I Table 6: Properties of High Impact Polystyrene (Chevron. 1986) *Test Property Method Value Units Melt Flow Condition D1238 2.7 gms/10 min Izod Impact — D256 Ft-lb/in 73°F (1/4" thick) 1.9 Notch Izod Impact — D256 Ft-lb/in 00F (1/4" thick) 1.3 Notch Dart Impact — Ft-lb/in 73°F NBS PS-31—70 440 ThiCk Dart Impact - Ft—lb/in 00F NBS PS—31-7O 415 Thick Vicant Softening Temp. D1525 214 0F Deflection Temp. 264 psi Unannealed D648 185 0F Tensile Strength — D638 Break at 2.0 in/min 3.200 psi Tensile Strength — D638 Yield at 2.0 in/min 2.800 psi Tensile Elongation - Break D638 70 % 82 Table 6: continued *Test Property Method Value Units Tensile Modulus D638 270.000 psi at .05 in/min Rockwell Hardness D785 70 ' L Scale Specific Gravity D792 1.03 *All Test Methods refer to ASTM Standards except those noted as NBS (National Bureau of Standards). This material complies with Food Additive Regulation 21 CFR 177.1640 for food contact. APPEND IX I I 83 APPENDIX II Table 7: Approximate Composition of Fresh and Dehydrated Onion (Qureshi et al.. 1968) (Farrell. 1985) Fresh Onion Dehydrated Onion (per 100 g) (per 100 g) Moisture 83.7 g 5.0 g Protein 1.406 g 10.1 9 Fat 0.256 g 1.1 g Fiber 0.623 g 5.7 g Ash 0.508 g 3.2 g Calcium 10.0 mg 363 mg Iron 0.366 mg 3.0 mg Magnesium 11.97 mg 122 mg Phosphorous 28.0 mg 340 mg Potassium 217.2 mg 943 mg Sodium 17.2 mg 54.0 mg Zinc 0.172 mg 2.0 mg Ascorbic Acid 9.6 mg 15.0 mg 84 Table 8: Physical Characteristics of Toasted. Chopped. Dehydrated Onions (Farrell. 1985) §_practeristic Maximum Moisture 3.25 (Dry Basis) Color Optical Index 90 Total Plate Count 300.000 organisms U.S. Standard size sieve Maximum Sieve % Size 2 (retained ON) 4 40 (pass through 8 90 (pass through) 12 APPENDIX I I I 85 APPENDIX III Table 9: Specifications for 8 02 High Impact Polystyrene and Polypropylene Tubs (Land O' Lakes Inc.. 1987) High Impact Dimension Polystyrene Tub Polypropylene Tub Wall thickness .013“-.017" .013"-.017" Height 2.3598"-2.4222" 1.870"-1.890" Top diameter 3.808"-3.838" 4.133"-4.157" Overflow capacity 7.7—7.9 fl. 02 10.14 fl. oz Piece weight 7.5—11.5 g 6.5—10.5 9 86 Table 10: Polypropylene and High Density Polyethylene Properties (Shell Chemical Co.. (Soltex Polymer Corp.. Property Density Melt Flow Tensile Yield Strength (2.0 in/min) Flexural Modulus Notched Izod Impact Strength Heat Deflection Temp. (66 psi) Vicat Softening Temp. Test HQLQQQ D1505 D1238 D638 D790 D256 D648 D1525 Polypropylene 0.903 g/cc 2.0 g/10 min 5.000 psi 200.000 psi 0.6 ft-lb/in 220°F 305°F 1987) 1987) High Density Polyethylene 0.960 g/cc 0.70 g/10 min 4.300 psi 212.000 psi 3.4 ft—lb/in 242°F 266°F Both materials comply with Food Additive Regulation 21 CFR 177.1520 for contact with foods. APPENDIX IV 87 uomuuxm HOUSOm coeao ooumuownoo mo zoom aoH Hmpoe mz\uo .mm enamflh figfigommoomomtoovomnoonommoomoma.8« 8730.»... < 13* . fi’(\€l))jllllJ .1 Juan” mmmo n mmm zoom mmoz u mm emom N >H xHomemd 88 APPENDIX IV Table 11: Probe Compound Properties Property Dipropyl Disulfide Dimethyl Disulfide Density .960 1.057 Molecular Weight 150.30 94.19 Solubility insoluble in H20 insoluble in H20 Boiling Range 195—196°C 107-111°C Refractive Index 1.4967 1.526 Flash point 66°C 14°C Chemical Formula (CH3CH2CHZS)2 CH3SSCH3 Typical Assay (GLC) 97% 99% APPENDIX V HREFI UNITS X 10 E-Z 89 1400 / 0 1200 r 1000 h 800 r 500 r 0 400 *- 200 - / /O OLE 4 I I I I I I I 0 I 2 3 4 5 6 7 8 gx IO E+8 Fioure 24. 