UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

The effects of microwave heating on the migration of contaminants from plastic into food simulants Lam, Andrea Yuen-Kwun 2006

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_2007-0148a.pdf [ 7.4MB ]
Metadata
JSON: 831-1.0100801.json
JSON-LD: 831-1.0100801-ld.json
RDF/XML (Pretty): 831-1.0100801-rdf.xml
RDF/JSON: 831-1.0100801-rdf.json
Turtle: 831-1.0100801-turtle.txt
N-Triples: 831-1.0100801-rdf-ntriples.txt
Original Record: 831-1.0100801-source.json
Full Text
831-1.0100801-fulltext.txt
Citation
831-1.0100801.ris

Full Text

THE EFFECTS OF M I C R O W A V E HEATING ON THE MIGRATION OF CONTAMINANTS F R O M PLASTIC INTO FOOD SIMULANTS B y A N D R E A Y U E N - K W U N L A M B . S c , The University of Toronto, 2001 A THESIS S U B M I T T E D I N P A R T I A L F U L F I L L M E N T O F T H E R E Q U I R E M E N T S F O R T H E D E G R E E O F M A S T E R O F S C I E N C E in T H E F A C U L T Y O F G R A D U A T E S T U D I E S (Occupational and Environmental Hygiene) T H E U N I V E R S I T Y O F B R I T I S H C O L U M B I A December 2006 ©Andrea Yuen-Kwun Lam, 2006 A B S T R A C T Plastic food packaging contains potentially toxic chemicals that are capable of diffusing to the'plastic surface through the process of migration, where they can contaminate food (Incarnato et a l , 2000). Thus, government regulations require food packaging migration testing to ensure public safety. Standardized migration test methods use conventional ovens to determine the maximum amount of migration that w i l l occur during heating. However, standardized migration tests may not accurately characterize the effects of microwave heating on migration. Studies by Galotto and Guarda (1999, 2004) showed that the amount of chemical migration released from polyvinyl chloride fi lm into aqueous and fatty food simulants during microwave heating was higher than the amount released during oven heating. The objective of this study was to determine i f the effects of microwave heating on migration were different from the effects of oven heating. The amount of chemicals migrating from one type of polypropylene plastic container into acetic acid and isopropanol food simulants during microwave and oven heating was measured using G C / M S analysis and compared. The results of this study showed that the effects of microwave and oven heating on migration are not comparable. The amount of chemical migration for most substances identified in acetic acid and isopropanol food simulant was significantly higher (p<0.05) in simulant exposed to microwave heated plastic compared to migration levels measured in simulant exposed to oven heated plastic. The number of migrants found in microwave heated acetic acid and isopropanol simulants (13 and 72, respectively) was greater than what was present in oven heated acetic acid and isopropanol simulants (4 and 70, respectively). The amount of identified chemical migrants in isopropanol simulant was also significantly higher (p<0.05) than the amount of migration measured in acetic acid simulant. The types of migrants identified in both simulants exposed to microwave heated plastic included: hydrocarbons, fatty acids and fatty acid amides/esters, antioxidant and antioxidant breakdown products, a monomer, an alcohol and unknown substances. The large number of unidentified migrants and the lack of toxicological data for many identified migrants emphasize the need for improved migration test methods and food packaging regulations, and more toxicity testing. T A B L E OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES . vii LIST OF FIGURES ix ACKNOWLEDGEMENTS xi 1 INTRODUCTION 1 1.1 Current North American and European food packaging regulations 1 1.2 Literature review of studies that examine the effect microwave heating on plastic migration 9 1.3 Literature review of studies that examine the effect of repeated heating or reuse on plastic migration 10 1.4 Current Knowledge Gaps 21 1.5 Research Questions 21 1.6 Study aims 22 2 METHODS 23 2.1 Overall migration 23 2.2 Development of GC/MS migration test method 23 2.3 Plastic migrants released into food simulants during microwave heating 26 2.3.1 Chemical migration into acetic acid food simulant 26 2.3.2 Chemical migration into isopropanol food simulant 29 2.3.3 Analysis of chemical standards 31 2.3.4 GC/MS analysis 32 2.4 GC/MS data analysis methodology for acetic acid and isopropanol simulant sample extracts 32 2.5 Quantification of identified chemical migrants in acetic acid and isopropanol simulant sample extracts 34 2.6 Statistical analysis to determine the differences in migration between food simulant type and experimental treatments 35 iii 2.7 Physical and chemical properties, usage information, regulatory and toxicological background of potential chemical migrants released during microwave heating.... 36 3 RESULTS 37 3.1 Plastic migrants released into acetic acid food simulant during microwave heating 37 3.2 Plastic migrants released into isopropanol food simulant during microwave heating 49 3.3 Comparison of plastic migrants released into acetic acid food simulant to isopropanol food simulant during microwave heating 64 4 DISCUSSION 68 4.1 Plastic migrants released into acetic acid food simulant during microwave heating 68 4.1.1 Negative controls 68 4.1.2 The effect of exposure to microwave heated plastic on migration 69 4.1.3 Statistical analysis of migrant peak areas and concentrations 70 4.2 Plastic migrants released into isopropanol food simulant during microwave heating 71 4.2.1 Negative controls 71 4.2.2 The effect of exposure to microwave heated plastic on migration 72 4.2.3 Statistical analysis of migrant peak areas and concentrations 77 4.3 Comparison of results of acetic acid food simulant and isopropanol food simulant experiments 79 4.4 Physical and chemical properties, usage information, regulatory and toxicological background of potential chemical migrants released during microwave heating.... 79 4.4.1 Physical and chemical properties of identified chemical migrants 79 4.4.2 Usage information and regulatory background of identified chemical migrants 82 4.4.3 Toxicological properties of identified chemical migrants 86 5 CONCLUSION 93 5.1 Study limitations 93 5.2 Recommendations 95 REFERENCES 99 iv APPENDIX A PRELIMINARY EXPERIMENTAL RESULTS FOR THE DEVELOPMENT OF GC/MS ANALYSIS METHODOLOGY FOR CHARACTERIZING CHEMICAL MIGRATION FROM MICROWAVE HEATED PLASTIC 112 A.l Introduction 112 A.2 Microwave power level determination 113 A.2.1 Objective 113 A.2.2 Protocol 113 A.2.3 Results 113 A.3 Microwave tray cooling time determination 114 A.3.1 Objective ....114 A.3.2 Protocol 114 A.3.3 Results 114 A.4 Determination of cooling time for food simulants between microwave heating cycles 115 A.4.1 Objective 115 A.4.2 Protocol 115 A.4.3 Results 116 A.5 Determination of heat and volume loss during transfer of heated food simulant from a glass crystallizing dish to a plastic container 116 A.5.1 Purpose 116 A.5.2 Protocol ....116 A.5.3 Results 117 A.6 Determination of toaster oven heating stability and target temperature setting.... 118 A.6.1 Purpose 118 A.6.2 Protocol 118 A.6.3 Results 121 A.7 Determination of internal surface area of polypropylene container 121 A.7.1 Purpose 121 A.7.2 Protocol 122 A.8 Verification of retention time and mass spectra of surrogate standards 123 A.8.1 Purpose 123 A.8.2 Protocol 123 A.8.3 Results 124 A.9 Determination of instrument detection limits for surrogate standards and identified chemical migrants 125 A.9.1 Purpose : 125 A.9.2 Protocol.... '. 125 A.9.3 Results 125 v A.10 Surrogate standard recovery for extraction solvent comparison 127 A. 10.1 Purpose 127 A. 10.2 Protocol 127 A.10.3 Results 128 A. l l Determination of surrogate standard and chemical migrant recovery for migration test methods 128 A. 11.1 Purpose 128 A.l 1.2 Protocol 128 A.l 1.3 Results 129 A.12 Determination of migrant recovery over time 131 A.12.1 Purpose 131 A. 12.2 Protocol 131 A.12.3 Results 132 APPENDIX B CALIBRATION CURVES FOR STANDARDS USED TO QUANTIFY MIGRANTS IN THE ACETIC ACID AND ISOPROPANOL SIMULANT MIGRATION EXPERIMENTS 134 APPENDIX C ADDITIONAL TOXICOLOGICAL AND BIOLOGICAL EFFECTS OF CHEMICAL MIGRANTS IDENTIFIED IN ACETIC ACID AND ISOPROPANOL FOOD SIMULANT 141 vi LIST OF TABLES Table 1.1 Comparison of regulatory requirements for health safety approval of food packaging in Canada, the United States, and the European Union 4 Table 1.2 Summary of studies on the effect of microwave heating on chemical migration from plastic 12 Table 1.3 Summary of studies on the effect of repeated heating or reuse on chemical migration from plastic food packaging 18 Table 2.1 Summary of preliminary GC/MS method development experimental results 25 Table 2.2 Summary of chemical standards analyzed for migrant identification confirmation..... 31 Table 3.1 The presence of specific chemicals identified in the control sample extracts and microwave heated acetic acid simulant sample extracts 43 Table 3.2 Results from the statistical analysis of chemicals unique to microwave and oven heated acetic acid sample extracts (n=4) 47 Table 3.3 Results from the statistical analysis of quantified chemicals unique to microwave and oven heated acetic acid sample extracts (n=4) 48 Table 3.4 The presence of specific chemicals identified in the control sample extracts and oven and microwave heated isopropanol simulant sample extracts 56 Table 3.5 Results from the statistical analysis of chemicals unique to microwave and oven heated isopropanol sample extracts (n=4) 62 Table 3.6 Results from the statistical analysis of quantified chemicals unique to microwave and oven heated isopropanol sample extracts (n=4) 63 Table 3.7 Summary of the results for the unpaired t-tests which analyzed the average mass ion peak areas of substances found in the microwave heated and oven heated isopropanol sample extracts 64 Table 3.8 Comparison of chemicals identified to microwave and oven heated acetic and isopropanol sample extracts. Separate columns show a comparison in microwave heated sample extracts (column 2) and oven heated sample extracts (column 4) 65 Table 3.9 Average measured concentration of each migrant detected per heating in acetic acid or isopropanol sample extracts (n=4) 67 Table 4.1 Physical and chemical properties of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic 81 vii Table 4.2 Usage information and regulatory background of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic 83 Table 4.3 Toxicological properties of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic 87 Table A.l Average internal temperatures of toaster oven after reaching a target temperature range of 91 to 100°C leaving oven door open for 1 minute 121 Table A.2 Average internal temperatures of toaster oven after reaching a target temperature range of 75 to 85°C leaving oven door open for 1 minute 121 Table A.3 Chemicals selected as surrogate standards and their solubility 124 Table A.4 Estimated instrument detection limits of surrogate standards and identified chemical migrants from polypropylene plastic 126 Table A.5 Percent recovery of spiked standards in acetic acid sample extracts 130 Table A.6 Percent recovery of spiked standards in isopropanol sample extracts 130 Table A.7 Percent recoveries of spiked standards in acetic acid food simulant at the beginning and end of a migration experiment 132 Table A.8 Percent recoveries of spiked standards in isopropanol food simulant at the beginning and end of a migration experiment 133 Table C.l Toxicological and biological effects of chemical migrants identified in acetic acid and isopropanol food simulant exposed to microwave heated polypropylene plastic 141 v i i i LIST OF FIGURES Figure 3.1 Chromatograph of all selected anticipated and known food contact polypropylene plastic migrant standards dissolved in toluene 38 Figure 3.2 Chromatographs of the toluene blank and method blank acetic acid simulant sample extract, (a) Toluene (media) blank shows the presence of column bleed, (b) Method blank shows an occurrence of column bleed and two unknown contaminants 39 Figure 3.3 Chromatographs of the oven and microwave heated glass control acetic acid simulant sample extract, (a) Oven heated glass control sample extract and (b) microwave heated glass control sample extract show the presence of column bleed and two unknown contaminants 40 Figure 3.4 Chromatographs of a room temperature control and an oven heated acetic acid simulant sample extract, (a) Room temperature control sample extract and (b) oven heated glass sample extract show the presence of column bleed, 2,4-di-tert-butylphenol and two unknown contaminants 41 Figure 3.5 Chromatograph of a microwave heated acetic acid simulant sample extract.... 42 Figure 3.6 Average select mass ion peak areas of substances identified in sample extracts of acetic acid solution that were exposed to polypropylene plastic containers 45 Figure 3.7 Chromatographs of the toluene blank and method blank isopropanol sample extract, (a) Toluene (media) blank shows the presence of column bleed, (b) Method blank shows an occurrence of column bleed and an unknown phthalate 50 Figure 3.8 Chromatographs of an oven and a microwave control isopropanol sample extract, (a) Oven heated glass control sample extract shows the presence of column bleed, (b) Microwave heated glass control sample extract shows the occurrence of column bleed, an unknown phthalate and a hydrocarbon (HC) contaminant series 51 Figure 3.9 Chromatograph of a room temperature control isopropanol food simulant sample extract 52 Figure 3.10 Chromatograph of an oven heated isopropanol simulant sample extract 53 Figure 3.11 Chromatograph of a microwave heated isopropanol simulant sample extract 54 Figure 3.12 Mass ion peak areas of chemicals positively identified in microwave heated isopropanol sample extracts ; 60 Figure A.l Calibration curve for paper surface area versus weight 122 ix Figure B.l Calibration curve for 2,4-di-tert-butylphenol standard analyzed for the acetic acid migration experiment 134 Figure B.2 Calibration curve for erucamide standard analyzed for the acetic acid migration experiment. 135 Figure B.3 Calibration curve for 2,4-di-tert-butylphenol standard analyzed for the isopropanol migration experiment 135 Figure B.4 Calibration curve for octadecanol analyzed for the isopropanol migration experiment 135 Figure B.5 Calibration curve for erucamide standard analyzed for the isopropanol migration experiment 136 Figure B.6 Calibration curve for Irganox 1076 standard analyzed for the isopropanol migration experiment 136 Figure B.7 Calibration curve for 2,6-di-tert-butylbenzoquinone standard analyzed for the isopropanol migration experiment 137 Figure B.8 Calibration curve for isopropal palmitate standard analyzed for the isopropanol migration experiment 137 Figure B.9 Calibration curve for stearic acid standard analyzed for the isopropanol migration experiment 138 Figure B.10 Calibration curve for tetradecane standard analyzed for the isopropanol migration experiment 138 Figure B.ll Calibration curve for pentadecane standard analyzed for the isopropanol migration experiment 139 Figure B.12 Calibration curve for heptadecane standard analyzed for the isopropanol migration experiment 139 Figure B.13 Calibration curve for octadecane standard analyzed for the isopropanol migration experiment 140 A C K N O W L E D G E M E N T S I would like to thank Dr. Chris van Netten and Dr. Hugh Davies for their guidance and encouragement throughout my complex and challenging thesis. I would also like to thank Dr. Ray Copes and Dr. Derek Gates for participating in my thesis committee - their feedback during the method development process is greatly appreciated. The Bridge Program and Health Canada provided funding for my research and education during my Master's program. I would like to thank Dr. Kay Teschke and the Bridge Program Coordinating Committee for providing me with the opportunity to participate in such an innovative interdisciplinary program. I would also like to thank Roger Sutcliffe and Bette Meek at Health Canada for their academic and financial support. The SOEH laboratory staff played an important role in my laboratory research. I would like to thank Dr. Winnie Chu for the use of laboratory space and equipment. I am infinitely indebted to Tim Ma for sharing his invaluable knowledge and time, his assistance with setting up and fixing equipment, and his patience. I would like also like to thank Tim Ma and Tom Barnjak for their great sense of humour which helped to make laboratory work go by quickly. I would like to thank my friends for their emotional support through some of the most difficult times in my life. I would like to especially thank Catherine, Rasha, Taskin, Nicole, Mary, Ed, Domenic, Daniela, Kathleen, Jennifer, and Steve. I would like to thank my family for their encouragement - I am extremely fortunate to have my sister and parents standing by me. I am eternally grateful for my father who kept me well-nourished, provided much-needed company during my long hours in the laboratory, and helped me find the strength to keep going despite all the seemingly endless obstacles. xi 1 INTRODUCTION Plastics are widely used in a variety of food packaging applications including polyvinyl chloride films, which are used to wrap or cover food during storage, and plastic containers, which are used for food storage and heating. However, plastic materials contain polymerization residues (such as monomers) and several types of additives (such as plasticizers) which are incorporated into the plastic matrix to improve the performance and durability of polymeric packaging materials (Lau and Wong, 2000). These substances are capable of undergoing a process called migration during which chemicals (known as migrants) present in plastic diffuse through the polymer matrix to the surface where they can contaminate food (Incarnato et al, 2000). The consumption of migrants in food can pose a potential threat to consumer health, if they are toxic. Examples of hazardous substances that have been identified in previous migration studies include: lead, styrene, and phthalates (e.g. di(2-ethylhexyl)phthalate) (Lau and Wong, 2000; Fordham et al, 1995). The potential endocrine disruptor, nonylphenol, was found in samples of bottled water in concentrations ranging from 180 to 300 ng/L (Loyo-Rosalas et al, 2004). This thesis will examine current North American and European food packaging regulations and review scientific literature that have studied the effects of repeated use and microwave heating on plastic migration. A specific migration test method will be developed to address knowledge gaps in our understanding of how microwave heating affects the migration of chemicals from plastics intended for food use. 1.1 Current North American and European food packaging regulations Concerns about the potential for public exposure to harmful migrants in plastic packaging has led to the establishment of legislation by government bodies such as the European Union that requires "migration testing" of food contact plastics and the establishment of migration limits for specific plastic additives (Nerin et al, 2002). Current standardized methods of migration testing involve the heating of food simulants in plastic packaging under various test conditions to identify potential migrants and determine the maximum amount of migration that will occur. The amount of migration from plastic into food is dependent on the physical and chemical properties of the polymer and food, as well as external factors such as the duration of testing and temperature. Exposure to high temperature is particularly important because the rate of 1 migration is proportional to temperature. Thermal breakdown products of the plastic, plastic residues and/or additives may also contaminate food when plastic is heated (Kim-Kang, 1990). Migration testing for food contact plastics in microwave ovens is critical for determining the extent of consumer exposure to potentially toxic substances. However, conventional ovens are presently used in standardized migration testing which may not accurately characterize the effects of microwave heating. The general population heats foods in containers not intended for microwave heating and although microwave ovens can heat foods to conventional oven temperatures, microwave heating usually occurs over short periods of time and the food contents are not heated uniformly. Castle et al. (1990) measured the temperature of ready-prepared foods heated in various microwaveable plastic containers and articles according to the manufacturers' instructions. The temperature of foods heated in the microwave ranged from 61°C to 121°C. However, the temperature measured at localized spots on the surface of microwave heated foods exceeded 200°C. Thus, the use of microwaves for migration testing may provide results that are different from those obtained from current standardized conventional oven tests. A review of the existing laws and regulations for ensuring the safety of food packaging materials in Canada, the US and the European Union was carried out by conducting a search on PubMed and government websites. A summary of the findings are described in Table 1.1. It was apparent from the review findings that the following issues were not adequately addressed: • Current regulations accept migration test methods that use conventional oven heating to determine the amount of chemical migration from food packaging. There is no explicit requirement for testing food packaging intended for microwave use in microwave ovens; • Standard methods for testing migration from food packaging during use in microwave ovens in Canada, the US and the EU do not exist; • Regulations inadequately address the effects of long term use and repeated heating on the type and amount of chemical migration. A full range of use conditions that are likely to affect the variability in chemical migration are also unaccounted for; • Canadian regulatory migration testing requirements are not well-defined or as thorough as American or European regulations. More specifically, current Canadian food packaging laws and regulations need to address: a food packaging definition that includes all types food contact materials; packaging material/chemical migrant stability, impurities, decomposition/transformation products, and reactions with food; toxicological testing of chemical migrant reaction by-products; the lack of more comprehensive toxicological testing; which migration tests are acceptable (i.e. standardized and non-standardized); the inclusion 2 of migration test modelling; and the regulatory status of the packaging material/chemical migrants in other countries. Thus, it is clear that current migration test requirements do not adequately characterize public exposure to plastic migrants. A standardized method for testing chemical migration from food packaging using microwave ovens also needs to be developed and incorporated into food packaging regulations to identify potentially toxic chemical migrants and protect public health. 3 Table 1.1 Comparison of regulatory requirements for health safety approval of food packaging in Canada, the United States, and the European Union Regulatory Requirements Canada1 United States2 European Union3 Definition of Food Packaging • A thing in which any food, drug, cosmetic or device is wholly or partly contained, placed or packed4 • Any substance that is intended for food use as a component of materials used in manufacturing, packing, packaging, transporting or holding food if the use is not intended to have any technical effect in the food5 • A l l materials and articles intended to come into contact with foodstuffs or water intended for human consumption, which includes: cutlery, processing machines, and containers. • Fixed public or private water supply equipment is excluded. Product Identity • Chemical name, formula and composition (i.e. impurities) • Chemical and physical properties • Manufacturing process • Chemical name, formula and composition • Chemical and physical properties • Manufacturing process including substances used to manufacture packaging materials • Chemical name, formula and composition including analytical methods used to determine the impurities and supporting documentation • Chemical and physical properties - an extensive list of data is required including stability, interaction with food, and unintentional decomposition or transformation products • Manufacturing process including substances used to manufacture packaging materials and any by-products ' (Health Canada, 2003) 2 (US F D A / C F S A N , 2002a; US F D A / C F S A N , 2002b; US F D A / C F S A N , 2003) 3 (Health and Consumer Protection Directorate-General of the European Commission, 2004b; Europa, 2006b; Europa, 2006c) 4 Canadian Food and Drugs Act, R.S. 1985, c. F-27, s. 1 5 (21 CFR 170.3(e)(3)) Table 1.1 (continued) Comparison of regulation requirements for health safety approval of food packaging in Canada, the United States, and the European Union Regulatory Requirements Canada United States European Union Intended Uses • Environmental conditions packing materials will be exposed to during packaging, transport and consumer use • Intended technical effect or purpose of each chemical component • Maximum concentration of each chemical component • Data to support that the chemical components will achieve its intended use at the levels present in the packaging material(s) • Types of foods packaging is used for • Range of possible uses and use conditions of packaging • Intended technical effect or purpose of each chemical component • Maximum concentration of each chemical component • Data to support that the chemical components will achieve its intended use at the levels present in the packaging material (s) • Types of food packaging is intended for • Maximum temperature and time the packaging is in contact with food • The technical function and percentage of the substance in the food contact material • The use conditions of the food packaging including the food the packaging will be in contact with, the maximum time and temperature the packaging will undergo during consumer use, and treatment of packaging prior to use (i.e. sterilization). Exisiting Authorization N / A N / A Information regarding the authorization for use of the substance, including restrictions on use conditions, in other countries should be submitted. Table 1.1 (continued) Comparison of regulation requirements for health safety approval of food packaging in Canada, the United States, and the European Union Regulatory Requirements Canada United States European Union Migration Data Migration study data are to be submitted detailing: • Migration levels • Study parameters/test conditions. 95% aqueous ethanol, 10% aqueous ethanol, traditional food simulants such as distilled water, 3% acetic acid, 15% ethanol, heptane and/or HB 307 (a synthetic triglyceride) are accepted. • Analytical method • Detection limit Studies must simulate actual use conditions, use simulants that closely reflect the food type the packaging is going to be exposed to. Standard migration test protocols which involve the use of conventional ovens are recommended and are described in detail. Each protocol is specific to the type of food packaging, the packaging use conditions and the type of food the package will exposed to. Results from non-standard and standardized migration test methods are accepted provided that the testing parameters reflect the most severe actual use conditions the packaging is subjected to and the US FDA Center for Food Safety and Applied Nutrition (CFSAN) recommended migration experimental design. Detailed documentation on the analytical methods used to identify and quantify chemical migrants is required. Standardized migration testing of the chemical migrant and its reaction by-products from the food contact article should be conducted using foods under the most extreme intended use conditions. Accepted standard migration test methods require the use of conventional ovens. Manufacturers must provide data on migrant concentrations in food and residual migrant levels in the food contact material. Alternative methods of migration testing and migration modeling are acceptable provided that a rationale is presented. A detailed description of the standardized analytical method used to identify and quantify the food contact substance and its reaction by-products should also be provided. Migration Testing in Microwave Ovens N / A The potential for variability in migration testing for microwavable containers is recognized. Thus, it is recommended that manufacturers consult the US FDA prior to migration testing. Migration testing of packaging materials during microwave heating may be addressed in the future. Migration Testing for Repeated Use Articles N / A Migration testing of "repeated use" packaging material using two types of aqueous and fatty food simulants for a maximum of 240 hours at the highest use temperature. Migration testing of "repeated use" packaging material additives is not required if the estimated migration concentration over the lifetime use of the article is "sufficiently low". Migration testing of "repeated use" packaging material must be conducted three times on a single sample. The migration level must not exceed migration limits in the first experiment and the amount of migration must not increase during the second or third test or further testing will be required. Table 1.1 (continued) Comparison of regulation requirements for health safety approval of food packaging in Canada, the United States, and the European Union Regulatory Requirements Canada United States European Union Toxicological Data The number and type of toxicological studies required depends on the amount of the chemical migrant expected to be consumed from exposure to contaminated food. Substances that have been estimated to pose the greatest health threat (i.e. the potential daily intake of the substance is greater than 25 mg/kg bw/day) the following studies are required: • data for structure activity of the substance; • two short term genotoxicity studies • a 90-day feeding study in rodents; • a multi generation study in rodents; • a teratology study in rodents; • a 1-year feeding study in non-rodents • a chronic toxicity/oncogenicity study in two rodent species A description of the methodology and results from a set of recommended standardized animal toxicity tests conducted on the food contact substance (FCS) and its breakdown products/impurities must be provided including: • estimates of the acceptable daily intake (ADI) • all adverse health effects • worst-case, upper-bound, and lifetime risk levels for carcinogenic FCS breakdown products. The extent of toxicological testing is dependent on the estimated migrant population exposure concentration the population. Toxicological data required for assessment is greatest for chemical migrants with a cumulative estimated daily intake (EDI) rate between 150 Ug and 3 mg/person/day; animal genetic toxicity tests and/or subchronic tests are recommended. If the estimated cumulative EDI migrant rate is greater than 3 mg/person/day the chemical is considered a food additive and must undergo a different regulatory screening process. Results from a set type and number of standardized toxicological studies must be provided on the food contact substance, including its breakdown/transformation products. The extent of toxicological testing is dependent on the amount of specific and/or overall migration of the substance into food. Contaminant migration levels greater than 5 mg/kg of food requires the largest toxicological data set which includes: • 3 in vitro mutagenicity tests; • 90-day oral toxicity studies; • toxicokinetic studies; • reproductive and developmental toxicity studies; • long-term toxicity and/or carcinogenicity tests. Environmental Data N / A The information that should be provided for the US FDA C F S A N to determine if the use of the FCS and/or its degradation by-products will have no adverse impacts on the environment includes: • the fate of the FCS and its degradation products in the environment during FCS use and disposal; • the toxic effects of the FCS and its degradation products on organisms in the environment; • proposed mitigation measures for reducing the negative effects of the use of the FCS on the environment. N / A Table 1.1 (continued) Comparison of regulation requirements for health safety approval of food packaging in Canada, the United States, and the European Union Regulatory Requirements Canada United States European Union Migration Limit(s) N / A N / A A specific migration limit (SML) and/or maximum permitted quantity for individual authorized substance(s) may be established. An overall migration limit of 60 mg of substance(s)/kg of food or food simulant also applies to all chemical migrants. Safety Assessment Exemption Criteria Migrants that have a potential daily intake of less than 0.025 ug/kg bw/day are considered to have no significant toxicological impacts and therefore no adverse health effects; migrants that meet this criteria are exempt from further regulatory screening. A l l migrant sources are not incorporated in the estimated daily intake calculations. A chemical migrant is exempt from a safety assessment by the US FDA C F S A N if supporting documentation shows that: • it is non-carcinogenic and has non-carcinogenic impurities6; • expected dietary consumption concentration of the FCS is <1.5 ug/person/day7; • The FCS does not negatively effect/alter the food that it migrates into8; • The use of the FCS has no adverse effects on the environment9. N / A 6 (21 CFR 170.39 (1)) 7 (21 CFR 170.39 (2)) 8 (21 CFR 170.39 (3)) 9 (21 CFR 170.39 (4)) 1.2 Literature review of studies that examine the effect microwave heating on plastic migration Due to the increase in domestic ownership of microwaves and the development of plastic containers designed for microwave use, a number of studies have been conducted to determine if microwave energy, reheating and/or reuse has an effect on chemical migration from food packaging. A general literature search was conducted to identify research studies that examined the effects of microwave heating, repeated heating and/or reuse on chemical migration from food packaging. Tables 1.2 and 1.3 summarize the main objectives, methods, findings and limitations of the studies selected from the literature search. Based on a critical review of six studies that examined the effects of microwave heating on chemical migration from food packaging, many variables, such as heating time and temperature, and food simulant and plastic type, were consistently shown experimentally to affect migration (See Table 1.2). However, the following common study limitations were identified: • There were no toxicological criteria for the target chemical migrants. Migrants were generally selected because they were known to be present in specific types of plastic or were previously measured in other studies. • The experimental designs of the studies were not robust. For example, many studies had few or no replicate samples, negative or positive controls, a disproportionate number of samples between comparison groups and/or large variability within comparison groups. For example, in the study by Castle et al. (1989) only 2 of the 12 types of food were cooked in the microwave and conventional oven. • The difference in the amount of migration from microwave heating compared to oven heating could not be determined because there was inconsistency between the microwave and oven cooking temperatures and times. For example, Rijk and De Kruijf (1993) compared overall migration in olive oil exposed to polypropylene that was heated in a microwave for 18 minutes (which reached 122- 125°C) to overall migration in olive oil heated in an oven for 2 hours maintained at 150°C. • The migration of individual potential contaminants into foods and/or food simulants was not analyzed, therefore the effect of microwave heating on a range of chemicals was not captured (Jickells et al., 1992; Inthorn et al., 2002). 