UTILIZATION OF CHLORINE DIOXIDE GAS IN FOOD PACKAGING APPLICATION By SIRIYUPA NETRAMAI A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Packaging 2011 ABSTRACT UTILIZATION OF CHLORINE DIOXIDE GAS IN FOOD PACKAGING APPLICATION By Siriyupa Netramai Chlorine dioxide (ClO2) in its gaseous form has been used in numerous studies for vapor-phase decontamination, both in treating produce before packaging, and decontaminating the products inside their packages. Yet, very little is known about its compatibilities with packaging materials or its performance as affected by food packaging systems. The overall goal of this dissertation was to evaluate potential use of ClO2 gas as an antimicrobial agent for food packaging applications. In the first study, mass transfer profiles (permeability, solubility and diffusion coefficients) of ClO2 for 10 types of polymeric packaging materials were determined by an isostatic method using a continuous system for measuring ClO2 concentration with an electrochemical sensor as a detector. Overall, PET, PLA, BOPP, nylon, and multilayer of EVA/EVOH/EVA had high ClO2 barrier, while PS, LLDPE, LDPE, HDPE, and PVC provided low barrier to ClO2. Effects of gaseous ClO2 on physical, mechanical, chemical, and barrier properties of polymeric packaging materials were then studied by exposing selected materials to ClO2 gas. After 14 days of exposure, significant changes, such as increases in barriers to O2 and CO2 of nylon, changes in permselectivity (PCO2/PO2) ratio of up to 46.8% in treated PE, PS, PET, and nylon films, and changes in FT-IR spectra of PET, PLA, and EVA/EVOH/EVA, indicate possible changes in chemical profiles and performance of the materials. Study on influences of packaging design on antimicrobial effect of ClO2 gas, on shredded Romaine lettuce, indicated that minimizing the distance between gas releasing location and target surfaces, as well as, maximizing the area of gas release could significantly improve antimicrobial activity of ClO2 gas in particular packaging system. Once the interior of the package was optimized, it was observed that the amount of ClO2 used per package could be reduced to half of its original concentration (from 8 to 4 mgClO2/kg lettuce per day), while still achieving the same level of log10 CFU reduction of Escherichia coli O157:H7 in packaged shredded lettuce. When in contact with food, ClO2 gas will decontaminate the surfaces, as well as being absorbed by the product. The latter amount could not be accounted for its antimicrobial capacity. Study on absorption behavior of Romaine lettuce showed that increasing ClO2 level and/or time of exposure increased residual ClO2 and chlorite - (ClO2 ) recovered from Romaine lettuce sample. The presence of cuts significantly increased the amount of ClO2 consumed, while exceed water did not increase ClO2 absorption by lettuce. This research approach could be of great importance when considering antimicrobial packaging with ClO2 gas as a safety measure. Information generated could also be used to generate parameters for computational modeling of packaging systems. Copyright by SIRIYUPA NETRAMAI 2011 ‘WE GROW GREAT by DREAMS’ - Woodrow Wilson – We grow great by dreams. All big men are dreamers. They see things in the soft haze of a spring day or in the red fire of a long winter's evening. Some of us let these great dreams die, but others nourish and protect them, nurse them through bad days till they bring them to the sunshine and light which come always to those who sincerely hope that their dreams will come true. v ACKNOWLEDGEMENTS I used to perceive a ‘Dr.’ prefix as something that was easily distinguished from other prefixes. The ‘becoming’ process, to the old me, was somewhat ‘abrupt’. Until 4 years into my Doctoral program at the School of Packaging, Michigan State University that I have come to realize the tremendous effort, time, and ‘heart’ one need to put in, in order to be recognized as ‘Dr.’. Since then my perception has changed; this becoming process is more like a ‘transition’ in which one gradually gather experience and learn knowledge new to them and the earned title is merely an acknowledgement that one had suffered enough research drawbacks and sufficiently academically ‘grown up’ to be able to solve those problems and manage to move on to learn new things. This reminds me of the thermal transition of semi-crystalline polymer. All other people typically recognizing are the change in the polymer properties, before and after processing. They tend to neglect the required time and applied energy needed to make that polymer useful. I regard this dissertation as something similar to ‘latent heat of fusion’. It is the work people might see as the stepping stone to the next phase of my life-long goals, which has only my name labeled on it. However, this dissertation and what has become of me now would not have been possible without the following individuals; they are the time and energy given to me that tailors me to what I am today. I would like to express my gratitude, firstly, to financial supports from the Royal Thai government and the U.S. Department of Agriculture, USDA (NIFSI 2008-01573). This gave me the opportunities to meet many wonderful people I now hold dear and the experiences I had during these 6 years journey. vi Secondly, to my supervisor, Dr.Maria Rubino for her mentorship, understanding, and most importantly, her friendship during my PhD program at the School of Packaging. Thank you for teaching me to be able to think critically and independently. I will miss our (mostly) lively conversations we shared together in her office, dearly. Thank you for being there, at the front seat, while I took the spotlight stage for each step further in my career, always. I also equally grateful for the presences of my dissertation committee members: Dr.Rafael Auras, Dr.Bassam Annous, Dr.Laura Bix, Dr.Harold Hugh, and Dr.Elliot Ryser. Thank you for their valuable input and discussion. I have learnt to respect other sciences and become more ‘multidisciplinary’ through their guidance. Dr.Auras, I thank you for your guidance and friendship. Without you pushing me to try working outside my comfort zone, to think differently, to look at things in other angle, I would not be able to understand how Archimedes felt when he proclaimed “Eureka !”. I would like to thank the people from Dr.Bassam Annous Research Group at Food Safety and Intervention Technologies Research Unit (ERRC-ARS, USDA, Wyndmoor, PA), Parvaneh Khosravi, Joseph Sites, Angela Burke, and Dr.Bassam Annous himself for all their help on my PKG study, during my stay at ERRC. Your support and friendship made my 4-month stay very fruitful and enjoyable. My thanks also go to the ‘Bag team’; Dr.Brajesh Tripathi, Thitisilp Kijchavengkul, Mahesh Khurana, and Yining Xia (Eric) for their team effort in preparing the designed bags and personally delivered them to me in Wyndmoor. I was told you worked tirelessly on producing those bags, thank you so much. vii I am thankful for the opportunity to become a part of the School of Packaging at Michigan State University. I am grateful for the help and support from all SoP faculty, especially Dr.Bruce Harte and Dr.Susan Selke, and supporting staff; Linda Estill, Colleen Wager, April Meersdom, Sherrie Lenneman, and Kelby Thayer. Dr.Bruce Harte, thank you, for being a teacher I wish to become, one day. Dr.Susan Selke, thank you, for her countless helps and for giving me the opportunity to serve on a selection committee for the SoP director. It was a unique experience which I have learned a great deal from working alongside with many SoP faculties. I also would like to extend my gratitude to Tom Palazzolo, Tom Hudson, and Barry Tigner from the Mechanical shop at Department of Physics & Astronomy, Elvet Potter from ORCBS, Joel Tenney and William Ernst from ICA TriNova LLC, Timothy Kearney from Biosystems, Myra Foster and Alexis Chalfant from CryoVac, Doug Soroka from ERRC-ARS, USDA, and Dr.Karl Matthews from Rutgers University for their help and guidance I have received during the course of my research work. I am grateful to Enyonam Quist, SoP colleagues and Thai friends, and Carol Flegler and Dr.Stanley Flegler for the tears and laughter we have shared. Your invaluable friendships make Michigan winter less unbearable. I thank Dr.Amaret Bhumiratana, Dr.Pornchai Matangkasombut, and all my FTMUKA students who inspired me to choose the path of academia. My sincere thanks to all of you, there is no doubt in my mind I am where I belong. Most importantly, I owe my deepest gratitude to my parents, Somsak Netramai and Yupa Netramai, my sister, Siripattra Netramai, as well as, Thitisilp Kijchavengkul, viii Apiradee Bhisanbut, and Hayati Samsudin, who I consider to be ‘the next best thing’ to family. I am in debt to all of your love, trust, faith, and support. ix TABLE OF CONTENTS LIST OF TABLES …………………………………………………………………... xiv LIST OF FIGURES ………………………………………………………………..... xvii LIST OF SCHEMES ………………………………………………………………... xx CHAPTER 1 INTRODUCTION …………………………………………………………………... 1 1.1. Rationale and significance ………………………………..………………... 1 1.2. Objectives and hypothesis …………………………………………………... 6 1.3. Content organization ………………………………………………...……… 10 Bibliography …………………………………………………………………...... 11 CHAPTER 2 LITERATURE REVIEW ………………………………………………………...…. 2.1. Food safety of fresh produce ……………………………………………...... 2.2. Safety strategies to minimize foodborne outbreaks associated with fresh produce consumption ……………………………………………………..... 2.2.1. Available packaging systems for fresh produce ………...……….……... 2.2.1.1. Antimicrobial packaging ……………………………………...….. 2.3. Chlorine dioxide (ClO2) …………………………………...……………….. 2.3.1. Physical and chemical characteristics ………………………...………... 2.3.2. Measurements of ClO2 concentration ………………………...………... 2.3.3. Microbial activity ………………………...…………………………….. 2.3.4. Applications …………………………………………………………….. 2.4. Effects of gaseous oxidizing agents on packaging materials ……...……….. 2.4.1. Mass transfer of packaging materials ……………………………...…… 2.4.1.1. Isostatic permeation technique ……………………………...……. 2.4.1.2. Factors affecting permeation ………………………………...…… 2.4.1.3. Mass transfer of gaseous ClO2 …………………………………… 2.4.2. Polymer degradation and its effects on packaging performance ……….. 2.4.2.1. Chemical degradation …………………………………………….. 2.4.3. Effects of gaseous ClO2 on packaging materials ………………...…….. Bibliography ……………………………………………………………...……... CHAPTER 3 CONTINUOUS DETECTION SYSTEM FOR MASS TRANSFER MEASUREMENT OF ClO2 ………………………………………………………… 3.1. Introduction ………………………………………………...………………. 3.2. Materials and Methods ……………………………………...……………… 3.2.1. Permeability cell ………………………………………………………... x 16 16 19 20 23 24 24 27 28 43 45 45 48 51 54 55 56 62 63 74 74 76 78 3.2.2. Electrochemical detector ……………….……………...……………….. 3.2.2.1. Calibration ………………………………………………...……… 3.2.2.2. Determination of noise and signal to noise ratio …...…………….. 3.2.3. Preparation of ClO2 solutions …………………………………...……... 3.2.4. Statistical analysis ………………………………………………...……. 3.3. Results and Discussion ……………………………………………...……… Bibliography ………………………………………………………………...…... CHAPTER 4 MASS TRANSFER OF ClO2 THROUGH POLYMERIC FILMS ………...………. 4.1. Introduction …………………………………………………………...……. 4.2. Materials and Methods ………………………………………………...…… 4.2.1. Polymeric packaging materials ……………………………………...…. 4.2.2. Polymer characterization ……………………………………………...... 4.2.2.1. Determination of the film’s density ………………………………. 4.2.2.2. Determination of physical properties ……………………...……... 4.2.2.3. Determination of free volume ………………………………...….. 4.2.2.4. Determination of cohesive energy density …………………...…... 4.2.3. Preparation of ClO2 solution ……………………………………...……. 4.2.4. Determination of ClO2 mass transfer parameters ………………...……. 4.2.5. Determination of activation energy for permeation …………..……….. 4.2.6. Statistical analysis ……………………………………………...………. 4.3. Results and Discussion …………………………………………...………… 4.3.1. Mass transfer of ClO2 in polymeric films ……………………...………. 80 81 81 82 82 83 85 87 87 88 88 89 89 89 90 91 91 92 96 96 97 97 4.3.2. Activation energy of ClO2 ……………………………………...……… 103 4.4. Conclusion ………………………………………………………………….. 111 Bibliography ……………………………………………………………...……... 112 CHAPTER 5 EFFECTS OF GASEOUS ClO2 EXPOSURE ON IMPORTANT PROPERTIES OF POLYMERIC FILMS ……………………………………………………...…… 5.1. Introduction …………………………………………………………...……. 5.2. Materials and Methods ………………………………………………...…… 5.2.1. Experimental design ……………………………………………...…….. 5.2.2. Polymeric packaging materials ………………………………...………. 5.2.3. Preparation of ClO2 solution …………………………………...………. 5.2.4. Evaluation of chemical structure of polymeric material ………...……... 5.2.5. Evaluation of physical properties ……………………………...……….. 5.2.6. Evaluation of mechanical properties …………………………...………. 5.2.7. Evaluation of barrier properties ………………………………………… 5.2.8. Color measurement of polymeric material …………………...………… 5.2.9. Statistical analysis …………………………………………...…………. 5.3. Results and Discussion …………………………………………...………… 5.3.1. Chemical structure of polymeric material …………………...…………. xi 115 115 116 116 118 118 120 120 120 121 122 123 123 124 5.3.2. Physical properties …………………………………………...………… 5.3.3. Mechanical properties ………………………………………...………... 5.3.4. Barrier properties ……………………………………………………….. 5.3.4.1. Barrier to moisture ………………………………………………... 5.3.4.2. Barrier to oxygen and carbon dioxide …………………...……….. 5.3.5. Color measurement ……………………………………………………... 5.4. Conclusion ………………………………………………………………….. Bibliography ……………………………………………………………...……... CHAPTER 6 ABSORPTION OF ClO2 GAS BY LETTUCE LEAF ……………………………… 6.1. Introduction ………………………………………………………...………. 6.2. Materials and Methods ………………………………………………...…… 6.2.1. Development of experimental setup for absorption study ……………… 6.2.2. Preparation of lettuce samples ………………………………………….. 6.2.2.1. Preparation of whole leaf and washed shredded lettuce samples … 6.2.3. Preparation of ClO2 solution ……………………………………...……. 6.2.4. Concentration range to be used in absorption study ……………………. 6.2.5. Quantification of residual ClO2 and ClO2 on shredded lettuce …...…... 6.2.5.1. Evaluation of factors affecting chlorine dioxide absorption …...… 6.2.6. Statistical analysis …………………………………………………...…. 6.3. Results and Discussion …………………………………………………...… 6.3.1. Absorption behavior of ClO2 into shredded lettuce ………………...….. 131 134 136 137 139 140 143 144 148 148 149 149 152 152 153 153 154 155 155 155 156 6.3.2. Factors affecting absorption behavior of ClO2 into lettuce ……………. 160 6.4. Conclusion …………...................................................................................... 167 Bibliography ………………………………………………………………...…... 168 CHAPTER 7 EFFECTS OF PACKAGING DESIGN ON GAS DISTRIBUTION WITHIN THE PACKAGE ………………………………………………………..…………… 7.1. Introduction ………………………………………………………………… 7.2. Materials and Methods ……………………………………………………... 7.2.1. Preparation of inoculums ……………………………………………... 7.2.2. Inoculation of shredded lettuce ………………………………………… 7.2.3. Packaging design ……………………………………………………….. 7.2.3.1. Study 1 ……………………………………………………………. 7.2.3.2. Study 2 ……………………………………………………………. 7.2.4. Chlorine dioxide treatment ……………………………………………... 7.2.5. Preparation of packaged shredded lettuce inoculated with E. coli O157:H7 ………………………………………….……..………. 7.2.6. Microbial enumeration …………………………………………………. 7.2.7. Statistical analysis ……………………………………………………… 7.3. Results and Discussion ……………………………………………………... 7.3.1. Study 1 ………………………………………………………………….. 7.3.2. Study 2 ………………………………………………………………….. xii 170 170 174 174 175 177 177 177 180 180 183 185 186 186 201 7.4. Conclusion ………………………………………………………………….. 209 Bibliography …………………………………………………………………….. 211 CHAPTER 8 OVERALL CONCLUSIONS AND FUTURE WORKS …………………………… 215 8.1. Overall conclusions ………………………………………………………… 215 8.1.1. Mass transfer of ClO2 gas ……………………………………………… 216 8.1.2. Impact of ClO2 gas on chemical, physical, mechanical, and barrier properties of packaging materials ………………………………………. 216 8.1.3. Antimicrobial activity of ClO2 gas on shredded lettuce ………...……... 217 8.1.4. Effects of packaging design on antimicrobial effectiveness …………… 218 8.2. Applications and future works ……………………………………………... 218 8.2.1. Mass transfer of ClO2 gas ……………………………………………… 218 8.2.2. Impact of ClO2 gas on properties and performance of packaging Materials ………………………………………………………………... 8.2.3. Improving antimicrobial capacity of ClO2 through packaging design … 8.2.3.1. Important factors in packaging design …………………………… 8.2.3.2. Computational modeling of antimicrobial packaging system ……. Bibliography …………………………………………………………………….. APPENDICES ………………………………………………………………………. APPENDIX 1: Antimicrobial efficacy of ClO2 gas against E. coli O157:H7 on lettuce ……………………….…………...………………….. APPENDIX 2: Thiosulfate titration for determination of ClO2 in solution …….. APPENDIX 3: Results for consistency test on isostatic method ………………... APPENDIX 4: Amperometric titration for determination of ClO2 in solution …. 220 220 220 223 228 231 232 236 239 240 APPENDIX 5: ClO2 degradation profile and recovery rate of the experimental setup for absorption study ……………………..……………...... 242 APPENDIX 6: Characterization of PD-961EZ bag used in packaging study …... 246 APPENDIX 7: Identification of ClO2 levels for packaging study ……………… 248 APPENDIX 8: Amount of ClO2 precursors to be used in packaging study …….. 250 APPENDIX 9: Preliminary study on sample portions for packaging study …….. 255 APPENDIX 10: Results on moisture content of shredded lettuce for packaging study ……………………………………………….... 258 APPENDIX 11: Statistical analysis result for packaging study ………………… 262 Bibliography …………………………………………………………………….. 263 xiii LIST OF TABLES TABLE 2-1. Analytical methods for ClO2 and related compounds ………………… 30 TABLE 2-2. Summary of research studies on antimicrobial efficiencies of ClO2 for food products and other surfaces ………………………...…... 35 TABLE 3-1. Permeation system mean square error (MSE) and amplitude of noise at 0.00 ppmV, and signal to noise (S/N) ratio at various concentrations of ClO2 ……………………...………………… 84 TABLE 4-1. Degradation profile of calculated ClO2 in the gas phase above o ClO2 solution (%) at 23, 30, and 40 C ……………………….……….. 94 TABLE 4-2. Permeability (P), diffusion (D), and solubility (S) coefficients of 10 mgClO2 /L (3600 ppmV) for selected polymers at 23°C …….…. 99 TABLE 4-3. Polymer film characteristics including density (ρ), fractional free volume (FFV), free volume (FV), and cohesive energy density (CED) ………….……………………………………………… 104 TABLE 4-4. Permeability coefficients (P) at various temperatures and activation energy for permeation (EP) of ClO2 for PET and PLA films ….…………………………………………………………... 110 TABLE 5-1. Amount of ClO2 consumed by film samples (mg) ……………………. 119 TABLE 5-2. Physical properties of selected polymeric packaging materials exposed to gaseous ClO2 ……………………………………………… 132 TABLE 5-3. Tensile properties of selected polymeric packaging materials exposed to gaseous ClO2 ……………………………………………… 135 TABLE 5-4. Barrier properties of selected polymeric packaging materials exposed to gaseous ClO2 ……………………………………………… 138 TABLE 5-5. Color properties of selected polymeric packaging materials exposed to gaseous ClO2 ……………………………………………… 141 - TABLE 6-1. ClO2 recovered from shredded lettuce after treatments with 3.0 xiv and 6.0 mg/L ClO2 gas ………………………….……………...……... 159 - TABLE 6-2. ClO2 recovered from whole leaf lettuce after treatments with 3.0 and 6.0 mg/L ClO2 gas ………………………………………………... 162 - TABLE 6-3. ClO2 recovered from washed shredded lettuce after treatments with 3.0 and 6.0 mg/L ClO2 gas ………………………………………. 163 - TABLE 6-4. Residual ClO2 and ClO2 on shredded lettuce after treatments with 6.0 mg/L ClO2 gas, for 15 min, and left at room temperature for 15 and 60 min …………………………………………………….... 165 TABLE 7-1. Randomized complete block design information for study 1 ………..... 185 TABLE 7-2. Log10 CFU/g of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in one-GR and two-GR design bags with 4 and 8 mgClO2/day ……………………………………………….…………... 188 TABLE 7-3. Log10 CFU/g of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in middle-GR design bags with 4 and 8 mgClO2/day …………………………………………………………… 189 TABLE A-1. Log10 CFU reduction of E. coli O157:H7 and total aerobic plate count (TAPC) of ClO2 treated lettuce sample ……………….……...... 235 2 TABLE A-2. K1, K2, and R for consistency test …………………………………... 239 TABLE A-3. Degradation profile (%) of ClO2 in solution according to time (min) ...................................................................................................... 243 TABLE A-4. Recovery rate (%) of titration procedure for residual ClO2 and - ClO2 ………………………………………………………………….. 244 TABLE A-5. Sample of raw data (shredded lettuce) for absorption study …………. 245 TABLE A-6. Important properties of PD-961EZ bag ………………………………. 247 TABLE A-7. Log10 CFU reduction of total plate count (TAPC) of ClO2 treated lettuce sample ……………………………..……………………..…… 249 xv TABLE A-8. Calculation of amount of precursors to be used in packaging study 1 ………………………………………………………………... 251 TABLE A-9. Moisture content (%) of shredded lettuce samples, for packaging study on one-GR and two-GR designs, after storage for 0, 1, 4, and 7 days ……………………………………………………………….…... 260 TABLE A-10. Moisture content (%) of shredded lettuce samples, for packaging study on middle-GR, after storage for 0, 1, 4, and 7 days …..………... 261 TABLE A-11. Statistical analysis, for Study 1 …………………………………….. 262 xvi LIST OF FIGURES FIGURE 2-1. General Supply Chain Flow for Lettuce/Leafy Greens ………….…… 18 FIGURE 2-2. Solubility of ClO2 in water …………………………...…….………... 26 FIGURE 2-3. Biofilm of E. coli O157:H7 on un-treated lettuce surfaces …………... 32 FIGURE 2-4. Biofilm of E. coli O157:H7 on ClO2-treated lettuce surfaces ……..… 32 FIGURE 2-5. Mass transfer profile as obtained by isostatic method ……………...... 49 FIGURE 3-1. Flow diagram of mass transfer study ……………………….………... 77 FIGURE 3-2. Schematic of permeation of ClO2 through the polymer film …..…….. 79 FIGURE 4-1. Calculated ClO2 concentrations in the gas phase above o ClO2 solution (ppmV) at 23, 30, and 40 C without the presence of film sample ……………………………………………... 93 FIGURE 4-2. Mass transfer of 10 mgClO2/L ClO2 gas (3600 ppmV) through polymeric material …………………………………………………... 98 FIGURE 4-3. Correlation between the free volume and cohesive energy density ratio (FV/CED) and the diffusion coefficient (D) of ClO2 gas ……... 106 FIGURE 4-4. Correlation between the free volume and cohesive energy density ratio (FV/CED) and the permeability coefficient (P) of ClO2 gas ….. 107 FIGURE 4-5. Correlation between the free volume (FV) and the diffusion coefficient (D) of ClO2 gas ………………………………………….. 108 FIGURE 4-6. Correlation between the free volume (FV) and the diffusion coefficient (D) of ClO2 gas ………………………………………….. 109 FIGURE 5-1. Flow diagram of ClO2 treatment of polymeric materials ………….… 117 FIGURE 5-2. Remaining amount of ClO2 (mg) in 200 mL solution for blank and film treatment …………………………………………….. 119 xvii FIGURE 5-3. FT-IR spectra of nylon ………………………………..……….…...… 125 FIGURE 5-4. FT-IR spectra of LLDPE ………………………………………….….. 127 FIGURE 5-5. FT-IR spectra of PET …………………………………………......….. 129 FIGURE 5-6. FT-IR spectra of EVA/EVOH/EVA ……………………..…...……… 130 FIGURE 6-1. Experimental setup for absorption study …………………………… 151 FIGURE 6-2. Absorption of ClO2 gas on shredded lettuce ………………...…...…... 158 - FIGURE 6-3. Residual ClO2 recovered from whole leaf and washed shredded lettuce samples as compared to those of unwashed shredded lettuce ……………………………………..…………...….. 164 FIGURE 6-4. Color profiles ………………………………………..…….…………. 166 FIGURE 7-1. Inoculation of shredded lettuce; shredded lettuce were submerged in E. coli O157:H7 inoculums for 5 min and then spin-dried …………………………………………………...….. 176 FIGURE 7-2. Bag designs for study 1 and 2 ……………………………………….. 178 FIGURE 7-3. Gas release profiles ……………………...………………………….... 181 FIGURE 7-4. Inoculated shredded lettuce in one-GR design and middle-GR design bags …………………………………………...………...……. 182 FIGURE 7-5. Sample portions ………………………………………………………. 184 FIGURE 7-6. Average log10 CFU/g reduction of E. coli O157:H7 of samples stored for 1, 4, and 7 days in one-GR design bag and two-GR design bag …………………………………………........ 190 o FIGURE 7-7. E coli O157:H7 inoculated shredded lettuce stored at 4 C for 0, 1, 4, and 7 days (Study 1) ……………………………………... 193 FIGURE 7-8. Changes observed in samples packaged with ClO2 gas, at 8 mgClO2/day level; bleached surfaces and watery and softer tissues …………………………………………...………..…… 198 FIGURE 7-9. Log10 CFU/g reduction of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in one-GR design xviii bag and two-GR design bag …………………………………………. 199 o FIGURE 7-10. E coli O157:H7 inoculated shredded lettuce stored at 4 C for 0, 1, 4, and 7 days (Study 2) …………………………………....... 204 FIGURE 7-11. Log10 CFU/g reduction of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in middle-GR design bag ………………………..................................................….. 207 FIGURE 7-12. Average log10 CFU/g reduction of E. coli O157:H7 of samples stored for 1, 4, and 7 days in middle-GR design bag …………………………………………..…………….…. 208 FIGURE 8-1. CFD modeling overview ……………………………………….…...... 225 FIGURE A-1. Chlorine dioxide gas treatment of shredded lettuce in an air-tight chamber …………………………………..…………….…………..... 233 FIGURE A-2. FT-IR spectrum of PD-961EZ bag …………………………………... 246 FIGURE A-3. Calculated amount of permeated ClO2 (mg) from LDPE tube with different permeability coefficients ……………..…….………… 254 FIGURE A-4. Gas release profile ………………………………...…………...…….. 255 FIGURE A-5. Log10 CFU/g reduction of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in one-GR design bag with 8 mgClO2/day ….…………….....………………………………...…. 256 xix LIST OF SCHEMES SCHEME 2-1. Photochemical dissociation pathways of ClO2 …...………….......…. 