5701/0000 0001/5 F00 01057011 01501 FIDE 0050 05500055 05 0 FHA/[710V OF 000ij 7r PREP UNITS X 10 E-2 90 800 700 ' 500P 500 r 400 300 200" 100* 0 I 2 3 4 5 6 7 8 9 gx10E+Q Figure 25. SIHNDHPI? CUPI/P PUP HIPPUPYI DI SUI PI IZP HPEH HPSPU/IISP HS H PU/VC I I 0N 0P PAPA/II I)” APPENDIX VI 91 APPENDIX VI CONSENT FORM FOR TASTE PANEL MEMBERS School of Packaging Michigan State University Lean Cream Ingredients: Cultured sour cream. whey protein concentrate. skim milk. lactic acid. water. food starch-modified. natural flavor. agar. potassium sorbate (a preservative). Vitamin A palmitate. I have read the above list of ingredients and find none that I know I am allergic to. I have also been informed of the nature of the research (including experimental materials and procedures) which will be used during the tasting session. I agree to serve on this taste panel. which is being conducted on this day of 1987. I understand that I am free to withdraw my consent and to discontinue participation in the panel at any time without penalty. Signature Date Figure 26. 92 QUESTIONNAIRE For the pair of samples. indicate whether the samples are the same or are different by placing an X in the appropriate space. If the samples are different. please describe (i.e.. type of off flavor detected). Taste as much or as little of each sample as you wish in order for you to answer these questions. SAMPLE PAIR 137 versus 264 Different Not different If different. please describe the difference: Figure 26: Cont'd. 93 QUESTIONNAIRE For the pair of samples. indicate whether the samples are the same or are different by placing an X in the appropriate space. If the samples are different. please describe (i.e.. type of off flavor detected).' Taste as much or as little of each sample as you wish in order for you to answer these questions. SAMPLE PAIR 538 versus 701 Different Not Different If different. please describe the difference: Figure 26: Cont'd. APPENDI X VI I APPENDIX VI I 9I'ZI SI 69 {9'99 652' ”It OI'III 2618 III 92 KB .213 ”II GC Analysis of Onion Flavored Sour Cream Extract Figure 27. APPENDIX VI I I Table 12: Known Probe Compound Concentration 18 ul M6252 in 50 ml Solvent 18 ul P282 in 50 ml Solvent Table 13: Known Probe Compound Concentration 10 ul M6282 in 22 ml Solvent 10 ul P232 in 22 ml Solvent 95 APPENDIX VIII Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide -- Total loss (extraction and concentration) Grams of Probe Before After Extraction and Extraction and % Total Concentration Concentration Loss 3.12 x 10—3 g 5.53 x 10—4 g 82% 1.36 g .56 g 58.8% Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide-—Loss due to extraction Grape of Probe % Loss Before After due to Extggction Extrection Extraction 1.035 x 10-3 g 2.74 x 10-4 g 73.5% .6075 g .352 g 42% 96 Table 14: Percent Recoveries for Dimethyl disulfide and Dipropyl disulfide--Loss due to concentration* % Loss due to Probe Concentration M8252 8.5% P282 