9 • Few studies examined a complete cross section of variables that would provide a more complete data set on the effects of a specific variable on migration i.e. instead of testing both simulant types and a range of plastics, only fatty food simulants or aqueous food simulants and one type of plastic were tested. • The methodology of some studies was poorly explained. For example, in the study by Galloto and Guarda (2004) there was no explanation regarding how microwave heating times and power levels were selected. It is difficult to conclude with confidence that past studies showed a difference between the effect of microwave and oven heating on migration. Thus, it is important that a scientifically sound study be conducted that systematically eliminates or controls factors that interfere with the comparability of the effects of microwave and oven heating. 1.3 Literature review of studies that examine the effect of repeated heating or reuse on plastic migration A critical review was also conducted on four studies that researched the effects of reheating and/or reuse on chemical migration (See Table 1.3). Overall, the results of the studies indicated that the amount of migration appears to decrease or increase with each successive heating depending on the chemical and certain characteristics of the plastic which change over time due to reuse (i.e. repeated washing and reheating). However, it is important to note the following limitations that were identified in the studies: • The duration of the experiment and number of heatings did not adequately capture the effects of repeated heating over the lifetime of plastic food containers. • Study designs were not systematic or rigorous i.e. there were no negative or positive controls and there were inconsistencies in the number of replicates for each experimental treatment. • Only one chemical was examined in studies by Castle et al. (1989) and Inthorn et al. (2002) which may not have adequately characterized the effect of repeated heating on other migrants. • Migrants were generally selected because they were known to be present in specific types of plastic (e.g. polyethylene terephthalate (PET) oligomers) or known to potentially affect the organoleptic properties of foods stored in plastic (e.g. acetaldehyde). Based on the findings of these studies, it is apparent that reheating and reuse affect the integrity of plastic food packaging. However, the cumulative effects of long-term use on chemical migration have not been effectively addressed which is imperative given that plastic food 10 containers are used over many years by the general public regardless of their intended use and life span. 11 Table 1.2 Summary of studies on the effect of microwave heating on chemical migration from plastic. Author(s) and Publication Year Objective(s) Method(s) Results and Conclusions Study Limitations Castle et al., 1989 • To compare the total amount of polyethylene terephthalate (PET) oligomer migration into food from P E T containers during microwave heating to the amount released during conventional oven heating. • A variety of foods were cooked in roasting bags or plastic trays in a conventional oven at 204°C for 30 to 90 minutes or in a microwave oven for 1.5 to 3 minutes at maximum power. • The amount of P E T oligomer migration was determined by gas chromatography with a mass spectrometer. • Foods cooked at a higher temperature and longer duration generally had higher amounts of P E T oligmer migration and therefore appear to have the greatest effect on P E T oligomer migration. • P E T oligomer migration was highest in fatty foods cooked in the conventional oven and microwave. • Although most foods cooked in the oven had higher levels of P E T oligomer migration it appears that microwave heating is also capable of influencing migration despite cooking foods for a shorter time and at a lower temperature. • No toxicological criteria for selecting P E T oligomers as the target migrant. • Only 2 of the 12 types of food analyzed were cooked in the microwave and the conventional oven. • Foods analyzed varied in composition and shape which affected cooking temperature and size of contact area with food packaging. • Cooking temperatures and times were not consistent between microwave and conventional oven and within foods that were microwave heated. • Temperature of foods prior to cooking was not indicated. • Type of P E T container tested varied depending on the food and cooking method. Only one type of food was cooked in the same type of P E T container in the microwave and conventional oven. • Only one set of replicate samples were analyzed. • A l l potential migrants were not analyzed. Table 1.2 (continued) Summary of studies on the effect of microwave heating on chemical migration from plastic. Author(s) and Publication Year Objective(s) Method(s) Results and Conclusions Study Limitations Castle etal., 1990 To compare the amount of PET oligomer and benzene migration from PET and thermoset polyester cookware, respectively, into various food simulants and foods during heating in a conventional oven compared to the amount released during heating in a microwave oven. • Pre-heated aqueous and fatty food simulants were heated in PET and thermoset polyester cookware in an oven for 30 to 120 mins at 100 to 175°C. • A variety of foods were cooked in PET trays in an oven at 204°C for 40 to 80 minutes and in thermoset polyester cookware in an oven for 40 to 90 minutes. • 4 types of food were cooked in PET trays and thermoset polyester cookware in a microwave for 2 to 4 minutes on high power. • The amount of PET . oligomer and benzene migration was determined by gas : chromatography with a mass spectrometer (GC/MS) and headspace GC/MS analysis, respectively. • PET oligomer and benzene migration into oven heated food simulants was consistently higher than the amounts detected in food cooked in the oven. • PET oligomer migration into olive oil was greater than 10 times the amount found in acetic acid. • PET oligomer and benzene migration was lower in microwave heated foods than oven heated foods. However, the lower amount of benzene migration in microwaved foods is likely due to evaporative loss. • The lower amount of PET migration in microwave heated food was likely due to the fact that the food was heated in the microwave oven for shorter time periods and generally lower temperatures (depending on the fat content of the food) than oven cooked food. No toxicological criteria for selecting PET oligomers as the target migrant. Only 1 of the 5 types of food cooked in PET trays was in the microwave and the conventional oven. Foods analyzed varied in composition and shape which affected cooking temperature and size of contact area with food packaging. Cooking temperatures and times were not consistent between microwave and conventional oven and within foods that were microwave heated. More food simulant samples and foods were heated in PET trays in an oven compared to what was microwave-heated. No food simulants were heated in the microwave. Temperature of foods prior to cooking was not indicated. No control or blank samples were analyzed. Only one set of replicate food simulant samples were analyzed; no replicate food samples were analyzed. All potential migrants were not analyzed. Table 1.2 (continued) Summary of studies on the effect of microwave heating on chemical migration from plastic. Author(s) and Publication Year Objective(s) Method(s) Results and Conclusions Study Limitations Jickellset al., 1992 • To determine if • 5 types of microwave • Some migrants measured in • No toxicological criteria for microwave heating pre-heated and non- olive oil and iso-octane were 3- selecting the target migrants. had an effect on the treated plastic were 11 % higher in microwave treated • No blank samples were rate of migration from filled with or plastic compared to non treated analyzed. food contact plastics immersed in olive oil plastic. • Microwave heating parameters into fatty-food and heated in an oven • There was no significant did not simulate actual use simulants. at temperatures difference (p = 0.05) between the conditions. ranging from 40°C to level of contaminant migration • Effects of microwave heating on 150°Cfor 0.5 to 24 for the majority of contaminants migration into olive oil was hours. tested into olive oil or iso-octane determined indirectly since • 4 types plastic were from microwave heated plastic migration was measured in immersed in iso- compared to non microwave simulant after it was heated in a octane and heated in a heated plastic. conventional oven. microwave oven and • There was a potential water bath at 40°C for evaporative loss of migrants 30 minutes. from plastic during the • A range of analytical microwave pre-heating therefore techniques were used the amount of migrants in olive to quantify the oil may be underestimated. following migrants in • Only one category of migrant the food simulants: was analyzed for each plastic volatile organic type. compounds, • Only fatty food simulants were antioxidants, PET tested. oligomers, and a • The effects of microwave plasticizer. heating on migration into iso-octane may have been underestimated since the plastics were heated at such a low temperature. Table 1.2 (continued) Summary of studies on the effect of microwave heating on chemical migration from plastic. Author(s) and Publication Year Objective(s) Method(s) Results and Conclusions Study Limitations Rijk and De Kruijf, 1993 • To compare the amount of overall migration of non-volatile contaminants from plastic containers into olive oil during microwave heating to migration during conventional oven heating. • A variety of plastic packaging materials were filled with or immersed in olive oil and were heated in the microwave and in a conventional oven according to standard test methods. • The standard test method was modified for microwave heating by selecting a microwave heating temperature for each type of packaging that heated the centre of the olive oil sample to greater or equal to 70°C for 2 minutes at a set power level. • The amount of overall migration was determined gravimetrically. • Overall migration during microwave heating compared to conventional oven heating was lower for all types of plastic tested except for one sample of PET coated board. • The amount of overall migration during microwave and conventional oven heating were in general agreement because migration values were within 3 mg/dm2 of each other. Thus, the overall migration into olive oil from packaging materials intended for food use may be determined by migration testing using conventional heating • No toxicological basis for targeting overall migration of non-volatile contaminants. • Gravimetric analysis did not allow for identification of specific migrants. • No blank samples and few replicates were analyzed. • Only one negative control for the microwave and conventional oven was analyzed. • Negative migration values were calculated and not addressed. • Only one type of food simulant was tested. • Cooking temperatures and times were not consistent between the microwave and conventional oven. • The microwave heating temperature was within ±20°C of the oven temperature for the same time of plastic; however, the microwave heating time period was approximately >50% shorter than standard oven heating times. • No explanation was provided for the large amount of migration from PET coated board during microwave heating compared to oven heating. Table 1.2 (continued) Summary of studies on the effect of microwave heating on chemical migration from plastic. Author(s) and Publication Year Objective(s) Method(s) Results and Conclusions Study Limitations Galotto and Guarda, 1999 • To compare the effect of microwave heating to conventional oven heating on overall migration of non-volatile contaminants into aqueous food simulants. • 3 types of aqueous food simulants were heated in 6 types of plastic food packaging in a microwave at maximum power for 3 minutes and in a conventional oven under standard test methods (i.e. 40°C for 10 days or 80 and 121°Cfor 30 minutes). • The amount of overall migration in the food simulants samples was determined gravimetrically. • Out of the 6 types of plastics tested, polyvinyl chloride (PVC) plastic was the only type that released higher amounts of migrants into all 3 food simulants during microwave heating than during conventional oven heating. • There was no consistency in the amount of overall migration in a specific food simulant or the type of migration test. • Microwave heating appears to have an effect on the amount and/or the rate of migration from P V C because the food simulants were heated for a very short period of time and at lower temperatures. • The amount of overall migration is highly dependent on the type of plastic, food simulant type, time and temperature conditions, and possibly type of heating. • N o toxicological basis for targeting overall migration of non-volatile contaminants. • Gravimetric analysis did not allow for identification of specific migrants. • N o blank or control samples were analyzed. ' -• Cooking temperatures and times were not consistent between the microwave and conventional oven. • No fatty food simulants were tested. • Limited description on the methodology used to determine overall migration and no discussion of potential errors. • N o description of how microwave heating time or power level was selected. • Temperature of food simulant during/after microwave heating was not measured therefore it is not known what the difference in the temperature of the food simulant was in the microwave compared to the oven. Table 1.2 (continued) Summary of studies on the effect of microwave heating on chemical migration from plastic. Author(s) and Publication Year Objective(s) Method(s) Results and Conclusions Study Limitations Galotto and Guarda, 2004 • To compare the effect of microwave heating to conventional oven heating on overall migration of non-volatile contaminants into fatty food simulants. • 4 types of fatty food simulants were heated in PVC and polypropylene (PP) in an oven under standard migration test conditions and in a microwave at the 800W power level for 2 different time periods. • Overall migration was analyzed using GC/MS with a flame ionization detector. • Overall migration from PVC film into all four fatty food simulants during microwave heating was consistently higher than oven heating. • Migration from PVC during microwave and oven heating into olive oil, isopropanol and 95% ethanol was generally 5 to 10 times higher than what was measured in n-heptane. • n-Heptane contained the lowest amount of migration from PP during oven and microwave heating. • Overall migration from PP into olive oil was the highest compared to the other simulants for both heating methods. • Microwave heating of isopropanol in a PP container for 1 minute and 30 seconds was the only heat treatment that resulted in a higher amount of migration from PP than oven heating. Oven heating resulted in the highest amount of migration from PP for all other food simulant and plastic types. • The heating temperature and duration, heating type combined with the type of food simulant may affect the amount or rate of chemical migration depending on the type of plastic. • No toxicological basis for targeting overall migration of non-volatile contaminants. • Gravimetric analysis did not allow for identification of specific migrants. • No blank or control samples were analyzed. • Cooking temperatures and times were not consistent between the microwave and conventional oven. • No aqueous food simulants were tested. • Limited description on the methodology used to determine overall migration and no discussion of potential errors. • No description of how microwave heating time or power level was selected. • Temperature of food simulant during/after microwave heating was not measured therefore the difference in the temperature of the food simulant in the microwave compared to the oven was unknown. Table 1.3 Summary of studies on the effect of repeated heating or reuse on chemical migration from plastic food packaging. Author(s) and Publication Year Objectives Method(s) Results and Conclusions Study Limitations Castle et al., 1989 • To examine the effect of repeated heat exposure on the amount of P E T oligomer migration into olive o i l . • Pre-heated olive oil was heated in P E T containers in a conventional oven for 2 hours at 175°C and removed for analysis • The experiment was repeated 5 times using new samples of olive oi l and the same tray. • After a 3 week resting period, the experiment was repeated 3 times using the same tray. • The amount of P E T oligomer migration was analyzed by gas chromatography with a flame ionization detector. • The amount of PET oligomer migration was highest during the first heating and gradually decreased with each subsequent heating. • P E T oligomer migration decreased the most dramatically (by approximately 50%) between the first and second heating whereas the amount of migration was more gradual between the 3rd and 8th heatings. • P E T oligomer migration decreased by approximately 80% by the 8th heating. • Oligomer diffusion to the surface of the plastic appears to have an effect on the amount of migration. • N o toxicological criteria for testing P E T oligomers. • P E T trays were heated 8 times; actual lifetime P E T tray use was not specified. • The time period of the experiment and number of heatings did not capture the effects of repeated heating over a lifetime of use. • Migration testing took place in a conventional oven. • N o control or blank samples were analyzed. • Only one set of duplicate samples were analyzed. • Only one type of chemical migrant and food simulant were tested. Inthorn et al., 2002 • To determine the effects of p H , temperature, heating duration, and repeated heating on the amount of lead that was leached from plastic ware containing lead pigments. • Three acetic acid solutions were microwave heated at three heat levels for three time periods three times per container. • Lead migration was quantified using an atomic absorption spectrophotometer. • The amount of lead leaching into the food simulant was inversely proportional to pH level, and the number of heating cycles, but was proportional to both temperature and duration of heating. • The decrease in the rate of lead migration with each heating appeared to be linear. • Only one migrant and food simulant type were tested. • No control or blank samples were analyzed. • The number of heatings did not capture the effects of repeated heating over a lifetime of use. Table 1.3 (continued) Summary of studies on the effect of repeated heating or reuse on chemical migration from plastic food packaging. Author(s) and Publication Year Objectives Method(s) Results and Conclusions Study Limitations Jetten and de • To determine the • PET and polycarbonate • Overall migration from PET and • No toxicological criteria for Kruijf, 2002 effects of reuse on (PC) bottles were PC appeared to increase between selecting target migrants. the surface washed commercially the 1st and 15th washing cycle • The number of washings characteristics (i.e. once and fifteen times. when exposed to specific fatty and heatings did not capture hydrophobicity, fat • PP cups were washed food simulants and decrease when the cumulative effects on absorption), once, twice or five exposed to aqueous food the amount of migration physical properties times. simulants. and physical integrity of and amount of • Overall and specific • Specific migration of various plastic over a lifetime of overall and specific migration were chemicals from PET and PC repeated use. migration from determined by heating increased or decreased slightly • Inconsistent number of three types of various food simulants between the 1st and 15 th replicate samples and plastic food in each type of plastic washings. controls for each containers. according to • Hydrophobicity of PC bottles experimental treatment and standardized test decreased significantly after 15 number of washings for methods. wash cycles. each type of plastic. • Differences in surface • Overall migration from PP was • No unwashed PC and PET characteristics of all only affected when exposed to samples were tested for plastic containers were 95% ethanol after washing; overall and specific determined using migration increased after the 1 st migration therefore it was several analytical washing and decreased after the not possible to determine techniques. 5th washing. the effect of one washing • Changes in the integrity • Specific migration of various cycle on migration. and vapour barrier of substances from and • A different form of each PET and PC were hydrophobicity of PP increased or type of plastic was tested. determined by decreased depending on the • The number and type of measuring the number of washing cycles. food simulants tested varied permeation of water • Thus, the amount of overall depending on the type of vapour, carbon dioxide migration varied depending on the plastic. and oxygen. type of plastic, simulant and number of washing cycles. Table 1.3 (continued) Summary of studies on the effect of repeated heating or reuse on chemical migration from plastic food packaging. Author(s) and Publication Year Objectives Method(s) Results and Conclusions Study Limitations Brede et al., 2003 • To determine the effects of repeated machine washing, boiling and brushing on the amount of bisphenol A (BPA) migration from polycarbonate (PC) baby bottles. • 12 brands of PC bottles were rinsed, washed repeatedly in a laboratory dishwasher with detergent, brushed and boiled. • PC bottles were then filled with hot water and heated in an oven at 100°C for 1 hour. • BPA migration was analyzed using solid phase extraction and GC/MS. • Average migration concentrations of BPA from all 12 bottles after 51 and 169 rounds of dishwashing and were 7 and 5.6 pg/L, respectively, compared to 0.18 pg/L released from unwashed bottles. • Repeated washing, boiling and brushing appears to significantly affect the amount of BPA migration from PC. • It is hypothesized that the increase in BPA migration may be a result of polymer degradation. • PC bottles were not heated in an oven instead of a microwave, which is more commonly practiced. • No replicates of each brand of PC bottles were tested. • No negative controls were tested. • Only one food simulant type and migrant were tested. Skjevrak et al., 2005 • To determine the amount of 2,4-di-tert butyl phenol migrating from plastic electric kettles into water after repeated boilings. • 1 portion of water was boiled once and 1 portion of water was boiled three times in 8 brands of polypropylene kettles • 1 portion of water was boiled once a week for 4 weeks and once every 2 weeks for 2 months in 2 brands of polypropylene kettles • 2,4-di-tert-butylphenol migration was analyzed using purge and trap or solid phase extraction and GC/MS. • The concentration of 2,4-di-tert-butylphenol was higher by ~1 to 175% in water that was boiled three times; The percentage difference in 2,4-di-tert-butylphenol concentration was dependent on the kettle brand. • The concentration of 2,4-di-tert-butylphenol in water exposed to 2 kettle brands over a 12 week period peaked within the first 3 weeks and decreased gradually to a stable concentration by the fourth week. • The volume of water boiled is unknown. • No replicate samples were tested. • No negative controls were tested. • Only one chemical migrant and one food simulant type was tested. to o 1.4 Current Knowledge Gaps Recent studies have not sufficiently addressed the following issues that are critical to understanding what potentially harmful substances the public is exposed to as a result of using plastic food containers: • The amount of migration is highly dependent on the type of plastic and food simulant, and heating temperature and duration. However, the majority of migration studies used test methods that determine the amount of migration during oven heating despite the fact that microwaves are more commonly used to heat foods stored in plastic containers. • Studies that compared the effects of oven and microwave heating on migration did not use consistent or realistic heating temperatures and times therefore whether microwave heating has a greater effect on chemical migration cannot be concluded with great certainty (Castle et al, 1989, 1990; Jickells et al., 1992; Rijk and De Kruijf, 1993). • Studies that examined the effects of repeated heating and/or reuse (i.e. machine washing) on migration showed contrasting results; migration increased or decreased with each successive heating and/or washing depending on which simulant and chemical migrant combination was tested (Jetten and De Kruijf, 2002; Brede et al., 2003; Skjevrak et al., 2005). In the majority of these migration studies, the number of times each plastic container was reheated was not representative of the number of times a container will be heated over its lifetime. • The chemicals that have been measured in the migration studies summarized in Tables 1.2 and 1.3 were not selected because they were known to have toxic health effects: All potential migrants were not analyzed in any of the migration studies, therefore the amount of measured overall or specific migration underestimates the range and amount of individual chemicals migrating from food contact plastics. Thus, the current knowledge gaps identified in scientific literature emphasize the need to further examine the effects of repeated microwave heating on the amount of migration of toxic substances from plastic containers. Migration must also be conducted under more realistic testing conditions and follow scientifically sound principles (i.e. incorporate the use of controls and compare samples from identical test conditions). 1.5 Research Questions Not all study limitations that were identified can be addressed in this study; thus, this study will focus on the effects of microwave heat exposure on migration from a reusable 21 polypropylene plastic food container. It is hypothesized that the amount of migration, number and types of chemicals migrating during microwave heating and oven heating are different. It is also hypothesized that the amount of chemical migration that occurs during microwave heating will depend on the type food simulant that is exposed to the plastic. Thus, the effect of microwave and oven heating on the amount of migration from one type of plastic container into acetic acid and isopropanol food simulants will be tested. 1.6 Study aims This study attempted to resolve the following knowledge gaps and previously discussed study design weaknesses: • To isolate the effects of microwave heating on plastic, migration was tested using the same heating temperature and duration in the oven and microwave. In contrast to certain previous study designs, the same number of food simulant samples was heated in the microwave and the oven. • Each plastic container was heated 10 times to simulate realistic repeated use conditions and to ensure that the maximum amount of chemical migrants was released into the food simulants. t • A general analytical method was used to qualitatively identify as many chemical migrants as possible from each sample of heated food simulant compared to previous studies that analyzed a few known migrants. • The usage information, toxicological properties and regulatory background of each migrant were researched to determine: o the origin of the migrant; o the potential health effects of ingesting migrants; and o if the measured migration levels violate any regulations. • Several negative controls were tested to ensure that migrants measured in food simulants during microwave heating are not contaminants from other sources. A positive control was also used in the study to determine the accuracy and limitations of the analytical methodology. 22 2 METHODS In this chapter, the following topics were covered: 1) Migration test method development and results; 2) Sample extraction methods; 3) GC/MS analysis methodology; 4) Data analysis methods including quantification of identified chemical migrants, statistical comparison of migration results between food simulants and between microwave and oven heated samples; 5) Method for researching the physical, chemical and toxicological properties of identified chemical migrants 6) Method for estimating population exposure to chemical migrants. 2.1 Overall migration The first part of this study originally consisted of testing overall migration released from plastic during microwave heating using gravimetric analysis by applying a modified version of the European Committee for Standardization Method ENV 1186 (Part 9) (CEN, 1994). However, several technical problems interfered with the accurate weighing of glass dishes used to measure the amount of overall migration. Static interference was identified as the cause of the unstable weight measurements after other sources of disruption were systematically eliminated. Many solutions to eliminate static charge were considered and some were implemented with limited success. Given the limited financial resources for and time constraints of this study, it was established that developing a GC/MS analysis method and using it to identify and quantify chemical migrants from microwave heated plastic was more practical. Overall migration is also a more crude method of measuring migration because it is based on a gravimetric measurement of all substances that migrate and does not require the identification of each chemical. The rationale behind the EU overall migration limit is that the total migration of substances above 60 mg/kg of food, regardless of the toxicological properties of individual migrants, will adversely affect the quality of food (Castle et al., 2004). 2.2 Development of GC/MS migration test method In the process of developing a GC/MS migration test methodology, a series of decisions were made to improve method efficiency and effectiveness including: 23 • The selection of 3% (w/v) acetic acid solution and isopropanol, which were food simulants that have been shown to contain the highest levels of migration when exposed to microwave heated in plastic in previous studies (Galatto and Guarda, 1999, 2004) • The selection of a commonly used type of short-term use plastic container to optimize the ability to identify and quantify a range of chemical migrants during microwave heating. • The selection of a liquid-liquid extraction method for dissolving migrants from food simulant samples because of its simplicity (i.e. it does not require specialized equipment) and low cost. A series of preliminary experiments were also conducted to develop the final methodology for extracting and analyzing extracts of microwave and oven heated food simulant samples. GC/MS analysis was initially carried out using an ion trap Varian Star 3400/Saturn 2000 system but an Agilent quadrapole GC/MS system (Model 5973N) was subsequently used for all experiments because it was found to be more accurate. The results of the key experiments were summarized in Table 2.1. 24 Table 2.1 Summary of preliminary GC/MS method development experimental results. Experiment Objective Result To select a microwave power level and heating time that results in the maximum heating temperature but minimal evaporative loss for each food simulant. A Panasonic microwave (Model NN-H664BF, manufactured by Matsushita Microwave Oven Ltd, purchased at Future Shop) was used for all microwave heating experiments. Power levels 6 and 4 were selected for 3% (w/v) acetic acid and isopropanol, respectively. Heating times of 3 and 2.5 minutes were selected for the acetic acid solution and isopropanol simulants, respectively. To determine the amount of time required for 200 mL of food simulant to cool to 40°C and the amount of time for the microwave tray to cool to room temperature after 200 mL of simulant was microwave heated. It took approximately 18 minutes for 3% (w/v) acetic acid and 9.57 minutes for isopropanol to cool to 40°C. It took approximately 3 and 2 minutes for the microwave tray to cool to room temperature in the freezer (-20°C) after heating 200 mL of acetic acid solution and isopropanol, respectively. To select the solvent type and amount that will extract a migrant concentration from each food simulant that can be quantified. 40 mL of methylene chloride was sufficient for extracting migrants from 3% (w/v) acetic acid. 10 mL of methylene chloride was used to extract migrants from isopropanol. To determine the difference in percent recovery of surrogate standards and known chemical migrants used to spike food simulant samples. Percent recovery of the fat soluble migrants from acetic acid solution and isopropanol were the highest. Erucamide and octadecanol had average recoveries ranging from 44 to 46% and 56 to 60%, respectively. 2,4-di-tert-butylphenol and dibutyl adipate had average recoveries ranging from 39 to 43%, and 52 to 54%, respectively. To determine the internal surface area of the plastic container in contact with the food simulant. The average internal surface area of the polypropylene container in contact with the food simulant was 169.63 ±2.71 cm2 (Relative Standard Deviation = 1.6%). 