24 SCHEME 2-2. Purposed mechanism for the initial reactions between ClO2 and CSH and subsequent decay of the cysteinyl-ClO2 adduct ……………...……………………..……………………..….. 34 SCHEME 2-3. Free radical mechanism of autoxidation in linear polymers ……...… 56 SCHEME 2-4. Solvolysis reaction of C-hetero bond in the polymers’ main-chain .... 57 SCHEME 2-5. Five-membered cyclic intermediates I and II yielded from the reaction of O3 with olefinic double bonds ………………..……...… 58 SCHEME 2-6. Direct reaction between O3 and saturated hydrocarbons …….......…. 59 xx CHAPTER 1: INTRODUCTION 1.1. Rationale and significance Each year, approximately one thousand reports of foodborne disease outbreaks (FBDOs) are collected by the Centers for Disease Control and Prevention (CDC), in the United States, resulting in around 20,000 cases of foodborne illness and 20 deaths (Centers for Disease Control and Prevention, 2010). The cost of foodborne illness in humans caused by bacterial pathogens is between $2.9 to $6.7 billion each year (Buzby, 1996). Among numerous confirmed foodborne disease cases reported, more than half were traced to consumption of raw or minimally processed foods, such as fresh spinach, shredded lettuce, salad, and deli meat (Department of Health and Human Services, 2008; Lynch et al., 2006). Some microorganisms reported to be the causes were Escherichia coli O157:H7, Listeria monocytogenes, Salmonella spp., and Staphylococcus aureus (Centers for Disease Control and Prevention, 2010; Department of Health and Human Services, 2008; FSIS, 2005; Nowak et al., 2006; Phillips, 1996; Sy et al., 2005b). To improve food safety, additional processing steps, e.g. washing with sanitizing solutions had been added to many food production lines (Huang et al., 2006; Kreske et al., 2006a). The use of packaging systems as an integral part of food processing is another possibility for enhancing food safety and prolonging product shelf-life. This approach is particularly useful when the foods will be consumed without any further processing or with minimal preparation, making contamination of pathogenic microorganisms a very important issue (Appendini and Hotchkiss, 2002; Department of 1 Health and Human Services, 2008; Ellis et al., 2006; Lynch et al., 2006; Nowak et al., 2006; Phillips, 1996; Sy et al., 2005a; USFDA, 2001), as in the case of ready-to-eat (RTE) food products, e.g. baby spinach leaves, and chicken salad (Appendini and Hotchkiss, 2002; Ellis et al., 2006). The consumer’s constantly growing demand for ‘near-fresh’ quality and shelfstable products has encouraged the development of several preservation techniques. Among others, modified atmosphere packaging (MAP) is, currently, one of the most promising and most extensively studied techniques. Modified atmosphere packaging (MAP) is defined as “the packaging of perishable product in an atmosphere which has been modified, so that its composition is other than that of the air” (Hintlian and Hotchkiss, 1986). Once sealed, the dynamic biological, chemical, and physical characteristics of both packages and products will alter the gaseous composition of the pre-set atmosphere inside the package (Hintlian and Hotchkiss, 1986; Ooraikul, 1991; Phillips, 1996). MAP application has found commercial success when applied to agricultural products, e.g. fruits, vegetable, meat, and seafood (Phillips, 1996). In the past few decades, new packaging strategies, such as including antimicrobial gas in the MAP packaging system, have also become a potential complementary approach to improve the safety of packaged fresh produce (Sy et al., 2005a). The antimicrobial agents can be integrated into the packaging system by means of adding sachets/pads, introducing the compound directly into the package headspace, or coating the antimicrobial compounds on the packaging surfaces (Appendini and Hotchkiss, 2002). Several chemical compounds are commercially used for such application and some show a strong potential for use as antimicrobial agents in the product/packaging 2 system, for example, silver substituted zeolite, glucose oxidase, sulfur dioxide, triclosan, allyl isothiocyanate, and essential oils (Appendini and Hotchkiss, 2002). In the past decade, chlorine dioxide (ClO2) is considered by many researchers as one of the most promising choices to be used in vapor-phase decontamination for food products, especially for vegetables and fruits (Gómez-López et al., 2009; Han et al., 1999; Sapers et al., 2003; Sy et al., 2005a; Yuk et al., 2006). Chlorine dioxide is a strong oxidizing agent and an effective surface disinfectant. Recognized as a disinfectant since the early 1900s, ClO2 was used initially used to treat water, as it causes fewer organoleptic problems than chlorine (Cl2) (USEPA, 1999). It is soluble in water and has equal, if not higher, inactivating capacity than that of Cl2, but is less effective than ozone (O3) (USEPA, 1999). Chlorine dioxide is gaining interest in the food and pharmaceutical industries, due to its broad antimicrobial effects (Sy et al., 2005a; Sy et al., 2005b). Its antimicrobial effect has been reported to be effective against many pathogenic microorganisms, such as Escherichia coli O157:H7, Listeria monocytogenes, and Salmonella spp (Ellis et al., 2006; Huang et al., 2006; Rodgers et al., 2004). In many food applications, ClO2 in solution can be substituted for chlorine (Cl2) solution, as it is known to be more effective and not to produce harmful end-products (Han et al., 1999; Kim et al., 1999; Kreske et al., 2006b; Rico et al., 2007; Ryu and Beuchat, 2005). The potential use of gaseous ClO2 for vapor-phase decontamination, during postharvest storage, in the processing line, and inside the product/package system has been 3 widely explored in the last decade (Ellis et al., 2006; Han et al., 2004; Selby et al., 2005). Adding ClO2 gas into the packaging system will extend the sanitizing time well beyond the processing line, without any time extension for the production. Introduction of ClO2 gas within a food packaging system is often by means of a sachet, either slow- or fast-release (Ellis et al., 2006; Shin, 2007), furthermore, in 2001, FDA approved the use of food packaging films incorporated with ClO2 to be used as packaging material for meats, poultry and seafood (USFDA, 2001). Most research regarding applications of ClO2, as a antimicrobial gas in the headspace of packaging systems, focuses on the appropriate dose for particular microorganisms and/or specific types of perishable food (Kaczur and Cawlfield, 1992; Sapers et al., 2003; Sy et al., 2005a). Only rarely, have these studies investigated the gas’s compatibility with polymeric packaging materials for food products (Ozen, 2000; Shin et al., 2006) or the effects of gas distribution on antimicrobial capacity of ClO2 gas. Different degrees of change in mechanical and barrier properties of several polymeric materials have been observed after ClO2 exposure, depending on gas concentration, relative humidity, and testing temperature (Ozen, 2000; Shin et al., 2006). o In 2001, Ozen reported permeability (P) coefficients of 100 mg ClO2/L of gas, at 20 C, for linear low-density poly(ethylene), LLDPE, oriented poly(propylene), OPP, and biaxially oriented nylon (BON). The experiment followed a quasi-isostatic method and used an amperometric titration method as a detection technique. However, the complete mass transfer profiles of the selected materials could not be obtained due to limitations of 4 the detection method, i.e. the steady state of permeation was reached too fast, and there was a delayed response in the detection system used, preventing the assessment of the unsteady region for the calculation of the diffusion coefficient (Ozen, 2000). If the package is to be considered as a strategy for the delivery and containment of ClO2 gas, more thorough study on mass transfer behavior of ClO2, as well as its effects on packaging integrities and performance are critical, as these could affect the polymers’ performance in packaging systems and possibly reduce the food product’s shelf-life. Also, the acquired knowledge could be used as a general guideline for material selection and packaging design for product/package systems that will include ClO2 gas as an antimicrobial agent. The study of how packaging design could affect gas distribution within the product/package system is another aspect of the research on antimicrobial packaging technology that was rarely performed. The uniformity of gas distribution inside the package is important, because the microbial population increases over time. The more effective the distribution, the sooner the microorganism will be inactivated (Ellis et al., 2006). Thus the impacts of packaging design on gas distribution as well as accessibility of antimicrobial compounds to hard-to-reach places should be explored. 5 1.2. Objectives and hypothesis The main goal of this research was to evaluate the potential use of gaseous ClO2 as an antimicrobial agent for food packaging applications. To achieve the overall goal, specific objectives and hypotheses of this study were set, as follows: 1. Develop a continuous detection method for mass transfer measurement of ClO2 2. Assess the mass transfer profile of ClO2 through different polymeric packaging materials Hypothesis: There are some differences in the mass transfer behavior of gaseous ClO2 through different types of polymeric films as indicated by their permeability (P), diffusion (D), and solubility (S) coefficients. The mass transfer behavior of ClO2 gas through each polymer is a unique characteristic which is influenced by chemical and physical properties of that particular polymeric material (Schnabel, 1992; Van Krevelen, 1997). The following hypotheses are set to be proved or disproved. Null hypothesis (H0): There is no difference in the mass transfer behavior of gaseous ClO2 through different types of polymeric films as indicated by their P, D, and S. Alternative hypothesis (H1): There are some differences in the mass transfer behavior of gaseous ClO2 through different types of polymeric films as indicated by their P, D, and S. 6 3. Determine the impact of ClO2 gas on the chemical, physical, mechanical, and barrier properties of selected polymeric packaging materials Hypothesis: There are some changes in the materials’ integrities and/or performance, after their exposure to ClO2 gas. By exposing the polymeric materials to reactive chemical compounds, like ClO2, several chemical changes, such as main-chain scission, cross-linking, and functional group formation, can occur. These changes can affect the material integrity and the package performance (Ozen, 2000; Schnabel, 1992; Selke et al., 2004). The following hypotheses are set to be proved or disproved. Null hypothesis (H0): There are no changes in the materials’ integrities and performance, after their exposure to ClO2 gas. Alternative hypothesis (H1): There are some changes in the materials’ integrities and/or performance, after their exposure to ClO2 gas. 4. Investigate the antimicrobial effect of ClO2 gas on fresh produce Hypothesis: There are some differences in terms of microbial load of fresh produce packed with air and those packed with ClO2 gas in the package. Since ClO2 demonstrates antimicrobial activity against many important microorganisms (Sy et al., 2005a; USEPA, 1999), adding it into the product/package system should help reduce microbial populations on the product surface (Huang et al., 2006; Sy et al., 2005b; Zhang and Farber, 1996). The following hypotheses are set to be proved or disproved. 7 Null hypothesis (H0): There is no difference in terms of microbial load of fresh produce packed with air and those packed with ClO2 gas in the headspace, during entire storage duration. Alternative hypothesis (H1): There are some differences in terms of microbial load of fresh produce packed with air and those packed with ClO2 gas in the headspace, during entire storage duration. 5. Identify the parameters in the packaging design responsible for improved gas distribution within the package, for future packaging development using CFD modeling, by: 5.1. Studying the effects of packaging design on antimicrobial effectiveness of ClO2 gas in product/package systems Hypothesis: There are some differences in terms of microbial load of fresh produce packed in bags of different designs. Since ClO2 is a very active surface disinfectant, the distance between the point of gas release and the food product surface should be minimized. If the package is designed to ensure fast and thorough distribution of a gas mixture throughout the interior of the packaging system, maximum exposure of the products’ surface can be achieved. This could result in the effective reduction in microbial loads which could lead to shelf life extension and improved safety of food products (Ellis et al., 2006). The following hypotheses are set to be proved or disproved. Null hypothesis (H0): There is no difference in terms of microbial load of fresh produce packed in bags with different internal designs. 8 Alternative hypothesis (H1): There are some differences in terms of microbial load of fresh produce packed in bags with different gas distribution pattern, as directed by different internal designs. 5.2. Measuring the absorption of ClO2 gas by lettuce leaf Hypothesis: There is a measurable amount of ClO2 gas being absorbed into lettuce and - there is a measurable amount of chlorite (ClO2 ) being generated from the oxidation reactions of ClO2 and organic matters in lettuce. Han et al (2004) detected up to 0.52 mg - ClO2/kg and 3.03 mg ClO2 /kg of strawberries after the fruits treatment with 3.0 mg ClO2 gas/L for 10 min (Han et al., 2004). Null hypothesis (H0): After ClO2 treatment, there is no ClO2 residual and ClO2 - detected on the lettuce samples Alternative hypothesis (H1): After ClO2 treatment, there is a measurable amount - of ClO2 residual and ClO2 found on the lettuce samples 9 1.3. Content organization A review of literature related to food safety, the roles of packaging in food applications, chlorine dioxide, product/packaging interactions, and polymer degradations is provided in Chapter 2. The work on mass transfer of ClO2 gas for polymeric packaging materials is reported in Chapter 3 (objective 1) and 4 (objective 2). Chapter 5 addressed the effect of ClO2-exposure on properties and performance of packaging materials (objective 3). The impact of packaging design on the antimicrobial effect of ClO2 gas is reported in Chapter 7 (objective 4 and 5.1) with ClO2-absorption by the lettuce leaf (objective 2.2) reported in Chapter 6. 10 BIBLIOGRAPHY 11 Appendini, P. and Hotchkiss, J. H. (2002). "Review of antimicrobial food packaging." Innovative Food Science & Emerging Technologies 3: 113-126. Buzby, J. C., Roberts, T., Lin, C.-T. J. and MacDonald, J. M. (1996). Bacterial foodborne disease: medical costs and productivity losses. USDA. Agricultural Economic Report No. 741. Centers for Disease Control and Prevention (2010) "Surveillance for Foodborne Disease Outbreaks --- United States, 2007." Retrieved August 13th, 2008, from http://www.cdc.gov/mmwr/pdf/wk/mm5931.pdf. Department of Health and Human Services (2008) "Summary Statistics for Foodborne Outbreaks, 2006." Retrieved November 1st, 2008, from http://www.cdc.gov/foodborneoutbreaks/documents/2006_line_list/2006_line_list .pdf. Ellis, M., Cooksey, K., Dawson, P., Han, I. and Vergano, P. (2006). "Quality of Fresh Chicken Breasts Using a Combination of Modified Atmosphere Packaging and Chlorine Dioxide Sachets." Journal of Food Protection 69(8): 1991 - 1996. FSIS (2005) "Poultry preparation." Retrieved November 22nd, 2009, from http://www.fsis.usda.gov/factsheets/Poultry_Preparation_Fact_Sheets/index.asp. Gómez-López, V. M., Rajkovic, A., Ragaert, P., Smigic, N. and Devlieghere, F. (2009). "Chlorine Dioxide for Minimally Processed Produce Preservation: A Review." Trends in Food Science & Technology 20(1): 17-26 Han, Y., Guentert, A. M., Smith, R. S., Linton, R. H. and Nelson, P. E. (1999). "Efficacy of chlorine dioxide gas as a sanitizer for tanks used for aseptic juice storage." Food Microbiology 16(53-61). Han, Y., Selby, T. L., Schultze, K. K., Nelson, P. E. and Linton, R. H. (2004). "Decontamination of Strawberries Using Batch and Continuous Chlorine Dioxide Gas Treatments." Journal of Food Protection 67(11): 2450-2455. Hintlian, C. B. and Hotchkiss, J. H. (1986). "The Safety of Modified Atmosphere Packaging: A Review." Food Technology 40: 70-76. Huang, T.-S., Xu, C., Walker, K., West, P., Zhang, S. and Weese, J. (2006). "Decontamination Efficacy of combined Chlorine Dioxide with Ultrasonication on Apples and Lettuce." Journal of Food Science 71(4): M134-M139. Kaczur, J. J. and Cawlfield, D. W. (1992). Chlorine Oxygen Acids and Salts, Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical Technology. Kroschwitz, J. I. New York, John Wiley & Sons Inc. 5. 12 Kim, J. M., Huang, T.-S., Marshall, M. R. and Wei, C.-I. (1999). "Chlorine dioxide treatment of seafoods to reduce bacterial loads." Journal of Food Science 64(6): 1089-1093. Kreske, A. C., Ryu, J.-H. and Beuchat, L. R. (2006a). "Evaluation of chlorine, chlorine dioxide, and a peroxyacetic acid-based sanitizer for effectiveness in killing Bacillus cereus and Bacillus thuringiensis spores in suspensions, on the surface of stainless steel, and on apples." Journal of Food Protection 69(8): 1892-1903. Kreske, A. C., Ryu, J. H., Pettigrew, C. A. and Beuchat, L. R. (2006b). "Lethality of Chlorine, Chlorine Dioxide, and a Commercial Produce Sanitizer to Bacillus cereus and Pseudomonas in a Liquid Detergent, on Stainless Steel, and in Biofilm." Journal of Food Protection 69(11): 2621-2634. Lynch, M., Painter, J., Woodruff, R. and Braden, C. (2006) "Surveillance for FoodborneDisease Outbreaks --- United States, 1998-2002." Retrieved November 1st, 2008, from http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5510a1.htm?s_cid=ss5510a1_e. Nowak, B., Sammet, K., Klein, G. and Mueffling, T. v. (2006). "Trands in the production and storage of fresh meat - the holistic approach to bacteriological meat quality." International Journal of Food Science and Technology 41: 303-310. Ooraikul, B., and Stiles, M. E. (1991). Modified atmosphere packaging of food. New York, Ellis Horwood. Ozen, B. F. (2000). Effect of Ozone and Chlorine Dioxide Treatments Used in Perishable Food Applications on Polymeric Materials: Changes in Mechanical, Thermal and Mass Transfer Properties. Food Science Department. West Lafayette, IN, Purdue University. Ph.D. Phillips, C. (1996). "Review: Modified atmosphere packaging and its effects on the microbiological quality and safety of produce." International Journal of Food Science and Technology 31: 463-479. Rico, D., Martín-Diana, A. B., Barat, J. M. and Barry-Ryana, C. (2007). "Extending and measuring the quality of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology 18(7): 373-386 Rodgers, S. L., Cash, J. N., Siddiq, M. and Ryser, E. T. (2004). "A comparison of different chemical sanitizers for inactivating Escherichia coli O157:H7 and Listeria monocytogenes in solution and on apples, lettuce, strawberries and cantaloupe." Journal of Food Protection 67(4): 721-731. Ryu, J. H. and Beuchat, L. R. (2005). "Biofilm Formation and Sporulation by Bacillus cereus on a Stainless Steel Surface and Subsequent Resistance of Vegetative Cells 13 and Spores to Chlorine, Chlorine Dioxide, and a Peroxyacetic Acid-Based Sanitizer." Journal of Food Protection 68(12): 2614-2622. Sapers, G. M., Walker, P. N., Sites, J. E., Annous, B. A. and Eblen, D. R. (2003). "Vapor-phase decontamination of apples innoculated with Escherichia coli." Journal of Food Science 68(3): 1003-1007. Schnabel, W. (1992). Polymer Degradation: Principles and Practical Applications. New York, Oxford University Press. Selby, T. L., Wong, E., Han, Y., Gerrard, D. E. and Linton, R. H. (2005). Effects of chlorine dioxide gas treatment on microbial safety and quality of fresh pork. 2005 IFT Annual Meeting, New Orleans, Louisiana, IFT. Selke, S. E. M., Culter, J. D. and Hernandez, R. J. (2004). Plastics Packaging: Properties, Processing, Applications, and Regulations. Cincinnati, OH, Hanser Gardner Publications, Inc. Shin, J. (2007). Evaluation of the Effect of Chlorine Dioxide and Allyl-isothiocyanate on the Growth of Salmonella Typhimurium and Listeria Monocytogenes on Fresh Chicken Breast and Effect of Chlorine Dioxide Exposure on the Physical Properties of Plastic Films. School of Packaging. East Lansing, Michigan State University. Ph.D. Shin, J., Ryser, E., Selke, S. and Harte, B. (2006). Evaluation of the Microbial Inhibition Performance of Gas Type Antimicrobials (Chlorine Dioxide and AllylIsothiocyanate) with Modified Atmosphere Packaging (MAP), and the Effct of Chlorine Dioxide Exposure on the Physical Properties of Plastic Films. The 15th IAPRI World Conference on Packaging. Tokyo, Japan. Sy, K. V., McWatters, K. H. and Beuchat, L. R. (2005a). "Efficacy of Gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries, and Raspberries." Journal of Food Protection 68(6): 1165 - 1175. Sy, K. V., Murray, M. B., Harrison, M. D. and Beuchat, L. R. (2005b). "Evaluation of gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes, and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection 68(6): 1176 - 1187. USEPA (1999) "4. Chlorine Dioxide." Retrieved September 11th, 2007, from http://www.epa.gov/ogwdw/mdbp/pdf/alter/chapt_4.pdf USFDA. (2001). "Agency Response Letter GRAS Notice No. GRN 000062." Retrieved September 29th, 2007, from http://www.fda.gov/Food/FoodIngredientsPackaging/GenerallyRecognizedasSafe GRAS/GRASListings/ucm153971.htm. 14 Van Krevelen, D. W. (1997). Properties of Polymers. New York, Elsevier Science B.V. Yuk, H.-G., Bartz, J. A. and Schneider, K. R. (2006). "The Effectiveness of Sanitizer Treatments in Inactivation of Salmonella spp. from Bell Pepper, Cucumber, and Strawberry." Journal of Food Science 71(3): M95-M99. Zhang, S. and Farber, J. M. (1996). "The effects of various disinfectants against Listeria monocytogenes on fresh cut vegetables." Food Microbiology 13: 311-321. 15 CHAPTER 2: LITERATURE REVIEW 2.1. Food safety of fresh produce Foodborne disease continues to be a public health threat in the United States (U.S.). The Centers for Disease Control and Prevention (CDC) reported that, in year 2007, 21,244 people was hospitalized, from 1,097 reported foodborne disease outbreaks (FBDOs), resulting in 18 deaths from foodborne illness (Centers for Disease Control and Prevention, 2010). The estimated cost of foodborne illness in terms of mental, financial, and medical expenses ranges from $10-83 billion each year (Department of Health and Human Services, 2008; Lynch et al., 2006; Mead et al., 1999; USFDA, 2004). Major disease carriers include dairy, poultry, beef, and leafy vegetables (Centers for Disease Control and Prevention, 2010; Lynch et al., 2006; USFDA, 2004). Vegetables and fruits are recognized as an important component of a healthy diet program as a source of fiber, vitamins, minerals, and antioxidants. However, they can also become a microbial carrier as they are: 1) grown in a natural environment and 2) usually consumed raw without any type of preparation (at the consumer level) that would reduce, control, or eliminate microorganisms prior to consumption (Gorny et al., 2006; USFDA, 2004). Each year (during 1998-2004), around 400 FBDOs are confirmed to be caused by consumption of produce, with around 56% attributed to mixed or unknown food commodities and the rest were related to consumptions of single food commodity, such 16 as (in the order from highest to lowest numbers of FBDOs) lettuce, sprouts, tomatoes, and berries (Tauxe, 2008). Microbial contamination onto fresh produce can occur at several points throughout an entire product production line. Figure 2-1 shows the general supply chain flow for lettuce/leafy greens (Gorny et al., 2006), as an example. Salmonella spp and Escherichia coli O157:H7 are the two pathogenic microorganisms most often associated with produce (Gorny et al., 2006; USFDA, 2004; 2006). Like in the case of any other fresh produce, both pathogens usually contaminate lettuce/leafy greens via fecal-oral route. At merely every step, produce comes in contact with human hands, water, or soil with special attention to control, reduce, or eliminate potential contamination from people and animal (Gorny et al., 2006; USFDA, 2004). Leafy greens are usually consumed uncooked or raw. These products are generally considered to be ready-to-eat (RTE) since the washing step is used in their production and protective package is employed in their distribution and marketing (Gorny et al., 2006; USFDA, 2006). During production, there are many ‘touch points’, e.g. handharvested and hand-sorted which can lead to cross-contamination (Gorny et al., 2006). Other factors include agricultural water quality, the use of manure as fertilizer, and the presence of animals nearby or in packing areas (Gorny et al., 2006; USFDA, 2004). 17 Field Production Raw Agricultural Commodity Fresh-cut Harvest Field Cored Cooling Cooling Value Added Processing and Cooling Transportation Cold Storage Transportation Retail or Foodservice Distribution Center Retail/ Foodservice Processing/Preparation Retail/Foodservice Outlet Consumer Figure 2-1. General supply chain flow for lettuce/leafy greens. Adapted from (Gorny et al., 2006). For interpretation of the references to color in this and all other figures, the reader is referred to the electronic version of this dissertation. 