16.8% *(% total loss—% extraction loss = loss due to concentration) Table 15: Concentration of Probe Compounds in Onion Probe M6282 P2'52 Flavored Sour Cream Grams Actual Grams % Probe ppm Probe Probe* Probe** in Product in Product 7.20 x 10-4 g 4.00 x 10-3 g .008% 80 ppm 1.83 x 10-4 g 4.48 x 10-4 9 .00092 9 ppm *Based on 50 9 sample dehydrated onion **Based on 50 9 sample dehydrated onion—~taking into account % recovery APPENDIX IX 97 oon.m1\uooe~.mu UO¢.0NI\Mo>vNImM EGO MIOHHNHH.I¢> goo oHH.u¢> and eIoHuoo.~I~> Baa «Ioanmv.mnnh Ina ana.u~> son bma.uu> Ina om~.u~> mudmmmm eon nuoax~aa.us son oHH.I> MOQFNIMN MOOFNIMK Ema oauax can anus» and oouex Ban OaIHN coo canax madmmMflmmmmw .Aoomv mousucuoaaou ooauouu Amino: on weapons Heaven on» as ouamuouo uoao> asauosascuo on» on uaoao>s=oo ma unaccouo ooouo ouao one we ouammouo uoac> assunsA«=oo on» scan: as ouououooaos n «a; “Oomvaa as weapons as ousmmouo uoom> sswunsdsaoo mocha I com I as ousuauoaaou co>uu on» an uosaoua as ouammouo uoao> asuuosduaao uooooua as casoaaoo onouo no aosumuuaooooo Hooves NaNIOHNHom.Hv+AvOH.mIvI> KAmOH.NV+ANOHuH¢m.mIVI> onIonane.eo+1nuoauoeo.~no.> MAMIonm~5.no+AHIoneee.one.» xxmuoaxnon.Ho+xeuoauao~.aous namuoaxeaa.~o+.euoaxmnn.aous xAnIoaxoam.H.+An1oexmne.~ous xxmuoaxeoe.Ho+.anermo.~IoI> xAMIoneom.no+anuoaxomm.euous madmmmmw NH Kunzummd oooua ouso ooouo ouaa Eon oa can an MOme MOQOM Moth “Chum MOQOM «me no casumcsauouoo wam ~m~ox umwo NmNOS ammo «mam Nmmoz NmNOS NmNOS mmmdemmmw mmmmmfimu mmmmwm NH I mN uv> “MK umh UH“ va ma ma NA ma HA ha OH ca a ma 0 ma 5 ma a ma m IIMMmmmz mooum enough aesumaom "0H munch B I BL I OGRAPHY BIBLIOGRAPHY Abraham. K. 0.. Schankaranarayana. M. L.. Raghavan. B.. and Natarajan. C. P. 1976. Alliums — Varieties. chemistry. and analysis. Lebensm. - Wiss. - Technol. 9:193. Amerine. M. A.. Pangborn. R. M.. and Roessler. E. B. 1965. Principles of sensory evaluation of food. Academic Press Inc.. New York and London. Bagley. E. and Long. F. A. 1958. Two stage sorption and desorption of organic vapor in cellulose acetate. J. Am. Chem. Soc. 77:2172. Becker. K.. Koszinowski. J.. and Piringer. O. 1983. Permeation of scenting and flavoring substances through polyolefins. Cited in Mohney. S. M. 1986. M.S. thesis. Mich. St. U. School of Pkg. Beckman. H. M.. Simms. W. C.. and Covington. J. 1979. Glossary of packaging terms. 5th Edition. The Packaging Institute U.S.A. Berens. A. R. 1977. Diffusion and relaxation in glassy polymer powders: 1. Fickian diffusion of vinyl chloride in poly (vinyl chloride). Polymer 18:697. Berens. A. R. 1978. Analysis of transport behavior in polymer powders. J. Membrane Science. 3:247. Bergen. R. L. 1968. Stress cracking of noncrystalline plastics. SPE J. 24(8):77. Bernhard. R. A. 1966. Separation and characterization of flavor components from vegetables. Advanc. Chem. Ser. No. 56,131. American Chemical Society Publications. Washington. D.C. Bernhard. R. A. 1968. Comparative distribution of volatile aliphatic disulfides derived from fresh and dehydrated onion. J. Food Sci. 33:298. Bhattacharyya. G. and Johnson. R. 1977. Statistical concepts and methods. John Wiley and Sons Inc.. New York. 98 99 Bhatti. M. B.. and Asghar. A. 1965. Certain dehydration studies on Pakistan onions. Agr. Pakistan 16:299. Boelens. M.. de Valios. P. J.. Wobben. H. J.. and van der Gen. A. 1971. Volatile flavor compounds from onion. J. Agr. Food Chem. 19 (5):984. Campbell. R. 1987. Personal communication. Group Leader. Land O' Lakes Inc. Minneapolis. MN. Chevron. 