25 2.3 Plastic migrants released into food simulants during microwave heating 2.3.1 Chemical migration into acetic acid food simulant 2.3.1.1 Equipment preparation All microwave heating took place in a fumehood. All glassware was washed in warm tap water with detergent and rinsed with distilled water or baked for an hour at 550°C in a muffle furnace. After being washed or baked, the glassware was then rinsed two times with methylene chloride prior to being used in this experiment to prevent contamination. All organic chemicals used in this experiment were analytical grade quality. All new plastic containers were rinsed with cool filtered water to remove any dust on the inner surface prior to being used. 2.3.1.2 Negative controls Media blank 1 mL of toluene was analyzed for the sample set to ensure that any chemicals identified in the sample extracts were not solvent contaminants. The effect of exposure to glassware The purpose of this negative control was to determine if any chemicals detected in any samples were a result of contamination due to contact between the food simulant and the glassware, plastic, equipment or the environment. A method blank sample extract was prepared by extracting 200 mL of acetic acid solution which was poured directly into a separatory funnel using the same method described for the microwave heated sample extracts. The effect of exposure to the inside of an oven on migration The purpose of this negative control was to ensure that chemicals detected in oven heated food simulant were not a result of contamination from inside the oven. This negative control experiment consisted of weighing a crystallizing dish (125 x 65 mm VWR Cat. No. 89000-292) filled with 200 mL of acetic acid solution and covered with a watchglass resting on top of 3 glass spacers. A crystallizing dish was selected because it had approximately the same dimensions and surface area as the polypropylene plastic container but was made of glass. The acetic acid solution was placed in a toaster oven (Betty Crocker Model BC-1660-C, Manufactured by E.F. Appliances Canada Ltd.) maintained at 90 to 100°C for 30 minutes, which is the equivalent of 10 microwave heating cycles. The crystallizing dish was then removed from the oven and cooled 26 for 180 minutes at room temperature. The mass of acetic acid solution lost due to evaporation was replaced based on the weight difference. A methylene chloride extract of the control sample was prepared and analyzed using the same method as the microwave heated samples. This control experiment was repeated three times using clean crystallizing dishes and fresh food simulant. The effect of exposure to the inside of a microwave on migration The purpose of this negative control was to ensure that any chemicals detected in microwave heated food simulant were not from inside the microwave oven. The methodology of this control experiment was the same as the one described for microwave heating experiment (described below); however, the acetic acid solution was heated in the microwave in glass crystallizing dish instead of a polypropylene container. This control experiment was repeated three times using clean crystallizing dishes and fresh food simulant. The effect of contact with plastic at room temperature on migration The purpose of this negative control was to ensure that the chemicals identified in microwave heated samples were not solely a result of the food simulant being in contact with plastic at room temperature. This control experiment consisted of pouring 200 mL of acetic acid solution in a polypropylene plastic container, covering the container with a watchglass and weighing it on an electronic balance. The acetic acid was then stored at room temperature for the same time period as 10 microwave heating and cooling cycles (i.e. approximately 3 hours and 29 minutes). The weight of the acetic acid was re-measured and recorded. A solvent extract of the control sample was prepared and analyzed using the same method as the microwave heated samples. This control experiment was repeated three times using new polypropylene containers and fresh food simulant. The effect of contact with non-microwave heated plastic on migration The purpose of this negative control was to ensure that the chemicals found in microwave heated samples were not a result of the food simulant being in contact with non-microwave heated plastic. This control experiment consisted of measuring and recording the weight of the watchglass and a new, empty 591 mL square plastic polypropylene container. A glass crystallizing dish filled with 200 mL of acetic acid solution and covered with a watchglass was heated on a hotplate (Corning Glass Works brand model PC351) until it reached 100°C. The 27 acetic acid solution was then poured into a polypropylene container, covered with a watch glass and placed in a toaster oven, with a temperature range of 90 to 100°C, for 30 minutes, which is the equivalent of 10 microwave heating cycles. The polypropylene container was then removed from the oven and cooled for 180 minutes at room temperature. The mass of acetic acid solution lost due to evaporation was replaced based on the weight difference before and after heating. It was assumed that 200 mL of acetic acid weighs 200 g; thus, the combined weight of the watchglass, container and acetic acid before oven heating could be determined. A methylene chloride extract of the oven heated control sample was prepared and analyzed using the same method as the microwave heated sample. This control experiment was repeated three times using new polypropylene containers and fresh food simulant. 2.3.1.3 Migrant extraction A new 591 mL square polypropylene container was filled with 200 mL of 3% (w/v) acetic acid, covered with a watch glass and weighed on an electronic balance. The food simulant was heated in a microwave for 3 minutes at power level 6. After heating, the food simulant was cooled outside the microwave in the fumehood for 18 minutes. The microwave tray was cooled in a freezer for 3 minutes and returned to the microwave. The door to the microwave was propped open halfway for 20 minutes for the microwave to cool to room temperature. The amount of food simulant evaporated during microwave heating was determined by reweighing the cooled food simulant in the plastic container covered with a watch glass. The difference in mass was replaced by adding fresh food simulant into the plastic container. The microwave heating and food simulant refilling process was repeated for a total of 10 heating cycles. After the heating cycles were completed, the food simulant sample was poured into a separatory funnel and extracted twice for 2 minutes with 20 mL of methylene chloride. The solvent extracts were transferred into a 50 mL Pyrex® brand culture tube and evaporated to dryness by applying a flow of nitrogen while it was heated in a hot water bath maintained at 35±5°C. The sample residue was reconstituted with toluene to 1 mL, mixed with a vortexer (Vortex Genie 2 Model G-560) for 30 seconds and transferred to a 15 mL scintillation vial. The sample residue was extracted twice; thus, the final extract volume was 2 mL. 1.5 mL of the final extract was transferred into a GC/MS vial with a glass 1 mL syringe. The entire microwave heating experiment was repeated three times with fresh 200 mL samples of acetic acid solution in new polypropylene plastic containers. After all samples were prepared they were analyzed using GC/MS to identify plastic migrants. 28 Overall, a total of 27 samples were analyzed using GC/MS (see description below) to determine the effect of microwave heating on migration from polypropylene plastic into an acetic acid solution. 2.3.2 Chemical migration into isopropanol food simulant 2.3.2.1 Equipment preparation All equipment was prepared identically to the way equipment was cleaned and organized for the acetic acid simulant migration experiment. 2.3.2.2 Negative controls Media blank 1 mL of toluene was analyzed for the sample set to ensure that any chemicals identified in the sample extracts were not solvent contaminants. The effect of exposure to glassware This control experiment was identical to the one described in the acetic acid food migration experiment except isopropanol was the food simulant. The effect of exposure to the inside of an oven on migration This control experiment was identical to the one described in the acetic acid food migration experiment except for the following changes: • Isopropanol was the food simulant; • The toaster oven was maintained at 75 to 85°C; • The cooling time for the isopropanol was 95.7 minutes. The effect of exposure to the inside of a microwave on migration This control experiment was identical to the one described in the acetic acid food migration experiment except isopropanol was the food simulant. 29 The effect of contact with plastic at room temperature on migration This control experiment was identical to the one described in the acetic acid food migration experiment except that isopropanol was the food simulant and the storage time was 2 hours and 40 seconds. The effect of contact with non-microwave heated plastic on migration This control experiment was identical to the one described in the acetic acid food migration experiment except for the following changes: • Isopropanol was the food simulant; • The initial heating temperature was 82°C; • The toaster oven was maintained at 75 to 85°C; • The isopropanol was heated for 25 minutes; • The cooling time for the isopropanol was 95.7 minutes. 2.3.2.3 M i g r a n t extraction The same volume of food simulant was heated and cooled in the same type of polypropylene container using the same method as the acetic acid simulant migration experiment. However, the following changes were applied: • Analytical grade isopropanol was used as the food simulant; • The food simulant was heated in a microwave for 2.5 minutes at power level 4; • After heating, the food simulant was cooled in the fumehood for 9.57 minutes; • The microwave tray was cooled in a freezer for 2 minutes and returned to the microwave; • The door to the microwave was propped open halfway during the 9.57 minutes to allow the microwave to cool to room temperature. Once the isopropanol sample was microwave heated in plastic, it was transferred to a 250 mL round bottom flask and attached to a rotor evaporator (Buchii Rotavapor-A). The flask was then submerged in a glass crystallizing dish filled with water and heated on a hotplate set at level 3. The rotor evaporator was turned on with a rotation speed level 3-5. The isopropanol sample was evaporated to dryness. The sample residue in the round bottom flask was extracted with 5 mL of methylene chloride by agitating the flask for 2 minutes and pipetting the methylene chloride into a 15 mL glass culture tube. The sample residue was extracted again; thus, the final extract volume was 10 mL. The sample extract was evaporated to dryness in a dry block heater (TECAN Driblock DB-3) maintained at 40±5°C with a nitrogen flow. 30 The sample extract was reconstituted, extracted and transferred into a GC/MS vial using the same method described in the acetic acid simulant migration experiment. The entire experiment was replicated three times with fresh isopropanol. Overall, a total of 27 samples were analyzed to determine the effect of microwave heating oh migration from polypropylene plastic into isopropanol. 2.3.3 Analysis of chemical standards Several chemicals standards were analyzed using GC/MS in order to positively confirm the identity of detected migrants. Chemical standards (Sigma Aldrich, Canada) were selected from a group of substances that were known general plastic migrants (See Appendix A for a more detailed description of the chemical selection process). Chemicals potentially identified in preliminary experiments conducted during this research study were also included. The chemical standards are summarized in Table 2.2 below: Table 2.2 Summary of chemical standards analyzed for migrant identification confirmation Chemical Name CAS# Diisobutyl phthalate 84-69-5 2,2'-Methylenebis(4-ethyl-6-tert-butylphenol) 88-24-4 Octadecanol 112-92-5 2,4-Dihydroxybenzophenone 131-56-6 2-Hydroxy-4-methoxybenzophenone 131-57-7 Adipic acid dibutyl ester 105-99-7 2-(2-Hydroxy-5-methyl-phenyl)benzotriazole 2440-22-4 Triacetin 102-76-1 2,2'-Methylenebis(6-t-butyl-4-methylphenol) 119-47-1 3,5-Di-tert-butyl-4-hydroxytoluene (BHT) 128-37-0 Sorbitan monolaurate 1338-39-2 2(3)-Tert-butyl-4-hydroxyanisole 25013-39-2 Erucamide 112-84-5 2,4-Di-tert-butylphenol 96-76-4 Each standard was analyzed using the same GC/MS conditions described in the next section. If a substance was uniquely identified in high concentrations in both microwave heated samples and the chromatogram matched one of the migration standards it was quantified (see data analysis method description below). 31 2.3.4 GC/MS analysis 1 fxL of each sample and standard was analyzed using an Agilent quadrapole GC/MS system (Model 5973N) equipped with an HP-5MS capillary column (30 m x 0.25 mm i.d.; film thickness 25 um; 5% phenyl methyl siloxane). The GC oven was programmed with the following conditions: initial temperature of 90°C for 1 min increased to a final temperature of 290°C at a rate of 8°C/min. The total run time for each sample was 45.5 minutes. The scanned mass ion size range was between 41 and 800 m/z. 2.4 GC/MS data analysis methodology for acetic acid and isopropanol simulant sample extracts A mass spectrum library search was conducted on each peak within each sample chromatograph using the of Agilent Enhanced ChemStation (Mass Selective Detector (MSD) D.02.00.275) software. The following peak selection criteria were used to search for chemical identities in microwave and oven heated acetic acid sample extracts: shoulder detection is off; initial threshold 12; initial peak width 0.05 and initial peak reject 0. The same peak selection criteria were applied to the mass spectrum library search conducted on microwave and oven heated sample extracts except for the peak initial threshold which were 16.5 and 15.5, respectively. The chemical identity for all peaks was manually searched in all acetic acid simulant control sample extracts. A mass spectrum library search uses an algorithm that assesses the similarity of the reference spectrum of known chemicals to the unknown spectrum. A quality index out of 100 is assigned to each spectrum match. A match between the unknown and known spectrum is assigned a high quality index if all significant peaks in the unknown spectrum with approximately the same ratios are also present in the library reference spectrum. Brief descriptions of mass spectrum libraries that were used in the search are described below: • Wiley 275 library has mass spectra for over 275000 compounds (John Wiley and Sons Inc, 2000). • NIST 98 library has mass spectra for over 100 000 compounds (US Department of Commerce, 1998). • NBS75K library has mass spectra for over 75 000 compounds (United States National Bureau of Standards, 1992).. 32 The library search strategy was stratified such that if the quality of the Wiley275 mass spectrum library match was less than 50, the NIST98 library was searched. In addition, if the quality of the NIST98 library match was less than 50 the NBS75K library was searched. Once a library search was complete, each set of matched reference mass spectra was graphically compared to the unknown spectra of each identified peak in all sample chromatographs. The quality index of each mass spectra match, the number of libraries that identified the same substance, the presence of ions that indicate column bleed and the pattern of the unknown mass spectra alone were criteria that were taken into consideration when assessing the potential identity of the unknown substance. The background noise was subtracted from the sample spectra in all cases to improve the mass spectra match. After all matched reference mass spectra were reviewed, each unknown spectra associated with each chromatograph peak was assigned a tentative chemical name, compound class or unknown identity. If no mass spectra match could be made with a reasonable degree of confidence (i.e. greater than 75 % quality index) than the peak was assigned an unknown chemical identity. The retention times, molecular mass, peak area and associated chemical identity (if known) of each peak in all sample chromatographs were then laid out in tables and compared. Any chemicals that had the same retention time but were assigned different chemical identities in different samples were reexamined. The most frequently matched chemical identity was retained if a chemical with a specific retention time was the same in the majority of the samples identified and the reference mass spectra visually matched the unknown spectra. If there was no consistency in the mass spectra matching of a chemical to a specific peak retention time among most sample chromatographs, then the chemical identity (i.e. chemical isomer) or compound class that most closely matched the unknown mass spectra was assigned to the peak retention time. If no chemical identity could be assigned but the mass spectra among replicate sample extracts matched visually, the three most intense and/or unique mass ions that had the same retention times were used as identifiers. To determine which chemicals may potentially be migrating from plastic and/or is affected by microwave heating, the following steps were taken: • Chemicals identified in any negative controls were only included for data analysis if a consistent association was observed between peak area size and experimental treatment (i.e. chemical A had the highest peak areas in all microwave heated sample extracts and the lowest peak areas in all room temperature control sample extracts). Chemicals that were 33 identified in any negative control sample extracts and did not meet the above criteria were excluded from further analysis based on the assumption that these substances were present in the solvent and/or residual contamination in the glassware. • Identified chemicals were searched for in all acetic acid simulant sample extracts if they met any of the following criteria: they had a mass spectra match quality index greater than 80%); they were found in all replicate samples that were exposed to plastic in an experimental treatment; and/or they had a peak area greater than 1000000. Chemicals that did not meet the previously described criteria were excluded from further analysis. The same criteria were applied to all isopropanol simulant sample extracts; however, the peak area cut-off was 5000000. The peak area cut-off was determined by estimating which peak area size was associated with the greatest number of visually identifiable and defined peaks in the microwave heated sample extracts. • Chemicals were searched for in all other samples by comparing chromatograph peaks with approximately the same retention time (i.e. by studying the average mass spectra across the peak apex and/or mass spectra at cross sections throughout the peak) and extracting specific ion chromatographs of the chemicals of interest to determine if the peaks with the same retention time are for the same chemical. • The peak areas of the most intense/common mass ion of each chemical positively identified in all sample extracts were plotted in a graph or summarized in a table. 2.5 Quantification of identified chemical migrants in acetic acid and isopropanol simulant sample extracts The purpose of this experiment was to quantify chemicals that were identified as potential plastic migrants in all samples from the acetic acid and isopropanol food simulant experiments. Erucamide and 2,4-di-tert-butylphenol were selected because they were identified in previous preliminary acetic acid and isopropanol food simulant experiments. Irganox 1076, 2,6-di-tert-butylbenzoquinone, isopropal palmitate, stearic acid, tetradecane, pentadecane, heptadecane, and octadecane were selected because they were identified in previous preliminary isopropanol food simulant experiments. All of the above-mentioned chemicals were also selected because a chemical standard for each substance was available for purchase; they were also identified in preliminary experiments or other studies as plastic migrants/migrant breakdown products (van Lierop et al, 1998). 34 A set of six working standards of each positively identified,chemical migrant with concentrations ranging from the estimated detection limit of each substance to approximately double or triple the maximum concentration detected in preliminary experiments. A 1 pL sub-sample of each working standard was analyzed using GC/MS at the beginning of the food simulant sample analysis sequences. All samples were analyzed using the same GC/MS conditions described previously in the methodology for analysis of chemical migrants. The amount of migrant present in each sample, if present, was back-calculated from the calibration curve generated from the working standards. 2.6 Statistical analysis to determine the differences in migration between food simulant type and experimental treatments After the average peak areas and/or concentrations of all identified potential polypropylene chemical migrants were calculated, the following additional data analyses were performed to determine the variability between replicate samples and the effect of food simulant type and type of heating on chemical migration: • The relative standard deviation and relative standard error was calculated using Microsoft Excel Version 10.2614.2625 (Microsoft Corp, 2001) for mass ion peak areas and/or concentrations of each chemical. The relative standard deviation was determined by dividing the standard deviation by the average peak area/concentration. The relative standard error was determined by dividing the standard error of the mean (which is the standard deviation divided by the square root of the number of samples) by the average peak area/concentration. • The percent difference in the average peak area or concentration of each chemical identified in oven heated compared to microwave heated experimental treatment sample extracts was calculated. • The percent difference in the average peak area or concentration of each chemical identified in acetic acid compared to isopropanol food simulant sample extracts (within the same experimental treatment groups) was also calculated. • An unpaired t-test (with a 95% confidence interval) was also performed on the average peak areas/concentrations of each identified potential migrant using S-Plus Version 7.0 (Insightful Corp, 2005) to determine if the difference in peak area/concentrations of chemicals between acetic acid and isopropanol simulant microwave or oven heated samples or room temperature control samples was significant. The unpaired t-test was also conducted to determine if the 35 difference in peak area/concentrations of chemicals between microwave and oven heated food simulant samples was significant. A p-value less than or equal to 0.05 was considered significant. 2.7 Physical and chemical properties, usage information, regulatory and toxicological background of potential chemical migrants released during microwave heating The following databases and resources were researched to find usage information (i.e. if the substance is a known antioxidant), toxicological data, and regulations for substances that were hypothesized to be plastic migrants that are released during microwave heating: Toxnet, PubMed, Medline, Web of Science, SciFinder, an internet European database of chemicals used in food contact plastics (Gilbert et al., 2000), US FDA and EC regulations, Registry of Toxic Effects of Chemical Substances (CCOHS, 2004), Kirk-Othmer Encyclopedia of Chemical Technology (Kroschwitz and Howe-Grant, 2005), Ullmann's Encyclopedia of industrial chemistry (Fritz, 2006) and Haw ley's Concise Chemical Dictionary (Lewis, 2002). 36 3 RESULTS 3.1 Plastic migrants released into acetic acid food simulant during microwave heating Based on a comparison between Figure 3.1 to Figures 3.2 through 3.5, 2,4-di-tert-butylphenol and erucamide were the only two chemicals identified in the acetic acid sample extracts were among the selected possible or known migrant chemical standards. Figure 3.1 shows a chromatograph of selected potential and known food contact polypropylene plastic migrants. Figures 3.2 to 3.5 show the chromatographs of the toluene blank sample, and the method blank, oven heated glass control, microwave heated glass control, room temperature control, oven heated and microwave heated acetic acid sample extracts. The chromatograph of one replicate sample extract per experimental treatment was presented. Substances that were positively identified or had mass ion peak areas greater than 10 000 in the microwave heated sample extracts were labelled. All peaks in the control sample extracts were labelled. 37 1 .Be+07 1.6e+07 1.4e+07 1.28+07 1e+07 : 8000000 : 6000000 4000000 2000000 2,4-bis( 1,1 -dimethylethyl)phenol 2(3)-Tert-butyl-4-hydroxyanisole Triacetin BHT Adipic acid dibutyl ester 2-(2-Hydroxy-5-methyl-phenyl) benzotriazole 2-Hydroxy-4-methoxybenzophenone Diisobutyl phthalate / t/ Octadecanol 2,4-Dihydroxy benzophenone - T - * 1-2,2'-Methylenebis(6-t-butyl-4-methylphenol) 2,2'-Methylenebis(4-ethyl-6-tert-butylphenol) 25.00 , Erucamide -r* r "ime-> Figure 3. 5.1 r~ 10.00 - i 1 1 1 1 1 - ^ r 15.00 20.00 1 Chromatograph of all selected anticipated and known food contact polypropylene plastic migrant standards dissolved in toluene. [Abundance 100000 60000 a. Toluene blank TIC: blk3.D\data.ms f) column bleed |Time-> 5.00 10.00 [Abundance 100000: 80000 ; 60000 : 40000 : 20000 b. Method blank sample extract TIC: Method blank.D\data.ms (*) column bleed unknown contaminants (m/z 161,229, 244) 20.00 rnme-> 5.00 10.00 15.00 25.00 30.00 35.00 40.00 Figure 3.2 Chromatographs of the toluene blank and method blank acetic acid simulant sample extract, (a) Toluene (media) blank shows the presence of column bleed, (b) Method blank shows an occurrence of column bleed and two unknown contaminants. lAbundance 100000 -I 50000 0 a. Oven heated glass control sample extract Column bleed TIC: oven control 1 .D\data.ms (*) Unknown contaminants (m/z 161,229, 244) 11 i 1 1 1 1 1 1 1 1 i 1 1 1 1 1 1—i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r rT ime-> 5.00 10.00 15.00 20.00 25.00 30.00 35.0 40.00 lAbundance 100000 -I 50000 0 b. Microwave heated glass control sample extract TIC: micro control 4.D\data.rns (*) Unknown contaminants (m/z 161,229,244) 11 I 1—i 1 — i — i — | 1—i—i 1—| 1—i 1 — i — | — r r i rne-> 5.00 10.00 15.00 20.00 Figure 3.3 Chromatographs of the oven and microwave heated glass control acetic acid simulant sample extract, (a) Oven heated glass control sample extract and (b) microwave heated glass control sample extract show the presence of column bleed and two unknown contaminants. Abundance 50000 a. Room temperature control sample extract TIC: control 3.D\data.msO Unknown contaminants Column bleed (m/z 161 229, 244) 2,4-Di-tert-butylphenol II J L L i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — |Tirne-> 5.00 10.00 15.00 20.00 . . 25.00 30.00 35.00 i i i | i i i r 40.00 ^Abundance 100000J 50000 b. Oven heated sample extract TIC: oven 1 .D\data .ms (*) Column bleed Unknown contaminants (m/z 161, 229, 244) 2,4-Di-tert-butylphenol —' ^ —• •• • ••! ii J-rA ii ~~| 1 1 1 1 1 1 1 1 1 p [Time-> 5.00 10.00 15.0 —i—i—i—i—|—r 20.00 25.00 ~\—i—|—i—i—r 30.00 i i i r 35.00 40.00 Figure 3.4 Chromatographs of a room temperature control and an oven heated acetic acid simulant sample extract, (a) Room temperature control sample extract and (b) oven heated glass sample extract show the presence of column bleed, 2,4-di-tert-butylphenol and two unknown contaminants. Abundance TIC: micro 3.D\deta.ms 2,6-Di-tert-butyl-4 methylene-2,5-cyclohexadiene-1 -one 2-Norbornene Column bleed \ 2,4-Di-tert-butylphenol / Unknown (m/z 117, 130, 371) Unknown (m/z 117, 130,399) if Unknown contaminants (m/z 161,229, 244) 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9 diene-2,8-dione) \ ^ ' - - •M. lmL, lUi.i i, ^ w i i — • * r t w > l Unknown (m/z 117, 130, 313) Erucamide Unknown (m/z 117, 341) 1 Unknown (m/z 261, 282) i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — i — | — i — i — i — r 10.00 15.00 20.00 25.00 30.00 35.00 i i i r T ime- ) 5.00 Figure 3.5 Chromatograph of a microwave heated acetic acid simulant sample extract. Table 3.1 is a summary of the chemicals that were qualitatively identified by GC/MS analysis in all sample extracts and met the selection criteria for further data analysis (i.e. were detected in all microwave heated sample extracts). The highlighted substances were excluded from data analysis because they were assumed to be contaminants since they were detected in the method blank and all oven and microwave heated glass control samples. Each symbol represents the presence (vO or absence (x) of a chemical in one sample. Table 3.1 The presence of specific chemicals identified in the control sample extracts and microwave heated acetic acid simulant sample extracts. Qualitatively Retention Experimental treatment Identified Chemicals time (min) Toluene Method Oven Microwave Room Oven Microwave blank blank heated heated temperature heated heated (n=l) (n=l) glass glass controls plastic plastic controls controls (n=4) (n=4) (n=4) (n=4) (n=4) 2-Norbornene 10.50 X X xxxx X X X X X X X X X X X X • • • • 2,6-Di-t-butyl-4- 10.67 X X xxxx xxxx xxxx xxxx methylene-2,5-cyclohexadiene-1 -one 2,4-Di-tert-butylphenol 11.08 X X xxxx xxxx 7,9-Di-tert-butyl-l- 17.04 X X xxxx xxxx xxxx oxaspiro[4.5]deca-6,9-diene-2,8-dione Unknown 17.70 X •/ (m/z 161,229, 244) Unknown 1 S.1(1 " X s (m/z 161,229,244) Unknown (m/z 55, 83) 22.15 X X xxxx xxxx xxxx xxxx Unknown 23.18 X X xxxx xxxx xxxx xxxx (m/z 43,55,267) Unknown 23.40 X X xxxx xxxx xxxx xxxx (m/z 117, 130, 371) Unknown (m/z 117, 130, 23.57 X X xxxx xxxx xxxx xxxx 313) Unknown 23.66 X X xxxx xxxx xxxx xxxx (m/z 57, 83,98) Unknown 25.40 X X xxxx xxxx xxxx (m/z 117, 130, 399) Unknown (m/z 117, 341) 25.55 X X xxxx xxxx xxxx • • • • Erucamide 26.43 X X xxxx xxxx xxxx xxxx Unknown (m/z 191,415) 26.59 X X xxxx xxxx xxxx xxxx * Not detected in a sample extract. x x x x Not detected in four sample extracts. •S Detected in a sample extract. •/•/•/S Detected in four sample extracts. Based on the results presented in Figures 3.2 to 3.5 and Table 3.1, it can be observed that no substances were qualitatively identified in the toluene blank with the exception of column bleed contaminants. The same column bleed contaminants found in the toluene blank occurred consistently throughout all control and microwave heated sample extracts. 43 Two unknown substances with retention times of 17.7 and 18.1 minutes (m/z 161, 229, 244) were identified in the method blank sample extract and all control and microwave heated sample extracts. Even though it appears that these two substances were present in higher concentrations in the microwave heated sample extract, they were excluded from further data analysis because they were also present in the method blank sample, the microwave and oven control samples, which had no direct exposure to plastic. Thus, it was unlikely that any change in peak height of these two chemicals was specifically associated with exposure to polypropylene plastic. Column bleed contaminants and the two unknown chemicals (m/z 161, 229, 244) identified in the method blank sample extract were the only substances present in the oven and microwave heated glass control sample extracts. Among all of the room temperature control replicate sample extracts, 2,4-di-tert-butyl phenol was the only chemical identified that was not a column bleed contaminant or present in any other controls or the method blank sample extract. 2,4-Di-tert-butylphenol, 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione, and two unknown substances with retention times of 25.40 minutes (m/z 117, 130, 399) and 25.55 minutes (m/z 117, 341) were the only chemicals identified in all replicate oven heated sample extracts that were not present in any microwave or oven controls or the method blank sample extract. 2,4-Di-tert-butyl phenol was the only chemical that was previously identified in all room temperature control sample extracts. The greatest number of and the largest peaks were identified in the microwave heated sample extract (See Figure 3.6). A total of thirteen substances were identified in all replicate microwave heated sample extracts; these chemicals were not present in any microwave or oven controls or the method blank sample extract. Four of the thirteen chemicals were previously identified in all oven heated sample extracts. Thus, 2-norbornene, 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-l-one, an unknown substance (m/z 55, 83) with a 22.15 minute retention time, an unknown substance (m/z 43, 55, 267) with a 23.18 minute retention time, an unknown substance (m/z 117, 130, 371) with a 23.40 minute retention time, an unknown substance (m/z 117, 130, 313) with a 23.57 minute retention time, an unknown substance (m/z 57, 83, 98) with a 23.66 minute retention time, erucamide and an unknown substance (m/z 191, 415) with a 26.59 minute retention time were unique to the microwave heated sample extracts. Figure 3.6 below shows the average selected mass ion peak areas of chemicals that were identified in more than one experimental treatment that involved the direct exposure of acetic 44 acid simulant to a polypropylene plastic container. The average mass ion peak area for each chemical measured in the microwave heated sample extracts was consistently larger than the average peak area measured in the oven heated sample extracts. 2,4-di-tert-butylphenol was the only chemical present in the room temperature control, oven and microwave heated sample extracts. The mass ion peak areas were plotted because no standards were available to quantify 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione or the unknown substances. cs cs 1400000 1200000 1000000 800000 600000 400000 4 200000 i • Room temperature control sample extracts* • Oven heated sample extracts 0 Microwave heated sample extracts I I 1 2,4-Di-tert- 7,9-Di-tert-butyl-l- Unknown Unknown butylphenol oxaspiro[4.5]deca- (m/z 117, 130, 399) (m/z 117, 341) 6,9-diene-2,8-dione Identified chemical migrants Figure 3.6 Average select mass ion peak areas of substances identified in sample extracts of acetic acid solution that were exposed to polypropylene plastic containers. Table 3.2 below is a summary of the average mass ion peak areas, relative standard deviation and relative standard error for each chemical identified in the acetic acid food simulant samples exposed to oven and microwave heated plastic. The mass ion peak areas of migrants in Table 3.2 were analyzed because no chemical standards were available to quantify them. Chemicals that were quantified were excluded and analyzed separately. The percent increase in the average mass ion peak areas in the microwave heated sample extracts compared to the oven heated sample extracts were included. The p-values from the unpaired t-tests comparing the 45 chemical mass ion peak areas in oven heated to microwave heated sample extracts were also summarized. 46 Table 3.2 Results from the statistical analysis of chemicals unique to microwave and oven heated acetic acid sample extracts (n=4). C h e m i c a l name Exper imen ta l Trea tment Average peak area Relat ive s tandard deviat ion Relat ive s tandard e r ro r % Difference between average migran t peak areas i n microwave heated compared to oven heated samples P-value 2-Norbornene Microwave heated plastic 54424 (±16317) 30% 15% - • 2,6-Di-tert-butyl-4-methylene-2,5-cyclohexadiene-1-one Microwave heated plastic 9652 (±1701) 18% 9% ; 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione Microwave heated plastic 30624 (±3347) 11% 5% + 1297% <0.0001 Oven heated plastic 2192 (±2803) 128% 64% Unknown (m/z 55, 83) Microwave heated plastic 8440 (±2772) 33% 16% Unknown (m/z 43, 55,267) Microwave heated plastic 4600 (±1522) 33% 17% Unknown (m/z 117, 130, 371) Microwave heated plastic 344282 (±36964) 11% 5% -Unknown (m/z 117, 130, 313) Microwave heated plastic 90641 (±15189) 17% 8% Unknown (m/z 57, 83, 98) Microwave heated plastic 6435 ( ± 3 6 9 9 ) ' 57% 29% Unknown (m/z 117, 130, 399) Microwave heated plastic 526026 (±23205) 4% 2% +5395% <0.0001 Oven heated plastic 9573 (±1600) 17% 8% Unknown (m/z 117,341) Microwave heated plastic 499785 (±71906) 14% 7% + 16262% <0.0001 Oven heated plastic 3055 (±524) 17% 9% Unknown (m/z 191,415) Microwave heated plastic 10568 (±1332) 13% 6% The majority of the substances identified in acetic acid samples microwave and/or oven heated in plastic had average mass ion peak areas with a relative standard deviation (RSD) and a relative standard error (RSE) that was less than 20% and 10%, respectively. The RSD was higher than the RSE for the average mass ion peak area of each chemical. 