18 2.2. Safety strategies to minimize foodborne outbreaks associated with fresh produce consumption The safety of food production, distribution, and marketing chain requires an integrated approach, by the manufacturers, to prevent/minimize foodborne illness, along with oversight by the federal food safety agencies (USFDA, 2004). To minimize the recurring outbreaks associated with fresh and fresh-cut produce, the Food and Drug Administration (FDA) released an action plan, in 2004, to minimize foodborne illness associated with fresh produce consumption (Gorny et al., 2006; USFDA, 2004). The plan suggests that each unit involved in producing, packing, processing, transporting, distributing, or preparing fresh produce (at the consumer level) has a responsibility to reduce, control, or eliminate microbial contamination of the produce (USFDA, 2004). To support the action plan’s goal, FDA developed the safety initiative for each specific commodity, e.g. ‘the Lettuce Safety Initiative’ (USFDA, 2006). The plan includes inspections of the farm and cooling and packing facilities, and focuses on good agricultural practices (GAPs) and good manufacturing practices (GMPs) (Gorny et al., 2006; USFDA, 2006). Under USDA rules, the programs of safety assurance and preventive process controls like the Hazard Analysis and Critical Control Point (HACCP) system are being implemented by producers and distributors to prevent food safety hazards which are categorized as biological, chemical, or physical hazards. Such systems focus on problem prevention and involve the identification, evaluation, and control of critical control points/steps in the food production and distribution, where failure to take appropriate action is most likely to cause potentially hazardous products to reach the consumer. 19 HACCP systems are already mandated in particular segments of the food industry, e.g. seafood, juice, and, meat and poultry products (Executive Secretariat, 1998; FSIS, 2006; 2009). Additional processing steps, e.g. sequential washing, fumigation (with antimicrobial gas), and irradiation are installed in the commercial food production lines and/or being developed at various research institutes, to strengthen end product safety (Han et al., 2004; Huang et al., 2006; Kreske et al., 2006a; Singh et al., 2002). Hurdle technology is a strategy many researchers recognize to be very efficient in reducing microbial risks and extending shelf-life of minimally processed foods (Allende et al., 2006; Rico et al., 2007). The technology involves the use of a series of preservation techniques, e.g. photochemical process, washing, thermal treatment, and low temperature storage, as well as the use of active packaging to treat the food product. The influences of combined techniques are also reported to be synergistic. In many cases, the food products are found to be of superior quality as compared to those subjected to single preservative technique, since each technique, in the series, is usually applied at a milder condition to achieve a comparable shelf-life/safety level (Huang et al., 2006; Rico et al., 2007; Sothornvit and Kiatchanapaibula, 2009). 2.2.1. Available packaging systems for fresh produce Using a packaging system as an extension from the production line to improve safety and economically extend shelf-life of fresh produce, is very practical when the food will be consumed without any further processing or with minimal preparation (Appendini and Hotchkiss, 2002; Ellis et al., 2006). 20 The function of packaging as a preservation tool for food products can be grouped into passive and active protections. When “packaging is independent of the foodstuff preparation and preservation techniques” (Multon, 1996), the package provides passive protection towards mechanical forces and microbial contaminations, as well as, protection against heat and mass transfers. In the case where packaging constitutes as an integral part of food preparation and preservation processes, thus considered as an indispensable element, it may be regarded as having an active role, for example, containers used in aseptic packaging and modified atmosphere packaging (MAP) technique, etc (Multon, 1996). In MAP technique, respiring produce is packed in polymeric packaging with the gas mixture in the headspace other than air, to reduce the product’s respiration rate, moisture loss, and/or decay, thus extending the product’s shelf life. Once packaged, gas composition in the package headspace is gradually altered as affected by several factors, such as the packaging materials’ permeability to O2 and CO2, the product’s biochemical rate, and the storage temperature (Ooraikul and Stiles, 1991; Phillips, 1996; Romig and Mir, 2000). MAP application has been successfully applied to commercial fresh produce and processed food products, e.g. fruits, vegetable, meat, seafood, etc, for decades (Phillips, 1996). The appropriate gas composition can prolong the product’s shelf-life; however, it cannot improve the food’s quality. Thus it is essential to start with raw material with high quality. Majority of studies in MAP for fresh produce involve 1) development of mathematic modeling, for example, the use of Michaelis-Menton type respiratory model to predict the effects of temperature and O2 on respiration rate of the product, the use of 21 Arrhenius equation to predict O2 partial pressure as a function of temperature, product surface area, and film permeability, and the application of growth model of pathogenic microorganisms under MAP condition and refrigerated temperature (Cameron et al., 1994; Harrison, 2000; Simpson, 2001; Yang, 2002), 2) effects of modified gaseous atmosphere on product’s properties (Cliffe-Byrnes, 2003; Fu, 1992; Lin, 2002), and 3) incorporation of preservation aids, e.g. disinfectants and oxygen scavengers (CegielskaRadziejewska, 2004; DeEll, 2006). Many innovations and developments have been done on MAP applications regarding improving its package efficiency, and several studies were conducted on the combination of MAP with additional preservation means, to further extend the product’s shelf-life, e.g. the use of blanching, acid treatment, and mild irradiation process to treat fresh food prior to packaging (Bagorogoza et al., 2001; Jimenez et al., 1999; Sawaya et al., 1995). One of the main concerns regarding using MAP as a packaging technique for food products is: the condition applied may be more effective in killing spoilage microorganisms which provides organoleptic indication of spoilage, giving the possibility for pathogenic microorganisms, e.g. Clostridium botulinum (anaerobe) and L. monocytogenes (psychotropic facultative anaerobe), to outgrow the former, and increase their populations to significant levels or produce toxin (Farber and Dodds, 1995). Thus, the addition of antimicrobial agents, in the form of organic acids, enzymes, spices, etc., into the product/package system has also been widely researched, in the last few decades (Appendini and Hotchkiss, 2002). 22 2.2.1.1. Antimicrobial packaging To improve the food products’ safety and prolong their shelf-life, there is the increasing use of antimicrobial packaging for minimally processed and RTE fresh products (Appendini and Hotchkiss, 2002; Ellis et al., 2006). The antimicrobial agents can be integrated into the packaging system by adding the gas providers, such as sachet and pad, coating the agent onto the packaging surface, and embedding the compounds into the polymer matrix (Appendini and Hotchkiss, 2002). The antimicrobial agents, once generated and/or released will reduce, inhibit and/or retard the microbial growth on the packaging interior and on the food product’s surface, where the majority of contamination and spoilage occurs (Appendini and Hotchkiss, 2002; Ooraikul and Stiles, 1991). Examples of antimicrobial agents incorporated into the packaging system, available commercially, are silver substituted zeolite, triclosan, and ethanol vapor (Appendini and Hotchkiss, 2002). Among others, chlorine dioxide (ClO2) is considered by many researchers as one of the promising choice to be used in vapor-phase decontamination within the package of the food products (Appendini and Hotchkiss, 2002; Kaczur and Cawlfield, 1992; Kim et al., 1999; Sapers et al., 2003). 23 2.3. Chlorine dioxide (ClO2) Chlorine dioxide (ClO2) is an oxidizing agent that is used primarily as a bleaching agent in pulp and paper production. Due to its broad antimicrobial spectrum, ClO2, both in liquid and gas forms, is also used in wide range of sanitizing application (Sy et al., 2005a; USEPA, 1999). 2.3.1. Physical and chemical characteristics Chlorine dioxide gas phase is greenish yellow in color and has pungent odor. It is highly soluble in water, particularly in chilled water, and reported to be stable at room temperature, in the absence of light, over a period of days (Mueller and Willner, 1993). Its solubilities in water, at various temperatures, are graphically shown in Figure 2-2. Chlorine dioxide does not chlorinate, but exists as a dissolved gas in the solution (USEPA, 1999). Photochemical dissociation of ClO2 is an issue of great concern, as some researches regarded it to be associated with Antarctic O3 depletion (Solomon, 1999). Photoreactivity of ClO2 is outlined, in Scheme 2-1, below:  ClO + O  ClOO  OClO + һυ Cl + O2 (a)  Cl + O2 (b) (c) Scheme 2-1. Photochemical dissociation pathways of ClO2. Adapted from (Vaida and Simon, 1995). 24 The excitation of OClO, in the near-UV region, lead to; (a) its photochemical dissociation into ClO + O; (b) its photoisomerization to form an unstable ClOO (chlorine superoxide), which will further dissociate into atomic Cl + O2 (asymmetrical process); or (c) its dissociation into Cl + O2 (symmetrical process) (Vaida and Simon, 1995). The dissociated ClO molecule is considered by many researchers to be an ‘active form’ of chlorine that can cause ozone depletion through catalytic cycle (Solomon, 1999). Its isomerized form, ClOO, is thermodynamically more stable, but very reactive in the gas phase (Richard and Vaida, 1991). Photoreactivity of OClO and its partitioning in pathway (a)-(c) depends on phase environments. It reported to be wavelength-dependent if OClO is in a gas phase, while, in condensed-phases, like liquid solutions or solid matrices, it is independent of excitation wavelength (Dunn and Simon, 1992; Mueller and Willner, 1993; Vaida and Simon, 1995). 25 ClO2 partial pressure, pClO2 (atmClO2·C-1·mg/L-1) 2.5e-5 y  5.52 E  6e3.49 E  2 x R 2  0.9979 2.0e-5 1.5e-5 1.0e-5 5.0e-6 0.0 0 10 20 30 Temperature (oC) Figure 2-2. Solubility of ClO2 in water. Adapted from (Ishi, 1958). 26 40 - The end-products of ClO2 treatment and degradation are chloride (Cl ), chlorite - - (ClO2 ) and chlorate (ClO3 ). The concentrations of remaining ClO2, chloride, chlorite and chlorate after the reaction depend on the applied dosage, pH, etc (USEPA, 1999). In - drinking water, up to 70% of ClO2 is reduced to ClO2 and the rest is found in the forms - - of ClO3 and Cl , after oxidation reaction (USEPA, 1999). Half reactions of key Redox reactions are as shown: ClO2 + e - - ClO2 + 2H2O + 4e - o = - ClO2 E = 0.954V = Cl + 4OH - - o E = 0.760V Examples of ClO2 reactions through an electron transfer mechanism will be described in section 2.3.3. 2.3.2. Measurements of ClO2 concentration The reactive nature of ClO2 gas makes it difficult to find a suitable detection method to accurately measure its concentration (Dunn and Simon, 1992; Vaida and Simon, 1995). Typical analytical methods used to monitor ClO2 concentration and some important background information is summarized in Table 2-1. As the main applications of ClO2 are in pulp production and waste water treatment, most of the detection techniques were developed for measuring ClO2 in the 27 solution forms. To determine the concentration of gaseous ClO2, several methods require that the gas be captured by an appropriate solution, e.g. potassium iodide (KI), or distilled water. Most methods also require large volume of gas to be used as a sample, e.g. 7.5 L of air for determination of ClO2 in workplace atmosphere (Agency for Toxic Substances and Disease Registry Division of Toxicology and Environmental Medicine, 2004; Kaczur and Cawlfield, 1992; Occupational Safety & Health Administration, 2010; USEPA, 1999). 2.3.3. Microbial activity ClO2 is recognized as a strong antimicrobial agent against bacterial, viral, and protozoan pathogens. Its inactivating efficiency is generally reported to be equal to or higher than that of Cl2 on a mass-dose basis, but less than that of ozone (USEPA, 1999). Several researches have reported the antimicrobial effects of ClO2 against important microorganisms as listed in Table 2-2 (last updated: August, 2010). Numbers of research also reported antimicrobial capacities of ClO2 on biofilm network of bacteria, such as B. cereus, L. monocytogenes, and E. coli O157:H7 (Table 22) (Annous et al., 2006; Annous et al., 2005; Kreske et al., 2006b; Lindsay et al., 2002; Ölmez and Temur, 2010; Ryu and Beuchat, 2005). A biofilm is an aggregation of microorganisms by a matrix of extracellular polymeric substances (EPS). Biofilm is an important survival mechanism for microorganisms, since inactivation of pathogenic microorganisms by antimicrobial agents is inevitably affected by the accessibility of the 28 surfaces where the microorganisms resided in. The film creates a protected environment, so that the bacteria living in the biofilm can tolerate harsh environmental conditions better (Ölmez and Temur, 2010; Simões et al., 2010). Netramai et al studies antimicrobial effects of ClO2 gas on biofilm of E. coli O157:H7 formed on Romaine lettuce leaf by treating inoculated lettuce leaf sample with 0.2 mg ClO2/L, for 30 and 60 min (see Appendix 1 for more detail). The treatments gave ≤ 1.57 ± 0.05 log CFU/g sample log reduction of pathogen. The comparison on topographic characteristics, between untreated and ClO2-treated surfaces, using Scanning Electron Microscope (SEM), showed the biofilm of E. coli O157:H7 being partially destroyed and/or disappeared, in many area of the surface as shown in Figure 2-3 and Figure 2-4 (Netramai et al., 2010). 29 Table 2-1. Analytical methods for ClO2 and related compounds (Agency for Toxic Substances and Disease Registry Division of Toxicology and Environmental Medicine, 2004; Henderson, 1999; Kaczur and Cawlfield, 1992; USEPA, 1999) Method 1 Principle DPD Colorimetric method (SM-4500-ClO2D) 1 DPD-glycine Colorimetric method (SM-4500-ClO2D) 1 DPD-FAS Titrimetric method (SM-4500-ClO2D ) 1 5-Step Amperometric method (4500-ClO2E) 1 Colorimetric method, in which, ClO2 reacts with DPD (N,N-diethyl-p-phenylenediamine) to form colored product. Standard method for chemical analysis of water and wastes. Interferences = Mn2 , Cl2, and other related oxidants + - Same as SM-4500-ClO2G, but interference from free Slowly reacts with ClO2 and other oxidants Cl is solved (up to 6 ppm Cl ) by adding glycine to 2 2 the sample. Standard method for chemical analysis of water and wastes. DPD color titration with standard ferrous ammonium sulfate (FAS). Interferences = Iron and other oxidants 1 Amperometric titration (4500-ClO2C) Disadvantage 1 - Interferences = Free halogens, organic 2+ chloramines, and Cu - Skilled analyst is preferred Iodide (I ) oxidation with pH control. Practical method. Iodide (I ) oxidation with pH control and gas purging step (more detail in Appendix 4). SM = Standard Method 30 Table 2-1. (cont’d) 1 Principle Disadvantage Thiosulfate titration Colorimetric method, in which, I2 reacts with sodium thiosulfate (more detail in Appendix 2). Skilled analyst is preferred Ion chromatography Ion chromatography with AS9 column and external standard Chloramines, ClO2 , OCl , and HOCl are undetectable Toxic gas vapor detector tube Reduction-oxidation (Redox) reaction Affected by temperature Electochemical gas sensor Reduction-oxidation (Redox) reaction Affected by humidity (condensation) Method (EPA Method 300.0 or 1 300.1) 1 SM = Standard Method 31 - - Figure 2-3. Biofilm of E. coli O157:H7 on un-treated lettuce surfaces; A – Visible bacteria cells; and B – Continuous layer of biofilm. Adapted from (Netramai et al., 2010). Figure 2-4. Biofilm of E. coli O157:H7 on ClO2-treated lettuce surfaces; C – Debris observed after ClO2 treatment; and D – Partially destroyed biofilm. Adapted from (Netramai et al., 2010). 32 Chlorine dioxide disinfects by oxidation through the one-electron transfer - mechanism in which it is usually reduced to ClO2 (USEPA, 1999). However, its primary mode of microorganism inactivation is not yet well established, and seems to depend on the type of target microorganism. Several studies reported two possible primary inactivation mechanisms to destroy microorganisms of ClO2 (USEPA, 1999). The first mechanism involves the reaction between ClO2 and, nucleic acid and/or peripheral structures, i.e. inactivates microorganisms primarily through the disruption of protein synthesis (Bernarde et al., 1967). In the second mechanism, microorganisms are inactivated mainly from the alteration of protein (in outer membrane) which alters the permeability of the cell membrane (Aieta and Berg, 1986). In 2006, Ison et al reported the first-order reaction between ClO2 and cysteine (CSH). Cysteine is one of the amino acid that maintains cellular redox potentials. The o reaction rate increased as the pH increases from 2.7 to 9.5 at 25 C. The purposed oxidation mechanism is through an electron transfer from CS- to ClO2. The reactive intermediates undergo two different pathways depending on the pH of the system. At low - pH, the major end-products are cysteic acid (CSO3H) and Cl , and, at high pH, the major - end-products are cystine (CSSC) and chlorite ion (ClO2 ), as shown in Scheme 2-2. Chlorite ion can further reacts with CSH at around neutral pH (6.7), but the rate of reaction is slower than the ClO2/CSH reaction by 6-order of magnitude. Chlorine dioxide - 8 can also react with glutathione anion (GS ) with a second-rate constant of 1.40 × 10 M 33 - 1 -1 s . Glutathione is a tripeptide of glycine, cysteine, and glutamic acid. It presents in animals, plants, and many bacteria, and functions in both chemical and enzymatic reactions (Ison et al., 2006). The studies on the reaction of ClO2 with other amino acid, i.e. tyrosine and tryptophan are also reported (Tan et al., 1987). - +  CS + H CS + ClO2  CS + ClO2 CS + ClO2  cysteinyl-ClO2 adduct CSH - (a) - (b) (c) Low pH pathway: cysteinyl-ClO2 adduct + H2O  ClOH + CSO2H CSO2H + ClOH  Cl + CSO3H  ClO2 + CSSC  CSSC + Cl + H2O - (d) (e) High pH pathway: - cysteinyl-ClO2 adduct + CS - ClO2 + CSH - - (f) (g) Scheme 2-2. Purposed mechanism for the initial reactions between ClO2 and CSH and subsequent decay of the cysteinyl-ClO2 adduct. Adapted from (Ison et al., 2006). 34 Table 2-2. Summary of research studies on antimicrobial efficiencies of ClO2 for food products and other surfaces. Microorganism Alicyclobacillus acidoterrestris (spore) Bacillus cereus Bacillus cereus (spore) Bacillus cereus (cell and spore) Bacillus thuringiensis (spore) Cryptosporidium parvum Encephalitozoon intestinalis (spore) Enterobacter sakazakii Escherichia coli 1 Food Product / Other surface Stainless steel Apple Stainless steel Stainless steel Stainless steel Apple Stainless steel Basil Lettuce Basil Lettuce Apple Lettuce Tomato Apple Treatment 1 Result (log10 reduction) 2 Reference 50-200 ppm (l) for 1-2 min at 40o 90 C 10-200 μm/mL (l) for 5 min 100-200 μm/mL (l) for 5 min 10-200 μm/mL (l) for 5 min 100-200 μm/mL (l) for 5 min 50-200 μm/mL (l) for 1-5 min ≤ 2.17 log CFU/cm (Podolak et al., 2009) ≤ 3.79 log CFU/fruit ≤ 3.71 log CFU/coupon ≤ 2.55 log CFU/coupon ≤ 4.48 log CFU/coupon ≤ 4.0 log CFU/coupon 10-200 μm/mL (l) for 5 min 10-200 μm/mL (l) for 5 min 4.1 mg/L (g) for 20 min 4.1 mg/L (g) for 20 min 4.1 mg/L (g) for 20 min 4.1 mg/L (g) for 20 min 10-100 μg/mL (l) for 1-5 min 10-100 μg/mL (l) for 1-5 min 10-100 μg/mL (l) for 1-5 min 0.03-0.30 ppm (g) for 1-20 hours ≤ 4.25 log CFU/fruit ≤ 2.71 log CFU/coupon 2.6 log CFU/g 3.3 log CFU/g 3.6 log CFU/g 3.6 log CFU/g ≥ 5.46 log CFU/fruit ≥ 4.81 log CFU/piece ≥ 5.31 log CFU/fruit ≤ 5 log CFU/g (Kreske et al., 2006a) (Kreske et al., 2006b) (Kreske et al., 2006a) (Kreske et al., 2006b) (Ryu and Beuchat, 2005) (Kreske et al., 2006a) (Kreske et al., 2006a) (Ortega et al., 2008) (Ortega et al., 2008) (Ortega et al., 2008) (Ortega et al., 2008) (Kim et al., 2006) (Kim et al., 2006) (Kim et al., 2006) (Sapers et al., 2003) (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 35 Table 2-2. (cont’d) Microorganism Food Product / Other surface Escherichia coli O157:H7 Alfalfa seed Alfalfa sprout Apple Apple (calyx, stem cavity and skin) Apple (fresh cut) Blueberries Cabbage (fresh cut) Cantaloupe Carrot Carrot (fresh cut) 1 1 Result (log10 reduction) Reference 10-50 mg/L (l) for 3-10 min 50 mg/L (l) for 1-10 min 50 mg/L (l) + 0.5 g/100 mL fumaric acid for 1-10 min 5-40 ppm (l) for 3-10 min + 170-kHz Ultrasonication 3-5 ppm (l) for 5 min + storage at o 4 C ≤ 1.22 log CFU/g ≤ 2.37 log CFU/g ≤ 3.96 log CFU/g (Singh, 2003) (Kim et al., 2009) (Kim et al., 2009) ≤ 3.9 log CFU/g (Huang et al., 2006) ≤ 5 log CFU/g (Rodgers et al., 2004) 3-12 mg/L (g) for 10-30 min ≤ 8 log CFU/site (Du et al., 2003) 3-5 ppm (l) for 5 min + storage at o 4 C 4 mg/L (g) for 12 hours 1.4-4.1 mg/L (g) for 6-21 min ≤ 5 log CFU/g 3-5 ppm (l) for 5 min + storage at o 4 C 5-20 mg/L (l) for 1-15 min 0.5-1.0 mg/L (g) for 5-15 min 1.4-4.1 mg/L (g) for 6-21 min ≤ 5 log CFU/g Treatment (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 36 4.5 log CFU/g ≤ 3.13 log CFU/g ≤ 1.39 log CFU/g ≤ 3.08 log CFU/g ≤ 5.62 log CFU/g (Rodgers et al., 2004) (Popa et al., 2007) (Rodgers et al., 2004) (Rodgers et al., 2004) (Singh et al., 2002) (Singh et al., 2002) (Sy et al., 2005b) Table 2-2. (cont’d) Microorganism Escherichia coli O157:H7 (cont’d) Food Product / Other surface Green pepper Lettuce Lettuce (fresh cut) Potato Strawberries Lactic acid bacteria 1 Rib eye steak Treatment 1 Result (log10 reduction) 0.15-1.2 mg/L (g) for 30 min 5-40 ppm (l) for 3-10 min + 170-kHz Ultrasonication 3-5 ppm (l) for 5 min + storage at o 4 C 4.3-8.7 mg/L (g) for 30-180 min 5-50 ppm (l) for 10 min 3-5 ppm (l) for 5 min + storage at o 4 C 1.4-4.1 mg/L (g) for 6-21 min 5-20 mg/L (l) for 1-15 min 0.5-1.0 mg/L (g) for 5-15 min 9 ppm (l) for 30-300 min with spray washing 3-5 ppm (l) for 5 min + storage at o 4 C 0.2-4.0 mg/L (g) for 15-30 min 0.6-3.0 mg/L (g) for 10 min 100-200 ppm (l) for 2 min 30-100 ppm (l) for 2 min + vacuum packaging for 4 weeks (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 37 Reference ≤ 7.3 log CFU/pepper ≤ 2.5 log CFU/g (Han et al., 2000) (Huang et al., 2006) ≤ 5 log CFU/g (Rodgers et al., 2004) ≤ 6.9 log CFU/g ≤ 1.44 log CFU/g ≤ 5 log CFU/g (Lee et al., 2004) (Kim et al., 2008) (Rodgers et al., 2004) ≤ 1.57 log CFU/g ≤ 0.90 log CFU/g ≤ 2.31 log CFU/g ≤ 1.4 log CFU/fruit (Sy et al., 2005b) (Singh et al., 2002) (Singh et al., 2002) (Rodgers et al., 2004) ≤ 5 log CFU/g (Park et al., 2008) ≤ 5 log CFU/fruit ≥ 5 log CFU/fruit ≤ 2 log CFU/fruit (Han et al., 2004) (Han et al., 2004) (Lukasik et al., 2003) (Unda et al., 2006) 2 ≥ 1.0 log CFU/cm Table 2-2. (cont’d) Microorganism Food Product / Other surface Treatment 1 Result (log10 reduction) Reference Lactobacillus buchneri Stainless steel coated with epoxy 2-14 mg/L (g) for 5-120 min ≤ 6 log CFU (Han et al., 1999) Leuconostoc mesenteroides Stainless steel coated with epoxy Alfalfa sprout 8-10 mg/L (g) for 30 min ≤ 6 log CFU (Han et al., 1999) 50 mg/L (l) for 1-10 min 50 mg/L (l) + 0.5 g/100 mL fumaric acid for 1-10 min 3-5 ppm (l) for 5 min + storage at o 4 C 3-5 ppm (l) for 5 min + storage at o 4 C 1.4-4.1 mg/L (g) for 10-29 min ≤ 2.36 log CFU/g ≤ 3.69 log CFU/g (Kim et al., 2009) (Kim et al., 2009) ≤ 5 log CFU/g (Rodgers et al., 2004) ≤ 5 log CFU/g (Rodgers et al., 2004) ≤ 5.88 log CFU/g (Sy et al., 2005b) ≤ 5 log CFU/g (Rodgers et al., 2004) ≤ 5.4 log CFU/g ≤ 5 log CFU/g (Lee et al., 2004) (Rodgers et al., 2004) ≤ 1.20 log CFU/g (Kim et al., 2008) Listeria monocytogenes Apple Cantaloupe Carrot (fresh cut) Lettuce Lettuce (fresh cut) 1 3-5 ppm (l) for 5 min + storage at o 4 C 4.3-8.7 mg/L (g) for 30-180 min 3-5 ppm (l) for 5 min + storage at o 4 C 5-50 ppm (l) for 10 min (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 38 Table 2-2. (cont’d) Microorganism Food Product / Other surface Listeria monocytogenes (cont’d) Lettuce (fresh cut) (cont’d) Mungbean sprout Strawberries Pseudomonas Salmonella spp. Stainless steel Apple Bell pepper Blueberries Cabbage (fresh cut) Carrot (fresh cut) Lettuce Lettuce (fresh cut) Onion 1 Treatment 1 Result (log10 reduction) Reference ≤ 1.53 log CFU/g ≤ 5 log CFU/g (Sy et al., 2005b) (Rodgers et al., 2004) ≤ 2.08 log CFU/g (Jin and Lee, 2007) ≥ 5 log CFU/g (Han et al., 2004) 1.4-4.1 mg/L (g) for 10-29 min 3-5 ppm (l) for 5 min + storage at o 4 C 100 ppm (l) for 5 min + MAP in 7 days 0.2-4.0 mg/L (g) for 15-30 min or 0.6-3.0 mg/L for 10 min 100-200 μm/mL (l) for 5 min 5-40 ppm (l) for 3-10 min + Ultrasonication 1.4-4.1 mg/L (g) for 6-25 min 100 mg (g) for 1 hour 4-8 mg/L (g) for 30-120 min 4 mg/L (g) for 12 hours 1.4-4.1 mg/L (g) for 10-31 min ≥ 6.15 log CFU/coupon ≤ 4.3 log CFU/g (Kreske et al., 2006b) (Huang et al., 2006) ≤ 4.21 log CFU/g ≤ 5.97 log CFU/fruit ≤ 3.67 log CFU/g 3.8 log CFU/g ≤ 4.42 log CFU/g (Sy et al., 2005b) (Yuk et al., 2006) (Sy et al., 2005a) (Popa et al., 2007) (Sy et al., 2005b) 1.4-4.1 mg/L (g) for 10-31 min ≤ 5.15 log CFU/g (Sy et al., 2005b) 5-40 ppm (l) for 3-10 min + 170-kHz Ultrasonication 1.4-4.1 mg/L (g) for 10-31 min up to 3.5 log CFU/g (Huang et al., 2006) up to 1.58 log CFU/g (Sy et al., 2005b) 1.4-4.1 mg/L (g) for 5-20 min up to 1.94 log CFU/g (Sy et al., 2005b) (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 39 Table 2-2. (cont’d) Microorganism Salmonella spp. (cont’d) Salmonella Montevideo Salmonella Typhimurium Food Product / Other surface Peach Raspberries Strawberries Tomato Strawberries Alfalfa sprout Chicken (breast) Lettuce Mungbean sprout Potato Tomato 1 Treatment 1 Result (log10 reduction) 1.4-4.1 mg/L (g) for 5-20 min 4-8 mg/L (g) for 30-120 min 4-8 mg/L (g) for 30-120 min 1.4-4.1 mg/L (g) for 5-20 min 5 ppm (l) + spray washing for 10-60 sec 100-200 ppm (l) for 2 min 50 mg/L (l) for 1-10 min 50 mg/L (l) + 0.5 g/100 mL fumaric acid for 1-10 min 2.25 mg (g) in 22 days and 6.6 mg (g) in 26 h. 4.3-8.7 mg/L (g) for 30-180 min 5-50 ppm (l) for 10 min 100 ppm (l) for 5 min + MAP in 7 days 9 ppm (l) for 30-300 min with spray washing Up to 1 mg/kg fruit (g) for 2 hours (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 40 Reference up to 3.23 log CFU/g up to 1.54 log CFU/g up to 3.76 log CFU/g up to 4.33 log CFU/g (Sy et al., 2005b) (Sy et al., 2005a) (Sy et al., 2005a) (Sy et al., 2005b) (Pao et al., 2009) 2 up to 5.6 log CFU/cm less than 2 log CFU/fruit (Lukasik et al., 2003) ≤ 2.23 log CFU/g (Kim et al., 2009) ≤ 3.57 log CFU/g (Kim et al., 2009) 1 log CFU/breast less than package without ClO2 after 15 days of storage ≤ 5.4 log CFU/g ≤ 1.95 log CFU/g ≤ 2.76 log CFU/g (Ellis et al., 2006) ≤ 1.9 log CFU/fruit (Park et al., 2008) ≤ 7 log CFU/fruit (Mahovic et al., 2009) (Lee et al., 2004) (Kim et al., 2008) (Jin and Lee, 2007) Table 2-2. (cont’d) Microorganism Food Product / Other surface Treatment Total aerobic plate count Crawfish Mungbean sprout Potato Rib eye steak Shrimp Molds Yeasts 1 Blueberries Potato Stainless steel coated with epoxy Blueberries Potato 1 Result (log10 reduction) Reference 10-40 mg/L (l) for 2 min ≤ 4 log CFU high-pressure water spray at 600 psi for 30 sec + 10-40 mg/L (l) for 2 min 100 ppm (l) for 5 min + MAP in 7 days 9 ppm (l) for 30-300 min with spray washing 30-100 ppm (l) for 2 min + vacuum packaging for 4 weeks ≤ 4.5 log CFU ≤ 1.35 log CFU/g (Andrews et al., 2002) (Andrews et al., 2002) (Jin and Lee, 2007) ≤ 1.1 log CFU/fruit (Park et al., 2008) ≥ 1.0 log CFU/cm 2 (Unda et al., 2006) high-pressure water spray at 600 psi for 30 sec + 10-40 mg/L (l) for 2 min 4 mg/L (g) for 12 hours 9 ppm (l) for 30-300 min with spray washing 8-10 mg/L (g) for 10-30 min ≤ 4 log CFU 3.0 log CFU/g ≤ 0.9 log CFU/fruit (Andrews et al., 2002) (Popa et al., 2007) (Park et al., 2008) ≤ 4 log CFU (Han et al., 1999) 3.2 log CFU/g ≤ 1.1 log CFU/fruit (Popa et al., 2007) (Park et al., 2008) 4 mg/L (g) for 12 hours 9 ppm (l) for 30-300 min with spray washing (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 41 Table 2-2. (cont’d) Microorganism Yeasts (cont’d) Yeasts and Molds Penicillium expansum (spore) Poliovirus 1 1 Food Product / Other surface 1 Result (log10 reduction) Reference 8-10 mg/L (g) for 10-30 min ≤ 4 log CFU (Han et al., 1999) 1.4-4.1 mg/L (g) for 6-25 min 4-8 mg/L (g) for 30-120 min ≤ 1.68 log CFU/g ≤ 2.78 log CFU/g (Sy et al., 2005b) (Sy et al., 2005a) 1.4-4.1 mg/L (g) for 5-20 min 1.4-4.1 mg/L (g) for 5-20 min 4-8 mg/L (g) for 30-120 min ≤ 0.22 log CFU/g ≤ 2.65 log CFU/g ≤ 3.18 log CFU/g (Sy et al., 2005b) (Sy et al., 2005b) (Sy et al., 2005a) Treatment Stainless steel coated with epoxy Apple Blueberries (calyx, stem cavity and skin) Onion Peach Raspberries (calyx, stem cavity and skin) Strawberries (calyx, stem cavity and skin) Tomato (Suspension) 4-8 mg/L (g) for 30-120 min ≤ 4.16 log CFU/g (Sy et al., 2005a) 1.4-4.1 mg/L (g) for 6-25 min 3 ppm (l) for 30-300 sec ≤ 1.16 log CFU/g ≤ 4 log CFU/mL (Sy et al., 2005b) (Okull et al., 2006) Strawberries 100-200 ppm (l) for 2 min ≤ 2 log CFU/fruit (Lukasik et al., 2003) (g) = ClO2 in a gas phase; (l) = ClO2 in a liquid/solution form. 42 Inactivation efficiency of ClO2 is not significantly affected by change of pH (USEPA, 1999), however, its antimicrobial capacities are influenced by change in temperature. Benarde et al investigated the effects of temperature on bactericidal activities of ClO2 by exposing E. coli suspension to 0.25-0.75 mgClO2/L solutions, for 5o 300 seconds, at 5-32 C. They reported increases in killing rates as ClO2 concentrations and/or testing temperatures increased, for example, to achieve 2 log10 CFU reduction o o (99% kill) at concentration of 0.25mg/L required 110 sec at 5 C, 74 sec at 10 C, 41 sec o o at 20 C, or 16 sec at 30 C, while, to obtain the same inactivation level, at 0.75 mg/L, o o required 60 sec at 5 C or 14 sec at 30 C (Benarde et al., 1967). 2.3.4. Applications Chlorine dioxide is used as a bleaching agent in many applications, especially for the paper, pulp, and textile industries (Department of the Environment Water Heritage and the Arts, 2005; Kaczur and Cawlfield, 1992). As registered bactericide, fungicide and algaecide biocide, ClO2 is approved by the U.S. Environmental Protection Agency (EPA) in both liquid and gas forms to be used as a sanitizing agent for food processing equipment, and pharmaceutical and factory tools, as well as, in potable- and waste water treatment (Department of Environment Food and Rural Affairs (DEFRA), 2003; Kaczur and Cawlfield, 1992; Kim et al., 1999; USEPA, 1999).One of the major uses of ClO2 is as a disinfectant for public water system as it causes less organoleptic effect and less 43 toxic by-products than those caused by Cl2 (Department of Environment Food and Rural Affairs (DEFRA), 2003; USEPA, 1999). Specifically, in food applications, ClO2 is reported to have antimicrobial effects against many important spoilage and pathogenic microorganisms (Table 2-2), such as E. coli O157:H7, Listeria monocytogenes, Salmonella spp., yeasts and molds, that reside on the surface of food products such as fresh and fresh-cut fruits and vegetables, as well as on fresh meats (Sy et al., 2005b; Taylor et al., 1999; Yuk et al., 2005). While ClO2 is often used as a sanitizing solution for meat, fruits, and vegetables, (Kim et al., 1999; Yuk et al., 2005), its gaseous form is gaining interest as an antimicrobial agent, for vapor-phase decontamination, in cleaning food-contact surface (Han et al., 1999), treating produce, such as green peppers, blueberries, and apples, before packaging them (Sapers et al., 2003; Sy et al., 2005a; Vermeiren et al., 1999), and sanitizing products, like chicken, directly inside their packages (Ellis et al., 2006). The introduction of gaseous ClO2 within the food packaging systems is often by means of sachet, either slow- or fast-release (Ellis et al., 2006; Shin, 2007); however, in 2001, the FDA approved the incorporation of ClO2 precursors within food packaging materials to be used as packaging material for meats, poultry, and seafood (USFDA, 2001). Most of the studies regarding food packaging applications of ClO2 gas focus on its sanitizing effects on particular microorganism and/or specific type of perishable food (Kaczur and Cawlfield, 1992; Sapers et al., 2003; Sy et al., 2005a). Only a few studies 44 consider its effects on the properties of the food packaging systems, as well (Ozen, 2000; Shin et al., 2006). 2.4. Effects of gaseous oxidizing agents on packaging materials Protection and containment are two of the most important roles of the package. The ability to protect and contain the product that the packaging provides depends on the stability of the material to the environmental factors, such as temperature, humidity, and exposure to chemical compounds. In the case where packaging is going to be considered as a strategy for the release and application of ClO2 gas, the interaction and compatibility between packaging materials and gaseous ClO2, such as the mass transfer properties of ClO2 and its effects on packaging integrities and performance are needed to be determined since these will impact the selection of materials that compatible with such gas, for particular application. The possible effects of exposing polymeric packaging materials to oxidizing agents will be discussed further in the next two sections. 2.4.1. Mass transfer of packaging materials Plastic packaging materials, unlike glass, metals, etc., are relatively permeable to small molecules such as permanent gases, water vapor, organic vapors, and liquids. Polymeric packaging materials provide a broad range of mass transfer characteristics for particular gas or vapor, from high to low barrier values. The specific barrier requirements of the package system will depend upon the product’s characteristics and the intended end-use application (Taub and Singh, 1998). 45 Permeation includes the transfer through the package of molecules from the product to the external storage environment or from the environment to the product. The mass transfer process provides the basis for further molecular activities within the package system, which can result in changes in the product, as well as physical damage to the package, or both (Selke et al., 2004; Van Krevelen, 1997). The permeation process starts with the sorption of the penetrant onto the polymer surface. Then, the penetrant travels through the polymer matrix by diffusion with the difference in chemical potential as a driving force, and desorp from the polymer at the other side of the film. Thus, permeation behavior of the polymer depends on its solubility and diffusion properties (Selke et al., 2004; Van Krevelen, 1997). The mass transfer phenomenon of ‘simple gases’, through a polymer film or sheet is a measure of the steady-state transfer rate of the permeant, and is normally expressed as the permeability coefficient (P). The relationship between P, solubility coefficient (S) and diffusion coefficient (D) usually follows Equation 1. P = D·S (1) The permeability coefficient defines as the amount of gas traveling through the polymeric material with particular thickness, per unit area, per second and at specific pressure gradient. The diffusion coefficient indicates how fast the permeant can move through the polymer matrix. It reported as the amount of gas, per unit area, traveling through a 46 material per second, at a particular concentration gradient as expressed by Fick’s first law: dm dt =  dc  D  A   dx  (2) Diffusion of simple gases usually depends only on temperature, not on concentration or time of exposure. This behavior is called Fickian (Selke et al., 2004; Van Krevelen, 1997). Solubility coefficient defines as the amount of gas per unit volume of the polymer at equilibrium for particular partial pressure according to Henry’s law: c = S·p (3) Several techniques are available for P, D, and S determinations. P and D are usually measured either by isostatic or quasi-isostatic methods (Selke et al., 2004; Van Krevelen, 1997). In isostatic procedure, both values are obtained from a continuous monitoring of the mass transfer process of the permeant through the material. While, in quasi-isostatic method, the accumulated amount of permeant permeated through the material is quantified at predetermined time intervals (Selke et al., 2004). S and D can be determined by gravimetric method, i.e. monitoring the weight change of the film during a sorption experiment. Generally, S values of simple gases in polymeric materials are low, thus it not easy to accurately determine (Matteucci et al., 2006). 47 The mass transfer of simple gases, especially at low concentration of permeant where there are no strong penetrant/polymer interaction, normally show this ideal permeation behavior, i.e. the solubility follows Henry’s law and diffusion follows Fick’s first law (Van Krevelen, 1997). However, if the significant interactions between penetrant and polymeric material, as in the cases of organic vapor, and/or between the penetrant themselves exist, the sorption behavior might be affected, i.e. divert from Henry’s law (Selke et al., 2004; Van Krevelen, 1997). 2.4.1.1. Isostatic permeation technique In this study, P, D, and S values of ClO2 for polymeric packaging materials were determined using a permeation approach that follows an isostatic method. The diffusion process of gaseous ClO2 through the materials was assumed to be independent of the permeant’s concentration and the polymer relaxation, which is known as Fickian diffusion (Matteucci et al., 2006; Selke et al., 2004). The diffusion coefficient was calculated from the mass transfer profile in the transient region that leads to the steadystate of the mass transfer, where the concentration gradient of the permeant across the film remains constant. The quantity of permeated ClO2 gas (kg) at time t (s) was then plotted to obtain the permeation curve (Figure 2-5). The P value was calculated from the flow rate of the permeated ClO2 gas at steady-state, Fss (kg/s). 48   Figure 2-5. Mass transfer profile as obtained by isostatic method The ratio of permeant flow as a function of time can be described as followed: Ft F   4   2       4 Dt      n 2 2    exp    4 Dt  n 1,3,5    (4) where Ft is a permeated gas flow rate at transient time t (kg/s), Fss is a permeated gas flow rate at steady state (kg/s),  is the film thickness (m), and D is diffusion. Equation (4) can be simplified to: ϕ = Ft F  4  1/ 2   exp x  X    2 X =  4 Dt (5) (6) 49 At steady state, ϕ = 1 and D is assumed constant, the P can be calculated from: P = Fss  p  A  p (7) = p2 – p1 (8) The diffusion coefficient, D, was obtained from the transient state according to the following calculations: D  2 7.2t1 / 2 (9) 2 where A is the surface area of the film sample (m ), and p is the partial pressure gradient of ClO2 between the lower half (p2) and upper half (p1) of the permeability cell. Since there is a constant removal of permeated ClO2 from the upper chamber, p1 is equal to 0.00 Pa. The partial pressure of gaseous ClO2 over the solution placed in the lower chamber was calculated from Henry’s solubility coefficient mentioned earlier and is 2 equal to 3.65 × 10 Pa (Ishi, 1958; USEPA, 1999). The t 1/2 is the time taken to reach a flow rate equivalent to half of Fss. The S values can be calculated using the “solution-diffusion” model, i.e. Equation 1 (Matteucci et al., 2006). 50 2.4.1.2. Factors affecting permeation The variables affecting mass transfer can be grouped into compositional variables and environmental and geometric variables. The examples of compositional variables are chemical composition of the packaging material and penetrant, morphology of the polymer, concentration of the penetrant, and presence of co-permeant. Important environmental and geometric variables are temperature, relative humidity (RH) and packaging geometry (Selke et al., 2004; Van Krevelen, 1997). P, D, and S are temperature dependent and can be described by a Van’t HoffArrhenius equation. P(T) D(T) S(T) = E  P P0 exp RT (10) = E  D D0 exp RT (11) = H s S 0 exp RT (12)  From Equation 1 and 10 –12: H S S0 × D0 × exp E D  P(T) = D(T) × S(T) = (13) where H S is molar heat of sorption, E D is activation energy of diffusion, and E P is apparent activation energy of permeation (Van Krevelen, 1997). 51 P0 , D0 , S 0 , H S , E D and E P can be estimated from the measurement of P(T), D(T), and S(T) at different testing temperature (Van Krevelen, 1997). Raising the testing temperature gives the studied polymer the energy and enables the segmental mobility within the polymer structure. This facilitates the diffusion of the penetrant molecule, thus increases the mass transfer of the system (Selke et al., 2004; Van Krevelen, 1997). Barrier properties of polymeric materials are strongly influenced by their chemical and physical characteristics, such as crystallinity, polarity, free volume, orientation, and cohesive energy density (Yang and Li, 1997). Polymer’s physical properties like crystallinity affect both diffusion and solubility. Since crystallization decrease amorphous area of the material, less volume is available for the gas to move through. Crystalline also impede the movement of the gas molecules by creating the torturous path they will have to travel (Schnabel, 1992; Selke et al., 2004; Van Krevelen, 1997). As diffusion coefficients are particularly affected by the polymer’s glass transition temperature (Tg) and free volume properties (Matteucci et al., 2006). The free volume (FV), the unoccupied spaces between molecules in the polymers’ structures (Selke et al., 2004), in the matrix of the semi-crystalline polymer, at particular temperature, can be calculated from its fractional free volume (FFV): 1 FFV = V  V0 V =   1.3Vw 1  52 (14) 3 where V is the specific volume (m /kg) of the amorphous polymer at the testing 3 temperature and V0 is a specific volume (m /kg) at 0 K. The FFV can also be estimated 3 3 from polymer’s density, ρ (m /kg) and its Van der Waals volume, Vw (m /mol) as shown in Equation 14. Vw values quoted from literature (Van Krevelen, 1997), is the space occupied by the polymer molecule, hence, a barrier for the penetrant (Recio et al., 2008; Van Krevelen, 1997). The crystalline region serves as the impermeable volume for the diffusion of the permeant, creates the torturous path for the molecules, and somewhat restricts the polymer chain mobility of the structure in the amorphous region, impeding the mass transfer process. Such regions are also excluded from the sorption process. Since there is a negligible amount of unoccupied space in the crystalline region of the material (Van Krevelen, 1997), the FV values of the semi-crystalline polymeric material can be corrected by: FV =  %Crystallinity    FFV 1  100   (15) Even though free volume property is normally regarded as an independent factor, it can also be considered that a formation/existence of free volume will be affected by the intermolecular forces within the polymer molecules described as cohesive energy density (CED). The material with high polarity has high CED (Alentiev and Yampolskii, 2002; Shimazu et al., 1999; Yang and Li, 1997). CED can be estimated based on cohesive energy ( E coh ) and molar volume ( Vm ) data as followed: 53 Ecoh Vm CED = (16) 3 where the E coh (kJ/mol) and Vm (m /mol) were estimated by mean of group contribution method (Van Krevelen, 1997). No confirmed relation was reported between CED and solubility coefficient (S), though the polymers with high FV values usually have high gas solubilities. This could be due to the fact that S values of a particular gas in different polymeric materials vary in a narrower range as compared to their D values (Alentiev and Yampolskii, 2002; Van Krevelen, 1997). Polymers with high CED and low FV tend to have higher barrier to a particular compound as compared to polymers with low CED and high FV, as they both influence molecular packing properties of the materials (Alentiev and Yampolskii, 2002; Van Krevelen, 1997).   2.4.1.3. Mass transfer of gaseous ClO2 Ozen (2000) studied mass transfer of gaseous ClO2 through polymeric materials, using a quasi-isostatic method, by exposing 100 mg/L ClO2 to the film sample in a threecompartment permeability cell, and utilizing an amperometric titration method as a o detection technique. The reported P values of ClO2 gas, at 20 C, for linear low-density poly(ethylene), LLDPE, oriented poly(propylene), OPP, and biaxially oriented nylon (BON) were 7.62 x 10 -16 -17 , 6.21 x 10 -17 , and 2.34 x 10 -2 -1 -1 KgClO2mm s Pa , respectively. Since i) the steady state was reached too fast and ii) the delay response of 54 the detection system used, prevented the assessment of the unsteady region for the calculation of the diffusion coefficient (Ozen, 2000). 2.4.2. Polymer degradation and its effects on packaging performance Polymer degradation can be initiated by several factors; thermal, radiation chemical, mechanical, photochemical, biological, and chemical degradations. The process causes the alterations in the materials’ functionality. More than one factor can simultaneously trigger the changes, for example, thermoplastic polymers are subjected to oxidative degradation during their processing due to high heat, applied forces and atmospheric oxygen (Schnabel, 1992). Chemical reactions normally found during the degradation process of macromolecules are categorized as 1) single step reaction in which the reaction rate is directly proportional to the initiation rate, and 2) chain reaction in which the reaction, once initiated, is self-propagating (Schnabel, 1992). Example of single step reactions are Norrish Type II reaction in ketone polymers initiated by photochemical process and enzymatic attack of glycosidic linkage caused by biological factor. This type of process can result in significant changes in physical and mechanical properties of the linear polymers as their properties are influenced by average molecular weight (MW) (Schnabel, 1992). In chain reaction process, the products from single initiation step can start several propagation reactions with other molecules, resulting in manipulation of deterioration process (Schnabel, 1992). Among others, autoxidation reactions initiated by heat, light, forces, or chemical exposure are quite common in polymers. The initiation steps result in 55 the formation of free radicals, which, in the presence of O2, further promote the chain reaction. Scheme 2-3 shows the free radical mechanism of autooxidation in linear polymers: R + PH  RH + P (a) P + O2  P-O-O (b)  P-O-O-H + P (c) P + P  products (d) R + P  products (e) Termination: R + P-O-O  products (f) P + P-O-O  products (g) 2P-O-O  2P-O + O2 (h) Initiation: Propagation: P-O-O + PH Scheme 2-3. Free radical mechanism of autoxidation in linear polymers; R is free radical generated by deterioration process of the polymer; and PH is the polymer. 2.4.2.1. Chemical degradation Chemical degradation defines as “processes which are induced under the influence of chemicals (e.g. acids, bases, solvents, reactive gases etc.) brought into contact with polymers” (Schnabel, 1992). A significant change usually occurs at elevated temperature due to the rather high activation energy of such processes. The degree of changes also depends on the polymers’ stabilities against particular chemical and the polymers’ characteristics, e.g. the presence of crystalline or unsaturated bonds in the structure (Schnabel, 1992). 56 For the polymers that contain hetero (non-carbon) atoms, e.g. O, N, or halogen, in their structure, the rupture of C-X (C-hetero) bonds (Scheme 2-4), especially in the mainchain, can be of great concern as it leads to a main-chain destruction. Common chemicals caused solvolysis are water, alcohols, ammonia, etc (Schnabel, 1992). ... R C ... ... X ... + YZ  R ... X C Z ... Y C + ... C ... R' R' Scheme 2-4. Solvolysis reaction of C-hetero bond in the polymers’ main-chain; YZ is a solvolysis agents In the case where a reactive C=C bond exists in the structure of the polymer, it is subjected to chemical reactions like metathesis where the degradation process is catalyzed by transition metal and ozonolysis where the reaction with ozone (O3) causes the scission through the formation of five-membered cyclic intermediates, which eventually will decompose into free radicals (Scheme 2-5). As mentioned previously, that ClO2 gas also demonstrates strong oxidizing capacity, so the exposure of polymeric materials to ClO2 might cause same type of chemical changes as those with O3 (Ozen, 2000; USEPA, 1999). 57 O O O O I ... ... O O O O . ... ... O  ... ... O II ... O  ... ... . C ... ... + O ... ... ... Scheme 2-5. Five-membered cyclic intermediates I and II yielded from the reaction of O3 with olefinic double bonds; I decomposes into carbonyl and biradical compounds; and II decomposes into free radicals Ozone can also slowly react with saturated hydrocarbon, at ambient temperature, to create free radicals (Scheme 2-6). The free radicals from the reaction between O3 with either unsaturated or saturated hydrocarbon can extract H atoms from the polymer structure, and, in the presence of O2, initiate the autoxidation reaction (Scheme 2-3 and Scheme 2-6). Ozonation of cellulose was studied by Lemeune et al. (2004). The significant reduction in the cellulose’s degree of polymerization (DPv) was observed, and the degradation increased as the ozone charge increased. The changes in chemical composition of the exposed cellulose sample confirmed that the primary reaction of ozone with cellulose fibers was glycosidic bond cleavage and that the oxidation mechanism followed a three-step process: 1) formation of the carbonyl groups, 2) oxidation to carboxyl groups, and 3) decarboxylation resulting in glycosidic bond cleavage (Lemeune et al., 2004). 58 RH + O O O HO  R O . O  O . R O OH   R + OH + O2 R-O-O-OH  RO + O-OH OH + RH OOH + RH RO + RH    H2O + R (a) H2O2 + R (b) ROH + R (c) In the presence of O2: R + O2  R-O-O Scheme 2-6. Direct reaction between O3 and saturated hydrocarbons When polymeric materials are exposed to strong oxidizing agents, such as O3 and ClO2, oxidative degradation can take place, as described previously. Such degradations are usually ‘selective’, either by reacting specifically with certain functional groups, or, taking place exclusively in the amorphous region of the semi-crystalline polymer that is more readily accessible (Kulshreshtha, 1992; Schnabel, 1992; Walzak et al., 1995). In most cases, the maximum interactions and changes occur on the surface of the materials. The typical changes caused by oxidative degradation are main chain scission, 59 depolymerization, cross-linking, and the changes in functional groups, such as formations of conjugated double bonds, carbonyl groups, etc. (Rivaton and Gardette, 1998; Rodriguez et al., 2003), and consequently, leading to the changes in polymer characteristics which may result in changes in mechanical properties, embrittlement, lack of transparency, as well as, the loss of additives, and the formation of toxic compounds. Bond scission in the polymer backbone causes the reduction in the polymer’s molecular weight. On the other hand, intermolecular cross-linking, which is the formation of the new chemical bonds between different polymer chains, results in the increase in molecular size (Kulshreshtha, 1992). The development of color in the oxidized polymers is a consequence of several possible degradation reactions, such as the formation of conjugated double bond, the oxidation of additives, etc. This change in color had been previously reported for some polymeric materials when exposed to oxidizing agents (Buchalla et al., 1993; Rodriguez et al., 2003; Walzak et al., 1995). Since the chemical changes caused by oxidative degradation could result in the alterations of mechanical, physical, and barrier properties, which eventually, lead to the changes in the polymers’ performance in packaging systems which could affect the product’s shelf-life or. It is critical to assess how ClO2 may impact the polymeric materials, especially when one potential application of ClO2 gas is to be used in combination with other gases, such as O2 and CO2 in a MAP system, the impact of ClO2 gas on the barrier to O2, CO2, and moisture of the different polymeric material is of great concern. 