1986. Chevron polystyrene extrusion grade resins — technical bulletin. Chevron Chemical Company. Houston. TX. Crank. J. and Park. G. S. 1968. Diffusion in polymers. Academic Press. New York. NY. Crank. J. 1975. The mathematics of diffusion. 2nd Edition. Clarendon Press. Oxford. England. Cruess. W. V. 1958. Commercial fruit and vegetable products - 4th Edition. McGraw Hill Book Company. New York. DeLassus. P. T. 1985. Transport of unusual molecules in polymer films. TAPPI Proceedings. Polymers. Laminations. and Coating Conference: 445. Durr. P.. Schobinger. U.. and Waldvogel. R. 1981. Aroma quality or orange juice after filling and storage in soft packages and glass bottles. Lebensm. — Verpack. 20:91. Farrell. K. 1985. Spices. condiments. and seasonings. AVI. Westport. CT. Fennema. O. 1976. Principles of food science - Part I - food chemistry. Marcel Dekker Inc.. New York. Freeman. G. G. and Whenham. R. J. 1974. Changes in onion (Allium cepa L.) flavour components resulting from some past-harvest processes. J. Sci. Fd. Agric. 25:499. Friedman. S. 1985. Packaging Encyclopedia and Yearbook. 30(4):305. Fujita. J. 1961. Diffusion in polymer-diluent systems. Fortsch. — Hochpolym. — Forsch. 3:1. 100 Gacula. M. C.. and Singh. J. 1984. Statistical methods in food and consumer research. Academic Press Inc.. Orlando. FL. Giacin. J. 1986. Packaging course (Pkg. 840). Mich. St. U. School of Packaging. Hanlon. J. 1984. Handbook of package engineering — 2nd edition. McGraw Hill Book Company. New York. Herman. M.. Gaylord. N.. and Bikales. N. 1964. Encyclopedia of polymer science and technology. Interscience Publishers. New York and London. Hernandez. R.. Giacin. J.. and Baner. A. 1986. Measuring the aroma barrier properties of polymeric packaging materials. Pkg. Technol. 16(4):12. Hirose. K. 1987. The influence of aroma compound absorption on mechanical and barrier properties of sealant films used in aseptic packaging. M.S. thesis. Mich. St. U. School of Pkg. IFT Sensory Evaluation Division. 1981. Sensory evaluation guide for testing food and beverage products. Food Technol. 35(11):50. Land O' Lakes. 1987. Packaging specification. Land 0' Lakes. Inc.. Minneapolis. MN. Larmond. E. 1977. Laboratory methods for sensory evaluation of food. Canadian Government Publishing Centre. Ottawa. Canada. Mazza. G. and LeMaguer. M. 1978. Volatiles retention during the dehydration of onion (Allium cepa L.). Lebensm. - Wiss. — Technol. 12(6):333. Mazza. G.. LeMaguer. M. and Hadziyev. D. 1979. Effect of cysteine on volatile development in dehydrated onions (Allium cepa L.). Can. Inst. Food Sci. Technol. J. 12:43. Mazza. G. LeMaguer. M.. and Hadziyev. D. 1980. Headspace sampling procedures for onion (Allium cepa L.) aroma assessment. Can. Inst. Food Sci. Technol. J. 13(2):87. 