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione and an unknown substance with mass ions 57, 83, and 98 were the only two substances with average mass ion peak areas that had a RSD greater than 50% and a RSE greater than 20%. Among the chemicals identified in microwave and oven heated sample extracts, the RSD and RSE were consistently higher in oven heated sample extracts. The increase in the average 47 mass ion peak areas for chemicals in microwave heated compared to oven heated sample extracts was significantly higher (p<0.01). Table 3.3 below is a summary of the average concentrations, relative standard deviation and relative standard error for each quantified chemical identified in the acetic acid food simulant sample extracts exposed to plastic. The percent increase in the average mass ion peak areas in the microwave heated sample extracts compared to the oven heated sample extracts were included. The p-values from the unpaired t-tests comparing the chemical mass ion peak areas in oven heated to microwave heated sample extracts were also summarized. Table 3.3 Results from the statistical analysis of quantified chemicals unique to microwave and oven heated acetic acid sample extracts (n=4). Chemical name Experimental Treatment Average concentration (ng/mL) Relative standard deviation Relative standard error % Difference between average migrant concentrations in microwave heated compared to oven heated samples P-value 2,4-Di-tert-butylphenol Microwave heated 19.55(±0.97) 5% 2% +355% <0.000T Oven heated 4.29(±0.7) 16% 8% Room temperature control 0.13(±0.02) 16% 8% Erucamide Microwave heated 261.52(±4.94) 2% 1% When the amount of 2,4-di-tert-butylphenol and erucamide was divided by the average food simulant contact surface area and the number of heating cycles (where applicable), the migration rate (per microwave heating and unit surface area) of 2,4-di-tert-butylphenol ranged 4 2 ^ ^ from 7.49 x 10" to 1.15 x 10" ng/cm"; the migration rate of erucamide was 0.15 ng/cnr. Both quantified substances identified in acetic acid samples that were microwave and oven heated in plastic had concentrations with RSDs and RSEs that were less than 20%. The average concentration for each chemical had RSDs that were consistently higher than the RSEs. The RSDs and RSEs of the average 2,4-di-tert-butylphenol concentrations in oven heated and room temperature control extracts were more than double the RSD and RSE of the average 2,4-di-tert-butylphenol concentration measured in microwave heated sample extracts. The increase in the average 2,4-di-tert-butylphenol concentration measured in microwave heated sample extracts was significantly higher (p<0.01) than the average 2,4-di-tert-butylphenol concentration in oven heated sample extracts. 48 3.2 Plastic migrants released into isopropanol food simulant during microwave heating A comparison between Figure 3.1 and Figures 3.7 through 3.11 shows that seven of the chemicals identified in the isopropanol sample extracts were among the selected possible or known migrant chemical standards. Figures 7 to 11 show the chromatographs of the toluene blank sample, and the method blank, oven heated glass control, microwave heated glass control, room temperature control, oven heated and microwave heated isopropanol sample extracts. The chromatograph of one replicate sample extract per experimental treatment is presented. Substances that were positively identified or had mass ion peak areas greater than 1 000 000 in the microwave heated sample extracts were labelled. All peaks in the control sample extracts were labelled. Figures 3.7 to 3.9 are presented using the same, but smaller scale than the one used by Figures 3.10 and 3.11 in order to make the peaks present in the chromatographs visible. 49 Abundance 40000-20000-n a. Toluene blank Column bleed JW TIC: blk3.D\data.ms U J Time--) ' 5.0 1 1 1 1 1 1 1 1 ] 10.00 1 1 1 1 15.00 1 - 1 1 20.0 1 1 1 1 j 1 1 1 1 1 1 1 1 1 j 1 1 1 1 j 1 1 1 1 J-3 25.00 30.00 35.00 40.00 Abundance 40000-b Method blank sample extract Column bleed / I ^ ^ T l l t ^ l ^ u ^ , ^ , . J , TIC: Method blank.D\data.ms(") Unknown phthalate (m/z 149, 167, 281) 20000-n u Time-) ' 5.0 1 1 I 1 j 1 1 1 3 10.00 i I i i 15.00 1 1 1 20.0 1 1 1 1 1 1 1 1 1 j 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ] 25.00 30.00 35.00 40.00 Figure 3.7 Chromatographs of the toluene blank and method blank isopropanol sample extract, (a) Toluene (media) blank shows the presence of column bleed, (b) Method blank shows an occurrence of column bleed and an unknown phthalate. |4bundance a Oven heated glass control sample extract C TIC: oven control 1.D\data.m$( olumn bleed -i 1 1 1 1 i 1 1 1 1 1 1 1 1 1 1 1 1 1— 10.00 15.00 20.00 25.00 - i 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 r 30.00 35.00 40.00 Tirne--> 5. [Abundance b. Microwave heated glass control sample extract Column bleed TIC: micro control 1.D\data.ms ["] 40000-Unknown phthalate (m/z 149, 167, 281) HC contaminants -1 1 1 r I I I I I I I I I I I i I I i | 15.00 20.00 25.00 30.01 i 1 1 1 1 1 1 1 1 1 1 1 1 1 r 35.00 40.00 Time-) 5. i.00 10.( Figure 3.8 Chromatographs of an oven and a microwave control isopropanol sample extract, (a) Oven heated glass control sample extract shows the presence of column bleed, (b) Microwave heated glass control sample extract shows the occurrence of column bleed, an unknown phthalate and a hydrocarbon (HC) contaminant series. lAbundance TIC control L D W a t a m i f ) Column bleed 2,4-Di-tert-butylphenol ~ \ — i — i — i — i — i — i — r - i — i — r — i — I — i — i — i — r n i r " i — i — i — I — i — i — i — i — — r JUL JUL ML Figure 3.9 Chromatograph of a room temperature control isopropanol food simulant sample extract. A b u n d a n c e TIC: oven 2 , D \ d a t a . m s f ) Unknown HC (m/z 69, 97, 168) 2,6-Di-tert-butylbenzo-quinone Unknown HC '(m/z 69, 97, 168) 2,6-Di-t-butyl-4-methylene-2,5-cyclohexadiene-1 -one Unknown (m/z 117, 130,399) 2,4-Di-tert-butylphenol Di-tert-butyl-1-oxaspiro[4.5]deca-6,9-diene-2,8-dione) unknown phthalate (m/z 149, 167,279) Unknown (m/z 117, 130,371) Unknown (m/z 117, 190, 341) Erucamide 1 -Octadecanol \ I ' I ' I ""i i : i " T r 15.00 2-Monostearin Irganox 1076 / t u n T i m e - ) n—rn—r 5.00 10.00 35.00 Figure 3.10 Chromatograph of an oven heated isopropanol simulant sample extract. lAbundance hlme-> TIC: micro 2.D\dQtQ.rns fl 2,4-Di-tert-butylphenol 2,6-Di-t-butyl-4-methylene-2,5-cyclohexadiene-1 -one Unknown HC (m/z 69, 97, 168) 2,6-Di-tert-butylbenzo-quinone Unknown HC (m/z 69, 97, 168) Unknown (m/z 117, 130, 399)| Unknown phthalate' Unknown HC (m/z 149, 167, 279) (m/z 43, 57, 71) 2-Monostearin 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione) v].Octadecanol Erucamide Irganox 1076 Figure 3.11 Chromatograph of a microwave heated isopropanol simulant sample extract. LA Table 3.4 is a summary of the chemicals that were qualitatively identified by GC/MS analysis in all sample extracts and met the selection criteria for further data analysis (i.e. were detected in all microwave heated sample extracts). A total of 72 substances were identified; however, only the positively identified substances and those that that had peak areas greater than 600 000 were summarized below. Each symbol represents the presence (V), possible presence (?), or absence (*) of a chemical in one sample. 55 Table 3.4 The presence of specific chemicals identified in the control sample extracts and oven and microwave heated isopropanol simulant sample extracts. Qualitatively Identified Chemicals Retention time (min) Experimental treatment Toluene blank (n=l) Method blank (n=D Oven heated glass controls (n=4) Microwave heated glass controls (n=4) Room temperature controls (n=4) Oven heated plastic (n=4) Microwave heated plastic (n=4) Unknown HC (m/z 69, 111, 168) 4.13 X X x x x x x x x x x x x x WW WW Unknown HC (m/z 69, 97, 168) 4.30 X X x x x x x x x x x x x x wvs WW Tetradecane 9.23 X X x x x x x x x x x x x x VWS WW 2,6-bis( 1,1 -dimethylethyl)-2,5-cyclohexadiene-1,4-dione 10.45 X X x x x x x x x x x x x x WW WW 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-1 -one 10.63 X X x x x x x x x x x x x x WW WW Pentadecane 10.83 X X x x x x x x x x x x x x WW WW 2,4-di-tert-butylphenol 11.06 X X x x x x x x x x ww WW WW Unknown (m/z 43, 56, 71) 12.42 X X x x x x x x x x x x x x WW WW Heptadecane 13.85 X X x x x x x x x x x x x x WW WW Octadecane 15.25 X X x x x x x x x x x x x x WW WW 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione 17.01 X X x x x x x x x x x x x x WW WW Unknown HC (m/z 43, 57, 69) 17.36 X X x x x x x x x x x x x x WW WW Unknown HC (m/z 43,57,71) 17.51 X X x x x x x x x x x x x x WW WW Unknown phthalate (m/z 149, 205,223) 17.60 X X x x x x x x x x x x x x WW WW Isopropal palmitate 18.20 X X x x x x x x x x x x x x WW WW Octadecanol 18.97 X X x x x x x x x x x x x x WW WW Methyl stearate 19.47 X X x x x x x x x x x x x x wvs WW Stearic acid 19.91 X X x x x x x x x x x x x x vwv WW Unknown HC (m/z 43,57,69) 20.00 X X x x x x x x x x x x x x WW WW Isopropyl stearate 20.59 X X x x x x x x x x x x x x WW vws Unknown phthalate (m/z 149,91,238) 22.20 X X x x x x x x x x x x x x WW WW Unknown HC (m/z 57, 69, 71) 22.40 X X x x x x x x x x x x x x WW WW Unknown HC (m/z 43,57,71) 22.48 X X x x x x x x x x x x x x WW iws Unknown HC (m/z 43,57,71) 22.52 X X x x x x x x x x x x x x x x x x •/xxx Unknown (m/z 130, 117,57) 23.39 X X x x x x x x x x x x x x wvs WW Unknown phthalate (m/z 149, 167,279) 24.15 X x x x x / x x x x x x x WW WW Unknown amide (m/z 41,59, 72) 24.37 X X x x x x x x x x x x x x WW WW 56 Table 3.4 (continued) The presence of specific chemicals identified in the toluene blank, negative control sample extracts and oven and microwave heated isopropanol simulant sample extracts. Qualitatively Identified Chemicals Retention time (min) Experimental treatment Toluene blank (n=l) Method blank (n=l) Oven heated glass controls (n=4) Microwave heated glass controls (n=4) Room temperature controls (n=4) Oven heated plastic (n=4) Microwave heated plastic (n=4) Unknown HC (m/z 43, 57,71) 24.56 X X X X X X X X X X x x x x X X X X WW Unknown HC (m/z 43, 57, 69) 24.60 X X x x x x x x x x x x x x //// //// Unknown (m/z 117, 130, 399) 25.40 X X x x x x x x x x //// //// ???? Unknown (m/z 117, 190, 341) 25.59 X X x x x x x x x x xxxx //// WW 2-Monostearin 25.83 X X x x x x x x x x xxxx, ///x WW Erucamide 26.43 X X x x x x x x x x xxxx //// WW Unknown amide (m/z 43, 59, 72) 26.60 X X x x x x x x x x xxxx WW WW Unknown HC (m/z 69, 83 111) 26.65 X X x x x x x x x x xxxx WW WW Nonacosane 27.36 X X x x x x x x x x xxxx V xxx //// Unknown HC (m/z 57, 71 85) 28.23 X X x x x x x x x x xxxx WW WW Unknown HC (m/z 57, 69, 83) 28.53 X X x x x x x x x x xxxx WW WW Unknown (m/z 57,415,620) 32.29 X X x x x x x x x x xxxx •/xxx •/xxx Unknown (m/z 441, 57, 646) 33.00 X X x x x x x x x x xxxx WW //// Irganox 1076 35.88 X X x x x x x x x x xxxx WW //// Unknown (m/z 316, 647,662) 36.05 X X x x x x x x x x xxxx WW //// Unknown (m/z 604, 287, 117) 41.21 X X x x x x x x x x xxxx V xxx /xxx * Not detected in a sample extract. x x x x Not detected in four sample extracts. / Detected in a sample extract. / / / • / Detected in four sample extracts. ???? Possibly detected in four sample extracts. ? v V • Possibly detected in one sample extract but > /xxx Detected in one sample extract but detected in 3 samples. not detected in three sample extracts. //Sx Detected in three sample extracts but not detected in one sample extract. Based on the results presented in Figures 3.7 to 3.11 and Table 3.4, it can be observed that no substances were qualitatively identified in the toluene blank with the exception of column bleed contaminants. The same column bleed contaminants found in the toluene blank occurred consistently throughout all control and microwave heated sample extracts. An unknown phthalate with mass ions 149, 167, and 279 was the only substance identified in the method blank sample extract in addition to column bleed contaminants. 57 Column bleed contaminants were the only substances identified in the oven heated glass control sample extracts. A hydrocarbon series and an unknown phthalate with mass ions 149, 167, and 279 were identified in all microwave heated glass control sample extracts. Column bleed contaminants previously identified in the toluene blank, method blank and oven heated glass control sample extracts were also identified. Among all of the room temperature control replicate sample extracts, 2,4-di-tert-butyl phenol and an unknown substance (m/z 117, 130, 399) with a 25.40 minute retention time were the only chemicals identified that were not previously identified in the toluene blank, the method blank, or oven or microwave heated glass control sample extracts. A total of seventy substances were identified in the oven heated sample extracts; three of the seventy substances were not identified in all replicate oven heated sample extracts. 2,4-Di-tert-butyl phenol, an unknown phthalate with mass ions 149, 167, and 279, and an unknown substance (m/z 117, 130, 399) with a 25.40 minute retention time were the only chemicals that were previously identified in other control sample extracts. The majority of the substances present in oven heated sample extracts were unidentified hydrocarbons. However, other positively identified chemicals included 2,6-bis(l,l-dimethylethyl)-2,5-cyclohexadiene-l,4-dione, 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-l-one, erucamide, 7,9-Di-tert-butylrl-oxaspiro[4.5]deca-6,9-diene-2,8-dione, octadecanol, Irganox 1076 and several fatty acid esters (e.g. methyl stearate). The greatest number of substances was identified in the oven and microwave heated sample extract (See Figures 3.10 and 3.11). However, the largest chemical peaks occurred in the microwave heated sample extracts. A total of seventy-two substances were identified among the microwave heated sample extracts; five of the seventy-two substances were not present in all replicate microwave heated sample extracts. 2,4-Di-tert-butyl phenol, an unknown phthalate with mass ions 149, 167, and 279, and an unknown substance (m/z 117, 130, 399) with a 25.40 minute retention time were the only chemicals that were previously identified in other control sample extracts. The types of chemicals that were identified in the microwave heated sample extracts were almost identical to those found in the oven heated sample extracts. Figure 3.12 below shows the average selected mass ion peak areas of chemicals that were positively identified in more than one experimental treatment that involved the direct exposure of isopropanol simulant to a polypropylene plastic container. The peak areas are plotted logarithmically to ensure that the oven heated peak areas are visible. The average mass ion peak 58 area of each chemical measured in the microwave heated sample extracts was consistently larger (i.e. generally by > 200%) than the average peak area measured in the oven heated sample extracts. 2,4-Di-tert-butylphenol was the only chemical present in the room temperature control, oven and microwave heated sample extracts. The mass ion peak areas were plotted because standards were not available to quantify all of the identified migrants. 59 S3 Da Identified chemical migrants • Room temperature control sample extracts* • Oven heated sample extracts Q Microwave heated sample extracts Figure 3.12 Mass ion peak areas of chemicals positively identified in microwave heated isopropanol sample extracts. ON o Table 3.5 below is a summary of the average mass ion peak area, relative standard deviation and relative standard error for each chemical positively identified in the isopropanol food simulant samples exposed to oven and microwave heated plastic. The mass ion peak areas of migrants in Table 3.5 were analyzed because no chemical standards were available to quantify them. Chemicals that were quantified were excluded and analyzed separately. The percent increase in the average mass ion peak areas in the microwave heated sample extracts compared to the oven heated sample extracts were included. The p-values from the unpaired t-tests comparing the chemical mass ion peak areas in oven heated to microwave heated sample extracts were also summarized. Approximately half of the substances identified in isopropanol samples microwave and/or oven heated in plastic had average peak areas with a relative standard deviation (RSD) and a relative standard error (RSE) that was less than 30%. The average peak areas for each substance had RSDs that were higher than the RSEs; the RSDs and RSEs were consistently higher in oven heated sample extracts compared to the microwave heated sample extracts. Overall, the calculated RSDs for each substance ranged from 3 to 83% whereas the RSEs ranged from 2 to 41%. The increase in the average mass ion peak areas for chemicals in microwave heated sample extracts compared to oven heated sample extracts was significantly higher (p<0.01) for only four substances. 61 Table 3.5 Results from the statistical analysis of chemicals unique to microwave and oven heated isopropanol sample extracts (n=4). C h e m i c a l name Exper imen ta l Treatment Average peak area Relat ive s tandard deviat ion Relat ive s tandard e r r o r % Difference between average migran t concentrations i n mic rowave heated compared to oven heated samples P-value 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-1-one Microwave heated 2505422 (±1587195) 63% 32% 233% 0.07 Oven heated 752762 (±180562) 24% 12% 7,9-Di-tert-butyl-1-oxaspiro[4.5]de ca-6,9-diene-2,8-dione Microwave heated 82332 -(±18345) 22% 11% 95% 0.0001 Oven heated 42130 (±14146) 34% 17% Unknown phthalate (m/z 149, 205,223) Microwave heated 452469 (±177604) 39% 20% 36% 0.24 Oven heated 332837 (±47598) 14% 7% Methyl stearate Microwave heated 126779 (±4139) 3% 2% 146% <0.0001 Oven heated 51486 (±2596) 5% • 3% Isopropyl stearate Microwave heated 499883 (±77685) 16% 8% 266% 0.0001 Oven heated 136749 (±4878) 4% 2% Unknown phthalate (m/z 149,91,238) Microwave heated 101281 (±63640) 63% 31% 35% 0.45 Oven heated 75098 (±12557) 17% 8% Unknown phthalate (m/z 149, 167,279) Microwave heated 722277 (±139393) 19% 10% 103% 0.002 Oven heated 356214 (±17547) 5% 2% 2-Monostearin* Microwave heated 4588988 (±921989) 20% 10% 1463% Oven heated 293606 (±241808) 82% 4 1 % Nonacosane* Microwave heated 586812 (±485035) 83% 4 1 % Oven heated 59721 A percentage difference and/or a t-test could not be determined/completed for these substances because they were not detected in the same number of oven heated and microwave heated sample extracts Table 3.6 below is a summary of the average concentration, relative standard deviation and relative standard error for each quantified chemical identified in the acetic acid food simulant samples exposed to plastic. Less.than half quantified substances identified in acetic acid samples that were microwave and oven heated in plastic had average concentrations with RSDs that were less than 25%. In contrast, only three chemicals had average concentrations 62 with RSEs that were greater than 20%. Each chemical had average concentrations with RSDs that were consistently higher than the RSE. However, the RSDs and RSEs were consistently higher in oven heated sample extracts compared to the microwave heated sample extracts. Table 3.6 Results from the statistical analysis of quantified chemicals unique to microwave and oven heated isopropanol sample extracts (n=4). C h e m i c a l Expe r imen ta l Average Relat ive Relat ive M i g r a t i o n name Treatment concentrat ion s tandard s tandard rate per (ng/mL) deviat ion e r ro r heating (ng/cm 2 ) 2,4-Di-tert- Microwave heated 347(±66) 19% 9% 0.20 butylphenol Oven heated I60(±10) 6% 3% 9.5E-02 Room temperature* 1.06(±0.1) 11% 6% 6.3E-03 Erucamide Microwave heated 1808(±38.2) 2% 1% 1.07 Oven heated 793(±65.2) 8% 4% 0.47 Octadecanol Microwave heated 72.1 (±7.3) 10% 5% 4.3E-02 Oven heated 14.2(±9.9) 70% 35% 8.4E-03 Irganox 1076 Microwave heated 546(±40) 7% 4% 0.32 Oven heated 299(±67) 23% 11% 0.18 2,6-Di-tert- Microwave heated 12.2(±4) 32% 16% 7.2E-03 butylbenzo- Oven heated 5.6(±4) 71% 35% 3.3E-03 quinone Isopropal Microwave heated 3.1 (±0.6) 21% 10% 1.8E-03 palmitate Oven heated 2.2(±0.8) 35% 18% 1.3E-03 Stearic acid Microwave heated 21.1(±1.3) 6% 3% 1.2E-02 Oven heated 9.0(±0.3) 4% 2% 5.3E-03 Tetradecane Microwave heated 1.1 (±0.3) 28% 14% 6.3E-04 Oven heated 0.2(±0.1) 47% 24% 1 .OE-04 Pentadecane Microwave heated 2.0(±0.4) 18% 9% 1.2E-03 Oven heated 0.4(±0.1) 24% 12% 2.4E-04 Heptadecane Microwave heated 1.1 (±0.3) 26% 13% 6.7E-04 Oven heated 0.6(±0.1) 16% 8% 3.7E-04 Octadecane Microwave heated 1.0(±0.3) 28% 14% 6.0E-04 Oven heated 0.7(±0.2) 33% 17% 4.1E-04 *Room temperature sample was not heated therefore the calculated migration rate does not take into account the number of heating cycles. Table 3.7 summarizes the unpaired t-test results and the percent increase in the average concentrations of substances identified in the microwave heated isopropanol sample extracts compared to the average concentrations of substances present in oven heated isopropanol sample extracts. 2,4-Di-tert-butylphenol was the only chemical identified in oven and microwave heated acetic acid and isopropanol simulant sample extracts whereas all other substances were identified in oven and microwave heated isopropanol sample extracts. The increase in the average migrant concentration in microwave heated compared to oven heated isopropanol sample extracts was significantly higher (p<0.01) in eight of the eleven substances. 63 Table 3.7 Summary of the results for the unpaired t-tests which analyzed the average mass ion peak areas of substances found in the microwave heated and oven heated isopropanol sample extracts. Chemical name % Difference between average migrant concentrations in microwave heated compared to oven heated samples P-value 2,4-Di-tert-butylphenol 116% 0.0013 Erucamide 128% <0.0001 Octadecanol 409% 0.0001 Irganox 1076 83% 0.0008 2,6-Di-tert-butylbenzoquinone 119% 0.055 Isopropal palmitate 39% 0.14 Stearic acid 134% <0.0001 Tetradecane 508% 0.0013 Pentadecane 386% 0.0001 Heptadecane 82% 0.016 Octadecane 45% 0.14 3.3 Comparison of plastic migrants released into acetic acid food simulant to isopropanol food simulant during microwave heating Table 3.8 shows the increase in average peak areas or concentrations of substances in isopropanol sample extracts compared to acetic acid sample extracts. Substances shown were not identified in any other control sample extracts. The average migrant peak areas/concentrations for all substances in microwave heated isopropanol simulant sample extracts was significantly higher (p<0.01) than average migrant peak areas/concentrations in microwave heated acetic acid simulant sample extracts. The average migrant peak areas/concentrations for all chemicals in oven heated isopropanol simulant sample extracts was also significantly higher (p<0.01) than average migrant peak areas/concentrations in oven heated acetic acid simulant sample extracts. 64 Table 3.8 Comparison of chemicals identified to microwave and oven heated acetic and isopropanol sample extracts. Separate columns show a comparison in microwave heated sample extracts (column 2) and oven heated sample extracts (column 4). C h e m i c a l name % Increase in average migrant concentrations/mass ion peak areas i n microwave heated isopropanol compared to acetic ac id s imulant samples P-value % Increase i n average migrant concentrations/mass ion peak areas i n oven heated isopropanol compared to acetic ac id s imulant samples P-value 2,4-Di-tert-butylphenol* 1677% 0.0001 3638% <0.0001 Erucamide 591% <0.0001 2,6-Di-tert-butyl-4-methylene-2,5-cyclohexadiene-1-one 259% 0.02 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione 588% <0.0001 4589% 0.0001 Unknown (m/z 117, 130, 371) 291% <0.0001 Unknown (m/z 57, 83, 98) 3372% 0.0003 Unknown (m/z 117, 130, 399) 1580% <0.0001 Unknown (m/z 117, 341) 1862% <0.0001 118475% 0.0008 Unknown (m/z 191, 415) 1577% 0.003 *The percent difference between average 2,4-Di-ter-butylphenol concentrations in acetic acid compared to isopropanol food simulant was 737%; the p-value was <0.0001. Table 3.9 shows the average concentration of each migrant detected per heating in acetic acid or isopropanol samples exposed to microwave or oven heated plastic. The amount of chemical migration from microwave heated samples was consistently higher than what was measured in oven heated samples. 2,4-Di-tert-butylphenol, erucamide and Irganox 1076 were among the quantified substances that occurred in the highest concentrations. Some tentatively identified substances such as the unknown phthalates did not have an available standard that could be used to calculate the concentration. If a standard or a structurally similar standard was available, a set of calibration standards were not analyzed when all of the migration test samples were analyzed because the presence of certain migrants (such as methyl stearate) were not identified in preliminary migration experiments; the use of structurally similar standards for quantification was also not considered. Based on the assumption that structurally similar chemicals have similar GC/MS responses (Skoog et al., 1990), the average concentrations of methyl stearate, isopropyl stearate, 65 2-monostearin were estimated using the calibration curve for stearic acid. The average concentration of 2,6-di-tert-butyl-4-methylene-2,5-cyclohexadiene-l-one was also estimated by using the calibration curve for 2,6-di-tert-butylbenzoquinone. A similar technique of estimating the concentration of migrants based on ratios of peak heights to concentrations of known standards compared unknown substances has been used by the Dutch National Institute for Health and Environment to enforce food packaging legislation (van Lierop, 1997). 66 Table 3.9 Average measured concentration of each migrant detected per heating in acetic acid or isopropanol sample extracts (n=4). Chemical name Experimental treatment Average concentration (ng/mL) measured in acetic acid simulant per heating Average concentration (ng/mL) measured in isopropanol simulant per heating 2,4-Di-tert-butylphenol Microwave heated 1.95(±0.10) 34.74(±6.56) Oven heated 0.43(±0.07) 16.05(±1.02) Erucamide Microwave heated 26.15(±0.49) 180.8(±3.8) Oven heated N/D 79.3(±6.5) Octadecanol Microwave heated N/D 7.2(±0.7) Oven heated N/D 1.4(±1.0) Irganox 1076 Microwave heated N/D 54.6(±4.0) Oven heated N/D 29.9(±6.7) 2,6-di-tert-butylbenzoquinone Microwave heated N/D 1.2(±0.4) Oven heated N/D 0.6(±0.4) Isopropal palmitate Microwave heated N/D 0.3(±0.06) Oven heated N/D 0.2(±0.08) Stearic acid Microwave heated N/D 2.1 (±0.13) Oven heated N/D 0.9(±0.03) Tetradecane Microwave heated N/D 0.11 (±0.03) Oven heated N/D 0.02(±0.01) Pentadecane Microwave heated N/D 0.20(±0.04) Oven heated N/D 0.04(±0.01) Heptadecane Microwave heated N/D 0.11 (±0.03 Oven heated N/D 0.06(±0.01) Octadecane Microwave heated N/D 0.10(±0.03) Oven heated N/D 0.07(±0.02) Methyl stearate* Microwave heated N/D 4.47(±0.15) Oven heated N/D 1.82(±0.09) Isopropyl stearate* Microwave heated N/D 17.52(±2.70) Oven heated N/D 4.82(±0.17) 2-Monostearin* Microwave heated N/D 151.75(±28.59) Oven heated N/D 10.30(±8.46) 2,6-Di-t-butyl-4-methylene-2,5-cyclohexadiene-1-one* Microwave heated N/D 23.41(±13.57) Oven heated N/D 7.91(±1.81) *Migrant concentration was estimated using t le calibration curve of a chemical with a similar structure. 67 4 DISCUSSION 4.1 Plastic migrants released into acetic acid food simulant during microwave heating 4.1.1 Negative controls 4.1.1.1 Media blank Column bleed contaminants were the only substances identified in the toluene blank which indicates that the source of any chemicals identified in all other sample extracts was not toluene. The repeated occurrence of column bleed contaminants in other samples was not a concern because they were easily identifiable by their characteristic mass ions and retention times and could be screened out from further analysis. 4.1.1.2 The effect of exposure to glassware The presence of the two unknown chemicals (m/z 161, 229, 244) in the method blank and all other control and microwave heated sample extracts was likely due to exposure of food simulant samples to residual environmental contamination on glassware even though all glassware was thoroughly cleaned between experiments. Other potential contaminant sources include: the handling of glassware with nitrile gloves, contaminants in the nitrogen gas or blow-down apparatus and contaminants in the fumehood. However, since only 2 substances with relatively small peak areas were identified in the method blank sample extract, the level of contamination during sample extraction or analysis was minimal. The two method blank contaminants were readily identified in other sample extracts by their mass spectra and retention times; thus, they were excluded from data analysis. 4.1.1.3 The effect of exposure to the inside of the oven and microwave There appeared to be no detectable sample contamination resulting from exposure of simulant to the inside of the microwave or toaster oven. 4.1.1.4 The effect of contact with plastic at room temperature on migration 2,4-Di-tert-butylphenol was the only substance identified in the room temperature control sample and was likely present as a result of contact between acetic acid simulant and plastic. 68 4.1.1.5 The effect of contact with non-microwave heated plastic on migration 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione, and two unknown substances with retention times of 25.40 minutes (m/z 117, 130, 399) and 25.55 minutes (m/z 117, 341) were identified in all replicate oven heated sample extracts. These substances were likely present as a result of acetic acid simulant exposure to heated plastic. Although 2,4-di-tert-butylphenol was identified in the room temperature control sample extracts, the average concentration in the oven heated sample extracts was approximately 4 times higher. Therefore, the increase in the amount of 2,4-di-tert-butylphenol migration was likely due to the heating of acetic acid simulant in plastic. 4.1.2 The effect of exposure to microwave heated plastic on migration The fact that the greatest number of and the largest peak areas were present in all replicated microwave heated sample extracts suggests that microwave heating had the greatest effect on the number of migrants and amount of chemical migration. Only four of the thirteen substances identified in all replicate microwave heated sample extracts were previously identified in oven heated sample extracts in this study. All thirteen substances were likely present as a result of exposure of acetic acid simulant to microwave heated plastic. Thus, it appears that microwave heating increased the number of chemicals migrating and rate of migration. 4.1.2.1 Identified migrants Among the thirteen substances that were present in microwave heated sample extracts, 2,4-di-tert-butylphenol, 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione and erucamide were previously identified and/or quantified in other migration studies (Svjevrak et al., 2005; van Lierop, 1997). 2-Norbornene was not identified in other studies but it is used as a monomer in the manufacture of olefin plastic; thus, it is likely a chemical migrant. 4.1.2.2 Unidentified migrants The presence of unidentified hydrocarbons and unknown substances in oven and microwave heated acetic acid sample extracts exposed to plastic was not unusual. Hydrocarbons are used as lubricants during the plastic manufacturing process, and as starting materials for plastics and resins (Ullmann, 2006). The vast majority of migration studies to date have focussed on overall migration, which does not require the identification of individual substances, 69 or measurement of specific migration of target chemicals. In studies where a range of all potential migrants were examined, the identity and/or quantity of identifiable chemicals were described whereas there was little or no mention of unknown substances that could not be positively matched. Studies by Skjevrak et al. (2005), Jenke et al. (2005), and Gramshaw et al, (1995) were among some of the few that described the presence of unidentified substances in simulant samples. Jenke et al. (2005) specifically mentioned the difficulty in positively confirming the identity of several chromatograph peaks due to the lack of reference standards for the proposed identity of specific chemicals. The small peak size of the unknown substances was also a problem because chemical identification is difficult if the chemical is present at levels close to its detection limit. It is important to note that the some of the unknown substances identified in the microwave and oven heated acetic acid sample extracts in this study had the largest peaks. Thus, the low concentration of some unknown substances likely did not affect the ability to identify them. 4.1.3 Statistical analysis of migrant peak areas and concentrations The results of the statistical analysis of the average mass ion peak areas or concentrations of substances identified microwave heated sample extracts showed that there was an acceptable amount of variation (i.e. relative standard deviation (RSD) <20% (US FDA/CFSAN, 2002a)) between replicate samples even though there was a relatively small sample size for each experimental treatment. The average chemical mass ion peak area or concentrations also did not vary greatly from the mean concentration expected for a large sample of plastic containers since the majority of relative standard errors (RSEs) were generally less than 10%. Substances that had average peak areas with RSDs and RSEs greater than 30% were possibly a result of the fact that the chemicals were present in levels close or to their detection limit. For example, the average peak area RSD and RSE for 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione in microwave heated sample extracts was 11% and 5%, respectively. However, the average peak area RSD and RSE for the same chemical in oven heated samples was 128% and 64%. Thus, the fact that the average peak area for 7,9-Di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione was over 1000 times higher in the microwave heated sample extracts compared to the levels measured in oven heated sample extracts likely increased the precision of the measured mass ion peak areas and reduced the RSD and RSE. Migrant average concentrations and/or mass ion peak areas with high RSD and RSE values may 70 also be a result of variation between plastic containers since chemicals with low RSD and RSE values occurred in the same replicate samples as the substances with the high RSDs and RSEs. Even though plastic containers from the same package and lot number were used for all replicate oven and microwave heated sample extracts, it appears that there may have been variation in the composition of plastic between individual containers. Overall, the number and concentration of substances migrating into sample extracts exposed to plastic increased with the addition of heat; microwave heating in particular appears to increase chemical migration of specific substances at a higher rate. The increase in concentration/mass ion peak area in microwave heated sample extracts for all substances measured in microwave and heated acetic acid simulant sample extracts was significantly different (p<0.0001). 4.2 Plastic migrants released into isopropanol food simulant during microwave heating 4.2.1 Negative controls 4.2.1.1 Media blank As with the acetic acid simulant media blank, column bleed contaminants were identified in the toluene blank and screened out from further data analysis. 4.2.1.2 The effect of exposure to glassware, the inside of the oven and microwave A hydrocarbon series and an unknown phthalate (m/z 149, 167, 279) were identified in the method blank and microwave heated glass control sample extracts that were not present in the toluene blank or oven heated glass control sample extracts. These substances had small peak areas therefore the level of contamination during sample extraction is minimal and the cleaning protocol appears to be effective. The hydrocarbon series found in microwave heated glass control sample extracts was not detected in other sample extracts therefore it did not interfere with data analysis. A small number of substances were identified in microwave heated glass control sample extracts whereas no contaminants were detected in oven heated glass sample extracts. Thus, contamination due to isopropanol simulant exposure to the inside of the microwave appears minimal. There also does not seem to be any sample contamination due to exposure to the inside of the toaster oven. 71 4.2.1.3 The effect of contact with plastic at room temperature on migration The presence of 2,4-di-tert-butylphenol and an unknown substance (m/z 117, 130, 399) in room temperature control sample extracts is likely a result of direct contact between isopropanol samples and the plastic containers because they were not identified in the method blank, oven or microwave heated glass control sample extracts. 