60 Barrier property of the polymeric materials is one of the most important performance of packaging system for food application, especially for fresh produce where respiration is still taking place during postharvest period (Martinez-Romero et al., 2003). The concentrations of O2 and CO2 accumulated in the package headspace affect the deterioration rate of fresh produce. Once the material is selected for a particular commodity, it is crucial that its permselectivity (PCO2/PO2) ratio be maintained through the shelf-life of the product, as in the case of MAP where the stable gaseous ratio in the package headspace is necessary (Martinez-Romero et al., 2003; Selke et al., 2004). The study on exposure of nylon to O3, which is considered to be a stronger oxidizing agent as compared to ClO2, showed an increase in tensile properties of the exposed polymeric materials, regardless of the applied conditions, but the barrier to oxygen of the material decreased as the time of exposure increased (Ozen, 2000; Ozen et al., 2002). The same trend on oxygen barrier was observed when exposed polyethylene (PE) film to O3. However, the changes PE film’s mechanical properties varied depending on the treatment conditions. IR spectra of the exposed nylon sample indicated the increase in –C-N- stretching, while, the exposure of polyethylene film to O3 caused the formation of oxygen-containing groups in the polymer main chain (Ozen et al., 2002). 61 2.4.3. Effects of gaseous ClO2 on packaging materials The exposure of polymeric material to ClO2 under different applied conditions could lead to changes in the overall performance of the polymeric materials (Ozen et al., 2002). Shin et al. (Shin et al., 2006) reported the changes in mechanical properties of PS, nylon, and LDPE after exposed to ClO2 gas at 20,000 ppmV or higher. The decreases in tensile strength and elongation (Kulshreshtha, 1992), a slight increase in oxygen permeability, of LDPE, LLDPE, oriented polypropylene (OPP), and biaxially oriented nylon (BON), after exposure of ClO2 gas, was reported by Ozen (Ozen, 2000). The degree of change varied depending on gas concentration (0.1–1.0 mg/L), o relative humidity (45–85% RH), and testing temperature (5-35 C). 62 BIBLIOGRAPHY 63 Agency for Toxic Substances and Disease Registry Division of Toxicology and Environmental Medicine (2004) "Toxicological Profile for Chlorine Dioxide and Chlorite." Retrieved October 11th, 2008, from http://www.atsdr.cdc.gov/toxprofiles/tp160.pdf. Aieta, E. and Berg, J. D. (1986). "A Review of Chlorine Dioxide in Drinking Water Treatment." Journal AWWA 78(6): 62 - 72. Alentiev, A. Y. and Yampolskii, Y. P. (2002). "Meares equation and the role of cohesion energy density in diffusion in polymers." Journal of Membrane Science 2006: 291–306. Allende, A., Tomás-Barberán, F. A. and Gil, M. I. (2006). "Minimal Processing for Healthy Traditional Foods." Trends in Food Science & Technology 17(9): 513519. Andrews, L. S., Key, A. M., Martin, R. L., Grodner, R. and Park, D. L. (2002). "Chlorine dioxide wash of shrimp and crawfish an alternative to aqueous chlorine." Food Microbiology 19: 261-267. Annous, B. A., Sites, J. E. and Burke, A. M. (2006). Inactivation of Salmonella on Cantaloupe Surfaces Using Hot Water or Chlorine Dioxide Gas Treatments. IFPA 18th Annual Conference Annous, B. A., Solomon, E. B., Cooke, P. E. and Burke, A. (2005). "Biofilm Formation by Salmonella Spp. on Cantaloupe Surfaces " Journal of Food Safety 25(4): 276 287. Appendini, P. and Hotchkiss, J. H. (2002). "Review of antimicrobial food packaging." Innovative Food Science & Emerging Technologies 3: 113-126. Bagorogoza, K., Bowers, J. and Okot-Kotber, M. (2001). "The effect of irradiation and modified atmosphere packaging on the quality of intact chill-stored turkey breast." Journal of Food Science 66(2): 367-. Benarde, M. A., Snow, W. B. and Olivieri, V. P. (1967). "Chlorine Dioxide Disinfection Temperature Effects." Journal of Applied Microbiology 30(1): 159-167. Bernarde, M. A., Snow, W. B. and Olivieri, V. P. (1967). "Kinetics and Mechanism of Bacterial Disinfection by Chlorine Dioxide " Journal of Applied Microbiology 15(2): 257-265. Buchalla, R., Schuttler, C. and Bogl, K. W. (1993). "Effects of Ionizing Radiation on Plastic Food Packaging Materials: A Review Part 1. Chemical and Physical Changes." Journal of Food Protection 56(11): 991 - 997. 64 Cameron, A. C., Beaudry, R. M., Banks, N. H. and Yelanich, M. V. (1994). "Modifiedatmosphere packaging of blueberry fruit: modeling respiration and package oxygen partial pressures as a function of temperature." Journal of the American Society for Horticultural Science 119(3): 534-539. Cegielska-Radziejewska, R., and Pikul, J. (2004). "Sodium Lactate addition on the quality and shelf life of refrigerated sliced poultry sausage packaged in air or nitrogen atmosphere." Journal of Food Protection 67(3): 601-606. Centers for Disease Control and Prevention (2010) "Surveillance for Foodborne Disease Outbreaks --- United States, 2007." Retrieved August 13th, 2008, from http://www.cdc.gov/mmwr/pdf/wk/mm5931.pdf. Cliffe-Byrnes, V., Laughlin, C. P. and O'Beirne, D. (2003). "The effects of packaging film and storage temperature on the quality of a dry coleslaw mix packaged in a modified atmosphere." International Journal of Food Science and Technology 38: 187-199. DeEll, J. R., Toivonen, P. M. A., Cornut, F., Roger, C. and Vigneault, C. (2006). "Addition of sorbitol with KMnO4 improves broccoli quality retention in modified atmosphere packages." Journal of Food Quality 29: 65-75. Department of Environment Food and Rural Affairs (DEFRA). (2003). "Sustainable Consumption and Production." Retrieved November 13, 2007, from http://www.defra.gov.uk/environment/business/scp/index.htm. Department of Health and Human Services (2008) "Summary Statistics for Foodborne Outbreaks, 2006." Retrieved November 1st, 2008, from http://www.cdc.gov/foodborneoutbreaks/documents/2006_line_list/2006_line_list .pdf. Department of the Environment Water Heritage and the Arts (2005) "Chlorine dioxide fact sheet." Retrieved October 17, 2008, from http://www.npi.gov.au/database/substance-info/profiles/21.html#health. Du, J., Han, Y. and Linton, R. H. (2003). "Efficacy of chlorine dioxide gas in reducing Escherichia coli O157:H7 on apple surfaces." Food Microbiology 20: 583-591. Dunn, R. C. and Simon, J. D. (1992). "Excited-State Photoreactions of Chlorine Dioxide in Water." Journal of the American Chemical Society 114: 4856 - 4860. Ellis, M., Cooksey, K., Dawson, P., Han, I. and Vergano, P. (2006). "Quality of Fresh Chicken Breasts Using a Combination of Modified Atmosphere Packaging and Chlorine Dioxide Sachets." Journal of Food Protection 69(8): 1991 - 1996. 65 Executive Secretariat (1998). "Hazard Analysis and Critical Control Point Principles and Application Guidelines." Journal of Food Protection 61(9): 1246-1259. Farber, J. M. and Dodds, K. L. (1995). Principles of Modified-atmosphere and Sous Vide Product Packaging. Boca Raton, FL, CRC Press, LLC. FSIS (2006) "NSS RTE/NRTE Introduction and Process Familiarization " Retrieved March 2nd, 2009, from http://www.fsis.usda.gov/PDF/RTE_Process_Familiarization.pdf. FSIS. (2009). "HACCP & Pathogen Reduction." Retrieved March 2nd, 2009, from http://www.fsis.usda.gov/Science/PR_&_HACCP_Guidance/index.asp. Fu, A.-H., Molins, R. A. and Sebranek, J. G. (1992). "Storage quality characteristics of beef rib eye steaks packaged in modified atmosphere." Journal of Food Science 57(2): 283-301. Gorny, J. R., Giclas, H., Gombas, D. and Means, K. (2006) "Commodity Specific Food Safety Guidelines for the Lettuce and Leafy Greens Supply Chain." Retrieved December 1st, 2008, from http://www.cfsan.fda.gov/~acrobat/lettsup.pdf. Han, Y., Guentert, A. M., Smith, R. S., Linton, R. H. and Nelson, P. E. (1999). "Efficacy of chlorine dioxide gas as a sanitizer for tanks used for aseptic juice storage " Food Microbiology 16(53-61). Han, Y., Linton, R. H., Nielsen, S. S. and Nelson, P. E. (2000). "Inactivation of Escherichia coli O157:H7 on surface-uninjured and -injured green pepper (Capsicum annuum L.) by chlorine dioxide gas as demonstrated by confocal laser scanning microscopy " Food Microbiology 17: 643-655. Han, Y., Selby, T. L., Schultze, K. K., Nelson, P. E. and Linton, R. H. (2004). "Decontamination of Strawberries Using Batch and Continuous Chlorine Dioxide Gas Treatments." Journal of Food Protection 67(11): 2450-2455. Harrison, W. A., Peters, A. C. and Fielding, L. M. (2000). "Growth of Listeria monocytogenes and Yersinia enterocolitica colonies under modified atmospheres at 4 and 8oC using model food system." Journal of Applied Microbiology 88: 3843. Henderson, R. E. (1999). Portable Gas Detectors used in Confined Space and Other Industrial Atmospheric Monitoring Programs. Safety and Health in Confined Spaces. McManus, N. Boca Raton, FL, Lewis Publishers. Huang, T.-S., Xu, C., Walker, K., West, P., Zhang, S. and Weese, J. (2006). "Decontamination Efficacy of combined Chlorine Dioxide with Ultrasonication on Apples and Lettuce." Journal of Food Science 71(4): M134-M139. 66 Ishi, G. (1958). "Solubility of chlorine dioxide." Chemical engineering (Japan) 22: 153 154. Ison, A., Odeh, I. N. and Margerum, D. W. (2006). "Kinetics and Mechanisms of Chlorine Dioxide and Chlorite Oxidations of Cysteine and Glutathione." Inorganic Chemistry 45(21): 8768–8775. Jimenez, S. M., Salsi, M. S., Tiburzi, M. C., Rafaghelli, R. C. and Pirovani, M. E. (1999). "Combined use of acetic acid treatment and modified atmosphere packaging for extending the shelf-life of chilled chicken breast portions." Journal of Applied Microbiology 87: 339-344. Jin, H. H. and Lee, S. Y. (2007). "Combined effect of aqueous chlorine dioxide and modified atmosphere packaging on inhibiting Salmonella Typhimurium and Listeria monocytogenes in mungbean sprouts." Journal of Food Science 72(9): M441-445. Kaczur, J. J. and Cawlfield, D. W. (1992). Chlorine Oxygen Acids and Salts, Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical Technology. Kroschwitz, J. I. New York, John Wiley & Sons Inc. 5. Kim, H., Ryu, J.-H. and Beuchat, L. R. (2006). "Survival of Enterobacter sakazakii on fresh produce as affected by temperature, and effectiveness of sanitizers for its elimination." International Journal of Food Microbiology 111: 134-143. Kim, J. M., Huang, T.-S., Marshall, M. R. and Wei, C.-I. (1999). "Chlorine dioxide treatment of seafoods to reduce bacterial loads." Journal of Food Science 64(6): 1089-1093. Kim, Y., Kim, M. and Song, K. B. (2009). "Combined treatment of fumaric acid with aqueous chlorine dioxide or UV-C irradiation to inactivate Escherichia coli O157:H7, Salmonella enterica serovar Typhimurium, and Listeria monocytogenes inoculated on alfalfa and clover sprouts." Lebensmittel-Wissenschaft undTechnologie 42: 1654-1658. Kim, Y. J., Lee, S. H., Park, J., Park, J., Chung, M., Kwon, K., Chung, K., Won, M. and Song, K. B. (2008). "Inactivation of Escherichia coli O157:H7, Salmonella typhimurium, and Listeria monocytogenes on Stored Iceberg Lettuce by Aqueous Chlorine Dioxide Treatment." Journal of Food Science 73(9): 18-22. Kreske, A. C., Ryu, J.-H. and Beuchat, L. R. (2006a). "Evaluation of chlorine, chlorine dioxide, and a peroxyacetic acid-based sanitizer for effectiveness in killing Bacillus cereus and Bacillus thuringiensis spores in suspensions, on the surface of stainless steel, and on apples." Journal of Food Protection 69(8): 1892-1903. 67 Kreske, A. C., Ryu, J. H., Pettigrew, C. A. and Beuchat, L. R. (2006b). "Lethality of Chlorine, Chlorine Dioxide, and a Commercial Produce Sanitizer to Bacillus cereus and Pseudomonas in a Liquid Detergent, on Stainless Steel, and in Biofilm." Journal of Food Protection 69(11): 2621-2634. Kulshreshtha, A. K. (1992). 3: Chemical Degradation. New York, Marcel Dekker, Inc. Lee, S. Y., Costello, M. and Kang, D. H. (2004). "Efficacy of Chlorine Dioxide Gas as a Sanitizer of Lettuce Leaves." Journal of Food Protection 67(7): 1371-1376. Lemeune, S., Jameel, H., Chang, H.-M. and Kadla, J. F. (2004). "Effects of ozone and chlorine dioxide on the chemical properties of cellulose fibers." Journal of Applied Polymer Science 93(3): 1219 - 1223. Lin, K.-W. a. L., S.-N. (2002). "Physicochemical properties and microbial stability of reduced-fat chinese-style sausage stored under modified atmosphere systems." Journal of Food Science 67(8): 3184-3189. Lindsay, D., Brozel, V. S., Mostert, J. F. and Holy, A. v. (2002). "Differential efficacy of a chlorine dioxide-containing sanitizer against single species and binary biofilms of a diary-associated Bacillus cereus and a Pseudomonas fluorescene isolate." Journal of Applied Microbiology 92: 352-361. Lukasik, J., Bradley, M. L., Scott, T. M., Dea, M., Koo, A., Hsu, W. Y., Bartz, J. A. and Farrah, S. R. (2003). "Reduction of poliovirus 1, bacteriophages, Salmonella montevideo, and Escherichia coli O157:H7 on strawberries by physical and disinfectant washes." Journal of Food Protection 66(2): 188-193. Lynch, M., Painter, J., Woodruff, R. and Braden, C. (2006) "Surveillance for FoodborneDisease Outbreaks --- United States, 1998-2002." Retrieved November 1st, 2008, from http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5510a1.htm?s_cid=ss5510a1_e. Mahovic, M., Bartz, J. A., Schneider, K. R. and Tenney, J. D. (2009). "Chlorine dioxide gas from an aqueous solution: reduction of Salmonella in wounds on tomato fruit and movement to sinks in a treatment chamber." Journal of Food Protection 72(5): 952-958. Martinez-Romero, D., Guillen, F., Castillo, S., Valero, D. and Serrano, M. (2003). "Modified armosphere packaging maintains quality of table grapes." Journal of Food Science 68(5): 1838-1843. Matteucci, S., Yampolskii, Y., Freeman, B. D. and Pinnau, I. (2006). Transport of Gases and Vapors in Glassy and Rubbery Polymers. Materials Science of Membranes for Gas and Vapor Separation. Yuri Yampolskii, I. P. B. F.: 1-47. 68 Mead, P. S., Slutsker, L., Dietz, V., McCaig, L. F., Bresee, J. S., Shapiro, C., Griffin, P. M. and Tauxe, R. V. (1999) "Food-Related Illness and Death in the United States." Retrieved November 1st, 2008, from http://www.cdc.gov/ncidod/eid/Vol5no5/mead.htm. Mueller, H. S. P. and Willner, H. (1993). "Vibrational and electronic spectra of chlorine dioxide, OClO, and chlorine superoxide ClOO, isolated in cryogenic matrixes." Journal of Physical Chemistry 97(41): 10589-10598. Multon, J. L. (1996). The Role of Packaging in Preserving Foodstuffs. New York, VCH Publishers, Inc. Netramai, S., Rubino, M., Auras, R., Annous, B. A. and Matthews, K. (2010). Surface study on antimicrobial effects of ClO2 gas on E. coli O157:H7 resided on lettuce leaf. IFT10 - IFT Annual Meeting & Food Expo, Chicago, IL, Institute of Food Technologists. Occupational Safety & Health Administration (2010). Determination of Chlorine Dioxide in Workplace Atmospheres. Sampling and Analytical Methods, United States Department of Labor. ID-202. Okull, D. O., Demirci, A., Rosenberger, D. and LaBorde, L. F. (2006). "Susceptibility of Penicillium expansum spores to sodium hypochlorite, electrolyzed oxidizing water, and chlorine dioxide solutions modified with nonionic surfactants." Journal of Food Protection 69(8): 1944-1948. Ölmez, H. and Temur, S. D. (2010). "Effects of different sanitizing treatments on biofilms and attachment of Escherichia coli and Listeria monocytogenes on green leaf lettuce " LWT - Food Science and Technology 43(6): 964-970. Ooraikul, B. and Stiles, M. E. (1991). Modified atmosphere packaging of food. New York, Ellis Horwood. Ortega, Y. R., Mann, A., Torres, M. P. and Cama, V. (2008). "Efficacy of Gaseous Chlorine Dioxide as a Sanitizer against Cryptosporidium parvum, Cyclospora cayetanensis, and Encephalitozoon intestinalis on Produce." Journal of Food Protection 71(12): 2410-2414. Ozen, B. F. (2000). Effect of Ozone and Chlorine Dioxide Treatments Used in Perishable Food Applications on Polymeric Materials: Changes in Mechanical, Thermal and Mass Transfer Properties. Food Science Department. West Lafayette, IN, Purdue University. Ph.D. Ozen, B. F., Mauer, L. J. and Floros, J. D. (2002). "Effects of ozone exposure on the structural, mechanical and barrier properties of select plastic packaging films." Packaging Technology and Science 15(6): 301 - 311. 69 Pao, S., Kelsey, D. F. and Long III, W. (2009). "Spray washing of tomatoes with chlorine dioxide to minimize Salmonella on inoculated fruit surfaces and crosscontamination from revolving brushes." journal of Food Protection 72(12): 24482452. Park, E. J., Gray, P. M., Oh, S. W., Kronenberg, J. and Kang, D. H. (2008). "Efficacy of FIT produce wash and chlorine dioxide on pathogen control in fresh potatoes." Journal of Food Science 73(6): M278-M282. Phillips, C. (1996). "Review: Modified atmosphere packaging and its effects on the microbiological quality and safety of produce." International Journal of Food Science and Technology 31: 463-479. Podolak, R., Elliott, P. H., Taylor, B. J., Khurana, A. and G., B. D. (2009). "Destruction of Alicyclobacillus acidoterrestris spores in apple juice on stainless steel surfaces by chemical disinfectants." Journal of Food Protection 72(3): 510-514. Popa, I., Hanson, E. J., Todd, E. C., Schilder, A. C. and Ryser, E. T. (2007). "Efficacy of chlorine dioxide gas sachets for enhancing the microbiological quality and safety of blueberries." Journal of Food Protection 70(9): 2084-2088. Recio, R., Lozano, A. E., Pra´danos, P., Marcos, A., Tejerina, F. and Herna´ndez, A. (2008). "Effect of Fractional Free Volume and Tg on Gas Separation Through Membranes Made with Different Glassy Polymers." Journal of Applied Polymer Science 107: 1039-1046. Richard, E. and Vaida, V. (1991). "The photochemical dynamic of the A2A2 state of chlorine dioxide." Journal of Chemical Physics 94: 163-171. Rico, D., Martín-Diana, A. B., Barat, J. M. and Barry-Ryana, C. (2007). "Extending and measuring the quality of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology 18(7): 373-386 Rivaton, A. and Gardette, J.-L. (1998). "Photo-Oxidation of Aromatic Polymers." Die Angewandte Makromolekulare Chemie 261/262: 173–188. Rodgers, S. L., Cash, J. N., Siddiq, M. and Ryser, E. T. (2004). "A comparison of different chemical sanitizers for inactivating Escherichia coli O157:H7 and Listeria monocytogenes in solution and on apples, lettuce, strawberries and cantaloupe." Journal of Food Protection 67(4): 721-731. Rodriguez, F., Cohen, C., Ober, C. K. and Archer, L. A. (2003). Principles of Polymer Systems. New York, Taylor & Francis. Romig, W. R. and Mir, N. (2000). Packaging Technology for Fresh Produce. The 15th International Congress for Plastics in Agriculture, Hershey, PA. 70 Ryu, J. H. and Beuchat, L. R. (2005). "Biofilm Formation and Sporulation by Bacillus cereus on a Stainless Steel Surface and Subsequent Resistance of Vegetative Cells and Spores to Chlorine, Chlorine Dioxide, and a Peroxyacetic Acid-Based Sanitizer." Journal of Food Protection 68(12): 2614-2622. Sapers, G. M., Walker, P. N., Sites, J. E., Annous, B. A. and Eblen, D. R. (2003). "Vapor-phase decontamination of apples innoculated with Escherichia coli." Journal of Food Science 68(3): 1003-1007. Sawaya, W. N., Elnawawy, A. S., Al-Zenki, S., Al-Otaibi, J., Al-Omirah, H. and AlAmiri, H. (1995). "Storage stability of chicken as affected by MAP and lactic acid treatment." Journal of Food Science 60(3): 661-. Schnabel, W. (1992). Polymer Degradation: Principles and Practical Applications. New York, Oxford University Press. Selke, S. E. M., Culter, J. D. and Hernandez, R. J. (2004). Plastics Packaging: Properties, Processing, Applications, and Regulations. Cincinnati, OH, Hanser Gardner Publications, Inc. Shimazu, A., Miyazaki, T., Sushita, T. M., Maeda, M. and Ikeda, K. (1999). "Relationships Between Chemical Structures and Solubility, Diffusivity, and Permselectivity of 1,3-Butadiene and n-Butane in 6FDA-Based Polyimides." Journal of Polymer Science: Part B: Polymer Physics 37: 2941–2949. Shin, J. (2007). Evaluation of the Effect of Chlorine Dioxide and Allyl-isothiocyanate on the Growth of Salmonella Typhimurium and Listeria Monocytogenes on Fresh Chicken Breast and Effect of Chlorine Dioxide Exposure on the Physical Properties of Plastic Films. School of Packaging. East Lansing, Michigan State University. Ph.D. Shin, J., Ryser, E., Selke, S. and Harte, B. (2006). Evaluation of the Microbial Inhibition Performance of Gas Type Antimicrobials (Chlorine Dioxide and AllylIsothiocyanate) with Modified Atmosphere Packaging (MAP), and the Effct of Chlorine Dioxide Exposure on the Physical Properties of Plastic Films. The 15th IAPRI World Conference on Packaging. Tokyo, Japan. Simões, M., Simões, L. C. and Vieira, M. J. (2010). "A review of current and emergent biofilm control strategies." LWT - Food Science and Technology 43(4): 573-583. Simpson, R., Almonacid, S. and Acevedo, C. (2001). "Development of a mathematical model for MAP systems applied to nonrespiring foods." journal of Food Science 66(4): 561-567. Singh, N., Singh R. K., and Bhunia, A. K. (2003). "Sequential disinfection of Escherichia coli O157:H7 inoculated alfalfa seeds before and during sprouting using aqueous 71 chlorine dioxide, ozonated water, and thyme essential oil." LebensmittelWissenschaft und-Technologie 36: 235-243. Singh, N., Singh, R. K., Bhunia, A. K. and Stroshine, R. L. (2002). "Efficacy of Chlorine Dioxide, Ozone, and Thyme Essential Oil or a Sequential Washing in Killing Escherichia coli O157:H7 on Lettuce and Baby Carrots " LebensmittelWissenschaft und-Technologie 35(8): 720-729 Solomon, S. (1999). "Stratospheric Ozone Depletion: A Review of Concepts and History." Reviews of Geophysics 37(3): 275-316. Sothornvit, R. and Kiatchanapaibula, P. (2009). "Quality and shelf-life of washed freshcut asparagus in modified atmosphere packaging." LWT - Food Science and Technology 42(9): 1484-1490. Sy, K. V., McWatters, K. H. and Beuchat, L. R. (2005a). "Efficacy of Gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries, and Raspberries." Journal of Food Protection 68(6): 1165 - 1175. Sy, K. V., Murray, M. B., Harrison, M. D. and Beuchat, L. R. (2005b). "Evaluation of gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes, and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection 68(6): 1176 - 1187. Tan, H. K., Wheeler, W. B. and Wei, C. I. (1987). "eaction of chlorine dioxide with amino acids and peptides." Mutation Research 188: 259-266. Taub, I. A. and Singh, R. P. (1998). Food Storage Stability. Boca Raton, FL, CRC Press LLC. Tauxe, R. V. (2008). Improving Foodbprne Disease Prevention: the Ecological Cpnnection. Michigan State University, Centers for Disease Control and Prevention. Taylor, J. H., Rogers, S. J. and Holah, J. T. (1999). "A comparison of the bactericidal efficacy of 18 disinfectants used in the food industry against Escherichia coli O157:H7 and Pseudomonas aeruginosa at 10 and 20oC." Journal of Applied Microbiology 87: 718-725. Unda, J. R., Molins, R. A. and Zamojcin, C. A. (2006). "Sanitization of Fresh Rib Eye Steaks with Chlorine Dioxide-Generating Binary Systems." Journal of Food Science 54(1): 7-10. USEPA (1999) "4. Chlorine Dioxide." Retrieved September 11th, 2007, from http://www.epa.gov/ogwdw/mdbp/pdf/alter/chapt_4.pdf 72 USFDA. (2001). "Agency Response Letter GRAS Notice No. GRN 000062." Retrieved September 29th, 2007, from http://www.fda.gov/Food/FoodIngredientsPackaging/GenerallyRecognizedasSafe GRAS/GRASListings/ucm153971.htm. USFDA (2004) "Produce Safety From Production to Consumption: 2004 Action Plan to Minimize Foodborne Illness Associated with Fresh Produce Consumption." Retrieved September 24th, 2008, from http://www.cfsan.fda.gov/~dms/prodpla2.html. USFDA (2006) "Lettuce Safety Initiative." Retrieved December 10th, 2008, from http://www.cfsan.fda.gov/~dms/lettsafe.html. Vaida, V. and Simon, J. D. (1995). "The Photoreactivity of Chlorine Dioxide." Science 268(5216): 1443-1448. Van Krevelen, D. W. (1997). Properties of Polymers. New York, Elsevier Science B.V. Vermeiren, L., Devleghere, F., Van Beest, M., De Krujif, N. and Debevere, J. (1999). "Developments in The Active Packaging of Foods." Trends in Food Science & Technology 10(3): 77-86. Walzak, M. J., Flynn, S., Foerch, R., Hill, J. M., Karbashewski, E., Lin, A. and Strobel, M. (1995). "UV and Ozone Treatment of Polypropylene and Poly(ethylene terephthalate)." Journal of Adhesion Science and Technology 9(9): 1229-1248. Yang, H., Wang, S., Li, Y. and Johnson, M. G. (2002). "Predictive models for the survival/death of Campylobacter jejuni and Salmonella Typhimurium in poultry scalding and chilling." Journal of Food Science 67(5): 1836-1843. Yang, Z. and Li, Y. (1997). "Barrier Properties and Structure of Acrylonitrile/Acrylic Copolymers." Chinese Journal of Polymer Science 15(3): 236-241. Yuk, H.-G., Bartz, J. A. and Schneider, K. R. (2005). "Effectiveness of individual or combined sanitizer treatments for inactivating Salmonella spp. on smooth surface, stem scar, and wounds of tomatoes." Journal of Food Science 70(9): M409-M414. Yuk, H.