101 Mohney. S. M. 1986. The aroma permeability and solubility of two cereal liner materials and their relationship to product quality. M.S. thesis. Mich. St. U. School of Pkg. Monte. W. C. and Landau—West. D. 1982. Solubility of polystyrene in certain vegetable oils. essential oils. and their constituents. J. Food Sci. 47:1832. Olafsdottir. G.. Steinke. J. A.. and Lindsay. R. C. 1985. Quantitative performance of a simple Tenax—GC adsorption method for use in the analysis of aroma volatiles. J. Food Sci. 50(5):143l. O'Mahoney. M. 1980. Statistics for sensory evaluation. Dept. of Food Science and TechnologY. University of California. Davis. CA. Prepared for IFT Sensory Evaluation Workshop. June 1980. Parker and Stabler 1913. Cited in Mohney. S. M. 1986. M.S. thesis. Mich. St. U. School of Pkg. Parliment. T. H. 1987. Sample analysis in flavor and fragrance research. American Lab. 19(1):51. Phillips. M. 1979. Lemon-tea drinkers -- a group at risk? New England J. Medicine 301(18):1005. Pruthi. J. S.. Singh. L. J.. and Girdhari. L. 1959. Some technological aspects of dehydrated garlic —- a study of some factors affecting the quality of garlic powder during dehydration. Food Sci. 8(12):441. Qureshi. E. I.. Chaudhry. M. I.. and Malik. M. A. 1968. Chemical composition and nutritive value of some varieties of onion. Pakistan J. of Medical Research 7:155. Rauch. J. 1985. The Rauch guide to the U.S. packaging industry. Rauch Associates. Inc. Bridgewater. NJ. Saghir. A. R.. Mann. L. K.. Bernhard. R. A.. and Jacobsen. J. V. 1964. Determination of aliphatic mono— and disulfides in allium by gas chromatography and their distribution in the common food species. Proc. Am. Soc. Hort. Sci. 84:386. Saimbhi. M. S.. Padda. D. H.. and Singh. Gurdalbir 1970. Evaluation of onion varieties for dehydration traits. J. Food Sci. Technol. (India) 8:149. 102 Shell Chemical Co.. 1987. Technical Bulletin. Shell Chemical Co.. Houston. TX. Soltex. 1987. Technical Bulletin. Soltex Polymer Corporation. Houston. TX. Stephenson. R. M. 1949. Aspects of modern onion and garlic dehydration. Food Technol. 3:364. Stoll. A. and Seebeck. E. 1951. Chemical investigations on allium. the specific principle of garlic. Advan. Enzymol. 11:377. Swett. R. M. 1986. Modern Plastics Encyclopedia 1986-1987. Vol 63. No. 10A. pg. 74. vom Bruck. C. G.. Figge. K.. and Rudolph. F. 1981. Interaction of fat-containing food with plastics packaging. JAOCS (58)8:811. Wahlroos. O. and Virtanen. A. I.. 1965. Acta Chem. Scand. 19:1327. Cited in Brodnitz. M. H.. Pollock. C. L. and Vallon. P. P. 1969. Flavor components of onion oil. J. Agr. Food Chem. (17)4:760. Westendorf. R. G. 1984. Trace analysis of volatile flavor compounds in foods by dynamic headspace gas chromatography. Presented at 35th Pittsburgh Conference on Analytical Chemistry and Applied Spectroscopy. Atlantic City. NJ. March 1984. Whitaker. J. R. 1976. Development of flavor. odor. and pungency in onion and garlic. Adv. Food Res. 22:73. Wyatt. D. M. 1986. Analytical analysis of tastes and odors imported to foods by packaging materials. J. Plastic Film Sheeting 12:144. Yagami. M.. Kawakishi. S. and Namiki. M. 1980. Identification of intermediates in the formation of onion flavors. Agr. Biol. Chem. 44:2533.