4.2.1.4 The effect of contact with non-microwave heated plastic in migration Sixty-seven of the substances identified in oven heated sample extracts were likely a result of exposure of isopropanol to heated plastic since they were not detected in the method blank, oven or microwave heated glass control sample extracts. Although 2,4-di-tert-butylphenol and an unknown substance (m/z 117, 130, 399) were identified in the room temperature control sample extracts, they were present in higher concentrations in all oven heated sample extracts. Therefore, the increase in the amount of 2,4-di-tert-butylphenol migration was likely due to the heating of acetic acid simulant in plastic. Although the unknown phthalate (m/z 149, 167, 279) was found consistently in all oven heated sample extracts, it was subsequently excluded as a chemical migrant (refer to discussion on phthalates below for explanation). 4.2.2 The effect of exposure to microwave heated plastic on migration The greatest number of large peaks was present in microwave heated sample extracts, which suggests that microwave heating had the greatest effect on the amount of chemical migration. The number and types of chemicals identified in the microwave heated sample extracts was similar to what was detected in oven heated sample extracts, which indicates that the effect of microwave heating on the amount of migration from plastic into isopropanol may have been greater than its effect on the number of migrants. 4.2.2.1 Phthalates The three phthalates that were identified in microwave heated sample extracts could not be positively identified due to GC/MS analysis method limitations despite the occurrence of high quality library mass spectra matches (i.e. greater than 90%). For example, bis(2-ethylhexyl)phthalate, dicyclohexyl phthalate, and diisoctyl phthalate were possible identity matches for the unknown phthalate (m/z 149, 167, and 279). However, the peak retention time for the standards bis(2-ethylhexyl)phthalate and dicyclohexyl phthalate was consistently different than the peak retention time of the unknown phthalate (m/z 149, 167, and 279). 72 Although a diisoctyl phthalate standard was not tested, the expected retention time for diisoctyl phthalate was also too far from the retention time for the unknown phthalate (m/z 149, 167, and 279). In this study, the use of GC/MS was inadequate for positively identifying phthalates due to the fact that the molecular ion was generally very weak or absent from the mass spectra which made it difficult to distinguish between phthalate isomers (George and Prest, 2001). The term "molecular ion" used in this study refers to the ion formed by the removal of or addition of electrons to a molecule without fragmentation of the molecular structure (Skoog et al, 1990). Alternative GC/MS analytical methods that better identify phthalates are available (i.e. positive chemical ionization with retention time locking (George and Prest, 2001); however, they were not explored or applied in this study due to time constraints and limited financial resources. An unknown phthalate with mass ions 149, 167, and 279 showed a significant increase in peak area in microwave heated sample extracts compared to oven heated sample extracts. However, it was decided that it was not scientifically sound to consider the unknown phthalate (m/z 149, 167, 279) a chemical migrant because it was detected in a microwave heated glass control and the method blank sample extract. This phthalate was likely present as a result of environmental contamination. The peak area of the unknown phthalate (m/z 149, 167, 279) in the microwave heated control sample extract was approximately the same size as the peak areas measured in the oven heated samples; thus it was difficult to determine whether or not the presence of this phthalate was a result of chemical migration or environmental contamination. The other two unknown phthalates ((m/z 149, 91, 238) and (m/z 149, 205, 223)) were considered chemical migrants because they were not identified in any of the previous controls. Phthalates are also commonly used as plasticizers in food contact plastics (Lau and Wong, 2000). 4.2.2.2 Identified migrants Among the substances that were present in microwave heated sample extracts, 2,4-di-tert-butylphenol, 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione, erucamide, methyl stearate, isopropyl stearate, stearic acid, 2-monostearin, isopropyl palmitate, octadecanol, nonacosane, tetradecane, pentadecane, heptadecane, octadecane, 2,6-di-tert-butybenzoquinone, Irganox 1076, and 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-l-one were related to the plastic manufacturing process or identified as plastic additives. In general, the measured concentrations of migrants in acetic acid and isopropanol simulant identified in this study could not be directly compared to results in other studies because different experimental conditions and methods of migrant extraction and analysis were 73 used. However, the fact that these identified migrants have known uses in different food contact plastics and were measured in other migration experiments confirms the results of this study. 2,4-Di-tert-butylphenol 2,4-Di-tert-butylphenol is used as starting material for production of UV stabilizer used plastics and agrochemicals; it is also a degradation product of antioxidants used in plastics. 2,4-Di-tert-butylphenol has been qualitatively identified in multi-layered polyolefin laminate, polycarbonate, polyolefin, high density polyethylene and polypropylene plastic (Salafranca et al., 1999; Jenke et al., 2005; Marque et al., 1998; Nerin et al., 2003; Svjevrak et al., 2005). The maximum concentration of 2,4-di-tert-butylphenol measured in samples of water exposed to polyolefin bottles at room temperature for 72 hours was 25 pg/L (Svjevrak et al., 2005). Marque et al. (1998) found concentrations of 2,4-di-tert-butylphenol ranging from 35 to 44 pg/dm2in isooctane exposed to polypropylene film at 25°C and 80°C for 24 hours. In this experiment, the average measured concentration of 2,4-di-tert-butylphenol in acetic acid simulant exposed to plastic at room temperature for -3.5 hours (-0.13 pg/L) was lower than the maximum concentration measured in the study by Svjevrak et al. (2005). The average measured concentration of 2,4-di-tert-butylphenol in isopropanol exposed to oven heated plastic (-9.5x10 3 pg/dm") was also lower than what was quantified in the study by Marque et al. (1998). The lower measured concentrations of 2,4-di-tert-butylphenol in this study compared to the studies by Svjevrak et al. (2005) and Marque et al. (1998) was likely due to the different types of plastic tested, different methods of testing (i.e. food simulants), extraction and analysis used, and the shorter testing periods. 7,9-Di-tert-butyl-l-oxaspiror4.51deca-6,9-diene-2,8-dione 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione is a degradation product of an antioxidant used in food contact plastics; it was previously identified in samples of water exposed to polyolefin bottles for 72 hours at room temperature (Svjevrak et al., 2005). In this study, 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione was only identified in acetic acid and isopropanol simulants exposed to microwave and oven heated plastic. It is possible that 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione was present in undetectable concentrations in both simulants exposed to plastic at room temperature since it was identified by Svjevrak et al. (2005) in water samples exposed to polyolefin plastic at room temperature. Overall, the levels of 7,9-di-tert-butyl-l-oxaspiro[4.5]deca-6,9-diene-2,8-dione in this study were 74 not directly comparable to the results reported by Svjevrak et al. (2005) because different plastics were tested and the food simulant was exposed to plastic for a longer period of time in the study by Svjevrak et al. (2005). Erucamide Erucamide is a slip and release agent used in manufacture of olefin plastics (EC JRC, 2006a) which has been qualitatively identified in samples of multi-layered polyolefin laminate, polycarbonate, polyolefin, high/low density polyethylene, polystyrene, polyvinyl chloride, polyethylene terephthalate, and polypropylene plastic (van Lierop, 1997; Salafranca et al, 1999; Jenke et al, 2005; Cooper and Tice, 1995). Erucamide has also been quantified in migration studies. Cooper and Tice (1995) oven heated samples of ethanol and acetic acid food simulant exposed to polypropylene for 10 days at 40°C; no erucamide was detected in the ethanol and acetic acid food simulant samples (<0.04 mg/kg of simulant) whereas a concentration of 1.5 mg/kg was measured in olive oil. In a study by Kawamura et al. (2000), levels of erucamide quantified in solvent extracts of polypropylene utensils and packaging ranged from <200 - 808 pg/g of simulant and <200 - 532 pg/g, respectively (additional method details were not described in English). In this study, the average concentration erucamide measured in acetic acid simulants exposed to microwave heated plastic was -0.03 mg/kg and the average concentrations measured in isopropanol simulant exposed to microwave and oven heated plastic were -0.23 and 0.10 mg/kg (or fxg/g), respectively. Although erucamide was not detected in acetic acid simulant in the study by Cooper and Tice (1995), erucamide was likely detected at higher levels in acetic acid simulant in this study because a higher heating temperature was used even though simulant exposure to microwave heated plastic took place over a shorter time period. Cooper and Tice (1995) also used a different GC detector which may have had a higher detection limit. The levels of erucamide measured in olive oil (Cooper and Tice, 1995) and other simulants (Kawamura et al, 2000) were likely higher than the amounts measured in microwave and oven heated isopropanol in this study because different plastics were tested and different experimental conditions were used (i.e. exposure to plastic over longer heating periods). Olive oil also likely has a higher water-octanol partition coefficient than isopropanol. Fatty acids, fatty acid esters, alcohols Methyl stearate, isopropyl stearate, stearic acid, 2-monostearin, isopropyl palmitate and octadecanol are primarily used as lubricants during the plastic manufacturing process or as a 75 plastic additive. Stearic acid has been qualitatively identified in samples of thermoset polyester, multi-layer polyolefin laminate, and polypropylene, polyethylene, polyvinyl chloride plastic (Jenke et al., 2005; Gramshaw et al., 1995; Stringer et al., 2000; Kawamura et al., 2000). . Octadecanol was the only other chemical previously quantified in a study by Kawamura et al. (2000). Measured concentrations of octadecanol that ranged from <50 to 272 pg/g (of simulant) and <50 to 451 pg/g in solvent extracts of polypropylene package and utensils, respectively (methodology details were not provided in English). This group of migrants was likely only occurred in isopropanol simulant exposed to microwave and oven heated plastic because they are highly fat soluble. The average measured octadecanol levels in this study (-0.002 to 0.009 pg/g) were likely lower than levels measured in the study by Kawamura et al. (2000) because different experimental conditions were used and different plastics were tested. Hydrocarbons Nonacosane, tetradecane, pentadecane, heptadecane, octadecane are used as lubricants during the plastic manufacturing process or as plastic additives and as starting material for plastics and resins. Linear hydrocarbons, including tetradecane, pentadecane, heptadecane and octadecane, have been previously qualitatively identified in samples of, or as migrants from nylon microwave and roasting bags, and polyethylene, polypropylene, polystyrene, polyvinyl chloride and polyethylene terephalate plastic (van Lierop et al., 1997; Salafranca et al., 1999; Soto-Valdez et al., 1997; Stringer et al., 2000). This group of migrants was likely only occurred in isopropanol simulant exposed to microwave and oven heated plastic because they have high fat solubility. Nonacosane has only been identified in polyvinyl chloride plastic (Stringer et al., 2000) whereas heptadecane was the only hydrocarbon previously identified in poly(ethersulphone) plastic (Gramshaw et al., 1995). The only quantitative migration study that examined non-volatile contaminants that included hydrocarbons was conducted by Kawamura et al. (1997b). The levels of octadecane measured in solvent extracts of polyethylene plastic ranged from < 50 to 340 pg/g of simulant (Kawamura et al., 1997b) which were higher than the concentrations measured in isopropanol exposed to microwave and oven heated plastic in this study (-8.9 x 10"5 to 1.3 x 10"4 pg/g of simulant, respectively). Different test materials and experiment methodology were likely the reasons for the dissimilar octadecanol migration levels. 76 Antioxidants and antioxidant breakdown products Irganox 1076 is an antioxidant used in food contact plastics whereas 2,6-di-tert-butybenzoquinone and 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-l-one are primarily breakdown products of Irganox-type antioxidants and BHT (which is also a plastic antioxidant), respectively. Irganox 1076 was the only migrant that has been previously qualitatively identified in samples of polycarbonate, polypropylene, high, and low density polyethylene, high impact polystyrene (HIPS), acrylonitrile butadiene styrene (ABS) (Cooper et al., 1998; Nerin et al., 2003). Irganox 1076 has also been quantified in previous migration studies. Cooper et al. (1998) detected levels of Irganox 1076 that ranged from 1.4 to 10.7 mg/kg in a variety of standardized heated food simulants exposed to polypropylene plastic. In a study by O'Brien and Cooper (2001), the measured concentration of Irganox 1076 ranged from 1.1-19.2 mg/kg in olive oil heated for 2 hours at 70°C and 121°C, and for 10 days at 40°C. Bieber et al. (1984) found concentrations of Irganox 1076 that ranged from 1.04 to 211 pg/dm2 in food simulants exposed to polypropylene and heated to 40°C for 10 days. Kawamura et al. (2000) quantified levels of Irganox 1076 in solvent extracts of polypropylene utensils and packaging which ranged from < 50 to 340 pg/g of simulant (detailed methodology was not described in English). In another study by Kawamura et al. (1997a), levels of Irganox 1076 in food simulants exposed to polyethylene products ranged from <0.2 to 2.11 pg/cm2. In this study, the average concentrations of Irganox 1076 measured in isopropanol simulant samples exposed to microwave and oven heated plastic ranged from -0.04 to 0.07 mg/kg or pg/g (or 0.018 to 0.032 pg/dm2) which were lower than measured levels in studies by Cooper et al. (1998), O'Brien and Cooper (2001), Bieber et al. (1984) and Kawamura et al. (2000). Different test materials and experiment methodology were likely the reasons for the lower Irganox 1076 migration levels measured in this study. 4.2.2.3 Unidentified migrants The fact that a large number of migrants were unidentified in the oven and microwave heated isopropanol samples exposed to plastic is not uncommon (See Discussion of plastic migrants released into acetic acid food simulant during microwave heating). 4.2.3 Statistical analysis of migrant peak areas and concentrations The results of the statistical analysis of the average mass ion peak areas or concentrations of each substance identified in all replicate microwave heated sample extracts show that there 77 was a reasonable amount of variation (i.e. RSD <25%) between replicate samples for approximately half of the substances. Thus, even though there was a relatively small sample size for each experimental treatment the measured concentrations/peak areas were fairly consistent. The average chemical mass ion peak area or concentrations for the majority of the substances did not vary greatly from the mean concentration expected for a large sample of plastic containers since the majority of RSEs were generally less than 20%. Substances that had average peak areas with RSDs and RSEs greater than 30% were potentially a result of the fact that the chemicals were present in levels close to their detection limit. The average peak area RSD and RSE for substances in microwave heated sample extracts were consistently lower than the average peak area RSD and RSE for the same chemical in oven heated samples. This observed trend may be due to the increase in average peak area/concentration of chemicals in the microwave heated sample extracts or variation between plastic containers. Overall, the number and concentration of substances migrating into sample extracts exposed to plastic increased with the addition of heat; microwave heating in particular appears to increase chemical migration of specific substances at a higher rate. The increase in concentration/mass ion peak area in microwave heated sample extracts for the majority of the substances measured in microwave and heated isopropanol simulant sample extracts was significantly different (p<0.05). Chemicals that did not show a significant increase in average peak area or concentration in replicate microwave sample extracts compared to oven heated extracts may not be as strongly affected by experimental heating conditions as other substances were (i.e. due to their physical and chemical properties). Chemicals for which peak areas were the only measure of concentration may have been affected by the fact that peak areas may not adequately reflect the actual concentration of substance i.e. the relationship between peak area and concentration may not be linear (See Appendix B). Thus, the actual difference in chemical concentration between experimental treatments may not be accurate. Another factor that may have compromised the ability to determine if the difference between microwave and oven heated sample extracts was significant was the amount of migrant present in the sample. If the migrant concentration was too low it would be difficult to determine the differences between samples and experimental treatments. Finally, variability in the concentration of migrants between plastic containers may have also affected the percent difference in the measured levels of migration within and between experimental treatments. 78 4.3 Comparison of results of acetic acid food simulant and isopropanol food simulant experiments The type of food simulant appears to have a significant impact on the amount of chemical migration and number of chemicals migrating. The amount of substance migration from plastic into isopropanol is significantly higher (p<0.05) than the levels found in acetic acid food simulant exposed to oven or microwave heated plastic. A total of seventy two substances were identified in all microwave heated isopropanol sample extracts whereas only thirteen chemicals were detected in all microwave heated acetic acid sample extracts. More migrants occurred in the highest concentrations in isopropanol samples because isopropanol is fat soluble and most monomers, additives and plastic materials are lipophilic (Feigenbaum et al, 1994). The results from this study are consistent with results from previous migration studies that compared migration in aqueous to fatty food simulants in terms of: the types of substances identified; the greater amount of migration from microwave heated compared to oven heated plastic into food simulants; and the higher levels of chemical migrants in acetic acid compared to isopropanol food simulant (Bieber et al, 1984; Bourges et al, 1993; Galotto and Guarda, 1999, 2004). 4.4 Physical and chemical properties, usage information, regulatory and toxicological background of potential chemical migrants released during microwave heating 4.4.1 Physical and chemical properties of identified chemical migrants Table 4.1 below summarizes the physical and chemical characteristics of the substances positively identified in all replicate acetic acid and isopropanol samples exposed to microwave heated plastic. The majority of the chemicals had a molecular weight greater than 200 g/mol and a low vapour pressure (i.e. < 0.01 mm Hg @ 25°C). All substances identified in acetic acid food simulant samples, except for erucamide, had a water solubility greater than 10 mg/L whereas all chemicals had a log K o w value10 that was less than 6. The majority of the chemicals that was present in isopropanol food simulant samples had log K o w values that ranged from 3 to 15 but low water solubility (i.e. <1 mg/L). The physical and chemical properties of the identified chemical migrants were generally consistent with the food simulants in which they occurred, and their measured concentration. The majority of the chemicals had a molecular weight greater than 200 g/mol and a low vapour pressure which was likely due to the fact that the migration test methods were selective for non-1 0 L o g K o w or the octadecanol-water partition coefficient is a measure of the ability of a chemical to dissolve in fat. 79 volatile migrants. 2-Norbornene has a moderate vapour pressure but a high water solubility which is likely why it was still detectable. Most substances identified in acetic acid food simulant samples were water soluble and had moderate to low log K o w values. Conversely, the majority of the substances identified in isopropanol samples had low water solubility and moderate to high log K o w values. More fat soluble migrants at higher concentrations were also identified in the isopropanol samples. Erucamide, 2,4-di-tert-butylphenol and 7,9-di-tert-butyl-l-oxaspiro[4.5]-deca-6,9-diene-2,8-dione were among the substances that had the highest levels in both food simulants even though their physical and chemical properties indicate that they were more soluble in one simulant or equally soluble in both simulants. This unexpected occurrence may be a result of the fact that high levels of these substances were initially present in the plastic container. 80 Table 4.1 Physical and chemical properties of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic. Chemical Name Molecular formula Molecular Weight (g/mol) Water Solubility @ 25°C (mg/L) Log K * Vapour pressure @ 25°C (mm Hg) Type(s) of food simulant migrant was identified in References 2-Norbornene 94.16 130 3.2 39.2 Acetic acid ACD/Labs, 2006 7,9-Di-tert-butyl-l-oxaspiro[4.5]-deca-6,9-diene-2,8-dione C17H24O3 276.37 83 3.1 1.79E-07 Acetic acid & Isopropanol ACD/Labs, 2006 Erucamide C 2 2 H 4 3 NO 337.58 0.2 5.3 5.82E-09 Acetic acid & Isopropanol SRC, 2004 2,6-Di-t-butyl-4-methylene-2,5-cyclohexadiene-1 -one C15H22O 218.33 110 4.3 7.89E-04 Acetic acid & Isopropanol ACD/Labs, 2006 2,6-Di-tert-butylbenzoquinone C14H20O2 220.31 11.6 4.4 4.11E-04 Isopropanol SRC, 2004 2,4-Di-tert-butylphenol C14H22O 206.33 35 5.2 4.48E-03 Acetic acid & Isopropanol SRC, 2004 Irganox 1076 C35H62O3 530.88 6.09E-09 13.4 3.38E-13 Isopropanol SRC, 2004 Stearic acid C i 8 H 3 6 0 2 284.49 0.597 8.2 7.22E-07 Isopropanol SRC, 2004 2-Monostearin C21H42O4 358.56 6.10E-01 7.46 1.79E-11 Isopropanol ACD/Labs, 2006; Kubow, 1996 Isopropyl stearate 0 2 iH4 2 02 326.57 1.34E-04 9.14 3.97E-05 Isopropanol SRC, 2004 Methyl stearate C19H38O2 298.51 1.17E-03 8.35 1.36E-05 Isopropanol SRC, 2004 Isopropyl palmitate C19H38O2 298.51 1.35E-03 8.16 5.59E-05 Isopropanol SRC, 2004 1 -Octadecanol C | 8 H 3 8 0 270.5 1.10E-03 7.72 2.70E-06 Isopropanol SRC, 2004 Tetradecane C14H30 198.4 2.20E-03 7.2 1.16E-02 Isopropanol SRC, 2004 Pentadecane Q 5 H 3 2 212.42 7.60E-05 7.71 3.43E-03 Isopropanol SRC, 2004 Heptadecane CnH 3 6 240.48 2.94E-04 8.69 2.28E-04 Isopropanol SRC, 2004 Octadecane C i 8 H 3 8 254.5 6.00E-03 9.18 3.41E-04 Isopropanol SRC, 2004 Nonacosane C29H60 408.8 2.76E-10 14.58 4.30E-10 Isopropanol SRC, 2004 *Log K o W or the octadecanol-water partition coefficient is a measure of the ability of a chemical to dissolve in fat. Substances with a higher Log K Q W value are more fat soluble. 4.4.2 Usage information and regulatory background of identified chemical migrants Table 4.2 below summarizes the usage and/or origin of the chemicals positively identified in all replicate acetic acid and isopropanol samples exposed to microwave heated plastic. The origin of all substances were related to the plastic manufacturing process (i.e. starting materials, processing aids) or additives used to improve the technical properties of plastic. All chemicals had no specific Canadian regulations whereas erucamide, Irganox 1076 and stearic acid were identified EU regulations; the use of stearic acid and octadecanol in food contact materials (i.e. how and where they may be used) was regulated in the US. The usage information for each of the substances detected and identified in all replicate microwave heated simulant sample extracts confirmed that these chemicals are plastic migrants. However, the lack of Canadian, US or EU regulations for the identified migrants is a concern given that many of these substances are commonly used in food contact plastics, have been measured previously in several migration studies and have unknown health effects. For example, in this study erucamide, Irganox 1067 and 2,4-di-tert-butylphenol were among the migrants that occurred in some of the highest concentrations whereas substances such as 7,9-di-tert-butyl-l-oxaspiro[4.5]-deca-6,9-diene-2,8-dione have little or no available toxicological data. These substances were not regulated in Canada, the US or Europe even though they are used in food packaging. Thus, the lack of exposure data on plastic migrants and exposure of the general population to unmonitored and potentially toxic substances needs to be addressed. 82 Table 4.2 Usage information and regulatory background of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic. Chemical Name Function/Origin Existing Regulations References 2-Norbornene • Monomer used in manufacture of olefin plastic N/A Kissin, 2005 7,9-Di-tert-butyl-l-oxaspiro[4.5]-deca-6,9-diene-2,8-dione • Degradation product of Irganox 1076 N/A Svjerak et al, 2005; Fischer et al., 1999 Erucamide • Slip and release agent used in manufacture of olefin plastics • Identified in the EU as an additive that may be incorporated in food contact plastics for "technical purposes" or as a medium in which food contact plastic polymerization occurs EC JRC, 2006a; EC, 2003 2,6-Di-t-butyl-4-methylene-2,5-cyclohexadiene-1-one • Breakdown product of butylated hydroxytoluene (BHT) which is an antioxidant used in food contact plastics N/A Lichtenthaler and Ranfelt, 1978 2,6-Di-tert-butybenzoquinone • Breakdown product of Irganox-type antioxidants N/A Bourges et al, 1993 2,4-Di-tert-butylphenol • Starting material for production of UV stabilizer used in plastics and agrochemicals; • Degradation product of antioxidants used in plastics (e.g. Irgafos 168, Irganox and bis(2,4-di-tert-butylphenyl)pentaerythrityldiphosphite) N/A Lorenc et al, 2003; Skjevrak et al, 2003; Tochacek and Sedlaf, 1995 Irganox 1076 • Antioxidant in food contact plastics • In the EU, Irganox 1076 has a Specific Migration Limit (SML) of 6 mg/kg EC JRC, 2006b; EC, 2005 Stearic acid • Lubricant used during the plastic manufacturing process or as a plastic additive to improve the flow and homogeneity of the plastic • Used to manufacture alkyd resins which are possibly used for coating plastic manufacturing machinery • Used to make polyvinyl stearate and stearyl palmitate, which are both surface lubricants • A copolymer with adipic acid • Identified in the EU as an additive that may be incorporated in food contact plastics for "technical purposes" or as a medium in which food contact plastic polymerization occurs • Under US FDA regulations, stearic acid: - may be present in foods as a lubricant or binder; - a component during manufacturing of other food grade additives; - or a defoaming agent component used in processing of beet sugar and yeast. Ahn and White, 2003; EC JRC, 2006c; EC, 2003; 21 CFR 172.860; 21 CFR 173.340 oo Table 4.2 (continued) Usage information and regulatory background of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic. Chemical Name Function/Origin Existing Regulations References 2-Monostearin • Lubricant used during the plastic manufacturing process or as a plastic additive to improve the flow and homogeneity of the plastic N/A Ullmann, 2006 Isopropyl stearate N/A Methyl stearate N/A Isopropyl palmitate N/A 1-Octadecanol • Under US FDA regulations, octadecanol may be used: as a component of food contact materials and to synthesize substances permitted for use as components of food contact materials. Examples of regulated food contact materials/material components include: laminates, cellophane, paper and paperboard, and resin and polymer coatings. Octadecanol as a direct food additive (i.e. as a lubricant, binder and defoaming agent) is also allowed. Ullmann, 2006; EC JRC, 2006d; 21 CFR 172.864; 21 CFR 175.300; 21 CFR 176; 21 CFR 177; 21 CFR 178 Tetradecane • Lubricant used during the plastic manufacturing process or as a plastic additive to improve the flow and homogeneity of the plastic • Starting material for plastics and resins N/A Ullmann, 2006 Pentadecane N/A Heptadecane N/A Octadecane N/A Nonacosane N/A oo 4^ The fact that most chemical migrants in food simulant are not regulated is a problem that has been previously identified by Skjevrak et al. (2005) and Grob (2002). Skjevrak et al. (2005) found that most of the migrants identified in food simulant samples exposed to a range of food contact plastics were not on lists of substances that were authorized for use as indirect food additives in the EU. The majority of the migrants were degradation products and impurities of other plastic additives/components such as adhesives, coatings and solvents. Grob (2002) noted the fact that current analytical techniques have a limited ability to detect or identify unknown migrants. 4.4.2.1 Erucamide The fact that there was no specific migration limit (SML) for erucamide is significant because it was one of the chemicals that were present in the highest concentrations and there is almost no toxicological data available for it. Erucamide is related to erucic acid which is regulated in the US and the EU most likely because of its well-documented range of toxicological effects (See Appendix C). In the EU, erucic acid is allowed as "a starting substance which may be used in the manufacture of plastic materials and articled" (EC, 2003). Although erucic acid does not have a SML, the Scientific Committee for Food, which consists of a group of scientist from a variety of backgrounds that advises the Commission of the European Communities on "any problem relating to the protection of the health and safety of persons arising or likely to arise from the consumption of food", has stated that the level of erucic acid in infant formula should not exceed 1% of total fatty acids due to the lack of available to determine safe erucic acid intake levels (SCF, 1995). In the US, rapeseed or canola oil, which contains erucic acid, is considered safe for consumption. However, rapeseed oil is not allowed in infant formula and all rapeseed oil must have low erucic acid content (i.e. less than 2% of total fatty acid content) (21 CFR 184.1555). 4.4.2.2 Irganox 1076 Irganox 1076 was the only EU regulated substance with a SML. This is likely due to its wide-spread use in food contact plastic, previously measured migration levels and toxic effects. Although there was limited toxicological data publicly available, it is assumed that there are more industry-based studies, and that they were used in establishing the SML. 85 4.4.2.3 Stearic Acid In the US and EU, stearic acid is allowed as an additive in food with no restrictions placed on its allowable concentration in food or plastic. However, given that there was no information available about stearic acid migration levels it would be prudent to monitor stearic acid in future migration studies. 4.4.2.4 Octadecanol Octadecanol is regulated in the US, but not in Canada or the EU; although no allowable concentration limit in foods has been established, the purity of octadecanol and how it may be used is described in great detail (21 CFR 172.864; 21 CFR 175.300; 21 CFR 176; 21 CFR 177; 21 CFR 178). However, more specific regulations on octadecanol in all jurisdictions should be developed given that the limited available toxicological data for octadecanol indicates that it has the potential to be toxic and it has been measured in migration studies. 4.4.3 Toxicological properties of identified chemical migrants Table 4.3 below summarizes the toxicological properties of chemicals positively identified in all replicate acetic acid or isopropanol samples exposed to microwave heated plastic. Oral toxicity studies conducted on mammals were specifically focused on since ingestion is the only route of exposure to chemical migrants. The health effects from other types of toxicity and/or biological studies were also summarized where there was limited oral toxicity data. Appendix C includes additional available toxicological and biological effects associated with other routes of exposure for some substances and for BHT and erucic acid, which are closely related to 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-l-one and erucamide. Overall there was very limited toxicological data available for most of the substances, especially information that related to chronic oral toxicity. Among the oral toxicity studies, few toxicological effects were studied. Only one toxicity study for 2,4-di-tert-butylphenol had no observed adverse effect level (NOAEL) values, which are required to determine the health risk associated with exposure to identified migrants. 