-G., Bartz, J. A. and Schneider, K. R. (2006). "The Effectiveness of Sanitizer Treatments in Inactivation of Salmonella spp. from Bell Pepper, Cucumber, and Strawberry." Journal of Food Science 71(3): M95-M99. 73 CHAPTER 3: CONTINUOUS DETECTION SYSTEM FOR MASS TRANSFER MEASUREMENT OF ClO2 3.1. Introduction Study on mass transfer of ClO2 gas through polymeric packaging materials, as well as understanding the effects of ClO2 gas on important properties of the materials are beneficial when designing a packaging system that will be used in conjunction with gaseous chlorine dioxide (ClO2), since presence of ClO2 gas within the food packaging system, at the correct concentration, is a crucial point of sufficient sanitization of the food product within the package. Due to the reactive nature of ClO2 gas, it is difficult to find a suitable detection method to accurately measure gaseous ClO2 concentration (Kaczur and Cawlfield, 1992; USEPA, 1999). Most of the devices and techniques currently available are designed to determine the concentration of ClO2 in liquid form as its most common applications are in waste- and portable water treatment, and pulp production (Kaczur and Cawlfield, 1992). Selecting a suitable detection technique for detecting ClO2 in the gaseous state is a very important step in the mass transfer study of ClO2 gas through polymeric films. The detection technique should have high sensitivity, repeatability, and accuracy, and, at the same time, be robust and economically feasible. 74 Ozen (2000) studied the mass transfer of gaseous ClO2 through polymeric 2 materials. In the study, two film samples, each with an area of 45 cm , were installed in a three-compartment permeation cell. The upper and lower chambers were flushed with 100 mg/L ClO2 for 5 min, while the middle compartment was filled with nitrogen. At predetermined time intervals, the specific volume of gas was sampled from the middle chamber and injected into distilled water. The concentration of permeated ClO2 in water was then determined using amperometric titration (Greenberg et al., 1992), and the permeability coefficient was calculated according to the quasi-isostatic technique. Since the steady state was reached too fast for the detection technique used, only the linear portion of the plot of cumulative amount of permeated ClO2 versus time was obtained from the experiment. Thus the diffusion coefficient which can be calculated using the data in the transient state could not be determined (Ozen, 2000). To obtain a complete mass transfer profile of ClO2 through packaging membranes, it is critical to develop a continuous detection technique that can provide information of the unsteady-state region. The objective of this study was to develop a continuous detection method for mass transfer measurement of gaseous ClO2 to be able to obtain its complete permeation profile, i.e. both the transient and steady state. 75 3.2. Materials and Methods The flow diagram and set up of the permeation system developed is shown in Figure 3-1. The system consisted of a nitrogen supply, a flow meter/controller, a permeability cell, and an electrochemical (EC) detector. Nitrogen gas from the generator picks up gaseous ClO2 that permeates through the film sample installed in the permeability cell, and carries ClO2 gas to the EC detector. The detector records the output of concentration of permeated ClO2 (in ppm or ppmV) every 30 seconds. The permeability cell (Figure 3-2) is made of grade 316 stainless-steel, which has excellent overall corrosion resistance properties (AK Steel Corporation, 2007). The two half-cells are sealed together using a Viton® O-ring made from fluoroelastomers which did not interact with ClO2 gas at the testing concentrations (Anchor Rubber Products, 2007). 76 Figure 3-1. Flow diagram of mass transfer study. Adapted from (Netramai et al., 2009). 77 3.2.1. Permeability cell The film sample is placed between the upper and lower chambers that have -5 3 volumes of 8.7 × 10 m and 20.3 × 10 -5 3 m , respectively. The permeated ClO2 collects in the upper chamber. A temperature control device, which consists of six heating wires fitted in the metal wall (two wires for the upper chamber, and four wires for the lower chamber), heats the permeability cell to the desired temperature. Temperature is monitored by two thermocouples located at the top and bottom of the unit, and data is transferred to the temperature controller. The temperature fluctuation inside the chamber varied ± 0.50°C from the testing condition. A relative humidity (RH) sensor installed in the upper and lower chambers monitored %RH of the chamber throughout the experiment. 78 Figure 3-2. Schematic of permeation of ClO2 through the polymer film. Adapted from (Netramai et al., 2009). 79 3.2.2. Electrochemical detector ® The detector used in this study was a ToxiPro electrochemical (EC) detector (Biosystems, Plano, TX) for ClO2. The detector is able to determine the concentration of a particular gas by generating an electrical signal that correlates the reaction of the electrode with the target gas. A typical EC sensor (two-electrode type), the main component of an EC detector, consists of the following components: (a) a hydrophobic membrane that protects and controls the amount of gas reaching the electrode; (b) a sensing (or working) electrode which the diffused gas reacts with; (c) a counter electrode that supplies or receives electrons for the sensing electrode; and, (d) an electrolyte that facilitates the reaction and enables ionic flow. The system produces an electric current proportional to the gas concentration. A three-electrode type EC sensor also has a third electrode, a reference electrode, in order to stabilize the sensing electrode (Emond and Kearney, 2007) The reaction that takes place in the sensor for ClO2 is (Emond and Kearney, 2007): + - - ClO2 + 4H + 5e  Cl + 2H2O (1)  For the ToxiPro detector, the operating temperature and humidity ranges are –20 to 50°C, and up to 95% RH, respectively, with no condensation (according to the manufacturer’s specifications) (Emond and Kearney, 2007). 80 The unit displayed on the detector is in ppm (or ppmV), i.e., μl of ClO2 gas per one liter of sampled gas mixture. The concentration in ppmV of ClO2 can also be converted to mg of ClO2/L of gas (mg/L), which is one of the common units used to describe the concentration of gaseous ClO2 (ICA TriNova LLC, 2006; Ishi, 1958). 3.2.2.1. Calibration The EC detector was calibrated every two weeks by exposing the detector to air (zero calibration, 0.00 ppmV of ClO2) and 1.00 ppmV of ClO2 gas generated by an EC gas generator (Cal 2000, ACD, Tucson, AZ). A bump test to monitor the response of the detector, i.e., exposing the detector to a known concentration of ClO2 at 0.00, 1.00, 2.50, 3.00, and 4.00 ppmV, was also performed every month by the local Office of Radiation, Chemical & Biological Safety (ORCBS) (East Lansing, MI). The detector gave a linear 2 response with a correlation factor (R ) of 0.9995. 3.2.2.2. Determination of noise and signal to noise ratio The background noise of the EC detector was determined by collecting mass transfer data in the absence of ClO2 solution in the lower chamber of the permeability cell over 24 hours for all temperature conditions used in this study. The test was repeated four times at each temperature. The data obtained were then used to calculate the amplitude of noise as 3s, where s is the square root of the mean squared error (MSE) (Freund and Wilson, 2003). The calculated amplitude of noise and the response of the 81 detector obtained from the bump test (mentioned previously) were then used to calculate the signal to noise (S/N) ratio of the EC detector at various concentrations of gaseous ClO2. 3.2.3. Preparation of ClO2 solutions A stock solution of 1000 mg ClO2 per liter was prepared by submerging a sachet containing the chemical precursors, i.e., sufficient sulfuric acid (H2SO4) and sodium -3 chlorite (NaClO2) to generate approximately 2 × 10 kg of ClO2 (z-series, ICA TriNova, Newnan, GA), in 2 liters of deionized, distilled water for 48 hours. The solution was titrated to determine the actual concentration using the titration procedure outlined by ICA Trinova (ICA TriNova LLC, 2006). The stock solution was diluted, as required, right before use. The final concentration was determined by titration and, according to Henry’s constant of gaseous ClO2 as reported in the literature (Ishi, 1958; Kaczur and Cawlfield, 1992), gave a concentration of 10 mg ClO2/L of gas (approximately 3600 ppmV) in the headspace. The selected concentration is considered to provide a high dose range of gaseous ClO2 for food applications (Kaczur and Cawlfield, 1992; Sy et al., 2005). 3.2.4. Statistical analysis The data obtained from four replicates of each sample were statistically analyzed by analysis of variance (ANOVA) using the Statistical Analysis System (SAS) software 82 (SAS Institute Inc., Cary, NC) at the confidence level of 95% ( = 0.05) with Tukey’s adjustment for comparison of means. 3.3. Results and Discussion The amplitudes of the background noise of the permeation system at 23, 30, and 40°C, as shown in Table 3-1, increased with increasing temperature. At 23°C, the S/N ratio of the system improved significantly as the concentration of ClO2 increased. In this study, the concentration of permeated ClO2 at the steady state varied, depending on the type of polymeric material being studied, and ranged from approximately 0.10 to 3.00 ppmV. The noise level was very low as compared to the signal; e.g., noise was only 1.29% at the signal of 0.10 ppmV. The permeation system developed in this study could be considered reliable and consistent for the measurements. This continuous system is also easy to set up and operate, as well as being space- and cost-effective. 83 Table 3-1. Permeation system mean square error (MSE) and amplitude of noise at 0.00 ppmV, and signal to noise (S/N) ratio at various concentrations of ClO2 Temperature o C -7 ClO2 ppmV MSE × 10 0.00 1.95 ± 1.09 s 2 a,1 Amplitude × 10 3s a -3 S/N × 10 1.29 ± 0.36 2 a 1.00 - - 7.82 ± 0.17 2.50 - - 19.7 ± 0.31 3.00 - - 23.4 ± 0.22 4.00 - - 30.7 ± 0.24 30 0.00 3.30 ± 0.74 40 0.00 5.06 ± 0.35 23 a b 1 b c d a - b - 1.71 ± 0.19 2.13 ± 0.07 Within columns, means sharing the same superscript letter are not significantly different (p > 0.05; n = 4). 84 BIBLIOGRAPHY 85 AK Steel Corporation (2007) "Product Data Sheet: 316/316L Stainless Steel." Retrieved August 27th, 2010, from http://www.aksteel.com/pdf/markets_products/stainless/austenitic/316_316L_Dat a_Sheet.pdf. Anchor Rubber Products (2007) "Material Properties: Viton® O-ring." Retrieved August 27th, 2010, from http://www.anchorrubber.com/pdf/fda_viton.pdf. Emond, J. and Kearney, T. (2007). Personal Communication: Electrochemical Sensor Technology. Netramai, S., Biosystems: 15. Freund, R. J. and Wilson, W. J. (2003). Statistical Methods. New York, Academic Press. Greenberg, A. E., Clesceri, L. S. and Eaton, A. D. (1992). 4500-ClO2 C. Amperometric method I. Standard Methods for the Examination of Water and Wastewater. Washington, D.C., American Public Health Association: 4–55, 56. ICA TriNova LLC (2006). Profiling gas release from ClO2 sachets. Standard Operation Procedure: Profiling gas release from ClO2 sachets, ICA Trinova. Ishi, G. (1958). "Solubility of chlorine dioxide." Chemical engineering (Japan) 22: 153 154. Kaczur, J. J. and Cawlfield, D. W. (1992). Chlorine Oxygen Acids and Salts, Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical Technology. Kroschwitz, J. I. New York, John Wiley & Sons Inc. 5. Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2009). "Mass Transfer Study of Chlorine Dioxide Gas through Polymeric Packaging Materials." Journal of Applied Polymer Science 114(5): 2929-2936. Ozen, B. F. (2000). Effect of Ozone and Chlorine Dioxide Treatments Used in Perishable Food Applications on Polymeric Materials: Changes in Mechanical, Thermal and Mass Transfer Properties. Food Science Department. West Lafayette, IN, Purdue University. Ph.D. Sy, K. V., McWatters, K. H. and Beuchat, L. R. (2005). "Efficacy of Gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries, and Raspberries." Journal of Food Protection 68(6): 1165 - 1175. USEPA (1999) "4. Chlorine Dioxide." Retrieved September 11th, 2007, from http://www.epa.gov/ogwdw/mdbp/pdf/alter/chapt_4.pdf 86 CHAPTER 4: MASS TRANSFER OF ClO2 THROUGH POLYMERIC FILMS 4.1. Introduction Due to its bactericidal effects, ClO2 gas is gaining significant interest in the food and pharmaceutical industries (Sy et al., 2005a; Sy et al., 2005b). One of the most recent applications of ClO2 is as a headspace gas in packaging systems where it acts in vaporphase decontamination to extend the shelf-life of perishable food products (Kaczur and Cawlfield, 1992; Sapers et al., 2003; Sy et al., 2005a). Understanding the interaction of gaseous ClO2 with polymeric materials is critical once ClO2 is included in the package headspace. The obtained data could lead to an improved packaging system that is able to deliver and maintain certain amounts of ClO2 to sanitize the product, without compromising package integrity. If a package is to be considered as a device for the release and application of ClO2 gas, then the mass transfer properties of ClO2 must be determined since this will impact the selection of material(s) with an appropriate gas barrier. The reported permeability (P) values of ClO2 gas, at 20°C, for linear low-density polyethylene (LLDPE), oriented poly(propylene) (OPP), and biaxially oriented nylon (BON), by Ozen (2000), were 7.62 × -16 10 -17 , 6.21 × 10 , and 2.34 × 10 -17 -2 -1 -1 kgClO2mm s Pa , respectively (Ozen, 2000). 87 However, the diffusion and solubility coefficients could not be determined due to the reasons discussed previously. The objectives of this study were to assess the mass transfer of ClO2 through various polymeric packaging materials by determining their permeability, diffusion and solubility coefficients (P, D, and S, respectively), using the developed continuous detection system described in Chapter 3. 4.2. Materials and Methods 4.2.1. Polymeric packaging materials The polymeric materials selected for this study were low-density polyethylene (LDPE) and linear low-density polyethylene (LLDPE) (Flexopack S.A., Attiki, Greece), high-density polyethylene (HDPE) (James River Corp. Flexible Packaging Group, Richmond, VA), biaxially-oriented poly(propylene) (BOPP) (Cryovac, Duncan, SC), polystyrene (PS) (TRYCITE™ 8001, Dow Chemical Company, Midland, MI), poly(ethylene terephthalate) (PET) (Mylar  A, DuPont, Wilmington, DE), poly(vinyl chloride) (PVC) (BEMIS, Shirley, MA), nylon 66 (Dartek F-101, DuPont, Wilmington,  DE), poly(lactic acid) (PLA) (EVLON , BI-AX International Inc., Wingham, Ontario, Canada), and a multilayer structure of ethylene vinyl acetate (EVA) and ethylene vinyl alcohol (EVOH) (EVA/EVOH/EVA; Cryovac, Duncan, SC). These materials are normally used in packaging systems for perishable food products and various nonperishable and pharmaceutical goods. Samples of each film were taken from the same lot 88 and were conditioned at 23°C and 50% RH for at least 24 hours before starting any experiment. 4.2.2. Polymer characterization 4.2.2.1. Determination of the film’s density 3 The density, ρ, in kg/m of the polymer film samples was measured using the flotation method at 23°C. A mixture of methanol and water, or a solution of calcium nitrate (Ca(NO3)2), was used as the flotation media for the films with ρ < 1 or ρ > 1, respectively (Kaczur and Cawlfield, 1992; Selby et al., 2005). 4.2.2.2. Determination of physical properties o The glass transition temperature, Tg ( C), and enthalpy of fusion, Hm (kJ/kg), of the polymeric films were determined using a Q-100 differential scanning calorimeter (DSC) (TA Instruments, New Castle, DE) according to the ASTM D3418-03 method (ASTM, 2003). At least 5 × 10 -6 kg of sample was used in each run, which consisted of two cycles (heat/cool/heat) with heating and cooling rates of 10°C/min. The experiment was repeated five times for each type of material. The Hm and Tg data were collected from the first and second heating, respectively, and data analyses were done using Universal Analysis Software (UAS Version 3.9A, TA Instruments, New Castle, DE). The crystallinity of the film samples was calculated by: 89 % Crystallinity  H m  100 0 H m (1) 0 where H m is the heat of fusion of the sample obtained by DSC, and H m is the heat of melting of 100% crystalline material. Values used were 288 kJ/kg (PE), 79 kJ/kg (BOPP), 138 kJ/kg (PET), 93.6 kJ/kg (PLA), 301 kJ/kg (nylon), 293 kJ/kg (EVA), and 163 kJ/kg (EVOH) (Auras et al., 2003; Rodriguez et al., 2003). 4.2.2.3. Determination of free volume The free volume (FV) of the semi-crystalline polymer, at a particular temperature, can be calculated from its fractional free volume (FFV) (Equation 2) and corrected as shown below: 1 FFV FV V  V0  V    1.3Vw (2) 1   %Crystallinity   1    FFV 100   (3) 3 where V is the specific volume (m /kg) of the amorphous polymer at the testing 3 temperature (23°C), and V0 is a specific volume (m /kg) at 0 K. The FFV can also be 3 estimated from a polymer’s density, ρ, and its van der Waals volume, Vw (m /mol), as shown in Equation 2. Vw values correspond to the space occupied by the polymer molecule that is a barrier for the permeant (Recio et al., 2008; Van Krevelen, 1997). 90 4.2.2.4. Determination of cohesive energy density The cohesive energy density (CED) can be estimated based on cohesive energy ( E coh ) and molar volume ( Vm ) data as follows: CED  Ecoh Vm (4) 3 where the E coh (kJ/mol) and Vm (m /mol) were estimated from the means obtained by group contribution method (Van Krevelen, 1997). 4.2.3. Preparation of ClO2 solution The same methodology as described in chapter 3 was used to prepare a final solution from a stock solution of 1,000 mgClO2/L. To assess of ClO2 stability in the permeability cell, the solution was placed in the lower chamber, with aluminum foil installed, instead of a film sample. Amounts of ClO2 (mg) in the solution were monitored, after which the ClO2 concentration (ppmV) was calculated in the gas phase, above the solution, which is assumed to be in equilibrium with the ClO2 solution. The experiments were repeated 4 times. The degradation profiles o of ClO2 in the gas phase, above ClO2 solution, at 23, 30, and 40 C are reported in Figure 4-1 and Table 4-1. 91 4.2.4. Determination of ClO2 mass transfer parameters The permeation experiment for each material was repeated four times. The amount of ClO2 gas permeated per second was calculated as follows: Permeated ClO2 measured at time t (kg/s) = permeated ClO2 L / L  67.5 g / mol  N 2 gas flow rateL / s  (5) 10 6 L / L 1 mol of ideal gas at particular temperatureL / mol 103 g / kg where 67.5 g/mol is the molecular weight of ClO2. The N2 gas flow rate is equal to 4.81 -3 × 10 L/s and represents the volume of N2 gas that flows through the upper chamber of the permeability cell in one second and carries the permeated ClO2 gas to the detector. The volumes of 1 mol of ideal gas at 23, 30 and 40°C are 24.3, 24.9, and 25.7 L/mol, respectively. 92 Calculated ClO2 in gas phase, ppmV 3.6e+3 23oC 30oC 3.0e+3 40oC 2.4e+3 0.0 0 4 8 12 16 20 24 28 Time, Hour Figure 4-1. Calculated ClO2 concentrations in the gas phase above ClO2 solution (ppmV) o at 23, 30, and 40 C without the presence of film sample 93 Table 4-1. Degradation profile of calculated ClO2 in the gas phase above ClO2 solution o (%) at 23, 30, and 40 C ClO2 degradation profiles (%) Time (Hour) o o 23 C o 30 C 1 40 C 3 4 nd - 8 - 3 - 12 - - 16 - 2.75 ± 0.73 20 - 24 28 1 2 - 3,4 a 1.54 ± 1.32 4 a,b 2.01 ± 1.12 a b 5.32 ± 1.22 a,b nd 4.83 ± 0.42 b nd nd nd 3.34 ± 1.15 a,4 2.49 ± 1.34 b 5.35 ± 1.31 Not determined. Within columns, means sharing the same superscript letter are not significantly different (p > 0.05; n = 4). 3 Time for the last polymer to reach steady state. 4 Time to terminate the experiment for the last polymer. 94 Following an isostatic method, the quantity of permeated ClO2 gas (kg) at time t (s) was then plotted to obtain the permeation curve. The P value was calculated from the flow rate of the permeated ClO2 gas at steady-state, Fss (kg/s). In this study, the steadystate was reached when the concentration of the permeated ClO2 detected did not increase over time and did not fluctuate more than ± 0.01 ppmV for the remainder of the experiment. The experiment was successfully terminated when steady-state was reached and held for four-fold the lag time of the particular material. The diffusion coefficient, D, was obtained by the following calculations: P  Fss  p  A  p = p2 – p1 D  (6) (7) 2 7.2t1 / 2 (8) 2 where  is the film thickness (m), A is the surface area of the film sample (m ), and p is the partial pressure gradient of ClO2 between the lower half (p2) and upper half (p1) of the permeability cell. Since there is a constant removal of permeated ClO2 from the upper chamber, p1 is equal to 0.00 Pa. The partial pressure of gaseous ClO2 over the solution placed in the lower chamber was calculated from Henry’s solubility coefficient (obtained 2 from the literature), as mentioned previously in Chapter 2, and is equal to 3.65 × 10 Pa 95 (Ishi, 1958; Kaczur and Cawlfield, 1992). The t1 / 2 is the time taken to reach a flow rate equivalent to half of Fss . 4.2.5. Determination of activation energy for permeation The temperature-dependence of the mass transfer phenomenon commonly follows an Arrhenius relationship. The activation energy of permeation (EP) can be calculated by: P E  P  P0 e RT (9) where P0 is the proportionality constant (pre-exponential term), R is the gas constant -3 -1 -1 (8.314472 x 10 kJK mol ), and T is temperature in Kelvin (K). The EP (kJ/mol) of ClO2 for PET and PLA were determined by running permeation tests at 23, 30, and 40°C and plotting the P data obtained on a graph of lnP -1 versus T . EP values were calculated from the slope which is equal to  Ep R (Auras et al., 2003; Matteucci et al., 2006; Selke et al., 2004). 4.2.6. Statistical analysis The data obtained from four replicates of each sample were statistically analyzed by analysis of variance (ANOVA) using the Statistical Analysis System (SAS) software (SAS Institute Inc., Cary, NC) at the confidence level of 95% ( = 0.05) with Tukey’s adjustment for comparison of means. 96 4.3. Results and Discussion 4.3.1. Mass transfer of ClO2 in polymeric films The mass transfer profiles of 10 mgClO2/L of gas (3600 ppmV) through various polymeric materials are shown in Figure 4-2. These permeation profiles represent the cumulative permeated gaseous ClO2 (kg) of each material versus time (s) (Rubino et al., 2001). Consistency tests were performed on obtained data and all values were reported in Appendix 3. The experiments assumed to follow the isostatic method (Gavara and Hernandez, 1993). The corresponding P, D, and S values of the materials as determined at 23°C are listed in Table 4-2. -17 The P value of gaseous ClO2 for LLDPE was 96.8 x 10 1 -2 -1 kgClO2mm s Pa - , and is comparable to the value previously reported by Ozen (Ozen, 2000). Permeability coefficients of BOPP and nylon determined in this study are also comparable to those of ® OPP and BON reported. Since the response of the particular EC detector (ToxiPro ) is typically within 90 seconds (Emond and Kearney, 2007), the lag time of the mass transfer phenomenon could be recorded, thus the D and S values were determined in this work. For an isostatic technique, such an immediate response of the detector also leads to the more precise identification of the time to reach steady state flow rate, Fss . Permeabilities of gaseous ClO2 through PET, nylon, BOPP, PLA and multilayer EVA/EVOH/EVA were at least one order of magnitude lower than those through PE, PVC, and PS, indicating that the former materials are better barriers for ClO2 gas. 97 2.5e-10 LLDPE Cumulative permeated ClO2, kg 2.0e-10 1.5e-10 LDPE 1.0e-10 PS HDPE PVC 5.0e-11 BOPP Nylon, PET PLA EVA/EVOH/EVA 0.0 0 5000 10000 15000 20000 25000 30000 Time of exposure, s Figure 4-2. Mass transfer of 10 mgClO2/L ClO2 gas (3600 ppmV) through polymeric material. Adapted from (Netramai et al., 2009). 98 Table 4-2. Permeability (P), diffusion (D), and solubility (S) coefficients of 10 mgClO2/L (3600 ppmV) for selected polymers at 23°C Polymer 2 -14  m    -17  KgClO2  m  P × 10    2   m  s  Pa  D × 10 a,1  s    a HDPE 24.1 ± 0.42 LDPE 66.0 ± 1.09 LLDPE 96.8 ± 1.34 BOPP 3.04 ± 0.16 PS 41.8 ± 0.82 PVC 23.5 ± 0.46 PET 1.26 ± 0.03 PLA 5.40 ± 0.13 Nylon 1.80 ± 0.07 EVA/EVOH/EVA Permeability is less than 0.07 3.14 ± 0.06 b b 26.2 ± 5.11 c c 40.4 ± 2.40 d d 0.39 ± 0.02 e e 9.15 ± 0.47 a e 7.65 ± 0.96 f d 0.32 ± 0.02 g a 2.86 ± 0.18 h f 0.53 ± 0.02 1 -3  S × 10   Kg   3   m  Pa  a 7.68 ± 0.27 b,d 2.59 ± 0.46 b 2.40 ± 0.17 a 7.84 ± 0.17 c 4.57 ± 0.02 d 3.11 ± 0.38 e 3.92 ± 0.19 f 1.90 ± 0.15 d 3.43 ± 0.18 KgClO2  m (24 hour of exposure) m 2  s  Pa Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 4). Adapted from (Netramai et al., 2009). 99 -19 No permeability (at detection limit of 7.32 × 10 -2 -1 -1 kgClO2mm s Pa ) was observed in EVA/EVOH/EVA multilayer film samples exposed to ClO2 for 24 h. Among all the selected materials under consideration, this material seemed to have the highest barrier against gaseous ClO2; the strong intermolecular forces from polar groups and hydrogen bonds may lower the chain mobility of the polymer and limit the available free volume as the molecules are held close together (Matteucci et al., 2006). The diffusion mechanism plays an important role in the mass transfer of ClO2 in PE films (Table 4-2). While the S values of PE films are in the same range, the D values of LDPE and LLDPE are approximately 8 and 13 times higher, respectively, than that of HDPE. As diffusion coefficients are particularly affected by the polymer’s Tg and free volume properties, (Matteucci et al., 2006) the differences in mass transfer profiles of ClO2 through different types of PE were probably due mainly to the differences in their free volumes, i.e., the unoccupied spaces between molecules in the polymers’ structures (Selke et al., 2004). The obtained FFV, FV, CED, crystallinity, and Tg of each material are listed in Table 4-3. The estimated FV of PE films have the following relationship: LLDPE > LDPE > HDPE. In HDPE, the polymer chains are closely packed, resulting in a denser structure and, consequently, less free volume and less active area for permeation, as compared to those of LDPE and LLDPE, which are branched and partial-branched structures (Rodriguez et al., 2003; Selke et al., 2004). For the selected PE samples, ClO2 permeability of the films increased, as their estimated FV increased. 