86 Table 4.3 Toxicological properties of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic. Chemical Name Study Type Route of Exposure Test Species Dose Toxic/Biological Effects References 2-Norbornene Acute toxicity Oral Rat L D 5 0 = 11.3 ml/kg (10147 mg/kg) Death Smythe et al., 1969 7,9-Di-tert-butyl-l-oxaspiro[4.5]-deca-6,9-diene-2,8-dione N/A N/A N/A N/A N/A • ... ; • - N/A Erucamide Metabolic pathway Intravenous injection Rat 1.2 ng/mL Erucamide was detected in blood, lungs, liver and spleen after injection. After 130 minutes, the majority of erucamide remained in the lung and spleen whereas the remaining amount was identified in the liver. Hamberger and Stenhagen, 2003 2,6-Di-t-butyl-4-methylene-2,5-cyclohexadiene-1-one Acute toxicity Oral Rat and mouse 75, 150 and 300 mg/kg • Decrease in blood coagulation factors in rats and mice • Haemorrhages in epididymis or thymus of rats Takahashi, 1988 Acute toxicity In vitro Rat 100 uM Alkylated similar immunoreactive proteins in rat liver as 2,6-di-t-butyl-4-methylene-2,5-cyclohexdien-1 -one Reed and Thompson, 1997 Genotoxicity In vitro Supercoiled pUC18 DNA 1 to 1 x 1 0 V M Caused DNA cleavage Nagai et al., 1993 2,6-Di-tert-butybenzoquinone Cytotoxicity In vitro Rat hepatocyte and PC 12 cell cultures Hepatocytes: L D 5 0 = 218 u.M GSH50=>800 uM PC 12 cells: L D 5 0 = 877(xM GSH50=>900(xM Cell death and glutathione depletion Siraki et al., 2004 Genotoxicity In vitro Burkett's lymphoma cell line BJAB and human myelogenous leukemic cell line HL 60 100 \iM Induced DNA fragmentation Oikawa et al., 1998 oo ^1 T a b l e 4.3 (cont inued) T o x i c o l o g i c a l p roper t i e s o f m i g r a n t s ident i f ied i n acetic ac id a n d i sop ropano l s imu lan t exposed to m i c r o w a v e heated p las t ic . Chemical Name Study Type Route of Exposure Test Species Dose Toxic/Biological Effects References 2,6-Di-tert-butybenzoquinone Genotoxicty In vitro Supercoiled pUC18 DNA 1 to 10>M DNA cleavage via generation of oxygen radicals Nagai et al, 1993 Acute toxicity Not specified Mouse LD50=3085 mg/kg Death Niculescu-Duvaz et al, 1991 Acute toxicity Intraperitoneal injection Mouse LD50=2270 mg/kg over 3 days Death Yamamoto et al, 1980 2,4-Di-tert-butylphenol Subchronic toxicity Oral Rat Dose 5 to 300 mg/kg/day for 18 to 28 days NOAEL = 5 mg/kg/day (newborn) 20 mg/kg/day (young) Toxic effects at 300 mg/kg/day include: • increased liver and kidney weight; • decreased spleen weight; • decrease in locomotor activity; • decrease in hemoglobin and hemacrit; • increased blood levels of total protein and bilirubin. Hirata-Koizumi et al, 2005 Acute toxicity In vitro Rat liver cells LC 5 0 = 150 u.M (2 hours) Cell death Moridani et al, 2003 Acute toxicity Oral Rat 5% of commercial feed over 6 days Increase in liver weight, cytochrome P-450 levels and microsomal monoxygenase activity Kawano et al, 1981 Irganox 1076 Estrogenic activity In vitro yeast two hybrid bioassay Genetically modified yeast cells 0.1 to 10J uM No estrogenic activity was detected Ogawa et al, 2006 Estrogenic activity In vitro bioassay Human osteoblast estrogen receptor alpha and beta cell lines 0 to 50 \ i M No estrogenic activity was detected ter Veld et al, 2006 Endocrine toxicity In vitro bioassay Genetically modified Chinese hamster ovary cells 0.1 to 10" ^M No androgen agonist or antagonisty activity was observed Araki et al, 2005 Subchronic toxicity Oral Rat 30-1000 mg/kg/day for 14 days Liver enlargement and induction of hepatic microsomal xenobiotic metabolism Lake et al, 1980 Table 4.3 (continued) Toxicological properties of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic. Chemical Name Study Type Route of Exposure Test Species Dose Toxic/Biological Effects References Stearic acid Subchronic toxicity Oral Rat 6% of dietary intake for 22 to 40 weeks Essential fatty acid deficiency and the proliferation of existing urinary tract tumors Cremonezzi et al., 2004 Subchronic toxicity Oral Human 0.8 to 16.5% of daily energy intake for 18 days to 6 weeks A diet high in stearic acid may have a lesser effect on the increase in high density lipoprotetin cholesterol compared to other fatty acids; stearic acid lowers low density lipoprotein cholesterol Kris-Etherton and Mustad, 1994; Mensink, 2005 Acute toxicity Oral Human 16.1 - 35.6g per meal in one day Stearic acid did not have any significant effect on the levels of blood fat or hemostatic function Sanders and Berry, 2005 Acute and Subchronic toxicity Oral Human 7% of dietary energy intake over 5 weeks A diet with a higher stearic acid content significantly decreased the mean blood platelet volume Thijssen et al., 2005 Oral Human Not specified in one day, after 3-4 weeks of fasting or over 3 weeks Tholstrup, 2005 Oral Human 8% of saturated fat intake over 40 days Schoene et al., 1994 Carcinogenicity Subcutaneous injection Mouse 10-114 injections of 0.5 to 10 mg/mL over 18 months in series of experiments 4 cases of subcutaneous sarcomas, 1 case of adrenal carcinoma, 1 case of leukemia-lymphoma, 3 cases of pulmonary tumors in 92 mice Swern, 1970 in Anonymous, 1987 Chronic toxicity Oral Rat 50 000 mg/kg/day (24 weeks) Foreign body-type reaction in perigonadal fat and occurrence of lipogranulomas Herting at al., 1959 in: Anonymous, 1987 Chronic toxicity Oral Rat 3000 ppm (or mg/kg) (30 weeks) Anorexia, severe pulmonary infection, high mortality Deichmann et al., 1958 in: Anonymous, 1987 2-Monostearin Effect of monoacylglycerols on the metabolism of lipoproteins Intravenous injection Rat 6 mg/mL Decreased the removal rate of cholesterol ester from plasma by the liver Mortimer et al., 1990 Isopropyl stearate Acute toxicity Oral Rat 8 mL/kg over'2 weeks 1 of 10 rats died; an average 21% weight gain was observed Annonymous, 1985 oo Table 4.3 (continued) Toxicological properties of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic. Chemical Name Study Type Route of Exposure Test Species Dose Toxic/Biological Effects References Methyl stearate Chronic toxicity Subcutaneous Mouse 5200 mg/kg (26 weeks) Caused growth of tumours Van Duuren et al, 1972 Chronic toxicity Subcutaneous Mouse 5200 mg/kg (26 weeks) Caused growth of tumours Swern et al, 1970 Isopropyl palmitate Acute toxicity Oral Rat LD50=>5 mg/kg Death Opdyke and Letizia, 1982 1-Octadecanol Subchronic toxicity Oral Rat 10-30%(w/w)in powdered diet for 1 week to 3 months Enlarged mitochondria and a decrease in the number of mitochondria per hepatocyte Wakabayashi et al, 1991 Acute toxicity Oral Rat LD50=20 000 mg/kg Death Grayson and Eckroth, 1978 Carcinogenicity Implant (bladder) Mouse Unknown # of 24 -27 mg pellets Tumorogenic Bryan et al, 1966 Tetradecane Mutagenicity In vitro Salmonella strains TA100, TA98, UTH8414 andUTH8413 50 to 2000 ^g/plate Not mutagenic Connor et al, 1985 Pentadecane Cytotoxicity In vitro Chinese hamster lung cells 50 mg/mL or 5 mM Not cytotoxic or Kusakabe et al, 2002 Acute toxicity Intravenous Rat LD 5 0 = 16.5 x 10" 3 mol/kg or 3.5 x 103 mg/kg Death Jeppsson, 1975 Heptadecane Metabolic pathway In vitro Rat, rabbit and chicken liver microsomes 50 |xL, 27 mCi" Rate of heptadecane metabolism was highest for chickens (approximately 10 times higher than all other species), followed by rats and then rabbits. Perdu-Durand and Tulliez, 1985 Metabolic pathway Oral Rat 1 mg, 10 \iC\ or 200 mg, 30 \iC\ over 7 days Majority (65%) was excreted in feces; approximately 7% was absorbed whereas the remainder was oxidized to heptadecanoic acid, incorporated into lipids, broken down and utilized to synthesize lipids and other cell components Tulliez and Bories, 1978 Octadecane Acute toxicity Oral Mouse L D 5 0 = 1229 mg/kg/day Death Schafer and Bowles, 1985 " The amount of radioactive material that disintegrates (decays) at the rate of 37 thousand atoms per second (US NRC, 2003). N O o Table 4.3 (continued) Toxicological properties of migrants identified in acetic acid and isopropanol simulant exposed to microwave heated plastic. Chemical Name Study Type Route of Exposure Test Species Dose Toxic/Biological Effects References Nonacosane Case study Oral Human Unknown Lipophilic crystalline material was present in lung and liver tissue. High concentrations of nonacosane were detected in lipid extracts of lung and liver tissue (1.2mg/g and 0.32 mg/g, respectively). Nonacosane also accumulated in lymph nodes, adrenal glands, heart, and kidney. Rocchiccioli et al., 1987; Salvayre et al., 1988 Metabolic pathway Oral Rat 30 mg, 20.5^Ci over 5 days Majority of nonacosane was excreted; the liver had the next highest amount whereas the remainder was distributed between the lung, heart, kidney and blood. Kolattukudy and Hankin, 1966 Overall the number of toxicological studies available for each identified migrant was severely limited or absent. Among the data that were available for most substances, the toxic effects presented were not comprehensive. Chronic low dose exposure oral toxicity data and data on how migrants are metabolized in humans were essentially non-existent. Substances such as the some of the fatty acids/fatty acid esters and hydrocarbons which had high LD50 values (i.e. greater than 1000 mg/kg) from oral toxicity animal studies may have low toxicity. However, since studies have not been conducted on the other adverse health effects of oral exposure to these substances (i.e. octadecane, isopropyl palmitate) it is difficult to determine their overall toxicity. Substances that have shown the demonstrated a range of toxic effects such as the antioxidants/antioxidant breakdown products Irganox 1076 and 2,4-di-tert-butylphenol (liver toxicity), stearic acid (carcinogenic), methyl stearate (tumorogenic), and octadecanol (tumorogenic) may be toxic. However, the small number of toxicity studies does not provide enough information on the full range of potential toxic effects, especially those related to long-term low-dose oral exposure, which is the most important and relevant route of exposure for chemical migrants identified in this study.' One must also take into account the uncertainty associated with relating animal toxic effects to human toxic effects due to factors such as: intraspecies and interspecies variability, route of exposure, toxicological dose, type of toxicity test (i.e. in vitro vs. in vivo testing) and the increased sensitivity of vulnerable populations. Thus, the available toxicity data was not extrapolated to reflect the potential adverse effects of migrant consumption by the general population. The lack of toxicity data on chemical migrants was previously identified in studies by Skjevrak et al. (2005) and Barlow (1994). It is estimated that among the 3200 monomers, starting materials and additives evaluated by the Scientific Committee for Food, approximately 80% and 60% of monomers and additives, respectively, completely lack or have inadequate toxicity data (Barlow, 1994). Thus, the limited available toxicological studies on plastic migrants in the scientific literature was not specific to the migrants identified in this study. 92 5 CONCLUSION The results of this study have shown that microwave and oven heating result in different levels of migration. The amount of chemical migration for the majority of substances identified in acetic acid and isopropanol food simulant under the same experimental treatments was significantly higher (p<0.05) in food simulant samples exposed to microwave heated plastic compared to what was measured in food simulant samples exposed to oven heated plastic. For example, the average concentrations of 2,4-di-tert-butylphenol in acetic acid and isopropanol simulant samples exposed to microwave heated plastic were 355% (p<0.0001) and 116% (p=0.0013) higher, respectively, than the average concentrations in acetic acid and isopropanol simulant samples exposed to oven heated plastic. The number of migrants found in microwave heated acetic acid and isopropanol simulants (13 and 72, respectively) was also greater than what was present in oven heated acetic acid and isopropanol simulants (4 and 70, respectively). In addition, the amount of migration in isopropanol was significantly higher (p<0.05) than the amount of migration in acetic acid simulant which was likely due the fact that more migrants in the tested polypropylene plastic were fat soluble. The types of migrants identified in acetic acid and isopropanol simulant samples exposed to microwave heated plastic included: hydrocarbons, fatty acids and fatty acid amides/esters, antioxidant and antioxidant breakdown products, a monomer, an alcohol and unknown migrants. The identified and quantified migrants present in the highest concentrations were 2,4-di-tert-butylphenol (an antioxidant breakdown product), erucamide (a slip and release agent) and Irganox 1076 (an antioxidant). The large number of unidentified migrants and the lack of toxicological data for most of the identified migrants underscore the need for improved migration test methods, food packaging regulations and toxicity testing. 5.1 Study limitations Although the results of this study show a strong difference between the effects of microwave and oven heating on migration from plastic, there were a number of limitations that are summarized below: • Migration testing method. . => Toaster oven stability. The temperature in the oven fluctuated within ±5°C therefore it is possible that the heating temperature was potentially higher or lower than the 93 microwave heating temperature. However, the method simulated real use conditions and the relative standard deviation between replicate samples was low. => The crystallizing dishes used in the glass control experiments were not the same shape as the polypropylene containers which may have affected the ability of contaminants to enter the food simulant. A uniform space was created with glass clips between the watchglass and the dish rim to replicate the spaces where the watchglass did not cover the corners of plastic container. Although the size of the space between the watchglass and dish was approximately the same as the space at the corners of the plastic container, the shape of the space was different, which may have affected the potential for environmental contamination in the glass control experiments. However, since contamination was minimal in both migration experiments, this discrepancy was not a concern. => Sample stability. There was a decrease in migrant concentration due to degradation even though samples and sample extracts were stored in the freezer during the migration experiments. To reduce the impact of chemical degradation the microwave and oven heated experiments were conducted consecutively. The effect of chemical degradation was likely minimal given the low relative standard deviation values for replicate samples and the fact that migrants in the control experiments were detected. => The number of heatings was small relative to the number of times a short term use plastic food is actually reheated. However, the highest levels of migration tend to occur over the first heatings (Castle et al, 1989; Svjevrak et al, 2005). The accumulation of migrants also helped improve the ability to use GC/MS to detect and quantify them. => Variability in plastic composition between containers was likely. However, it was minimized by ensuring that all oven heated and microwave heated sample replicates were from the same package and/or had the same lot number. Extraction solvent. Methylene chloride is a highly fat soluble solvent, therefore it would be more selective for fat soluble migrants. However, water soluble migrants were extracted successfully. Based on preliminary experiments, the liquid-liquid extraction method was more effective with methylene chloride. Future experiments may use a mixture of solvents with different solubility characteristics to extract more water soluble migrants. GC/MS analysis. => A method detection limit was not determined; however, the relative standard error for the majority of the identified migrants in both simulants was acceptable (less than 25%). 94 Thus, the migration test methods had fairly high precision and a method detection limit was not required. => An internal standard was not added to each sample to verify instrument response and retention time stability. However, given that the selected internal standard should have a similar retention behavior and volatility as the target analyte, several internal standards would have been required in order to make use of an internal standard effective (Skoog et al., 1990). The type of target analytes were variable and unpredictable, therefore selecting appropriate internal standards that were also not chemical migrants or did not co-elute with a chemical migrant was unlikely. => The ability of GC/MS analysis to differentiate between phthalate isomers or positively identify all of the substances was limited. 5.2 Recommendations Based on the limitations described above, the following recommendations can be made to improve the current study methods: • Migration test method. => Increase the number of sample replicates to reduce relative standard error; => Conduct the migration test over shorter duration of time, freeze samples at a lower temperature or spread out replicate samples of each experimental treatment to reduce the impact of chemical degradation; => Improve the method recovery by testing a greater number of substances, increasing the extraction time, lowering the heating temperature used to evaporate samples, and/or using different equipment/methods to extract samples (i.e. Soxhlet extraction); => Increase the number of heatings to determine the effects of long term heating over the lifetime of the container and to improve the ability to detect and identify chemical migrates by increasing migrant concentrations. => Test foods instead of food simulants and/or a mixture of food simulants to reproduce actual use conditions. => The composition of additives may change between batches of plastic (Feigenbaum et al., 1994). Thus, replicate samples of plastic within the same lot number and package should be tested and compared against samples from other lots to ensure that the variation of migrants and migrant levels between batches of plastic is minimal. 95 • GC/MS data analysis method. => Use a lower peak cut-off threshold to increase the number of migrants analyzed; => Freeze sample extracts until they are analyzed because the time required to analyze each sample accumulates and results in samples sitting at room temperature for hours. In addition to the recommendations for improving the migration test method, the following changes to Canadian Food and Drugs Act Regulations are proposed: • Canadian Food and Drugs Act regulations do not explicitly cover containers/materials intended for food use; however, the food industry voluntarily submits information for food contact plastics to Health Canada for approval which is assessed in the same manner as food packaging materials (M.A. Pelletier, Food Packaging Materials and Food Additives, Health Canada, Ottawa, pers. commun, August 24, 2004). It is recommended that the current definition of packaging in the Canadian Food and Drugs Act be expanded to include all food contact materials in order to ensure that containers/packaging intended for food use undergo health hazard assessments. Broadening the definition of food packaging is more protective of human health and is consistent with the definitions used by the US FDA and the EU. • Migration and toxicological testing of residual substances in food contact materials and their breakdown/transformation by-products are currently not required for food packaging industry submissions. However, the chemical identity of many migrants is unknown and many migrants are degradation products and impurities of other plastic additives/components (Skjevrak et al, 2005). It is strongly recommended that Health Canada requires the identification, migration and toxicological testing of all migrants in order to make current regulatory requirements more protective of human health. • The results of this study and other studies (i.e. Galotto and Guarda, 1999, Inthorn et al, 2002) have demonstrated that migration is largely dependent on many factors including how the food simulant is heated, the type of plastic and the type of food simulant. Current Canadian regulations require migration testing under "actual use conditions" using simulants that reflect the type of food the packaging will be exposed; the effects of repeated use and microwave heating on migration are not addressed and migration testing during oven heating is acceptable. It is recommended that the migration testing under "worst-case" actual use conditions over the lifetime use of the food contact material is incorporated into food packaging regulations. This will ensure that the migration of chemicals in food contact materials is not underestimated. "Worst-case" actual use conditions should be clearly 96 defined by a set of criteria such that the amount of migration over the lifetime of the product is adequately characterized. It is also recommended that a "worst-case" testing protocol be developed to capture a range of variables that may impact migration such as a range of food simulant types and incorporating freezing, refrigeration and dishwashing as part of the simulated use conditions. Microwave heating must be required for testing migration from food contact materials intended for microwave use. • Regulations need to be established for known migrants, especially for chemicals that have been detected and measured in migration studies and/or have potential toxic health effects. In addition to making migration testing and food packaging regulation recommendations, the author of this study also had several opinions about how to improve the safety of Canadian food packaging. The opinions were based reviewing literature and EU and US food packaging regulations; they are summarized below: • The current toxicological threshold for excluding substances from further health assessment is not adequate for protecting human health because it is based on the assumption that below a certain exposure concentration, any chemical is non-toxic. This assumption does not take into account that there are substances that are considered non-threshold toxicants. Non-threshold toxicants are substances have an adverse health effect regardless of the dose. US FDA regulations have a similar toxicological threshold cutoff; however, carcinogenic chemicals are excluded. In accordance with the "precautionary principle" which states that "Where there are threats of serious or irreversible damage, lack of full scientific certainty shall not be used as a reason for postponing cost-effective measures to prevent environmental degradation." (Environment Canada, 2001), it is strongly recommended that the toxicological threshold is removed or modified to account for migrants that may be carcinogenic or have toxic effects at low doses. • The amount of toxicological data required for migrants is insufficient compared to the comprehensive range of toxicity testing required by the EU and the US FDA. Canadian food packaging regulatory toxicological data requirements lack crucial types of toxicity tests and toxicity testing of migrant transformation/breakdown products. It is particularly important to understand what the developmental effects of ingesting migrants are since expectant mothers may be exposed to potentially toxic migrants from plastic packaging during pregnancy which may result in unknown adverse health effects in their children. Young children also tend to be highly exposed to foods stored or heated in plastic dishware and packaging. It is 97 recommended that the criteria for toxicological data is expanded to include endocrine disruption potential testing, and developmental and long-term chronic oral toxicity tests for all potential migrants and migrant breakdown/transformation products. • The information requirements for pre-market hazard assessments do not explicitly address issues such as migration modeling or the criteria for scientifically sound analytical methods of migration testing (i.e. the criteria for selecting migration test methods, specific types of packaging and/or use conditions). It is recommended that migration models and testing methods that are US FDA and E U approved be reviewed and/or incorporated such that food packaging manufactures use internationally recognized testing methods and/or models (i.e. required number of replicate samples, controls, analytical methods). This will help ensure the quality and consistency of migration data submitted. • It is recommended that Health Canada consider the authorization status of food contact substances in other countries. Relevant data (i.e. use restrictions, toxicity testing data, and migration limits) obtained from other internationally recognized government regulatory bodies on the food contact substance may facilitate the safety assessment of food packaging materials. • It is strongly recommended that Health Canada review the scientific literature for methods of microwave migration testing and select/adopt interim methods that may be appropriate to follow for food contact materials that are intended for microwave use. It is also recommended that Health Canada collaborate with international regulatory agency counterparts to develop a migration testing protocol for food contact materials intended for microwave use. 98 REFERENCES ACD/Labs, 2006. Advanced Chemistry Development Software V.8.14 for Solaris. Ahn, S, White, JL, 2003. Influence of low molecular weight additives on die extrusion rheometer flow of thermoplastic melts. International Polymer Processing 28(3):243-251. Anonymous, 1985. Final report on the safety assessment of butyl stearate, cetyl stearate, isocetyl stearate, isopropyl stearate, myristyl stearate, and octyl stearate. Journal of the American College of Toxicology 4(5): 107-46. Anonymous, 1985. Final report on the safety assessment of stearyl alcohol, oleyl alcohol, and octyl dodecanol. Journal of the American College of Toxicology 4(5): 1-29. Anonymous, 1987. Final report on the safety assessment of oleic acid, lauric acid, palmatic acid, myristic acid, and stearic acid. Journal of the American College of Toxicology 6(3):321-401. Araki, N, Ohno, K, Nakai, M, Takeyoshi, M, Iida, M, 2005. Screening for androgen receptor activities in 253 industrial chemicals by in vitro reporter gene assays using AR-EcoScreen cells. Toxicology In Vitro 19(6):831-42. Barlow, SM, 1994. The role of the Scientific Committee for Food in evaluating plastics for packaging. Food Additives and Contaminants 11(2): 249-259. Bieber, WD, Freytag, W, Figge, K, von Bruck, CG, Ross, L, 1984. Transfer of additives from plastic materials into foodstuffs and into food simulants - a comparison. Food and Chemical Toxicology 22(9):737-742. Bingham, E, Nord, PJ, 1977. Cocarcinogenic effects of n-alkanes and ultraviolet light on mice. Journal of the National Cancer Institute 58(4): 1099-101. Borg, K, 1975. Physiopathological effects of rapeseed oil: a review. Acta medica Scandinavica. Supplementum 585:5-13. Bourges, F, Bureau, G, Pascat, B, 1993. Effects of electron beam irradiation on the migration of antioxidants and their degradation products from commercial polypropylene into food simulating liquids. Food Additives and Contaminants 10(4):443-452. Bryan, GT, Springberg, PD, 1966. Role of vehicle in the genesis of bladder carcinomas in mice by pellet implantation technique. Cancer Research 26:105-109. Buttke, TM, Cuchens, MA, 1984. Inhibition of lymphocyte proliferation by free fatty acids. II. Toxicity of stearic acid towards phytohaemagglutinin-activated T cells. Immunology 53(3):507-14. Canadian Food and Drug Act, R.S. 1985, c. F-27, s. 1. 99 Carroll, KK, Noble, RL, 1957. Influence of a dietary supplement of erucic acid and other fatty acids on fertility in the rat; sterility caused by erucic acid. Canadian journal of biochemistry and physiology 35(11): 1093-105. Castle, L, Macarthur, R, Mead, EM, Read, WA, 2004. Measurement uncertainty associated with overall migration testing. Food Additives and Contaminants 21(3):256-264. Castle, L, Jickells, SM, Gilbert, J, Harrison, N, 1990. Migration testing of plastics and microwave-active materials for high temperature food-use applications. Food Additives and Contaminants 7(6):779-796. Castle, L, Mayo, A, Crews, C, Gilbert, J, 1989. Migration of polyethylene terephthalate) (PET) oligomers from PET plastics into foods during microwave and conventional cooking and into bottled beverages. Journal of Food Protection 52(5):337-342. CEN (European Committee for Standardization), 1994. ENV 1186 Part 7: Methods for overall migration into aqueous food simulants by article filling. CEN (European Committee for Standardization), 2000. European Standard EN 1186-3, Materials and articles in contact with foodstuffs - plastics - Part 3: Test methods for overall migration into aqueous simulants by total immersion. Colovai, AI, Herdan, J, Popovici, C, Mihai, C, Grigoras, C, Voiculetz, N, Nicolescu-Duvaz, I, 1993. In vitro inhibition of DNA adduct formation by hindered hydroquinols and quinones. Carcinogenesis 14(6): 1137-41. Connor, TH, Theiss, JC, Hanna, HA, Monteith, DK, Matney, TS, 1985. Genotoxicity of organic chemicals frequently found in the air of mobile homes. Toxicology Letters 25(1): 33-40. Cooper, I, Tice, PA, 1995. Migration studies on fatty acid amide slip additives from plastics into food simulants. Food Additives and Contaminants 12(2):235-244. Cooper, I, Goodson, A, O'Brien, A, 1998. Specific migration testing with alternative fatty food simulants. Food Additives and Contaminants 15(1): 72-78. Cosmetic Ingredient Review Expert Panel, 2002. Final Report on the Safety Assessment of BHT. International Journal of Toxicology 21(Suppl. 2): 19-94. Cremonezzi, DC, Diaz, MP, Valentich, MA, Eynard, AR, 2004. Neoplastic and preneoplastic lesions induced by melamine in rat urothelium are modulated by dietary polyunsaturated fatty acids. Food and Chemical Toxicology 42(12): 1999-2007. Deichmann, WB, Radomski, JL, MacDonald, WE, Kascht, RL, Erdmann, RL, 1958. The chronic toxicity of octadecylamine. AMA Archives of Industrial Health 18(6):483-7. EFSA (European Food Safety Authority), 2004a. About EFSA. Downloaded: August 20, 2004. Last Updated: February 19, 2004. http://www.efsa.eu.int/about_efsa/catindex_en.html 100 E F S A (European Food Safety Authority), 2004b. Members of the Panel on food additives, flavourings, processing aids and materials in contact with food [ A F C ] . Downloaded: August 20, 2004. Last Updated: February 20, 2004. http://www.efsa.eu.int/science/afc/afc_members/catindex_en.html Engfeldt, B , Brunius, E , 1975a. Morphological effects of rapeseed oi l in rats. I. Short-term studies. Acta medica Scandinavica. Supplementum 585:15-26. Engfeldt, B , Brunius, E , 1975b. Morphological effects of rapeseed o i l in rats. II. Long-term studies. Acta medica Scandinavica. Supplementum 585:27-40. Engfeldt, B , Gustafsson, B , 1975. Morphological effects of rapeseed o i l in rats. III. Studies in germ-free rats. Acta medica Scandinavica. Supplementum 585:41-6. Environment Canada, 2001. A Canadian Perspective on the Precautionary Approach/Principle Proposed Guiding Principles. Downloaded December 1, 2006. Updated November 23, 2001. http://www.ec.gc.ca/econom/booklet_e.htm Europa, 2006a. Farm Safety: From the Farm to the Fork - Scientific Committee on Food Background. Last Updated: 2003. Downloaded: August 20, 2004. http://europa.eu.int/comm/food/fs/sc/scf/background_en.html Europa, 2006b. Food and Feed Safety: Food contact materials - E U Legislation. Last updated: 2006. Downloaded: October 4, 2006. http://europa.eu.int/comm/food/food/chemicalsafety/foodcontact/eu_legisl_en.htm Europa, 2006c. Food Contact Materials - Introduction. Last updated: 2006. Downloaded: October 4, 2006. http://ec.europa.eu/food/food/chemicalsafety/foodcontact/index_en.htm E C (European Communities), 2003. Corrigendum to Commission Directive 2002/72/EC of 6 August 2002 relating to plastic materials and articles intended to come into contact with foodstuffs. Official Journal of the European Union V o l . 46 http://ec.europa.eu/food/food/chemicalsafety/foodcontact/2002-72_en.pdf E C (European Communities), 2005. Food Contact Materials: Substances listed in E U Directives on plastics in contact with food. Downloaded: June 10, 2006. http://ec.europa.eu/food/food/chemicalsafety/foodcontact/eu_substances_en.pdf E C J R C (European Commission Joint Research Centre), 2006a. Erucamide. Last updated: 2006. Downloaded: June 10, 2006. http://www.jrc.cec.eu.int E C J R C (European Commission Joint Research Centre), 2006b. Octadecyl 3-(3,5-di-tert-butyl-4-hydroxyphenyl)propionate. Last updated: 2006. Downloaded: June 10, 2006. http://www.jrc.cec.eu.int E C J R C (European Commission Joint Research Centre), 2006c. Stearic acid. Last updated: 2006. Downloaded: June 10, 2006. http://www.jrc.cec.eu.int 101 E C J R C (European Commission Joint Research Centre), 2006d. 1-Octadecanol. Last updated: 2006. Downloaded: June 10, 2006. http://www.jrc.cec.eu.int Feigenbaum, A E , Bouquant, J, Ducruet, V J , Ehret-Henry, J, Marque, D L , Riquet, A M , Scholler, D , Wittmann, JC, 1994. Guidelines of the Commission of the European Communities: a challenge for the control of food packaging. Food Additives and Contaminants 11(2): 141-154. Fermor, B F , Masters, JR, Wood, C B , Mi l le r , j , Apostolov, K , Habib, N A , 1992. Fatty acid composition of normal and malignant cells and cytotoxicity of stearic, oleic and sterculic acids in vitro. European Journal of Cancer 28A(6-7): 1143-7. Fischer, K ; Norman, V S ; Freitag, D , 1999. Studies of the behaviour and fate of the polymer-additives octadecyl-3-(3.5-di-t-butyl-4-hydroxyphenyl)propionate and tri-(2.4-di-t-butylphenyl)phosphite in the environment. Chemosphere 39(4):611-625. Fordham, PJ, Gramshaw, JW, Crews, H M , Castle, L , 1995. Element residues in food contact plastics and their migration into food simulants, measured by inductively-coupled plasma-mass spectrometry. Food Additives and Contaminants 12(5):651-669. Fulton (Jr), JE, Pay, SR, Fulton (III), JE, 1984. Comedogenicity of current therapeutic products, cosmetics, and ingredients in the rabbit ear. Journal of the American Academy of Dermatology 10(1):96-105. Galotto, M J , Guarda, A , 1999. Comparison between thermal and microwave treatment on the overall migration of plastic materials intended to be in contact with foods. Packaging Technology and Science 12:277-281. Galotto, M G , and Guarda, A , 2004. Suitability of alternative fatty food simulants to study the effect of thermal and microwave heating on overall migration of plastic packaging. Packaging Technology and Science 17: 219-223. George, C , Prest, H , 2001. A new approach to the analysis of phthalate esters by G C / M S . Agilent Technologies. Gilbert, J, Simoneau, C, Cote, D , and Boenke, A , 2000. A n internet compendium of analytical methods and spectroscopic information for monomers and additives used in food packaging plastics. Food Additives and Contaminants 17(10):889-893. Gramshaw, JW, Vandenburg, H J , Lakin, R A , 1995. Identification of potential migrants from samples of dual-ovenable plastics. Food Additives and Contaminants 12(2): 211-222. Grayson, M , Eckroth, D , (Ed) 1978. Kirk-Othmer Encyclopedia of Chemical Technology. 3 r d Edition. New York: Wiley. Grob, K , 2002. Comprehensive analysis of migrates from food packaging materials: a challenge. Food Additives and Contaminants 19(Supplement):185-191. Halse, K , Solheim, E , Nordstoga, K , 1993. Pathological hepatic accumulation of long-chain n-alkanes ("paraffin liver") in cows (Harbitz and Foil ing, 1940). A n overlooked discovery. 102 Description of lesions and identification of alkanes. APMIS: acta pathologica, microbiologica, et immunologica Scandinavica 101(6):430-6. Hamberger, A, Stenhagen, G, 2003. Erucamide as a modulator of water balance: new function of a fatty acid amide. Neurochemical Research 28(2): 177-185. Health and Consumer Protection Directorate-General of the European Commission (Unit D3 Chemical and physical risks; surveillance), 2004a. Questions and Answers on Food Contact Materials Legislation: A Quick Reference Booklet. Downloaded: August 20, 2004. Last Updated: June 3, 2004. http://europa.eu.int/comm/food/food/chemicalsafety/foodcontact/qanda_booklet_en.pdf Health and Consumer Protection Directorate-General of the European Commission (Unit D3 Chemical and physical risks; surveillance), 2004b. Food Contact Materials: Note for Guidance -Note for Guidance of Petitioner When Presenting an Application for Safety Assessment of a Substance to be Used in Food Contact Materials Prior to its Authorisation. Last Updated: January 6, 2004. Downloaded: August 20, 2004. http://europa.eu.int/comm/food/food/chemicalsafety/foodcontact/note_guidance_en.pdf Health and Consumer Protection Directorate-General of the European Commission (Unit D3 Chemical and physical risks; surveillance), 2003. Food Contact Materials: Practical Guide - A Practical Guide for Users of European Directives. Downloaded: August 20, 2004. Last Updated: April 15, 2003. http://europa.eu.int/comm/food/food/chemicalsafety/foodcontact/practical_guide_en.pdf Health Canada, 2002. Health Canada Decision-Making Framework for Identifying, Assessing, and Managing Health Risks - August 1, 2000. Last Updated: December 13, 2002. Downloaded: August 19,2004. http://www.hc-sc.gc.ca/hpfb-dgpsa/hcrisk_ll_e.html. Health Canada, 2003. Information requirements for food packaging submissions. Downloaded: February 23, 2004. Last Updated: July 11,2003. http://www.hc-sc.gc.ca/food-aliment/cs-ipc/chha-edpcs/e_guide_packaging01.html. Health Canada, 2003a. Canadian Guidelines for Body Weight Classification in Adults. Ottawa: Minister of Public Works and Government Services Canada. Downloaded: June 10, 2006. Last Updated: September 19, 2003. http://www.hc-sc.gc.ca/fn-an/nutrition/weights-poids/guide-ld-adult/bmi_chart Java-graph_imc_java_e.html Health Canada, 2004. Food packaging materials. Downloaded: August 15, 2004. Last Updated: February 24, 2004. http://www.hc-sc.gc.ca/food-aliment/cs-ipc/chha-edpcs/3_food_packaging.html Herting, DC, Harris, PL, Crain, RC, 1959. Lipogranuloma from dietary saturated fats: production and reversal. Toxicology and Applied Pharmacology 1:505-14. Hirata-Koizumi, M , Hamamura, M , Furukawa, H, Fukuda, N, Ito, Y, Wako, Y, Yamashita, K, Takahashi, M , Kamata, E , Ema, M , Hasegawa, R, 2005. Elevated susceptibility of newborn as compared with young rats to 2-tert-butylphenol and 2,4-di-tert-butylphenol toxicity. Congenital Anomalies 45(4): 146-53. 