100 Materials considered to be a good barrier usually have low FV and high CED (small D and S) (Alentiev and Yampolskii, 2002). Even though the material’s free volume is regarded as an independent property, it is implied that the free volume will be more abundant in material with weak intermolecular interactions, i.e., low CED (Alentiev and Yampolskii, 2002). Numbers of researches also reported positive empirical correlations between the FV/CED ratio and mass transfer properties of materials for many permeant/polymer pairs (Alentiev and Yampolskii, 2002; Li et al., 1996; LopezRubio et al.). In this work, the plot of the FV/CED ratio against D and P values of ClO2 for various polymeric materials at 23°C (Figure 4-3 and Figure 4-4, respectively) showed a positive correlation between FV/CED and mass transfer coefficients. FV and CED were then separately plotted against D values to further pinpoint the major factor affecting mass transfer of ClO2 gas, in Figure 4-5 and Figure 4-6, respectively. Positive correlation between FV and D values was observed, while there was no significant correlation between CED and D values. Also, from the plots in Figure 4-5, Figure 4-3, and Figure 4-4), a polymer tested in the glassy stage with a high barrier to ClO2 has a low FV/CED ratio resulting from low FV and/or high CED values. Thus, the glassy or rubbery stage (Tg) of a material and the FV has an important effect on the permeability of ClO2. Even though the FV of the PE films were less than those of most of the remaining materials, the D values of PE films were very high, resulting in high P values. The PE films were tested under rubbery stages (the testing temperature is above the Tg), which 101 enables the polymer’s chain mobility and facilitates diffusion of the permeant (Van Krevelen, 1997). BOPP was also tested above its Tg, but its low P value could be due to the decreased mobility of the oriented material (Dias et al., 2008) and the highly crystalline nature of isotactic PP, which reduces the availability of free volume in its structure (Selke et al., 2004). The tightly packed polymer chains create a tortuous path for the permeant to move through, thus limiting the diffusion of ClO2 gas and resulting in its low permeability. PS had a high permeability value due to high D and fairly high S values. This result, similar to the case of PE films, was due to the abundant free volume created by the bulkiness of the benzene ring side group and the chain stiffness of the PS molecule (Matteucci et al., 2006). Also, the PS commonly used for packaging purposes is atactic, having a noncrystalline structure (Selke et al., 2004). For PVC, its high D value resulted in a rather high permeability of ClO2. Since most PVC available for packaging purposes is usually highly plasticized to improve processability (Selke et al., 2004), its poor barrier to ClO2 could be due to the presence of plasticizers that act as lubricants, reducing the intermolecular forces between polymer molecules, increasing the chain mobility, and facilitating the movement of ClO2 molecules through the polymer’s matrix (Matteucci et al., 2006). The high CED and low FV values that lead to the low S and D values likely accounted for the low permeability of ClO2 in nylon and PET. The incompatibility of 102 both materials to ClO2 was possibly due to their high intermolecular forces, generating a chemically rigid compact structure to ClO2. For PLA film, the lower CED value and higher FV make the polymer a much poorer barrier to ClO2 as compared to PET, even though the barrier properties to gases and vapors of the two polymers are normally comparable (Auras et al., 2003; Matteucci et al., 2006). 4.3.2. Activation energy of ClO2 The EP data of ClO2 for PET and PLA films obtained from plotting P data -1 obtained at various temperatures (lnP versus T ) are shown in Table 4-4. The Ep value for PET is significantly lower than that for PLA indicating that the permeation of ClO2 through PLA is less temperature dependent, which could be considered beneficial in a packaging system that is occasionally subjected to temperature abuse during transportation or distribution. 103 Table 4-3. Polymer film characteristics including density (ρ), fractional free volume (FFV), free volume (FV), and cohesive energy density (CED) Polymer Thickness × 10 m  -5 3 ρ × 10  kg     3 m  Crystallinity (%) Tg (oC) 1 HDPE 5.59 0.931 38.63 ± 3.11 LDPE 2.79 0.924 23.04 ± 3.31 nd nd LLDPE 4.57 0.917 24.13 ± 1.52 nd BOPP 1.78 0.917 33.08 ± 2.01 3.00 ± 1.36 PS 2.54 1.03 na 92.65 ± 0.20 PVC 2.79 1.30 na 57.96 ± 0.49 PET 1.27 1.39 22.38 ± 1.36 81.66 ± 0.50 PLA 3.81 1.21 31.44 ± 0.55 69.07 ± 0.27 Nylon 2.03 1.13 24.35 ± 0.46 nd 1.78 na 4 EVA/EVOH/EVA: EVA EVOH 1 2 17.25 ± 3.66 2.25 ± 0.20 nd FV and FFV calculations were based on data at 23°C. CED values were calculated according to the group contribution method at 25°C (Van Krevelen, 1997). 3 Not determined due to equipment limitations. 4 Not available due absence of a particular attribute, or test not performed. 104 Table 4-3. (cont’d) 2 Polymer 1 -2 FFV × 10 1 -2 FV × 10 5 CED × 10  kJ     3 m  -7 FV/CED × 10  m3     kJ    HDPE 11.7 7.21 3.00 2.40 LDPE 12.5 9.58 3.00 3.19 LLDPE 13.1 9.93 3.00 3.31 BOPP 13.1 8.78 2.66 3.29 PS 19.3 19.3 4.11 4.69 PVC 21.0 21.0 4.41 4.76 PET 11.2 8.67 5.43 1.60 PLA 21.9 15.0 5.18 2.89 Nylon 15.5 11.7 6.47 1.81 na na na na EVA/EVOH/EVA: EVA EVOH 1 2 FV and FFV calculations were based on data at 23°C. CED values were calculated according to the group contribution method at 25°C (Van Krevelen, 1997). 3 Not determined due to equipment limitations. 4 Not available due absence of a particular attribute, or test not performed. 105 5e-13 LLDPE 4e-13 2 -1 D, m -s 3e-13 LDPE 2e-13 PS 1e-13 HDPE PET PVC Nylon PLA 0 PP 1e-7 2e-7 3e-7 4e-7 5e-7 3 FV/CED, cm /J Figure 4-3. Correlation between the free volume and cohesive energy density ratio (FV/CED) and the diffusion coefficient (D) of ClO2 gas 106 1e-15 LLDPE -2 -1 -1 P, KgClO2-m-m -s -Pa 1e-15 8e-16 LDPE 6e-16 PS 4e-16 HDPE 2e-16 PVC PET Nylon PLA 0 1e-7 2e-7 PP 3e-7 4e-7 5e-7 3 FV/CED, cm /J Figure 4-4. Correlation between the free volume and cohesive energy density ratio (FV/CED) and the permeability coefficient (P) of ClO2 gas 107 5e-13 LLDPE 4e-13 3e-13 2 -1 D, m -s LDPE 2e-13 PS 1e-13 HDPE PVC PP PLA PET 0 0.06 0.08 Nylon 0.10 0.12 0.14 0.16 0.18 0.20 0.22 FV Figure 4-5. Correlation between the free volume (FV) and the diffusion coefficient (D) of ClO2 gas 108 1.2e-13 1.0e-13 PS 2 -1 D, m -s 8.0e-14 PVC 6.0e-14 4.0e-14 PLA 2.0e-14 0.0 PP 2e+5 Nylon PET 3e+5 4e+5 5e+5 6e+5 7e+5 -3 CED, kJ/m Figure 4-6. Correlation between the free volume (FV) and the diffusion coefficient (D) of ClO2 gas 109 Table 4-4. Permeability coefficients (P) at various temperatures and activation energy for permeation (EP) of ClO2 for PET and PLA films PET Temperature o ( C) 23 PLA -17 -17 P × 10 P × 10  KgClO2  m     2   m  s  Pa  EP (kJ/mol) a,1 b 2.67 ± 0.04 40 EP (kJ/mol) a 1.26 ± 0.00 30  KgClO2  m     2   m  s  Pa  5.40 ± 0.13 51.05 ± 4.35 3.98 ± 0.05 c b 21.5 ± 0.74 129.03 ± 2.82 c 94.4 ± 0.52 1 Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 4). Adapted from (Netramai et al., 2009). 110 4.4. Conclusion A continuous system for measuring mass transfer of ClO2 through different polymeric materials was developed utilizing an electrochemical detector (details in Chapter 3). The permeability, diffusion, and solubility coefficients (P, D and S, respectively) of 10 mg ClO2/L of gas (3600 ppmV) for 10 types of selected polymeric materials were determined by an isostatic method. The results ranged from a permeability -17 value below 0.07 × 10 -17 96.8 × 10 -2 -1 kgClO2mm s Pa -2 -1 kgClO2mm s Pa -1 -1 for multilayer EVA/ EVOH/EVA to for LLDPE. BOPP, PET, PLA, nylon, and multilayer EVA/EVOH/EVA are good barriers for gaseous ClO2 as compared to PEs, PVC, and PS, which have higher permeability coefficients than the former group. Calculated cohesive energy density and free volume values can play an important role in predicting the mass transfer behavior of ClO2 through the tested film samples. The activation energy of permeation, in the temperature range of 23 to 40°C, for ClO2 gas for PET and PLA were found to be 51.05 and 129.03 kJ/mol, respectively. 111 BIBLIOGRAPHY 112 Alentiev, A. Y. and Yampolskii, Y. P. (2002). "Meares equation and the role of cohesion energy density in diffusion in polymers." Journal of Membrane Science 206: 291306. ASTM (2003). ASTM D3418-03 Standard Test Method for Transition Temperatures and Enthalpies of Fusion and Crystallization of Polymers by Differential Scanning Calorimetry. West Conshohocken, PA, ASTM International. Auras, R. A., Harte, B., Selke, S. and Hernandez, R. H. (2003). "Mechanical, Physical, and Barrier Properties of Poly(lactide) Films." Journal of Plastic Film & Sheeting 19(2): 123 - 135. Dias, P., Lin, Y. J., Hiltner, A., Baer, E., Chen, H. Y. and Chum, S. P. (2008). "Effect of Chain Architecture on Biaxial Orientation and Oxygen Permeability of Polypropylene Film." Journal of Applied Polymer Science 107: 1730–1736. Emond, J. and Kearney, T. (2007). Personal Communication: Electrochemical Sensor Technology. Netramai, S., Biosystems: 15. Gavara, R. and Hernandez, R. J. (1993). "Consistency Test for Continuous Flow Permeability Experimental Data " Journal of Plastic Film and Sheeting 9(2): 126138. Ishi, G. (1958). "Solubility of chlorine dioxide." Chemical engineering (Japan) 22: 153 154. Kaczur, J. J. and Cawlfield, D. W. (1992). Chlorine Oxygen Acids and Salts, Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical Technology. Kroschwitz, J. I. New York, John Wiley & Sons Inc. 5. Li, Y., Weng, Z., Huang, Z. and Pan, Z. (1996). "Barrier Properties of Vinylidene Chloride Copolymers." Chinese Journal of Polymer Science 14(2): 105-110. Lopez-Rubio, A., Cava, D., Giménez, E., Cabedo, L., Catalá, R., Gavara, R. and Lagarón, J. M. "How the fields of synchrotron X-Ray analysis and nanotechnology cam help increase packaged food quality " Retrieved January 12th, 2009, from http://europe.ilsi.org/NR/rdonlyres/4F673305-60CF-424F-87D958FC44DE5985/0/24Lagaron.pdf. Matteucci, S., Yampolskii, Y., Freeman, B. D. and Pinnau, I. (2006). Transport of Gases and Vapors in Glassy and Rubbery Polymers. Materials Science of Membranes for Gas and Vapor Separation. Yuri Yampolskii, I. P. B. F.: 1-47. Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2009). "Mass Transfer Study of Chlorine Dioxide Gas through Polymeric Packaging Materials." Journal of Applied Polymer Science 114(5): 2929-2936. 113 Ozen, B. F. (2000). Effect of Ozone and Chlorine Dioxide Treatments Used in Perishable Food Applications on Polymeric Materials: Changes in Mechanical, Thermal and Mass Transfer Properties. Food Science Department. West Lafayette, IN, Purdue University. Ph.D. Recio, R., Lozano, A. E., Pra´danos, P., Marcos, A., Tejerina, F. and Herna´ndez, A. (2008). "Effect of Fractional Free Volume and Tg on Gas Separation Through Membranes Made with Different Glassy Polymers." Journal of Applied Polymer Science 107: 1039-1046. Rodriguez, F., Cohen, C., Ober, C. K. and Archer, L. A. (2003). Principles of Polymer Systems. New York, Taylor & Francis. Rubino, M., Tung, M. A., Yada, S. and Britt, I. J. (2001). "Permeation of Oxygen, Water Vapor, and Limonene through Printed and Unprinted Biaxially Oriented Polypropylene Films." Journal of Agricultural and Food Chemistry 49(6): 30413045. Sapers, G. M., Walker, P. N., Sites, J. E., Annous, B. A. and Eblen, D. R. (2003). "Vapor-phase decontamination of apples innoculated with Escherichia coli." Journal of Food Science 68(3): 1003-1007. Selby, T. L., Wong, E., Han, Y., Gerrard, D. E. and Linton, R. H. (2005). Effects of chlorine dioxide gas treatment on microbial safety and quality of fresh pork. 2005 IFT Annual Meeting, New Orleans, Louisiana, IFT. Selke, S. E. M., Culter, J. D. and Hernandez, R. J. (2004). Plastics Packaging: Properties, Processing, Applications, and Regulations. Cincinnati, OH, Hanser Gardner Publications, Inc. Sy, K. V., McWatters, K. H. and Beuchat, L. R. (2005a). "Efficacy of Gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries, and Raspberries." Journal of Food Protection 68(6): 1165 - 1175. Sy, K. V., Murray, M. B., Harrison, M. D. and Beuchat, L. R. (2005b). "Evaluation of gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes, and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection 68(6): 1176 - 1187. Van Krevelen, D. W. (1997). Properties of Polymers. New York, Elsevier Science B.V. 114 CHAPTER 5: EFFECTS OF GASEOUS ClO2 EXPOSURE ON IMPORTANT PROPERTIES OF POLYMERIC FILMS 5.1. Introduction New packaging strategies, such as including antimicrobial gas in the headspace have become a potential complementary approach to improve the safety of packaged fresh produce (Sy et al., 2005). Chlorine dioxide (ClO2), with its high oxidizing capacity and broad disinfecting property, is considered by many researchers as a promising choice for such applications (Kaczur and Cawlfield, 1992; Kim et al., 1999; Sapers et al., 2003; Sy et al., 2005), however, when exposing polymeric materials to oxidizing agents, chemical changes can occur. These changes can impact the integrity and performance of the packaging material, consequently, resulting in a reduction of the product shelf-life. Thus, the specific objectives of this work were to determine and evaluate the effects of exposure of gaseous ClO2 on a) chemical, b) physical, c) mechanical, and d) barrier properties of different polymeric materials. 115 5.2. Materials and Methods 5.2.1. Experimental design Film samples were randomly placed in glass jars (2 films, with random polymer type, per jar) and exposed to ClO2 gas generated by a ClO2 solution in the jar bottom, which released 10 mgClO2/L (approximately, 3,600 ppmV) of ClO2 gas into the headspace. Film samples were periodically removed from the ClO2 treatment in order to characterize their chemical, physical, mechanical, and barrier properties as outlined in Figure 5-1. o The samples were conditioned at 23 C, 50% RH for at least 24 hours, before ClO2 exposure. The films were then exposed to gaseous ClO2 for 1, 7, and 14 days (‘Day 1’, ‘Day 7’, and ‘Day 14’ samples, respectively). After removing the sample at the o specific time, the treated films were conditioned by storing at 23 C, 50% RH for at least 24 hours. The unexposed films or ‘control’ (Day 0) were also similarly conditioned. The chemical, physical and mechanical properties of the conditioned control (Day 0) and treated (Day 1, Day 7, and Day 14) polymeric materials were then evaluated. 116 Figure 5-1. Flow diagram of ClO2 treatment of polymeric materials. Adapted from (Netramai et al., 2010). 117 5.2.2. Polymeric packaging materials The same polymeric materials used in a previous study (Netramai et al., 2009) were considered for this study, i.e. LDPE (Flexopack S.A., Attiki, Greece), LLDPE (Flexopack S.A., Attiki, Greece), HDPE (James River Corp. Flexible Packaging Group, Richmond, VA), BOPP (Cryovac, Duncan, SC), PS (TRYCITE™ 8001, Dow Chemical Company, Midland, MI), PET (Mylar  A, DuPont, Wilmington, DE), PVC (BEMIS,  Shirley, MA), nylon (Dartek F-101, DuPont, Wilmington, DE), PLA (EVLON , BI-AX International Inc. Wingham, ON, Canada), and multilayer structure EVA/EVOH/EVA (Cryovac, Duncan, SC). 5.2.3. Preparation of ClO2 solution The same methodology as described in chapter 3 was used to prepare a final solution from a stock solution of 1,000 mg/L ClO2. To monitor ClO2 concentration over time, ClO2 solution in 8 random glass jars containing film samples were periodically sampling (another set of 8 glass jars for control). Figure 5-2 shows mg of ClO2 in the solution after 1, 7, and 14 days of film exposure. The differences between ClO2 remaining in control glass jars and those in jars with samples (Table 5-1) were assumed to be the amount of ClO2 that reacted with film samples. 118 Remaining ClO2 in solution, mg 70 Remaining ClO2 in blank solution, mg Remaining ClO2 in treatment solution, mg 60 50 40 30 20 10 0 1 7 14 Time of Exposure, Day Figure 5-2. Remaining amount of ClO2 (mg) in 200 mL solution for blank and film treatment Table 5-1. Amount of ClO2 consumed by film samples (mg) Time of Exposure (Day) Amount of ClO2 consumed by film samples (mg) 1 0.56 ± 0.26 7 4.50 ± 0.30 14 8.03 ± 0.07 119 5.2.4. Evaluation of chemical structure of polymeric material Fourier transform infrared (FT-IR) spectroscopy was used to evaluate the chemical changes of the polymer after ClO2 exposure using a FT-IR spectrophotometer (Shimadzu IR Prestige-21, Shimadzu Scientific Instruments, Columbia, MD). In addition to comparing the IR spectrum of the control, exposed, and conditioned films, IR spectra of exposed samples without conditioning were also collected, in order to evaluate if the changes were transient or permanent. 5.2.5. Evaluation of physical properties o Glass transition (Tg) and melting (Tm) temperatures ( C), and enthalpy of fusion, Hm (J/g) of the control and exposed polymeric films were determined using a differential scanning calorimeter (DSC Q-100, TA Instruments, New Castle, DE) according to ASTM D3418-03 (ASTM, 2003b). The analyses were done using Universal Analysis Software (UAS Version 3.9A, TA Instruments, New Castle, DE). 5.2.6. Evaluation of mechanical properties -2 -2 Tensile strength, TS (Nm ) and modulus of elasticity, MoE (Nm ), in both the machine direction (MD) and transverse direction (TD) of the control and exposed polymeric films, were measured using a universal tensile tester machine (Instron 5565, Instron Inc., Canton, MA), according to ASTM D 882-97 (ASTM, 2002). Each film sample was run by five replicates. The predetermined parameters for measuring TS and MoE of the samples were as follows: 120  -1 -1 4  - - Initial grip separation of 1.25 x 10 m with a rate of grip separation of 2.08 x 10 m/s were used for PS and PLA. Initial grip separation of 1.00 x 10 m with a rate of grip separation of 8.33 x 10 4 m/s were used for BOPP, PVC, PET, and multilayer structure of EVA/EVOH/EVA.  -2 Initial grip separation of 5.00 x 10 m with a rate of grip separation of 8.33 x 10 3 - m/s were used for PEs and nylon. 5.2.7. Evaluation of barrier properties Barrier characteristics of the control and exposed polymeric films were determined in accordance to ASTM D1434-82 (2003) (ASTM, 2003a). Water vapor transmission rates (WVTR) were evaluated using a water vapor permeability analyzer  100, (Permatran-W Model 3/11, Mocon, Minneapolis, MN). Carbon dioxide permeation rates (CO2TR) were evaluated using a carbon dioxide permeability analyzer (Permatran-C TM Model 4/41, Mocon, Minneapolis, MN). Oxygen transmission rates (O2TR) were measured using an oxygen permeability analyzer (O2TR 8001, Illinois Instruments, Johnsburg, IL). The experimental conditions for each of the measurement were as follows: • -2 -1 Water vapor transmission rate (kgm s ): relative humidity difference across the film 100%, temperature 23°C. 121 • -2 -1 Carbon dioxide transmission rate (kgm s ): permeant 101325 Pa of CO2, carrier gas N2, temperature 23°C, and 0% relative humidity. • -2 -1 Oxygen transmission rate (kgm s ): permeant 21278.25 Pa of O2, carrier gas N2, temperature 23°C, and 0% relative humidity. -5 The tested film areas were between 7.90 x 10 -3 to 5.00 x 10 2 m , depending on the barrier characteristic of each sample. The obtained WVTR, CO2TR, and O2TR were then used to calculate PH 2O , PCO 2 , and PO2 , respectively, of the film samples by multiplying the thickness of the samples and dividing by the partial pressure gradients. -2 -1 -1 The final values were expressed as Kgmm s Pa . 5.2.8. Color measurement of polymeric material Surface color of the films was measured from two random locations per piece of film, four films for each sample type, for L*, a* and b* values using a reflectometer (Integrating Reflectometer JY 9800, TMI Testing Machines INC, Ronkonkoma, NY). The overall color difference ( E ) was calculated using: E  L2  a 2  b 2 (1) where L , a , and b are the differences between the L, a, and b values, respectively, of the control sample, and those of the exposed sample. 122 5.2.9. Statistical analysis With the exception of mechanical testing (5 replicates), all other tests were based on 4 replicates. The data obtained were statistically analyzed by analysis of variance (ANOVA) using the Statistical Analysis System (SAS) software (SAS Institute Inc., Cary, NC) at a confidence level of 95% ( = 0.05) with Tukey’s adjustment for comparison of means. 5.3. Results and Discussion To monitor the changes in the materials’ integrities and performance, the film samples were exposed to gaseous ClO2 at a concentration of 3,600 ppmV, for up to 14 days, which is near the upper range of ClO2 gas used in treating food produce. The exposure times were selected according to the shelf-life of fresh produce (Shin et al., 2006). A one day exposure was considered short-term exposure, at which the mass transfer of ClO2 at this particular concentration is at steady state in all of the testing polymeric materials, with the exception of multilayer EVA/EVOH/EVA (Netramai et al., 2009). The 7 and 14 days exposures were regarded as long-term exposure which is referred to as ‘persistent gas exposure’ (PGE). A PGE could be encountered when a sustained release device is included in the packaging system. According to this scenario, ClO2 gas could be present at a specific concentration within the package atmosphere up to the end of the product shelf-life. The chemical, physical and mechanical properties were evaluated in function of the short and long-term exposure. 123 5.3.1. Chemical structure of polymeric material Changes in the IR spectra were attributed to possible interactions between ClO2 gas and the polymeric film samples (Figure 2-5). In general, the types of changes observed from persistent ClO2 exposure agreed with those for short-term. As the time of exposure increased, the degree of change also increased. In order to determine if the chemical changes were transient or permanent, IR spectra of the unconditioned exposed samples were also collected. Generally, the changes in the absorbance intensities of most of the short-term exposed samples seemed to be somewhat temporary, as the absorbance of the exposed samples, after conditioning was equivalent to those of the controls. In the long-term exposure, however, most of the changes tended to be permanent, i.e. after conditioning, the absorbance of the exposed samples differed from the control. The most notable result was that of the nylon (Figure 5-3). In this case, an additional IR spectrum was also obtained at 10 hours of exposure, in which the mass transfer of ClO2 in the nylon film was at steady state (Netramai et al., 2009). Changes in the IR spectrum at exposure time of 10 hours was equivalent to those seen at 1 and 7 days when compared to the control (Day 0) sample. 124 3 O R' ... R NH ... NH n 2 Absorbance, O.D. Day 0 10 hours Day 1 Day 7 Day 14 O 0.5 0.0 1000 N-H Stretch 1 0 4000 3500 3000 2500 2000 1500 1000 -1 Wavenumber, cm Figure 5-3. FT-IR spectra of nylon; () day 0; () 10 hours; (- - -) day 1; (-  -) day 7; and ( ) day 14. Adapted from (Netramai et al., 2010). 125 Generally, for film samples affected by ClO2, the changes in peak intensity of the 14 days had been the most dramatic, followed by the 7 days and those of the short-term exposure. However, the peak intensities changed of a slower rate during longer exposure. This could be attributed to the availability of functional groups that react with ClO2, as oxidative degradation is normally a surface phenomenon (Rivaton and Gardette, 1998; Rodriguez et al., 2003). The IR spectra for HDPE, LDPE, and LLDPE (Figure 5-4) showed minor changes -1 in the intensity of the peaks in the 2700-3000 cm region, which indicates possible changes in the C-H bond in the methyl or methylene group. The IR spectra of PVC, and BOPP materials (not shown) showed only slight changes in their absorbance intensity. -1 The shifts of the peaks in the fingerprint area, i.e. 750-1400 cm region, to the higher wavenumbers in the exposed PEs and PS samples, indicate the possible presence of a CCl bond in the exposed samples (Robinson et al., 2005). The increased absorbance intensities in the fingerprint area of the exposed nylon (Figure 5-3) and EVA/EVOH/EVA films (Figure 5-6) also suggest partial chlorination (Robinson et al., 2005) in the exposed samples. If the degree of chlorination is significantly high, this alteration should increase the polymers’ polarity, leading to a film with improved barrier properties to gases and other organic compounds, (Kharitonov, 2000; Selke et al., 2004). 126 3 H C-H Stretch (methylene group) ... H Absorbance, O.D. 2 Day 0 Day 1 Day 7 Day 14 H H n ... 2.4 0.2 1.6 2800 1 0.0 1000 0 4000 3500 3000 2500 2000 1500 1000 -1 Wavenumber, cm Figure 5-4. FT-IR spectra of LLDPE; () day 0; () day 1; (- - -) day 7; and (-  -) day 14. Adapted from (Netramai et al., 2010). 