103 Hoak, JC, 1994. Stearic acid, clotting, and thrombosis. American Journal of Clinical Nutrition 60(6 Suppl):1050S-1053S. Horton, A W , Eshleman, D N , Schuff, A R , Perman, W H , 1976. Correlation of cocarcinogenic activity among n-alkanes with their physical effects on phospholipid micelles. Journal of the National Cancer Institute 56(2):387-91. Incarnato, L , Scarfato, P, Milana, M R , Feliciani, R, Acierno, D , 2000. Migrational behaviour of food grade PP (polypropylene). Journal of Polymer Engineering 20(2): 121-141. Insightful Corp., 2005. S-Plus Version 7.0 for Microsoft Windows. Inthorn, D , Lertsupochavanich, O, Silapanuntakui, S, Sujirarat, D , Intaraprasong, B , 2002. Factors affecting the lead leaching from microwavable plastic ware made with lead-containing pigments. Journal of Food Protection 65(7): 1166-1171. Jenke, D R , Jene, J M , Poss, M , Story, J, Tsilipetros, T, Odufu, A , Terbush, W , 2005. Accumulation of extractables in buffer solutions from a polyolefin plastic container. International Journal of Pharmacy. 297(l-2):120-33. Jeppsson, R, 1975. Parabolic relation between lipophilicity and biological activity of aliphatic hydrocarbons, ethers, and ketones after intravenous injections of emulsion formulations into mice. Acta Pharmacologica et Toxicologica 37(1): 56-64. Jetten, J., de Kruijf, N . , 2002. Quality and safety aspects of reusable plastic food packaging materials: influence of reuse on intrinsic properties. Food Additives and Contaminants 19(1):76-88. Jickells, S M , Gramshaw, JW, Castle, L , Gilbert, J, 1992. The effect of microwave energy on specific migration from food contact plastics. Food Additives and Contaminants 9(1): 19-27. John Wiley and Sons Inc., 2000. Wiley Registry of Mass Spectral Data, Seventh Edition Database, with NIST Spectral Data. Kawamura, Y , Miura, M , Sugita, T, Yamada, T, 1997a. Residue and release of antioxidants and ultraviolet stabilizers in polyethylene products in contact with foodstuffs. Journal of the Food Hygienic Society of Japan 38(l):27-33. Kawamura, Y , Watanabe, K , Sayama, K , Takeda, Y , Yamada, T, 1997b. Simultaneous determination of polymer additives in polyethylene by G C / M S . Journal of the Food Hygienic Society of Japan 38(5):307-318. Kawamura, Y , Yonezawa, R, Maehara, T, Yamada, T, 2000. Determination of food additives in food contact polypropylene. Journal of the Food Hygienic Society of Japan 41(2): 154-161. Kawano, S, Nakao, T, Hiraga, K , 1981. Induction of hepatic microsomal monooxygenases in female rats given various substituted phenols and hydroquinones. Japan Journal of Pharmacology 31(3):459-62. 104 Khalil, MH, Marcelletti, JF, Katz, LR, Katz, DH, Pope, LE, 2000. Topical application of docosanol- or stearic acid-containing creams reduces severity of phenol burn wounds in mice. Contact Dermatitis 43(2):79-81. Kim-Kang, H, 1990. Volatiles in packaging materials. Critical Reviews in Food Science and Nutrition 29(4):255-271.' Kissin, YV, 2005. Polymers of higher olefins. In: Kirk-Othmer Encyclopedia of Chemical Technology. 5th Edition. New York: John Wiley & Sons, Inc. http://www.mrw.interscience.wiley.com/kirk/articles/polykiss.a01/sect2.html Kolattukudy, PE, Hankin, L, 1966. Metabolism of a plant wax paraffin (n-nonacosane) in rat. Journal of Nutrition 90(2):l67-74. Kris-Etherton, PM, Mustad, VA, 1994. Chocolate feeding studies: a novel approach for evaluating the plasma lipid effects of stearic acid. American Journal Clinical Nutrition 60(6 Suppl): 1029S-1036S. Kroschwitz, Jl, and Howe-Grant, M, (ed.) 2005. Kirk-Othmer Encyclopedia of Chemical Technology. 5lh Edition. New York: John Wiley & Sons, Inc. Kubow, S, 1996. The influence of positional distribution of fatty acids in native, interesterified and structure-specific lipids on lipoprotein metabolism and atherogenesis. The Journal of Nutritional Biochemistry 7(10):530-541. Kusakabe, H, Yamakage, K, Wakuri, S, Sasaki, K, Nakagawa, Y, Watanabe, M, Hayashi, M, Sofuni, T, Ono, H, Tanaka, N, 2002. Relevance of chemical structure and cytotoxicity to the induction of chromosome aberrations based on the testing results of 98 high production volume industrial chemicals. Mutation Research 517(1-2): 187-198. Lake, BG, Gangolli, SD, Schmid, K, Schweizer, W, Staubli, W, Waechter, F, 1980. The induction of rat hepatic microsomal xenobiotic metabolism by n-octadecyl beta-(3',5'-di-tert-butyl-4'-hydroxyphenyl)-propionate. Food and Cosmetics Toxicology 18(l):47-54. Lau, OW, Wong, SK, 2000. Contamination in food from packaging material. Journal of Chromatography 822:255-270. Lemercier, JN, Meier, BW, Gomez, JD, Thompson, JA, 2004. Inhibition of glutathione S-transferase Pl-1 in mouse lung epithelial cells by the tumor promoter 2,6-di-tert-butyl-4-methylene-2,5-cyclohexadienone (BHT-quinone methide): protein adducts investigated by electrospray mass spectrometry. Chemical research in toxicology 7(12): 1675-83. Lewis, MA, Yoerg, DG, Bolton, JL, Thompson, JA, 1996. Alkylation of 2'-deoxynucleosides and DNA by quinone methides derived from 2,6-di-tert-butyl-4-methylphenol. Chemical Research in Toxicology 9(8): 1368-1374. Lewis, RJ, 2002. Hawley's Condensed Chemical Dictionary. 14th Edition. New York: John Wiley & Sons, Inc. 105 Lichtenthaler, RG, Ranfelt, F, 1978. Determination of antioxidants and their transformation products in polyethylene by high-performance liquid chromatography. Journal of Chromatography 149:553-560. Lorenc, JF, Lambeth, G, Scheffer, W, 2003. Alkylphenols. In: Kirk-Othmer Encyclopedia of Chemical Technology. 5th Edition. New York: John Wiley & Sons, Inc. http://www.mrw.interscience.wiley.com/emrw/9780471238966/kirk/article/alkylore.a01/current/ html#alkylore.a01-sec 1-0009 Marque, D, Feigenbaum, A, Dainelli, D, Riquet, A, 1998. Safety evaluation in an ionized multilayer plastic film used for vacuum cooking and meat preservation. Food Additives and Contaminants 15(7):831-841. MDL Information Systems, 2006. Registry of Toxic Effects of Chemical Substances. Canadian Centre for Occupational Health and Safety, http://ccinfoweb2.ccohs.ca/rtecs/ Meier, BW, Gomez, JD, Zhou, A, Thompson, JA, 2005. Immunochemical and proteomic analysis of covalent adducts formed by quinone methide tumor promoters in mouse lung epithelial cell lines. Chemical Research in Toxicology 18(10): 1575-1585. Mensink, RP, 2005. Effects of stearic acid on plasma lipid and lipoproteins in humans. Lipids 40(12): 1201-5. Microsoft Corp., 2001. Microsoft Excel Version 10.2614.2625. Ministry of National Health and Welfare, 1995. Investigating Human Exposure to Environmental Contaminants: A Handbook for Exposure Calculations. Downloaded: June 11, 2006. http://dsp-psd.pwgsc.gc.ca/Collection/H49-96-l-1995E-l.pdf, http://dsp-psd.pwgsc.gc.ca/Collection/H49-96-l-1995E-2.pdf, http://dsp-psd.pwgsc.gc.ca/Collection/H49-96-l-1995E-3.pdf. Mitchell, CA, Davies, MJ, Grounds, MD, McGeachie, JK, Crawford, GJ, Hong, Y, Chirila, TV, 1996. Enhancement of neurovascularization in regenerating skeletal muscle by the sustained release of erucamide from a polymer matrix. Journal of Biomaterials Applications 10(3): 230-247. Mitra A, Govindwar S, Joseph P, Kulkarni A, 1992. Inhibition of human term placental and fetal liver glutathione-S-transferases by fatty acids and fatty acid esters. Toxicology Letters 60(3):281-8. Moridani, MY, Siraki, A, O'Brien, PJ, 2003. Quantitative structure toxicity relationships for phenols in isolated rat hepatocytes. Chemico-biological interactions 145(2):213-23. Mortimer, BC, Simmonds, WJ, Cockman, SJ, Stick, RV, Redgrave, TG, 1990. The effect of monostearoylglycerol on the metabolism of chylomicron-like lipid emulsions injected intravenously in rats. Biochimica et Biophysica Acta 1046(l):46-56. 106 Muhammad, F, Monteiro-Riviere, NA, Riviere, JE, 2005. Comparative in vivo toxicity of topical JP-8 jet fuel and its individual hydrocarbon components: Identification of tridecane and tetradecane as key constituents responsible for dermal irritation. Toxicologic Pathology 33(2): 258-266. Nagai, F, Ushiyama, K, Kano, I, (1993). DNA cleavage by metabolites of butylated hydroxytoluene. Archives of Toxicology 67(8):552-7. Nerin, C, Acosta, D, Rubio, C, 2002. Potential migration release of volatile compounds for plastic containers destined for food use in microwave ovens. Food Additives and Contaminants 19(6):594-601. Nerin, C, Fernandez, C, Domeno, C, Salfranca, J, 2003. Determination of potential migrants in polycarbonate containers used for microwave ovens by high-performance liquid chromatography with ultraviolet and fluorescence detection. Journal of Agricultural and Food Chemistry 51(19):5647-5653. Niculescu-Duvaz, I, Herdan, J, Stoica, G, Lupu F; Arnautu, M ; Cinca, S, Eliescu, R, Voiculetz, N, 1991. A new class of potential carcinogenesis inhibitors: hindered p-benzoquinones. Neoplasma 38(6):545-57. O'Brien, A, Cooper, I, 2001. Polymer additive migration into foods - a direct comparison of experimental data and values calculated from migration models for polypropylene. Food Additives and Contaminants 18(4): 343-355. Ogawa,Y, Kawamura, Y, Wakui C, Mutsuga M , Nishimura, T, Tanamoto, K, 2006. Estrogenic activities of chemicals related to food contact plastics and rubbers tested by the yeast two-hybrid assay. Food Additives and Contaminants 23(4):422-30. Oikawa, S, Nishino, K, Oikawa, S, Inoue, S, Mizutani, T, Kawanishi, S, 1998. Oxidative DNA damage and apoptosis induced by metabolites of butylated hydroxytoluene. Biochemical Pharmacology 56(3):361-70. Opdyke, DL, Letizia, C, 1982. Monographs on fragrance raw materials. Food and Chemical Toxicology Suppl:633-852. Otton, R, Curi, R, 2005. Toxicity of a mixture of fatty acids on human blood lymphocytes and leukaemia cell lines. Toxicology In Vitro 19(6):749-55. Perdu-Durand, EF, Tulliez, JE, 1985. Hydrocarbon hydroxylation system in liver microsomes from four animal species. Food and Chemical Toxicology 23(3):363-6. Rantuccio F, Sinisi D, Coviello C, Conte A, Scardigno A, 1984. Histological changes in rabbits after application of medicaments and cosmetic bases (III). Contact Dermatitis 10(4):212-9. Reed, M , Thompson, DC, 1997. Immunochemical Visualization and Identification of Rat Liver Proteins Adducted by 2,6-Di-tert-butyl-4-methylphenol (BHT). Chemical Research in Toxicology 10(10): 1109-1117. 107 Reichmuth, A , Wunder l i , S, Weber , M , M e y e r , V R , 2004. T h e uncertainty of weighing data obtained with electronic analytical balances M i c r o c h i m i c a A c t a 148 (3-4): 133-141. Ri jk , R , D e Kruij f , N , 1993. Migrat ion testing with olive oil in a microwave oven. F o o d Addit ives and Contaminants 10(6):631-645. Rocch icc io l i , F , Lageron, A , Duboucher , C , 1987. A b n o r m a l n-nonacosane storage in humans: detection by gas chromatography/mass spectrometry of tissue extracts. B iomed ica l and Environmental M a s s Spectrometry 14(9), 481-5. Safirman, C , Stoica, G , L u p u , F , Voicule tz , N , N icu lescu-Duvaz , I, 1987. In vitro screening of carcinogenesis inhibitors acting by inhibition of microsomal po lycyc l ic aromatic hydrocarbon activation. Neoplasma 34(3):261-8. Salafranca, J , Cacho , J , Ner in , C , 1999. Composi t ional analysis and feasibility study of recycled high density polyethylene for direct food contact applications. Polymer R e c y c l i n g 4(1): 13-25. Salvayre, R , Negre, A , Rocch icc io l i , F , Duboucher , C , Maret , A , V i e u , C , Lageron, A , Polonovski , J , Douste-Blazy , L , 1988. A new human pathology with visceral accumulation of long-chain n-alkanes; tissue distribution of the stored compounds and pathophysiological hypotheses. B i o c h i m i c a et B iophys ica Acta . 958(3):477-83. Sanders, T A , Berry , S E , 2005. Influence of stearic acid on postprandial l ipemia and hemostatic function. L i p i d s 40(12): 1221-7. S C F (Scientific Committee for Food) , 1995. Reports o f the Scientific Committee for F o o d (Thirty-fourth series). Luxembourg: European C o m m i s s i o n . Schafer (Jr), E W , Bowles (Jr), W A , 1985. Acute oral toxicity and repellency of 933 chemicals to house and deer mice. Arch ives of environmental contamination and toxicology 14(1): 111-29. Schoene, N W , A l l m a n , M A , Doughtery, R M , Iacono, J M , 1994; Diss imi lar responses of platelets to dietary stearic and palmitic acids. A m e r i c a n Journal o f C l i n i c a l Nutrit ion 60(6 Suppl): 1059S. Shields, M , 2005. Measured obesity: Overweight Canadian children and adolescents. Nutrit ion Findings from the Canadian C o m m u n i t y Health Survey, Issue N o . 1. Catalogue no. 82-620-M W E 2 0 0 5 0 0 1 . Ottawa: Statistics Canada. Singletary, K W , Nelshoppen, J M , Scardefield, S, W a l l i g , M , 1992. Inhibition by butylated hydroxytoluene and its oxidative metabolites of D M B A - i n d u c e d mammary tumorigenesis and of mammary D M B A - D N A adduct formation in v ivo in the female rat. F o o d and C h e m i c a l T o x i c o l o g y 30(6):455-65. Siraki , A G , C h a n , T S , O'Brien , P J , 2004. Appl icat ion of quantitative structure-toxicity relationships for the comparison of the cytotoxicity of 14 p-benzoquinone congeners in primary cultured rat hepatocytes versus P C 1 2 cells. Tox ico log ica l Sciences 81(l):148-59. 108 Skjevrak, I, Brede, C , Steffensen, IL, Mikalsen, A , Alexander, J, Fjeldal, P, Herikstad, H , 2005. Non-targeted multi-component analytical surveillance of plastic food contact materials: Identification of substances not included in E U positive lists and their risk assessment. Food Additives and Contaminants 22(10): 1012-22. Skoog, D A , West, D M , Holler, FJ , 1990. Analytical Chemistry: A n Introduction. 5 t h Edition. Philadelphia: Saunders College Publishing. Smyth (Jr), H F , Carpenter, C P , Wei l , CS,Pozzani, U C , Striegel, J A , Nycum, JS, 1969. Range-finding toxicity data: List VII . American Industrial Hygiene Association journal 30(5):470-6. Soto-Valdez, H . , Gramshaw, JW, Vandenburg, H J , 1997. Determination of potential migrants present in Nylon 'microwave and roasting bags' and migration into olive oi l . Food Additives and Contaminants 14(3):309-318. S R C , 2004. The Physical Properties Database ( P H Y S P R O P ) . Syracuse Research Corporation. Stringer, R, Labunska, I, Santillo, D , Johnston, P, Siddorn, J, and Stephenson, A , 2000. Concentration of phthalate esters and identification of other additives in P V C children's toys.Environmental Science and Pollution Research 7(1): 27-36. Swern, D , Wieder, R, McDonough, M , Meranze, D R , Shimkin, M B , 1970. Investigation of fatty acids and derivatives for carcinogenic activity. Cancer Research 30(4): 1037-46. Takahashi, O, 1988. 2,6-di-tert-butyl-4-methylene-2,5-cyclohexadienone (BHT quinone methide): an active metabolite of B H T causing haemorrhages in rats. Archives of Toxicology 62(4): 325-7. ter Veld , M G , Schouten, B , Louisse, J, van Es, D S , van der Saag, PT, Rietjens, TM, Murk, A J , 2006. Estrogenic potency of food-packaging-associated plasticizers and antioxidants as detected in ERalpha and ERbeta reporter gene cell lines. Journal of Agricultural and Food Chemistry 54(12):4407-16. Thijssen, M A , Hornstra, G , Mensink, R P , 2005. Stearic, oleic, and linoleic acids have comparable effects on markers of thrombotic tendency in healthy human subjects. Journal of Nutrition 135(12):2805-11. Tholstrup, T, 2005. Influence of stearic acid on hemostatic risk factors in humans. Lipids 40(12): 1229-35. Tjepkema, M , 2005. Measured Obesity: Adult obesity in Canada. Measured height and weight. Nutrition Findings from the Canadian Community Health Survey, Issue No. 1. Catalogue no. 82-620-MWE2005001. Ottawa: Statistics Canada. Tochacek, J, Sedlaf, J, 1995. Hydrolysis and stabilization performance of bis(2,4-di-t-butylphneyl(pentaerythrityl diphosphite in polypropylene. Polymer Degradation and Stability 50:345-352. 109 Tornier, C, Rosdy, M , Maibach, HI, 2006. In vitro skin irritation testing on reconstituted human epidermis: reproducibility for 50 chemicals tested with two protocols. Toxicology In Vitro 20(4):401-16. Tulliez, JE, Bories, GF, 1978. Metabolism of a n-paraffin, heptadecane, in rats. Lipids. 13(2): 110-5. Ullmann, F, 2006. Ullmann's Encyclopedia of industrial chemistry. 7 t h Edition. New York: Wiley-VCH Verlag GmbH& Co. KGaA. US Federal Food, Drug and Cosmetic Act, Title 21, Volume 3, CFR (Code of Federal Regulations) §170.3(e)(3), §170.39(1-4), §172.860, §172.864, §173.340, §175.300, §176.180, §176.200, §176.210, §177.1200, §177.1390, §178.3480, §184.1555 (2003). US FDA/CFSAN (US Food and Drug Administration/Center for Food Safety and Applied Nutrition), 2002a. Guidance for Industry: Preparation of Food Contact Notifications and Food Additive Petitions for Food Contact Substances - Chemistry Recommendations Final Guidance. Downloaded: June 9, 2004. Last updated: April, 2002. http://www.cfsan.fda.gov/~dms/opa2pmnc.html US FDA/CFSAN, 2002b (US Food and Drug Administration/Center for Food Safety and Applied Nutrition). Guidance for Industry: Preparation of Food Contact Notifications for Food Contact Substances - Toxicology Recommendations Final Guidance. Downloaded: August 15, 2004. Last Updated: April, 2002. http://www.cfsan.fda.gov/~dms/opa2pmnt.html US FDA/CFSAN, 2003 (US Food and Drug Administration/Center for Food Safety and Applied Nutrition). Guidance for Industry: Preparing a Claim of Categorical Exclusion or an Environmental Assessment for Submission to the Center for Food Safety and Applied Nutrition DRAFT GUIDANCE. Downloaded: August 15, 2004. Last Updated: September 2003. http://www.cfsan.fda.gov/~dms/opa-eg.html US FDA/CFSAN, 2004a (US Food and Drug Administration/Center for Food Safety and Applied Nutrition). Food Contact Substance Notification Program's Cumulative Estimated Daily Intake/Acceptable Daily Intake Database. Downloaded: August 20, 2004. Last Updated: February 11,2004. http://www.cfsan.fda.gov/~dms/opa-edi.html US FDA/CFSAN, 2004b (US Food and Drug Administration/Center for Food Safety and Applied Nutrition). The List of "Indirect" Additives Used in Food Contact Substances. Last Updated: August 26, 2004. Downloaded: August 28, 2004. http://www.cfsan.fda.gov/~dms/opa-indt.html US NRC (United States Nuclear Regulatory Commission), 2003. Microcurie. Updated June 23, 2003. Downloaded November 7, 2006. http://www.nrc.gov/reading-rrn/basic-ref/glossary/microcurie.html Van Duuren, BL, Katz, C, Shimkin, MB, Swern, D, Wieder, R, 1972. Replication of low-level carcinogenic activity bioassays. Cancer Research 32:880-881. 110 van Leirop, JBH, 1997. Enforcement of food packaging legislation. Food Additives and Contaminants 14(6-7):555-560. van Leirop, B, Castle, L, Feigenbaum, A, Boenke, A, 1998. Spectra for the Identification of Additives in Food Packaging. Dodrecht: Kluwer Academic Publishers. Wakabayashi, T, Adachi, K, Popinigis, J, 1991. Effects of alkyl alcohols and related chemicals on rat liver structure and function. I. Induction of two distinct types of megamitochondria. Acta Pathologica Japonica 41(6):405-13. Wakamatsu, K, Masaki, T, ltoh, F, Kondo, K, Sudo, K, 1990. Isolation of fatty acid amide as an angiogenic principle from bovine mesentery. Biochemical and Biophysical Research Communications 168(2):423-9. Wang, ZJ, Li, GM, Tang, WL, Yin, M, 2006. Neuroprotective effects of stearic acid against toxicity of oxygen/glucose deprivation or glutamate on rat cortical or hippocampal slices. Acta Pharmacologica Sinica 27(2): 145-50. Wei, Y, Wang, D, Topczewski, F, Pagliassotti, MJ, 2006. Saturated fatty acids induce endoplasmic reticulum stress and apoptosis independently of ceramide in liver cells. American Journal of Physiology Endocrinology and Metabolism 291(2):E275-81. Yamamoto, K, Tajima, K, Misutani, T, 1980. The acute toxicity of butylated hydroxytoluene and its metabolites in mice. Toxicology Letters 6(3): 173-176. Yoon, MA, Jeong, TS, Park, DS, Xu, MZ, Oh, HW, Song, KB, Lee, WS, Park, HY, 2006. Antioxidant effects of quinoline alkaloids and 2,4-di-tert-butylphenol isolated from Scolopendra subspinipes. Biological and Pharmaceutical Bulletin 29(4):735-9. I l l A P P E N D I X A PRELIMINARY EXPERIMENTAL RESULTS FOR THE DEVELOPMENT OF G C / M S ANALYSIS METHODOLOGY FOR CHARACTERIZING CHEMICAL MIGRATION FROM MICROWAVE HEATED PLASTIC A . l Introduction A method for testing chemical migration from plastic food containers during microwave heating was developed using the acetic acid food simulant and modified for the isopropanol food simulant. The following is a list of experiments that were conducted to select experimental parameters, verify that different parts of the method were functioning and the equipment used was working properly: • Test to determine the microwave heating time and power level for the microwave heating experiment for both food simulants; • Test to determine the cooling time for the microwave tray between microwave heating cycles; • Test to determine the cooling time for food simulant between microwave heating cycles; • Test to determine toaster oven heating stability and target temperature setting; • Test to determine the internal surface area of the polypropylene container; • Test to verify retention time and mass spectra of surrogate standards; • Test to determine the instrument detection limits for surrogate standards and identified chemical migrants; • To compare the surrogate standard recovery between two extraction solvents; • Test to determine surrogate standard and chemical migrant recovery for migration test methods; • Test to determine the change in migrant recovery over time. 112 A.2 Microwave power level determination A.l.I Objective The purpose of this experiment was to determine the optimal microwave power level for heating 200 mL of food simulant based on evaporative loss and heating temperature. The optimal criteria is the microwave power level that heats the food simulant to the highest temperature with the least amount of evaporative loss. A.2.2 Protocol Square short term use polypropylene containers (591 mL) were filled with 200 mL of 3% (w/v) acetic acid or isopropanol, covered with a watchglass and systematically heated in the microwave. Acetic acid simulant was subjected to each combination of 3-minute microwave heating durations and 4 through 10 power settings in power level increments of 2. A 3-minute heating duration was selected because it was assumed that the average person will heat their food for 3 minutes. Isopropanol simulant was systematically subjected to 2, 2.5 and 3-minute microwave heating durations at power setting 4. A lower power level was selected because isopropanol has a lower boiling point and is flammable. Thus, a shorter heating time and lower heating temperature was considered safer. After each heating the temperature and volume of the food simulant was measured and recorded. The power level that resulted in the minimum amount of food simulant loss (approximately less than 20% of the total volume) and reached the highest temperature was selected for all subsequent microwave heatings. A.2.3 Results A power level of 6 was selected for the acetic acid simulant because it heated the acetic acid solution to the highest temperature (91±1.2°C) but resulted in evaporative loss of 14±0 mL. A heating time of 2.5 minutes and power level 4 was selected for isopropanol because a near boiling point temperature was obtained (80±0.9°C) and the average amount of evaporative loss of 24±1.5 mL. Both microwave power levels and heating times were incorporated in the microwave migration experiments. 113 A.3 Microwave tray cooling time determination A.3.1 Objective The purpose of these experiments was to determine how long it takes for the microwave tray to cool to approximately room temperature after 200 mL of acetic acid solution has been heated for 3 minutes at power level 6. Consecutive microwave heatings without cooling may result in heating food simulants at increasing and/or inconsistent baseline temperatures, which could affect the amount of evaporative loss. Thus, it is important to cool the microwave tray to approximately room temperature between microwave heatings. A.3.2 Protocol This experiment consisted of measuring and recording the temperature of the microwave tray after 200 mL of acetic acid solution was heated and placed the tray in a freezer for a set time period. The temperature of the microwave tray was first measured by holding the thermometer bulb against the center of the tray, then reading and recording the temperature after 1 minute. 200 mL of acetic acid solution was poured into a short-term use polypropylene plastic container covered with a watchglass and heated in the microwave at the predetermined power level and duration. The plastic container was removed and the microwave tray was placed in the freezer (maintained at -25°C) for 3 minutes. The microwave tray was then removed and the temperature of the centre of the tray was measured and recorded in 5-minute increments for a maximum of 20 minutes. The entire process of measuring the temperature of the microwave tray centre after heating 200 mL of acetic acid solution and cooling the microwave tray in the freezer was repeated with the same plastic container after it was rinsed with cool tap water. However, fresh acetic acid solution was used for each new heating. The entire experiment was repeated twice. The experiment was then repeated using isopropanol food simulant (which was heated for 2.5 minutes at power level 4). A.3.3 Results The average temperature of the microwave tray before heating 200 mL of acetic acid solution was 21°C. After 3 minutes in the freezer and 21 minutes at room temperature, the centre of the microwave tray cooled to the initial temperature. Thus, 3 minutes was determined as freezing time for the microwave tray during the acetic acid simulant migration experiment. The average tray temperature after microwave heating 200 mL of isopropanol and freezing the tray for 2 minutes was 19±0.3°C. Thus, 2 minutes in the freezer was sufficient to cool the 114 microwave tray during the isopropanol simulant migration experiment. The microwave tray cooling times were incorporated in the migration experiments for both food simulants. A.4 Determination of cooling time for food simulants between microwave heating cycles A.4.1 Objective The purpose of this experiment was to determine the time it takes to cool 200 mL of acetic acid solution and isopropanol to 40°C in the fumehood after it has been heated in the microwave at a predetermined power level for a set time period. 40°C was selected because it was assumed to be a suitable approximation of the temperature at which people consume their soup. A lower target temperature would also require a longer cooling period which would add to the total amount of time required to complete each experiment. A.4.2 Protocol This experiment consisted of measuring and recording the temperature of the acetic acid solution while it is being cooled in the fumehood after it was heated in the microwave. 200 mL of acetic acid solution was poured into a short-term use polypropylene plastic container covered with a watchglass. The temperature of the acetic acid solution was measured with a thermometer and recorded after allowing the thermometer to acclimatize in the centre of the container for 1 minute. The acetic acid solution was then heated in the microwave at power level 6 for 3 minutes. The container with the acetic acid solution was then taken out of the microwave and left in the fumehood. The time at which the temperature of the acetic acid solution reached 40°C was measured and recorded. This experiment was repeated twice with fresh acetic acid solution using the same container and watchglass after the inside of the microwave, microwave tray and plastic container returned to room temperature. The second part of this experiment consisted of following the same method described previously; however, 200 mL of ispropanol was heated for 2.5 minutes at power level 4 in the microwave. The time at which the temperature of isopropanol reached 40°C was measured and recorded. This experiment was repeated twice with fresh isopropanol using the same container and watchglass after the inside of the microwave, microwave tray, and plastic container returned to room temperature. 115 A.4.3 Results The average amount of time required to cool 200 mL of acetic acid solution to 40°C after being microwave heated was 17 minutes and 56 seconds. The cooling time ranged from 17 minutes and 2 seconds to 18 minutes. The average amount of time required to cool 200 mL of isopropanol to 40°C after being microwave heated was 9 minutes and 34 seconds. The cooling time ranged from 8 minutes and 50 seconds to 10 minutes and 24 seconds. The food simulant cooling times were incorporated in the migration experiments for both food simulants. A .5 Determination of heat and volume loss during transfer of heated food simulant from a glass crystallizing dish to a plastic container A. 5.1 Purpose The purpose of this experiment was to determine the amount of food simulant evaporative and heat loss after 200 mL of food simulant were heated in a glass crystallizing dish (125 x 65 mm VWR Cat. No. 89000-292) on a hot plate to a target temperature and transferred to a square plastic polypropylene container (591 mL). If the temperature difference and/or amount of evaporative loss is too great, it can be determined if the amount of food simulant heated or target temperature should be adjusted. An evaporative loss greater than 25 mL is considered unacceptable whereas a temperature difference greater than 15°C is considered too great. A.5.2 Protocol Distilled water was tested as the aqueous simulant in this experiment because it was assumed that it would boil at the same rate as the acetic acid solution. A target temperature of 100°C was selected for water because in a preliminary experiment, water that was heated to 90°C and transferred to a plastic container had a temperature decrease that ranged from 7 to 10°C. It was assumed that the temperature decrease would be approximately the same if water was heated to 100°C; thus, the final temperature of water after it was heated and transferred to a plastic container would be ~90°C which is the temperature required for the oven heating control experiment. Isopropanol was the fatty food simulant tested in the second part of this experiment. A target temperature of 82°C was selected because the boiling point of isopropanol is 82.4°C and the temperature required for the oven heating control experiment is 80°C. Although a higher 116 heating temperature would result in a final fatty food simulant temperature that was closer to 80°C, it is unsafe to heat isopropanol above its boiling point. The food simulant temperature was measured with a bulb thermometer throughout this experiment. The first part of this experiment consisted of heating 200 mL of distilled water to 100°C in a glass crystallizing dish covered with a watchglass on a hot plate (Corning Glass Works Model PC-351) adjusted to heat setting 5. The water was then transferred to a polypropylene container and covered with a watchglass. The maximum water temperature was measured and recorded. After the water had cooled to room temperature, the water volume was measured with a graduated cylinder and recorded. This experiment was repeated twice with fresh distilled water. The second part of this experiment consisted of heating 200 mL of isopropanol to 82°C in a glass crystallizing dish covered with a watchglass on a hotplate adjusted to heat setting 4. A heating temperature of 82°C was selected because it was just below the boiling point of isopropanol and therefore would not pose a fire hazard or boil. The isopropanol was then transferred to a polypropylene container and covered with a watchglass. The maximum temperature of isopropanol was measured and recorded. After the isopropanol had cooled to room temperature, the volume of isopropanol was measured with a graduated cylinder and recorded. This experiment was repeated twice with fresh isopropanol. If the final average amount of evaporative loss and temperature decrease of water exceeded the limits stated previously, the target heating temperature and the amount of water initially heated was increased by a maximum of 5°C and mL, respectively. If the average amount of evaporative loss and temperature decrease of isopropanol exceeded the limits stated previously, the amount of water initially heated was increased by a maximum of 25 mL. The heating temperature of isopropanol was not adjusted for safety concerns. A.5.3 Results The average temperature loss for distilled water was 11±1.3 °C whereas the average evaporative loss was 18±0 mL. The average temperature loss for isopropanol was 6 ±0.5°C whereas the average evaporative loss was 19±1.2 mL. The target food simulant heating temperatures were incorporated in the oven heating and oven heating glass control migration experiments for both food simulants. 117 A.6 Determination of toaster oven heating stability and target temperature setting A.6.1 Purpose The purpose of this experiment was to determine the temperature stability of the toaster oven (Betty Crocker Model BC-1660-C, Manufactured by E.F. Appliances Canada Ltd.) for a target heating temperature specific to a food simulant. A toaster oven was selected for the final oven heating experiment because the drying oven used in previous experiments did not heat consistently. The toaster oven is also commercially available and physically smaller than a drying oven therefore it was able to regulate small changes in temperature more quickly. A. 6.2 Protocol For this experiment, three thermometers were used to measure the internal temperature to capture the spatial variability inside the toaster oven: a mechanical thermometer (Accu-Temp 40 - 310°C) and two bulb thermometers. The mechanical thermometer was placed approximately in the centre of the toaster oven rack whereas the red bulb thermometer was placed lengthwise across the centre of the rack with the bulb facing the left corner. The blue bulb thermometer was placed in the same orientation as the red bulb thermometer; however, the blue bulb thermometer was placed closer to the oven door. Both bulb thermometers did not have contact with the oven wall. The first part of the experiment consisted of establishing the temperature setting required for 200 mL of acetic acid and isopropanol simulant to heat to approximately the same maximum temperature as 200 mL of microwave heated acetic acid of isopropanol food simulant. The average maximum temperature of acetic acid solution after being microwave heated for 3 minutes at power level 6 was 91°C. Thus, the target temperature range for the toaster oven was between 91°C and 100°C because it was above the maximum temperature of microwave heated acetic acid solution but below the boiling point of water. A temperature range greater than 91°C was also selected because it was assumed that the actual microwave heated the acetic acid solution approaches temperatures close to or exceeds its boiling point inside the microwave (i.e. the temperature of the heated food simulant measured outside the microwave is an underestimate of the temperature inside the microwave). The temperature setting was selected by initially choosing a temperature setting and allowing the toaster oven to stabilize for 1 hour. If the temperature remained the same after an additional hour, the measured temperature for the selected temperature setting was established. The temperature setting was then adjusted to be 118 higher or lower depending on how close the measured temperature was to the target temperature. The process of allowing the oven temperature to stabilize and changing the temperature setting was repeated until the temperature setting for the target temperature range ±5°C was established. The average maximum temperature of isopropanol after being microwave heated for 2.5 minutes at power level 2.5 was 80°C. Thus, the target temperature range for the toaster oven was between 75°C and 85°C because it included the maximum temperature of microwave heated isopropanol but was below the boiling point of isopropanol. A temperature range maximum of 85°C was also selected because it was assumed that the actual microwave heated isopropanol approaches temperatures close to its boiling point inside the microwave. The temperature setting was selected by initially choosing a temperature setting and allowing the toaster oven to stabilize for 1 hour. If the temperature remained the same after an additional hour, the measured temperature for the selected temperature setting was established. The temperature setting was then adjusted to be higher or lower depending on how close the measured temperature was to the target temperature. The process of allowing the oven temperature to stabilize and changing the temperature setting was repeated until the temperature setting for the target temperature range ±5°C was established. The second part of this experiment consisted of testing the stability of both simulant temperature settings over set periods of time. The ability of the toaster oven to maintain the set temperature range for isopropanol was tested by first allowing the target temperature range of 91°C to 100°C to stabilize for an hour. The toaster oven door was then left open for one minute. The internal oven temperature was then measured and recorded 10, 20, 30, and 60 minutes after the oven door was closed. The first three measurement time periods were selected to determine how fast the temperature increased in even time increments during the 30 minute heating period required for the acetic acid solution oven heating control experiment. The 60 minute measurement time period was selected because it was double the required heating time and to show longer term heating stability. This experiment was repeated twice and the temperatures from each thermometer were averaged. The ability of the toaster oven to maintain the set temperature range for acetic acid solution was tested by first allowing the target temperature range of 75°C to 85°C to stabilize for an hour. The toaster oven door was then left open for one minute. The internal oven temperature was then measured and recorded 8:20, 16:40, 25:00, and 50:00 minutes after oven door was closed. The first three measurement time periods were selected to determine how fast 119 the temperature increased in even time increments during the 25 minute heating period required for the acetic acid solution oven heating control experiment. The 50 minute measurement time period was selected because it was double the required heating time and to show longer term heating stability. This experiment was repeated twice and the temperatures from each thermometer were averaged. Once the stability of the toaster oven was established, the heat setting was applied in both oven heating control experiments. Prior to placing any food simulant in the toaster oven, the heating temperature was allowed to stabilize for at least one hour between heating cycles. 120 A.6.3 Results Table A.l Average internal temperatures of toaster oven after reaching a target temperature range of 91 to 100°C leaving oven door open for 1 minute. Thermometer type Time elapsed after closing oven door 10 minutes 20 minutes 30 minutes 60 minutes Temperature (°C) Mechanical oven 90 (±1.5) 90 (±1.0) 90 (±0.0) 90 (±0.6) Red bulb 102 (±3.3) 98 (±2.5) 100 (±2.0) 99 (±2.3) Blue bulb 100 (±3.8) 97 (±3.1) 98 (±2.6) 98 (±2.9) Table A.2 Average internal temperatures of toaster oven after reaching a target temperature range of 75 to 85°C leaving oven door open for 1 minute. Thermometer type Time elapsed after closing oven door 8:20 minutes 16:40 minutes 25:00 minutes 50:00 minutes Temperature (°C) Mechanical oven 80 (±2.0) 79 (±3.1) 80 (±2.5) 80 (±1.5) Red bulb 84 (±1.7) 81 (±3.3) 81 (±1.8) 81 (±1.5) Blue bulb 83 (±1.6) 80 (±4.4) 80 (±2.3) 80 (±1.3) Based on the experimental results presented in Tables A. l and A.2, the toaster oven heating temperature appears to stabilize within the target temperature range in less than 10 minutes after opening the oven door. Thus, the toaster oven was considered acceptable for use in the oven heating and oven heating glass control migration experiments for both food simulants. A.7 Determination of internal surface area of polypropylene container A.7.1 Purpose The objective of this experiment was to determine the internal surface area of a square polypropylene container (591 mL). This measurement was required for calculating of a chemical migration rate which is expressed as a mass per unit surface area (mg/dm2). 121 A.7.2 Protocol Six pieces of paper with a known surface area (25, 50, 75, 100, 150, 250 cm") were cut from a single sheet. Each piece of paper was weighed and recorded. A calibration curve and an equation relating surface area and weight were generated using a spreadsheet application. 3 short-term use polypropylene 591 mL containers were filled with 200 mL of water. The 200 mL fill line was traced on the outside of the container. The portion above the 200 mL fill line was removed from each container. The 200 mL fill portion of each container was cut into small pieces, flattened and traced onto pieces of the same paper used to generate the calibration curve. The outlines on each piece of paper were cut out. The total weight of the paper cut outs for each container were weighed. The surface area of each container was back calculated using the calibration curve equation. The average surface area of the container was then calculated. Figure A. 1 below shows the plot of the paper weight by surface area. The calibration curve shows a strong linear relationship. 300 Figure A.l Calibration curve for paper surface area versus weight. The average internal surface area of the polypropylene container samples was 169.63 ±2.71 cm2 (Relative Standard Deviation = 1.6%). The method used to determine the internal surface area of the plastic containers examined in this study appears to be fairly accurate given that the standard deviation and relative standard deviation were small. This method may not be suitable for rigid containers because they are not easily cut into pieces and flattened. 122 Potential sources of error for this experiment include the weighing technique, the accuracy of the electronic balance and variability in tracing and cutting out the traces by hand due to the tendency of the plastic to slide or revert to its original shape. A.8 Verification of retention time and mass spectra of surrogate standards A.8.1 Purpose The purpose of selecting and analyzing standards for a range of potential migrants was to confirm the identity of substances and quantify what migrants may be present in food simulants heated in plastic. Chemical surrogate standards were selected based on the following criteria: • Molecular weight. Chemicals with a molecular weight less than 45 g/mol were excluded because they would not be detected by the mass spectrometer. Chemicals with a molecular weight greater than 400 g/mol were excluded because they would not fit through the GC column. • General use in a range of plastics. Chemicals that were previously identified as being a non-specific plastic additive in plastic food packaging were chosen (van Lierop et al., 1998). • Potential to be environmental contaminants. Chemicals that have been identified ubiquitously in the environment were excluded. • Availability for purchase. Chemicals that were not sold in a range of catalogues were excluded. A.8.2 Protocol Approximately 1 to 5 mg of each chemical was transferred using a metal scupula (for powders) or glass pipette (for liquids) into a 5 mL glass vial. 4 mL of toluene was added to vials that contained water insoluble substances whereas 4 mL of ethanol was added to vials that contained water soluble substances. 100 pL of each standard was transferred to a 2 mL GC/MS vial. Each GC/MS vial was then filled with 400 uL of toluene. All standards were analyzed using GC/MS under the same conditions as those used in the migration experiments. 123 A.8.3 Results A total of 12 chemicals were selected as surrogates for substances that were expected to migrate from plastic (See Table A.3 below). The GC/MS library software identified almost every substance by matching the mass spectrum to the one stored for the specific substance in the database. Sorbitan monolaurate (CAS # 1338-39-2) was the only substance that was not identified. Sorbitan monolaurate had approximately three dominant peaks and several moderate to small peaks. The chromatograms of the other surrogate chemicals had only one large distinct peak. Table A.3 Chemicals selected as surrogate standards and their solubility. Chemical Name CAS# Solubility Diisobutyl phthalate 84-69-5 Soluble in toluene 2,2'-Methylenebis(4-ethyl-6-tert-butylphenol) 88-24-4 Soluble in toluene Octadecanol 112-92-5 Soluble in toluene 2,4-Dihydroxybenzophenone 131-56-6 Soluble in toluene 2-Hydroxy-4-methoxybenzophenone 131-57-7 Soluble in toluene Adipic acid dibutyl ester 105-99-7 Soluble in toluene 2-(2-Hydroxy-5-methyl-phenyl)benzotriazole 2440-22-4 Soluble in toluene Triacetin 102-76-1 Soluble in ethanol 2,2'-Methylenebis(6-t-butyl-4-methylphenol) 119-47-1 Soluble in ethanol 3,5-Di-tert-butyl-4-hydroxytoluene (BHT) 128-37-0 Soluble in ethanol Sorbitan monolaurate 1338-39-2 Soluble in ethanol 2(3)-Tert-butyl-4-hydroxyanisole 25013-39-2 Soluble in ethanol 124 A.9 Determination of instrument detection limits for surrogate standards and identified chemical migrants A.9.1 Purpose The purpose of this experiment is to determine the instrument detection limit for selected surrogate standards and chemical migrants identified in preliminary migration experiments. A.9.2 Protocol Approximately 0.05 g of each standard was accurately weighed, recorded and dissolved into 50 mL of analytical grade ethanol or toluene to create stock solutions. 4 dilution solutions for each standard was created by transferring 5, 10, 50 and 200 uL of each stock solution into separate 5 mL volumetric flasks and filling each flask to the 5 mL mark with ethanol or toluene. 1 mL of each dilution solution was analyzed in triplicate using GC/MS according to the conditions applied in the previous chemical migration experiment. The estimated instrument detection limit (DDL) was calculated as the nanograms detectable in the lowest measureable dilution solution of each standard multiplied by the target signal-to-noise ratio (3:1) divided by the determined mean signal-to-noise ratio. The estimated limit of detection was calculated by multiplying the DDL by the sample factor (SF). The sample factor (SF) (mL1) is equal to the final extract volume (2 mL) divided by the injection volume (1 uL), all divided by 200 mL ((1000 uL final extract volume/1 ul injection volume)/200mL = lOmL"1). A.9.3 Results Table A.4 shows the average mass ion peak areas, peak-to-peak signal-to-noise ratios and estimated instrument detection limits for each surrogate standard and chemical migrant that was identified in preliminary migration experiments. The relative standard deviation for the mass ion peak area of each substance was less than 20% which indicates that there was an acceptable amount of variation among the replicate samples analyzed per substance. The estimated instrument detection limits were generally higher (i.e. greater than 10 ng/mL) for large, polar chemicals such as stearic acid which is likely due to the selectivity of the stationary phase in the HP-5MS capillary column for non-polar/semi-volatile compounds. In contrast, the estimated instrument detection limits for semi-volatile and non polar chemicals were less than 5 ng/mL. 125 Table A.4 Estimated instrument detection limits of surrogate standards and identified chemical migrants from polypropylene plastic. Chemical Standard Mass detectable (ng) Peak Ion Average Peak Area Relative Standard Deviation Average Peak to Peak Signal to Noise Ratio Relative Standard Deviation Estimated Instrument Detection Limit (ng/mL) Diisobutyl phthalate 1.04 149 468520 3.8% 25 12.8% 1.25 2,2'-Methylenebis(4-ethyl-6-tert-butylphenol) 0.91 191 21174 6.8% 0.9 22.2% 30.33 Octadecanol 8.10 83 260099 17.7% 5.5 12.0% 43.92 2,4-Dihydroxybenzophenone 45.68 213 3736294 17.4% 54 17.4% 25.38 2-Hydroxy-4-methoxybenzophenone 11.00 227 1865593 5.6% 14.6 3.8% 22.55 Dibutyl adipate 1.84 185 80791 16.0% 3 29.1% 18.40 2-(2-Hydroxy-5-methyl-phenyl)benzotriazole 12.20 225 3175506 8.0% 47.1 8.2% 7.77 Erucamide 43.84 59 2874249 10.5% 22.6 8.6% 58.11 2,4-Di-tert-butylphenol 0.82 191 489500 7.0% 59.2 12.5% 0.42 Triacetin 1.09 43 314277 11.0% 12.7 14.2% 2.57 2,2'-Methylenebis(6-t-butyl-4-methylphenol) 1.25 177 47504 15.5% 2.0 14.2% 18.44 2(3)-Tert-butyl-4-hydroxyanisole 0.92 165 114105 15.8% 3.9 18.7% 7.14 3,5-Di-tert-butyl-4-hydroxytoluene (BHT) 2.72 205 1462789 2.4% 171.1 24.9% 0.48 2,6-di-tert-butylbenzoquinone 1.06 177 38954 10.4% 7.1 31.5% 4.48 1 -octadecene 1.09 55 62711 4.6% 2.2 7.1% 15.15 isopropyl palmitate 2.27 43 208200 1.3% 5.8 11.5% 11.82 bis(2-ethylhexyl)phthalate 2.27 149 499010 1.9% 22.0 13.9% 3.09 dicyclohexyl phthalate 10.16 149 2621157 3.2% 163.8 12.6% 1.86 stearic acid 89.44 43 320584 11.0% 4.5 14.2% 591.88 Irganox 1076 10.84 530 378974 6.0% 16.8 22.0% 19.32 n-undecane 1.48 43 610862 1.1% 31.1 9.0% 1.43 dodecane 1.50 57 610862 1.1% 29.8 5.1% 1.51 n-tetradecane 1.53 57 743100 1.2% 24.5 19.0% 1.87 n-pentadecane 1.54 57 703747 3.3% 34.9 6.9% 1.32 hexadecane 3.09 57 1550844 2.2% 53.8 7.8% 1.73 heptadecane 1.56 57 772427 2.1% 43.8 5.7% 1.07 126 Table A.4 (continued) Estimated instrument detection limits of surrogate standards and Chemical Standard Mass detectable (ng) Peak Ion Average Peak Area Relative Standard Deviation Average Peak to Peak Signal to Noise Ratio Relative Standard Deviation Estimated Instrument Detection Limit (ng/mL) n-octadecane 2.81 . 57 1648788 2.6% 52.0 4.7% 1.62 eicosane 1.58 57 726521 3.2% 15.1 6.9% 3.14 n-tetracosane 1.55 57 664806 4.7% 23.6 3.1% 1.96 hexacosane 1.56 57 624512 5.5% 16.7 5.4% 2.80 n-triacontane 3.88 57 .1264455 0.6% 23.9 6.9% 4.86 hexatriacontane 15.00 57 1757533 6.3% 7.4 6.8% 60.81 A.10 Surrogate standard recovery for extraction solvent comparison A.10.1 Purpose A solvent recovery comparison was conducted between methylene chloride and ethyl acetate to determine if the results obtained using methylene chloride as the extraction solvent were comparable to those obtained using ethyl acetate. Methylene chloride is a more preferable extraction solvent because it has a lower boiling point and it is denser than water which would require the use of fewer separatory funnels during the acetic acid simulant migration experiment and less time to concentrate the sample extracts for analysis. A.10.2 Protocol A surrogate standard solution was created by dissolving phenol, acenaphthalene, naphthalene, and methyl stearate in toluene. 100 uL of the internal standard was injected into six 200 mL samples of acetic acid simulant. Three samples were extracted twice with methylene chloride whereas the remaining three were extracted twice with ethyl acetate. Once the extraction process was complete, the solvent extracts were transferred to culture tubes, evaporated to dryness, reconstituted with 1 mL of toluene, vortexed and transferred to a GC/MS vial. 100 uL of the internal standard was dissolved in adding 900 pL of analytical grade toluene in a GC/MS vial. A solvent blank sample was prepared by adding 1 mL of toluene to a GC/MS vial. All samples were analyzed using GC/MS. To determine the percent recovery of naphthalene, acenaphthene, phenol and methyl stearate, the peak area for each substance in the ethyl acetate and methylene chloride extracts were divided by the peak areas in the internal standard sample and multiplied by 100. 127 A. 10.3 Results Recovery of the surrogate standards in methylene chloride sample extracts was fairly consistent; however, methyl stearate was the only substance that was consistently identified. All of the chemicals in the surrogate standards in the ethyl acetate sample extracts were identified; however, the occurrence of each chemical in all sample extracts was not consistent and the concentration of each chemical was variable between samples. It is hypothesized that the variation in recovery of the internal standard was primarily a result of the variability in heating temperature combined with the molecular weight of the chemicals in the internal standard (i.e. the lighter chemicals evaporated at higher temperatures). Although more chemicals from the internal standard were identified in the ethyl acetate sample extracts, methylene chloride was selected as the extraction solvent because the recovery for methyl stearate was consistently two times higher in all methylene chloride sample extracts. A . l l Determination of surrogate standard and chemical migrant recovery for migration test methods A.ll.l Purpose The purpose of this experiment was to determine the percent recovery of surrogate and migration standards in food simulant samples. A.11.2 Protocol A -50 ng/pL stock solution of 2,4-di-tert-buylphenol, dibutyl adipate, octadecanol, and erucamide was created by measuring and accurately recording - 50 mg of each substance and dissolving it in 100 mL of toluene. 2,4-di-tert-butylphenol and erucamide were selected as standards for this recovery experiment because they were previously identified as plastic migrants in migration experiments. 2,4-Di-tert-butylphenol also has physical and chemical properties that greatly contrast erucamide. Dibutyl adipate was selected as a standard because it was identified as a potential plastic chemical migrant; it was also assumed to be representative of acids and had a moderate retention time. Octadecanol was chosen as a standard because it was identified as a potential plastic chemical migrant; it was also assumed to be representative of alcohols and had a moderate retention time. Six working standards with concentrations ranging from 20 to 50 ng/pL were created from the stock solution. -1.5 mL of each standard was transferred into a GC/MS vial. 128 Three 200 mL samples of acetic acid simulant and three 200 mL samples of isopropanol were spiked with 100 uL of the stock solution of 2,4-di-tert-butylphenol, dibutyl adipate, octadecanol, and erucamide. 100 uL of toluene was spiked into a fourth flask. All flasks were then shaken for 2 minutes. All samples of acetic acid simulant were extracted twice with 20 mL of methylene chloride by agitating the flask for 2 minutes and transferring the methylene chloride to a culture tube. The acetic acid sample extracts were then evaporated to dryness in a hot water bath with a flow of nitrogen. All samples of isopropanol were evaporated to dryness, extracted twice with 5 mL of methylene chloride by agitating the flask for 2 minutes and pipetting the methylene chloride into a 15 mL glass culture tube. The isopropanol sample extracts were evaporated to dryness in a dry block heater with a flow of nitrogen. All sample extracts were reconstituted with 1 mL of toluene, vortexed for 30 seconds and pipetted into a glass vial. The sample extract reconstitution process was repeated. Thus, the final extract volume for each sample was 2 mL. 1 mL of each final sample extract was transferred with a syringe into a GC/MS vial. All samples, including the working standards, were then analyzed using the same GC/MS conditions used in the acetic acid food simulant migration test method. Once all samples were analyzed, the peak area ratio to concentration was plotted for each working standard. The amount of each standard that was detected in each sample was divided by the expected concentration to determine the percent recovery. A.11.3 Results Table A.5 and A.6 show the average percent recovery of the three spiked acetic acid and isopropanol simulant samples. Among the acetic acid simulant samples, the average percent recovery for octadecanol was the highest at 56%. Dibutyl adipate had the next highest percent recovery at 52%. Both erucamide and 2,4-di-tert-butylphenol had the lowest percent recoveries which were 44% and 43%, respectively. The average recovery for all substances ranged from 43 to 56% which was low; however, the relative standard deviation of the average recovery for each substance was less than 15% which shows that the ability of the acetic acid migration test method to extract chemicals is consistent. 129 Table A.5 Percent recovery of spiked standards in acetic acid sample extracts. Surrogate standard or identified migrant Average % recovery Average % recovery relative standard deviation Erucamide 44% 15% 2,4-di-tert-butylphenol 43% 6% Dibutyl adipate 52% 2% Octadecanol 56% 4% Among the isopropanol samples, the average percent recovery for octadecanol was the highest at 43%. Erucamide had the next highest recovery at 39% whereas dibutyl adipate (36%) and 2,4-di-tert-butylphenol (34%) had the lowest recoveries. The average recovery for all substances ranged from 34 to 43% which was lower than the percent recovery of standards from the acetic acid migration test. The lower percent recovery is likely due to the additional evaporation and transfer steps used to concentrate the isopropanol samples before extracting them with methylene chloride. However, the relative standard deviation of the average recovery for each substance was less than 5% which shows that the ability of the isopropanol migration test method to extract chemicals isvery consistent. Table A.6 Percent recovery of spiked standards in isopropanol sample extracts. Surrogate standard or Average % Relative standard deviation identified migrant recovery Erucamide 39% 3% 2,4-di-tert-butylphenol 34% 3% Dibutyl adipate 36% 1% Octadecanol 43% 4% 130 A.12 Determination of migrant recovery over time A. 12.1 Purpose The objective of this experiment was to determine how much migrant loss occurs at the beginning compared to the end of the acetic acid and isopropanol simulant microwave migration test. A. 12.2 Protocol On the first day of the microwave heating experiment, three 200 mL samples of acetic acid simulant were spiked with 100 pL of a -0.5 pg/pL mixture of the following chemicals: erucamide, 2,4-di-tert-butylphenol, dibutyl adipate, and octadecanol. Each sample was shaken for 2 minutes, extracted twice with 20 mL of methylene chloride, transferred to a culture tube and frozen at -20°C. Once more samples were prepared, the methylene chloride sample extracts were evaporated to dryness using a water bath and a flow of nitrogen. Each sample extract was reconsituted with 1 mL of toluene and vortexed for 30 seconds twice and transferred to a scintillation vial. 1.5 mL of each sample was then transferred to GC/MS vial and frozen at -20°C until all samples were prepared. On the last day of the microwave migration experiment, three 200 mL samples of acetic acid simulant were spiked with 100 pL of the standard mixture, extracted and prepared using the method described above. All previously frozen samples were sonicated. A set of 6 working standards for each of the spiked chemicals were prepared. All samples and working standards were then arranged in a sequence and analyzed using the GC/MS conditions described for the acetic acid simulant microwave migration test method. Once all samples were analyzed, the peak area ratio to concentration of was plotted for each working standard. The amount of each standard that was detected in each sample was divided by the expected concentration to determine the percent recovery. The same experiment was conducted using isopropanol and the isopropanol migration test extraction and sample preparation method. However, all frozen sample extracts were dried and reconstituted with toluene on the same day. 131 A. 12.3 Results Table A.7 shows the percent recovery of each chemical spiked and extracted at the beginning and end of the acetic acid microwave migration test experiment. Only one of the three samples that were prepared at the end of the experiment was valid because the chromatograph files for the other two samples were overwritten by accident during the GC/MS analysis. However, the calculated standard recovery in the valid sample was overestimated because the spiking amount was higher than what was spiked at the beginning of the experiment. The percent recovery of erucamide and dibutyl adipate in the spiked samples prepared at the beginning of the migration experiment was lower by 15% whereas the percent recovery of 2,4-di-tert-butylphenol was lower by 12%. The difference in percent recovery was lowest for octadecanol (4%). The relative standard deviation of the average percent recovery for the standards in samples prepared at the beginning of the experiment was less than 10% which shows that the migrant extraction and sample preparation method is consistent. Table A.7 Percent recoveries of spiked standards in acetic acid food simulant at the beginning and end of a migration experiment. Surrogate standard or identified migrant Average % recovery for samples spiked at the beginning of the experiment Relative standard deviation % Recovery for samples spiked at the end of the experiment Relative standard deviation Erucamide 52% 3% 67% N/A 2,4-di-tert-butylphenol 32% 7% 44% N/A Dibutyl adipate 42% 3% 57% N/A Octadecanol 57% 3% 61% N/A Table A.8 shows that average recovery of erucamide, 2,4-di-tert-butylphenol, dibutyl adipate, and octadecanol were 23%, 8%, 13% and 32% lower, respectively, in samples that were spiked and extracted at the beginning of the migration experiment compared to samples prepared on the last day of the experiment. The relative standard deviation of the average percent recovery for erucamide, 2,4-di-tert-butylphenol and dibutyl adipate in samples prepared at the beginning and end of the experiment was less than 10% which shows that the migrant extraction and sample preparation method is consistent. The large relative error for the average recovery of octadecanol in samples spiked at the beginning of the experiment was a result of poor recovery in one sample since two of the three spiked samples had an average recovery of 41% and a relative standard error of 9%. Thus, the difference in octadecanol recovery was likely closer to 20%. The low octadecanol recovery in the third sample was unlikely due to the amount of 132 standard spiked or systematic error since none of the other substances in the same sample were present in very low concentrations. However, the low recovery may be due to random error since only 3 samples were tested. Table A.8 Percent recoveries of spiked standards in isopropanol food simulant at the beginning and end of a migration experiment. Surrogate standard or identified migrant Average % recovery for samples spiked at the beginning of the experiment Relative standard deviation Average % recovery for samples spiked at the end of the experiment Relative standard deviation Erucamide 23% 2% 45% 1% 2,4-di-tert-butylphenol 31% 4% 39% 6% Dibutyl adipate 41% 7% 54% 0.5% Octadecanol 28% 21% 60% 0.2% Overall the estimated amount of erucamide, 2,4-di-tert-butylphenol and dibutyl adipate lost from samples prepared at the beginning of the experiment was acceptable (i.e. less than 25%). Octadecanol also appears to have been fairly stable if the spiked sample with the lowest concentration was not included. Both migration test methods were designed to minimize the amount of migrant loss during the preparation and storage of samples by preparing simulant samples exposed to microwave and oven heated plastic at the end of the experiment. Thus, the actual migrant loss during the acetic acid and isopropanol microwave migration experiment was less than what was calculated above and did not affect the migration test results significantly. 133 A P P E N D I X B CALIBRATION CURVES FOR STANDARDS USED TO QUANTIFY MIGRANTS IN THE ACETIC ACID AND ISOPROPANOL SIMULANT MIGRATION EXPERIMENTS Figures B.l and B.2 show the calibration curves for the substances that were quantified in the acetic acid migration experiment. Figures B.3 to B.13 show the calibration curves for the substances that were quantified in the isopropanol migration experiment. Although six standards were analyzed for every substance, each calibration curve does not have six points. A maximum of two data points were excluded if they were outside the actual measured concentration range and resulted in a better regression line fit. 50000000 -| , 0 10 20 30 40 50 60 Concentration (ng/uL) Figure B.l Calibration curve for 2,4-di-tert-butylphenol standard analyzed for the acetic acid migration experiment. 134 5000000 4000000 S 3000000 S 2000000 OH 1000000 0 10 20 30 40 Concentration (ng/uL) 50 60 Figure B.2 Calibration curve for erucamide standard analyzed for the acetic acid migration experiment. c s 140000000 120000000 100000000 80000000 _ 60000000 ^ 40000000 20000000 0 50 100 150 Concentration (ng/uL) 200 250 Figure B.3 Calibration curve for 2,4-di-tert-butylphenol standard analyzed for the isopropanol migration experiment. CB M <U OH 21000000 18000000 15000000 12000000 9000000 6000000 3000000 0 20 40 60 80 100 120 Concentration (ng/uL) 140 160 180 Figure B.4 Calibration curve for octadecanol analyzed for the isopropanol migration experiment. 135 ca OH 80000000 70000000 60000000 50000000 40000000 30000000 20000000 10000000 0 0 100 500 200 300 400 Concentration (ng/uL) Figure B.5 Calibration curve for erucamide standard analyzed for the isopropanol migration experiment. 600 30000000 Concentration (ng/uL) Figure B.6 Calibration curve for Irganox 1076 standard analyzed for the isopropanol migration experiment. 136 7000000 6000000 5000000 CO „ 4000000 | 3000000 OH 2000000 H 1000000 0 0 10 20 30 40 50 60 Concentration (ng/uL) Figure B.7 Calibration curve for 2,6-di-tert-butylbenzoquinone standard analyzed for the isopropanol migration experiment. 10000000 8000000 a 6000000 <P 4000000 2000000 10 20 30 40 50 60 Concentration (ng/uL) Figure B.8 Calibration curve for isopropal palmitate standard analyzed for the isopropanol migration experiment. 137 250000000 3500 Concentration (ng/uL) Figure B.9 Calibration curve for stearic acid standard analyzed for the isopropanol migration experiment. 8000000 7000000 . 6000000 2 5000000 -j Z 4000000 £ 3000000 2000000 1000000 0 0 10 15 20 25 Concentration (ng/uL) Figure B.10 Calibration curve for tetradecane standard analyzed for the isopropanol migration experiment. 138 9000000 Concentration (ng/uL) Figure B.ll Calibration curve for pentadecane standard analyzed for the isopropanol migration experiment. 9000000 -7500000 -« 6000000 -<u Z 4500000 -CO <o ^ 3000000 -1500000 -0 -• 0 5 10 15 20 25 Concentration (ng/uL) Figure B.12 Calibration curve for heptadecane standard analyzed for the isopropanol migration experiment. 139 03 20000000 17500000 15000000 12500000 10000000 7500000 5000000 2500000 0 10 15 20 25 Concentration (ng/uL) 30 35 40 Figure B.13 Calibration curve for octadecane standard analyzed for the isopropanol migration experiment. A P P E N D I X C ADDITIONAL TOXICOLOGICAL AND BIOLOGICAL EFFECTS OF CHEMICAL MIGRANTS IDENTIFIED IN ACETIC ACID AND ISOPROPANOL FOOD SIMULANT. Table C.l Toxicological and biological effects of chemical migrants identified in acetic acid and isopropanol food simulant exposed to microwave heated polypropylene plastic. Chemical Name Other toxic/biological effects References Erucamide • Angiogenic in chicken fetal membrane, rat cornea and skeletal muscle • Regulate peripheral organs and tissues • Inhibits intestinal diarrhea and regulates fluid volume in other organs • The toxicity of erucamide is likely similar to erucic acid which has oral toxic effects such as: fibrosis/lipoidosis of myocardium; reduction of respiratory capacity; myocardial lesions; reduction in spermatogenesis; and deficiency in mammary development and lactation. Wakamatsu et al., 1990; Mitchell etal., 1996 Hamberger and Stenhagen, 2003 Carroll and Noble, 1957; Engfeldt and Brunius, 1975a, 1975b; Engfeldt and Gustafsson, 1975; Borg, 1975 2,6-Di-t-butyl-4-methylene-2,5-cyclohexadiene-1-one • May inhibit glutathione-S-transferase Pl-1 (GSTP1) activity in mice lung cells which contributes to decreased cellular protection and interferes with GSTP1 regulation of stress kinases • Capable of damaging calf thymus DNA by binding to DNA at positions 1 and N 2 of deoxyguanosine, N 6 of deoxyadenosine and 7 of deoxyguanosine • Potentially promotes lung tumour formation in mice; forms adducts on immunoreactive proteins, cytoskeletal and stress-related proteins. • The toxicity of 2,6-di-t-butyl-4-methylene-2,5-cyclohexadiene-l-one is likely similar to BHT which has oral toxic/health effects such as: increase in liver and kidney weight, decreased liver enzyme activity, liver hemorrhaging, suppression and inhibition of immune response, liver adenomas and tumours. Lemercier et al., 2004 Lewis et al., 1996 Meier et al., 2005 Cosmetic Ingredient Review Expert Panel, 2002 2,6-Di-tert-butybenzoquinone • Has carcinogenic inhibition potential for: benzo[a]pyrene metabolism and DNA adduct formation in mice hepatic microsomes; 1,12-dimethylbena[a]anthracene induced mammary tumorigenesis and DNA adduct formation inhibition in rats Colovai etal., 1993; Niculescu-Duvaz et al., 1991; Singletary etal., 1992 2,4-Di-tert-butylphenol • Has carcinogenic inhibition potential of approximately 31.8% and antioxidant potential Safirman et al., 1987; Yoon et al., 2006 141 Table C.l (continued) Toxicological and biological effects of chemical migrants identified in acetic acid and isopropanol food simulant exposed to microwave heated polypropylene plastic. Chemical Name Other toxic/biological effects References Stearic acid • . Cytotoxic to carcinogenic and normal human cells, and normal animal cells; possible increased susceptibility of human leukaemic cell mitochondria compared to human circulating lymphocyte mitochondria to cytoxicity of a mixture of fatty acids • Has the potential to cause hypercoagulation of blood in dogs • Has anti-inflammatory effects when topically applied in low concentrations to chemically induced burns in mice • Potential skin irritant and low skin sensitization potential; eye irritant • Inhibits glutathione-S-transferase activity in human term placenta and fetal liver • Potential to protect against rat neurotoxicity induced by oxygen-glucose deprivation or glutamate • Disrupts endoplasmic reticulum homeostasis and induces apoptosis in liver cells in rats Fermor et a l , 1992; Buttke andCuchens, 1984;Otton and Curi, 2005 Hoak, 1994 Khalil, 2000 Anonymous, 1987 Mitraetal, 1992 Wang et a l , 2006 Wei et a l , 2006 Isopropyl stearate • Not an eye irritant; mild to moderate skin irritant Annonymous, 1985 Isopropyl palmitate • Possible skin irritant Tornier et a l , 2006; Fulton et a l , 1984; Rantuccio et a l , 1984 1-Octadecanol • Skin irritant Annonymous, 1985 Tetradecane • Skin irritant • Co-carcinogenic with U V light exposure (in mice Muhammad et a l , 2005 Bingham and Nord, 1977 Pentadecane • Skin irritant Muhammad et a l , 2005 Octadecane • Co-carcinogenic with benzo[a]pyrene (in mice) Horton et a l , 1976 Nonacosane • Bioaccumulates in cow liver Halseetal, 1993 142 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0100801/manifest

Comment

Related Items