127 Increased absorbance intensities of the exposed PLA, and nylon (Figure 5-3) -1 samples in the 3300-3700 cm region coincided with changes in the hydroxyl group, and -1 N-H bond, respectively. The increase in absorbance intensities in the 1100-1200 cm region for the exposed nylon could indicate changes in the C-N bond (Ozen et al., 2002; Robinson et al., 2005). In the spectra shown in Figure 5-5, slight changes can be seen in the peak -1 intensities in 2800-3100 cm region for the exposed PET films that might represent changes in the C-H bond in its methylene group and/or benzene ring. Such changes were not observed in exposed PS samples where the absorbance signals, at that particular region, were at the noise level. The surface oxidations for PET are reported to be complex, and can lead to the formation of many functional groups, e.g. carboxylic acid, terminal vinyl groups, and phenols (Walzak et al., 1995), however, there seemed to be no significant formation of such groups in the ClO2-exposed samples, under these testing conditions. The spectrum of the exposed multilayer EVA/EVOH/EVA (Figure 5-6) showed -1 an increase in the intensity of the peaks within the 3000-3700 cm region, which indicates a change in the hydroxyl group. The minor increase in absorbance of the peak in -1 the 1600-1700 cm region could be due to the formation of a carbonyl group. 128 3 Day 0 Day 1 Day 7 Day 14 ... O O O O ... Absorbance, O.D. 2 n 0.2 0.0 1 3000 C-H Stretch (methylene group and benzene ring) 0 4000 3500 3000 2500 2000 1500 1000 -1 Wavenumber, cm Figure 5-5. FT-IR spectra of PET; () day 0; (- - -) day 1; (-  -) day 7; and ( ) day 14. Adapted from (Netramai et al., 2010). 129 3 OH Day 0 Day 1 Day 7 Day 14 ... ... ... m Absorbance, O.D. n CH3 2 O O ... ... ... m n 1 O-H Stretch C=O Stretch 0 4000 3500 3000 2500 2000 1500 1000 -1 Wavenumber, cm Figure 5-6. FT-IR spectra of EVA/EVOH/EVA; () day 0; () day 1; (- - -) day 7; and (-  -) day 14. Adapted from (Netramai et al., 2010). 130 5.3.2. Physical properties Tg of PS and PET, and the Tm of nylon and EVOH (Table 5-2), were significantly lower than those of the control samples. An increase in the heat of fusion of exposed nylon was also observed, indicating an increase in crystallinity after exposure. The physical properties of other ClO2-exposed samples remained. The shift in Tm and the increase in crystallinity of ClO2-exposed nylon could result from an increase in molecular ordering, frequently observed when exposing nylon to strong oxidizing compounds, which, in turn, can increase crystallinity (Ozen et al., 2002). The small decrease in Tg for ClO2-exposed PS and PET, together with changes in the C-H bond in the methyl group observed through IR spectra mentioned previously, suggests that the main chain scission reaction could have taken place, promoting slight degradation of the exposed materials which increased polymer chain mobility (Buchalla et al., 1993a; Ozen et al., 2002; Walzak et al., 1995). In light of these changes in chemical and physical properties after ClO2-exposure, a change in the material’s performance as a packaging film could reasonably be expected (Buchalla et al., 1993a; b; Ozen et al., 2002). Thus, the barrier, and mechanical properties of the exposed samples were characterized. 131 Table 5-2. Physical properties of selected polymeric packaging materials exposed to gaseous ClO2. Adapted from (Netramai et al., 2010). Tg Tm ( C) ( C) 0 na 133.55 ± 0.26 14 na 133.86 ± 0.82 0 na 112.44 ± 0.35 14 na 112.31 ± 0.23 0 na 122.52 ± 0.24 14 na 122.27 ± 0.12 62.85 ± 5.90 0 57.96 ± 0.49 a na na 14 58.15 ± 0.73 a na na 0 92.65 ± 0.20 a na na 14 91.64 ± 0.51 b na na 0 57.40 ± 2.11 14 Sample Time of Exposure (day) 59.55 ± 0.24 HDPE o 2 a a a LDPE a a LLDPE PVC PS PP 1 2 Heat of Fusion (J/g) o a a a a 161.51 ± 0.82 a 160.35 ± 0.58 a 111.27 ± 8.96 a 111.10 ± 10.41 a 66.35 ± 9.53 a 65.97 ± 2.33 a 69.48 ± 4.37 a a 26.14 ± 1.59 a 25.99 ± 3.77 Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 5). Not available due absence of a particular attribute, or test not performed. 132 Table 5-2. (cont’d) Tg (oC) 0 69.07 ± 0.27 14 69.20 ± 1.21 0 81.66 ± 0.50 14 80.37 ± 0.22 248.38 ± 0.18 0 na 261.35 ± 0.15 14 na 259.81 ± 0.30 0 na 123.74 ± 1.22 14 na 122.82 ± 0.68 0 na 159.39 ± 0.46 14 Sample Time of Exposure (day) na 147.81 ± 1.14 PLA PET Tm (oC) a a a b Nylon Heat of Fusion (J/g) a 167.00 ± 0.22 a 166.72 ± 0.32 a 248.94 ± 0.67 a a b a 29.24 ± 0.51 a 28.12 ± 0.51 a 31.33 ± 1.91 a 33.99 ± 3.23 a 56.01 ± 1.06 b 62.91 ± 1.81 EVA/EVOH/EVA: EVA EVOH 1 2 a a a b a 50.55 ± 10.73 a 52.54 ± 7.14 a 54.35 ± 11.54 a 56.49 ± 7.68 Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 5). Not available due absence of a particular attribute, or test not performed. 133 5.3.3. Mechanical properties Tensile strength (TS) and/or modulus of elasticity (MoE) of the exposed PE films (Table 5-3) decreased significantly due to oxidative degradation of the polymer chains which was confirmed by the increase in methyl groups as seen in the IR spectra (Figure 5-4) (Ozen et al., 2002). Such degradation can reduce film structure’s rigidity, and consequently alter the tensile characteristics of the oxidized polymer (Selke et al., 2004). Mechanical performance of the control and other ClO2-exposed films remained statistically unchanged, even though the formations of polar groups were observed in most of the exposed samples. Theoretically, increased polarity should improve the intermolecular forces between the chains, leading to an increase in tensile strength and a decrease in elongation at break, as it limits the mobility of the polymer chain (Selke et al., 2004). 134 Table 5-3. Tensile properties of selected polymeric packaging materials exposed to gaseous ClO2. Adapted from (Netramai et al., 2010). Sample HDPE Time of Exposure (day) 0 7 2 8 Tensile strength × 10 (N/m ) MD TD a,1 MD a 2.71 ± 0.06 4.15 ± 0.12 b a 14 PS PET PVC Nylon EVA/EVOH/EVA 1 0 3.08 ± 0.12 14 2.85 ± 0.12 0 8.48 ± 0.44 14 8.65 ± 0.23 0 19.2 ± 1.36 14 18.8 ± 0.95 21.4 ± 0.68 0 9.13 ± 0.26ª 9.87 ± 0.62 14 9.17 ± 0.38ª 9.34 ± 0.55 0 6.86 ± 0.38 14 6.42 ±0.52 0 LDPE 3.45 ± 0.38 11.2 ± 0.34 14 10.0 ± 2.31 2.24 ± 0.52 a a 1.99 ±0.03 b a 1.88 ± 0.10 a a 7.98 ± 0.32 a a 7.56 ± 0.13 a a 21.5 ± 0.41 a a a a a a 6.80 ± 0.53 a a 5.83 ± 1.06 a a 9.53 ± 0.39 a a 8.59 ± 0.69 TD a 6.09 ± 0.14 b 5.28 ± 0.21 a 0.89 ± 0.06 a 0.83 ± 0.02 a 33.6 ± 1.35 a 34.4 ± 0.09 a 43.2 ± 1.92 a 43.8 ± 0.33 a 11.5 ± 0.38 a 11.6 ± 0.23 a 9.68 ± 0.42 a 9.50 ± 0.21 a 5.51 ± 0.15 a 6.14 ± 0.91 a 6.68 ± 0.28 a 6.45 ± 0.51 a 0.84 ± 0.06 a 0.76 ± 0.04 a 31.9 ± 1.53 a 31.3 ± 0.10 a 47.2 ± 1.03 b 44.4 ± 0.67 a 11.8 ± 0.32 a 11.8 ± 0.24 a 10.1 ± 0.55 a 9.78 ± 0.44 a 4.92 ± 0.75 a 5.96 ± 0.63 Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 5). 135 2 Modulus (secant) × 10 (N/m ) 5.3.4. Barrier properties One potential application of ClO2 gas is to be used in combination with other gases, such as O2 and CO2 in a modified atmosphere packaging (MAP) system, so the impact of ClO2 gas on the barrier properties to O2, CO2 and moisture of different polymeric materials is of concern. Barrier properties of these polymeric materials are one of the most important performance concerns in food packaging systems, especially for fresh produce that is respiring after harvest (Martinez-Romero et al., 2003). The concentrations of O2 and CO2 that accumulate in the package headspace affect the deterioration rate of fresh produce. Once the material is selected for a particular commodity, it is crucial that its permselectivity (PCO2/PO2) ratio be maintained through the shelf-life of the product, as in the case of MAP where a stable gaseous ratio in the package headspace is necessary (Exama et al., 1993; Martinez-Romero et al., 2003; Selke et al., 2004). By exposing polymeric materials to reactive chemical compounds such as ClO2, their barrier properties might be altered (Ozen, 2000). The comparisons of PH 2O , PCO , and PO , between the control and the exposed samples, indicated significant 2 2 changes in barrier characteristics of some polymeric materials commonly used in packaging fresh produce. 136 5.3.4.1. Barrier to moisture After ClO2 exposure, the moisture barrier of PVC and PET significantly decreased, while PH 2O of the other materials remained statistically unchanged (Table 54). It is important to note the shift to lower Tg in the exposed PET samples (Table 52), which implies oxidative degradation of the material after ClO2 exposure, could increase polymer chain mobility and decrease intermolecular forces. This plays an important role in the significant increase in PH 2O of the material, since it will accelerate diffusion process (Kharitonov, 2000). The IR spectra for ClO2- exposed PLA sample also showed an increase in hydroxyl groups, possibly indicating an increase in film polarity, which could increase intermolecular forces between the polymeric chains, and promote rigidity in the structure. However, the hydroxyl group also will favor the interaction between polymer and moisture (Ozen et al., 2002; Walzak et al., 1995). Thus, the barrier properties of PLA may remain unchanged after exposure as a result of these antagonistic interactions. 137 Table 5-4. Barrier properties of selected polymeric packaging materials exposed to gaseous ClO2. Adapted from (Netramai et al., 2010). -18 Sample Time of Exposure (day) PH O × 10 2  Kg  m     2   m  s  Pa  a,1 -18 PCO × 10 2  Kg  m     2   m  s  Pa  a 0 0.49 ± 0.00 14 0.48 ± 0.00 0 10.5 ± 0.74 14 9.84 ± 1.06 0 0.24 ± 0.00 14 0.25 ± 0.01 0 28.2 ± 1.18 14 26.9 ± 1.56 0 1.64 ± 0.01 14 1.94 ± 0.03 1.53 ± 0.04 0 na 1.23 ± 0.01 14 na 1.21 ± 0.03 0 0.40 ± 0.01 14 0.41 ± 0.01 HDPE PS BOPP PLA PET b a a a a a a a b 1 b 77.6 ± 0.40 a 111 ± 6.63 b 123 ± 6.37 a 5.30 ± 0.27 a 5.38 ± 0.25 a 37.3 ± 17.2 a 43.4 ± 13.9 a 1.54 ± 0.04 a a Nylon EVA/EVOH/EVA 58.3 ± 8.08 a a a a 0.35 ± 0.04 b 0.71 ± 0.04 -18 PO × 10 2  Kg  m     2   m  s  Pa  a 9.26 ± 1.24 a 8.40 ± 0.29 a 16.8 ± 0.57 a 15.8 ± 0.64 a 3.86 ± 0.20 a 3.91 ± 0.18 a 21.2 ± 3.25 a 30.8 ± 12.3 a 0.22 ± 0.01 a 0.26 ± 0.04 a 0.24 ± 0.01 b 0.22 ± 0.00 a 0.10 ± 0.01 b 0.13 ± 0.01 PCO PO 2 2 % Change a 6.30 ± 0.32 b 9.25 ± 0.28 a 6.58 ± 0.17 b 7.77 ± 0.19 + 18.0 a 4.75 ± 0.15 a 4.91 ± 0.25 - a 1.73 ± 0.64 a 1.41 ± 0.05 - a 6.88 ± 0.13 b 5.85 ± 0.65 - 14.9 a 5.08 ± 0.08 b 5.63 ± 0.06 + 10.9 a 3.86 ± 0.04 b 5.30 ± 0.24 Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 4). 138 + 46.8 + 37.2 5.3.4.2. Barrier to oxygen and carbon dioxide Based on the results of PO 2 listed in Table 5-4, only the nylon films showed an increase in barrier to O2, after the exposure to ClO2. This could be due to an increase in crystallinity (Table 5-2), since the exposed material had a lower available amorphous region in which permeation can take place (Kulshreshtha, 1992; Schnabel, 1992). This coincides with a study on the effects of ozone treatment on biaxial oriented nylon mentioned previously (Ozen et al., 2002). An increase of the O2 barrier in ClO2-exposed materials was speculated to be due to a possible partial chlorination which promotes polarity within the polymer matrix, however, barrier to O2 of LDPE (data not shown) and multilayer EVA/EVOH/EVA films decreased after being exposed to ClO2 gas. These results suggest that the main chain scission as shown by FT-IR spectroscopy results of PEs samples could be the dominant degradation reaction, due to ClO2 exposure, in both materials, as this would increase the chain mobility and facilitate the transfer of gas throughout the polymer structure (Table 5-4). The same explanation can reasonably be given for the increase in PCO 2 of the treated PEs, PS and EVA/EVOH/EVA films. The PCO 2 of polymeric material is usually higher than the PO 2 of the same material. Given that both gases are of the same penetrant type, i.e. non-reactive gas (Selke et al., 2004), a similar trend would be expected for ClO2-exposed polymeric films. However, the degree of change varied, depending on the particular polymer/penetrant pair (Matteucci et al., 2006). This led to the alteration of the PCO2/PO2 ratio of the 139 materials after they have been exposed to ClO2. With the exception of 7-day exposed HDPE film sample, a comparison of the modified PCO2/PO2 ratio to the original value showed no significant change for the other materials, after 1 and 7 days of treatment. However, after 14 days of persistent exposure, the PCO2/PO2 ratios of many materials have changed ranging from 10.9% to 46.8%. The most notable change occurred in the exposed HDPE sample. This is critical information when designing MAP systems for fresh produce where the ratio of CO2 and O2 within the packaging atmosphere will impact the shelf-life of the product. 5.3.5. Color measurement Overall lightness (L*) of the exposed LLDPE, PVC, PS, PET and nylon increased as compared to the unexposed samples (Table 5-5). The changes in b* values of the day 1 and day 7 samples of exposed PEs, PVC, PS, PET, and nylon films indicated that the films became more yellow in color, however, after 2 weeks of exposure, the color shifted toward being more bluish. Significant overall color differences ( E ) were found in exposed PVC and PS films. From visual observations on discoloration of the exposed samples, in general, the film samples changed from opaque white or transparent to yellowish dull and became darker in color at the end of exposure. 140 Table 5-5. Color properties of selected polymeric packaging materials exposed to gaseous ClO2. Adapted from (Netramai et al., 2010). Time of Exposure (day) L* 0 38.34 ± 0.98 1 39.24 ± 0.72 7 39.02 ± 1.04 14 39.44 ± 0.38 0 28.27 ± 0.57 1 29.28 ± 1.54 7 28.95 ± 1.12 14 32.87 ± 0.54 0 25.96 ± 0.83 1 34.55 ± 1.23 7 35.74 ± 1.81 14 Sample 38.05 ± 2.18 LDPE PVC PS 1 a* a a -2.69 ± 0.28 a a -2.67 ± 0.26 a b -3.09 ± 0.31 a a -2.75 ± 0.18 a a -2.82 ± 0.17 a a -2.37 ± 0.21 a a -2.53 ± 0.84 b a -2.37 ± 0.28 a a -2.29 ± 0.51 b a -3.14 ± 0.83 b a -3.28 ± 0.87 c E * b* a -3.59 ± 0.43 a -6.06 ± 0.22 a -5.91 ± 0.24 b,c -5.68 ± 0.18 b,d -6.53 ± 0.20 a 0.91 ± 0.85 a 0.88 ± 0.80 a 1.20 ± 0.69 a -2.10 ± 0.19 a,c -2.48 ± 0.51 a,d -2.03 ± 0.35 b -3.79 ± 0.30 a 1.17 ± 1.01 b 0.83 ± 0.81 c 4.92 ± 0.53 a -1.90 ± 0.29 b -1.06 ± 0.63 b -1.01 ± 0.19 a -2.16 ± 0.24 a 8.67 ± 1.04 b 9.87 ± 1.40 c 12.2 ± 1.64 Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 4). 141 Table 5-5. (cont’d) Time of Exposure (day) L* 0 34.91 ± 1.37 1 Sample 33.12 ± 1.52 PET a* a b a,b 7 33.71 ± 0.81 14 36.38 ±0.97 0 33.68 ± 0.79 a,c a b,c 1 38.01 ± 2.77 7 37.50 ± 0.77 14 37.42 ± 2.25 Nylon 1 b,c a,c E * b* a -3.13 ± 0.64 a -2.96 ± 0.61 a -2.96 ± 0.54 a -2.59 ± 0.49 a -2.82 ± 0.59 b -2.67 ± 0.74 b -2.72 ± 0.62 b -2.61 ± 0.39 a -3.27 ± 0.16 a,b -2.95 ± 0.31 a,b -2.93 ± 0.22 a,c -3.98 ± 0.28 a 1.83 ± 1.42 a 1.26 ± 1.08 a 1.72 ± 1.03 a -6.06 ± 0.25 b -4.97 ± 0.87 b -4.91 ± 0.32 a -5.75 ± 0.62 a 4.47 ± 1.98 a 3.99 ± 0.75 a 3.76 ± 1.68 Within columns, means (± S.D.) sharing the same superscript letter are not significantly different (p > 0.05; n = 4). 142 5.4. Conclusion The effects of ClO2 gas on properties and performance of 10 types of polymeric packaging materials, including HDPE, LDPE, LLDPE, BOPP, PS, PVC, PET, PLA, nylon, and a multilayer structure of EVA/EVOH/EVA were studied by assessing their IR spectrum, physical, mechanical, barrier, and color properties. The samples were exposed o to 3,600 ppmV ClO2 gas, at 23 C for 1, 7, and 14 days. The IR spectra of the ClO2 treated samples indicated many possible changes in their chemical characteristics, such as formation of polar groups in polyolefins, changes in functional groups, and main chain scission degradation, as well as, possible chlorination of several sample types. The ClO2 treated PE samples showed a decrease in tensile properties as compared to the untreated control films. Decreases in barrier to moisture, oxygen, and/or carbon dioxide were found in treated HDPE, PVC, PET, PLA, and multilayer EVA/EVOH/EVA. A significant increase in barrier to O2 was observed in ClO2 treated nylon film, which could be the result of molecular re-ordering, as implied through the increase in the crystallinity of the material. After 14 days of exposure, the permselectivity of several polymer films was altered, up to 46.8%. 143 BIBLIOGRAPHY 144 ASTM (2002). ASTM D882-02 Standard Test Method for Tensile Properties of Thin Plastic Sheeting. West Conshohocken, PA, ASTM International. ASTM (2003a). ASTM D1434-82(2003) Standard Test Method for Determining Gas Permeability Characteristics of Plastic Film and Sheeting. West Conshohocken, PA, ASTM International. ASTM (2003b). ASTM D3418-03 Standard Test Method for Transition Temperatures and Enthalpies of Fusion and Crystallization of Polymers by Differential Scanning Calorimetry. West Conshohocken, PA, ASTM International. Buchalla, R., Schuttler, C. and Bogl, K. W. (1993a). "Effects of Ionizing Radiation on Plastic Food Packaging Materials: A Review Part 1. Chemical and Physical Changes." Journal of Food Protection 56(11): 991 - 997. Buchalla, R., Schuttler, C. and Bogl, K. W. (1993b). "Effects of Ionizing Radiation on Plastic Food Packaging Materials: A Review Part 2. Global Migration, Sensory Changes and the Fate of Additives." Journal of Food Protection 56(11): 998 1005. Exama, A., Arul, J., Lencki, R. W., Lee, L. Z. and Toupin, C. (1993). "Suitability of Plastic Films for Modified Atmosphere Packaging of Fruits and Vegetables." Journal of Food Science 58(6): 1365-1370. Kaczur, J. J. and Cawlfield, D. W. (1992). Chlorine Oxygen Acids and Salts, Chlorous Acid, Chlorites, and Chlorine Dioxide. Kirk-Othmer Encyclopedia of Chemical Technology. Kroschwitz, J. I. New York, John Wiley & Sons Inc. 5. Kharitonov, A. P. (2000). "Practical applications of the direct fluorination of polymers." Journal of Fluorine Chemistry 103(2): 123-127. Kim, J. M., Huang, T.-S., Marshall, M. R. and Wei, C.-I. (1999). "Chlorine dioxide treatment of seafoods to reduce bacterial loads." Journal of Food Science 64(6): 1089-1093. Kulshreshtha, A. K. (1992). 3: Chemical Degradation. New York, Marcel Dekker, Inc. Martinez-Romero, D., Guillen, F., Castillo, S., Valero, D. and Serrano, M. (2003). "Modified armosphere packaging maintains quality of table grapes." Journal of Food Science 68(5): 1838-1843. Matteucci, S., Yampolskii, Y., Freeman, B. D. and Pinnau, I. (2006). Transport of Gases and Vapors in Glassy and Rubbery Polymers. Materials Science of Membranes for Gas and Vapor Separation. Yuri Yampolskii, I. P. B. F.: 1-47. 145 Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2009). "Mass Transfer Study of Chlorine Dioxide Gas through Polymeric Packaging Materials." Journal of Applied Polymer Science 114(5): 2929-2936. Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2010). "Effect of Chlorine Dioxide Gas on Physical, Thermal, Mechanical, and Barrier Properties of Polymeric Packaging Materials." 115(3): 1742-1750. Ozen, B. F. (2000). Effect of Ozone and Chlorine Dioxide Treatments Used in Perishable Food Applications on Polymeric Materials: Changes in Mechanical, Thermal and Mass Transfer Properties. Food Science Department. West Lafayette, IN, Purdue University. Ph.D. Ozen, B. F., Mauer, L. J. and Floros, J. D. (2002). "Effects of ozone exposure on the structural, mechanical and barrier properties of select plastic packaging films." Packaging Technology and Science 15(6): 301 - 311. Rivaton, A. and Gardette, J.-L. (1998). "Photo-Oxidation of Aromatic Polymers." Die Angewandte Makromolekulare Chemie 261/262: 173–188. Robinson, J. W., Frame, E. M. S. and Frame II, G. M. (2005). Undergraduate Instrumental Analysis. New York, Marcel Dekker. Rodriguez, F., Cohen, C., Ober, C. K. and Archer, L. A. (2003). Principles of Polymer Systems. New York, Taylor & Francis. Sapers, G. M., Walker, P. N., Sites, J. E., Annous, B. A. and Eblen, D. R. (2003). "Vapor-phase decontamination of apples innoculated with Escherichia coli." Journal of Food Science 68(3): 1003-1007. Schnabel, W. (1992). Polymer Degradation: Principles and Practical Applications. New York, Oxford University Press. Selke, S. E. M., Culter, J. D. and Hernandez, R. J. (2004). Plastics Packaging: Properties, Processing, Applications, and Regulations. Cincinnati, OH, Hanser Gardner Publications, Inc. Shin, J., Ryser, E., Selke, S. and Harte, B. (2006). Evaluation of the Microbial Inhibition Performance of Gas Type Antimicrobials (Chlorine Dioxide and AllylIsothiocyanate) with Modified Atmosphere Packaging (MAP), and the Effct of Chlorine Dioxide Exposure on the Physical Properties of Plastic Films. The 15th IAPRI World Conference on Packaging. Tokyo, Japan. Sy, K. V., McWatters, K. H. and Beuchat, L. R. (2005). "Efficacy of Gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Yeasts, and Molds on Blueberries, Strawberries, and Raspberries." Journal of Food Protection 68(6): 1165 - 1175. 146 Walzak, M. J., Flynn, S., Foerch, R., Hill, J. M., Karbashewski, E., Lin, A. and Strobel, M. (1995). "UV and Ozone Treatment of Polypropylene and Poly(ethylene terephthalate)." Journal of Adhesion Science and Technology 9(9): 1229-1248. 147 CHAPTER 6: ABSORPTION OF ClO2 GAS BY LETTUCE LEAF 6.1. Introduction The absorption behavior of gaseous ClO2 by shredded lettuce was investigated in - this chapter. The study assessed 1) residual ClO2; and 2) chlorite (ClO2 ) in ClO2exposed lettuce pieces. The amount of ClO2 absorbed by the lettuce, as well as, identifying the factors affecting this behavior, will help determine the optimal dose of ClO2 gas to be included in specific product/package system. When absorbed into produce, ClO2 can react with organic matter, such as plant cells and pigments (Richardson et al., 1998; USEPA, 1999), and also with normal microflora residing on the plant’s surface. These reactions typically generate chlorite - (ClO2 ) as a major by-product, along with numerous other compounds produced in trace amounts (Han et al., 2004). Han et al. (2004) reported measurable levels of residual ClO2 - and ClO2 on strawberries after ClO2-treatment (3.0 mg/L for 10 min), with small o amounts of ClO2- present even after one week of storage at 4 C (Han et al., 2004). The later form of ClO2 absorbed by produce does not possess antimicrobial activity, and, if 148 determined to be significantly high; should be taken into consideration when determining the appropriate ClO2 dose for the target food product. Other factors that must be considered when studying ClO2-absorption behavior are those associated with processing steps in the food production chain, i.e. the presence of cuts and/or bruises, and excess water on the produce surface. Fresh-cut and ready-toeat products are typically subjected to a wide range of food processing steps, however processing of these products is typically regarded as ‘less severe’ than other preparation and preservation techniques used for other product types. After sorting, trimming, cutting, washing and drying, the plant cells become damaged which, in turn, accelerates physiological deterioration, biochemical changes, and microbial degradation (Allende et al., 2006; Rico et al., 2007). In addition, the intracellular plant compounds released during processing will react with ClO2. Chlorine dioxide which has high solubility (Ishi, 1958; USEPA, 1999) could also attracted to intracellular liquid in the damaged cells, and/or the presence of water on the target surfaces, resulting in more ClO2 being absorbed. 6.2. Materials and Methods 6.2.1. Development of experimental setup for absorption study To be able to effectively study ClO2-absorption in lettuce, the experimental set up was designed to provide stable testing environment with constant concentration of ClO2 3 gas in the head space. The experimental setup used (Figure 6-1) consisted of 1) a 740 in 149 (approximately 12 L) glass chamber (custom made at Glassblowing Facility, Department of Chemistry, MSU, East Lansing, MI) with a removable glass lid providing access to lettuce samples; 2) low-density poly(ethylene) (LDPE) screen for supporting the lettuce ® sample; (3) Viton O-ring (Anchor Rubber Products, 2007); and (4) metal clamp ring (together with the O-ring) for hermetic sealing of the glass chamber. A chlorine dioxide sampling port was installed at the bottom of the glass chamber. Chlorine dioxide solution with known concentration was placed in the bottom of the chamber to provide 3.0 and 6.0 mg/L of ClO2 in the gas phase (details on selected ClO2 level in section 6.2.4). Even though the glass chamber, O-ring, and metal clamp ring provided air-tight conditions and minimize the loss of ClO2 gas to the environment, small amount of ClO2 gas was lost through reactions with the system, photo-degradation (Vaida and Simon, 1995) and sorption into the LDPE screen (Netramai et al., 2009). The chlorine dioxide degradation profiles for particular system were determined by monitoring amount of ClO2 in solution (Appendix 2), without lettuce sample as described in Appendix 5. 150 Figure 6-1. Experimental setup for absorption study 151 6.2.2. Preparation of lettuce samples Romaine lettuce (Lactuca sativa L. var. longifolia) was purchased from a local supermarket in East Lansing, MI, during August – November, 2010, and stored at 4oC, until 1 h before use. To avoid variations of the samples’ textures, each batch of lettuce was purchased and used within 5 days, and then discarded if they were not used in the experiments. For each experiment, 3 lettuce leaves with no visible or minor bruises, were selected from the inner, middle, and outer layers of the head, and cut to measure 1.25 × 2 1.5 in . Cut pieces included the midrib and leaf areas, but excluded the midrib connected to the stem. Twenty-five pieces of shredded lettuce weighting between 40-50 g were then used in each experiment. 6.2.2.1. Preparation of whole leaf and washed shredded lettuce samples To study the factors that could affect ClO2 absorption, i.e. cuts and moisture on the leaf surface, the following samples were prepared; (1) Three whole lettuce leaves, from the inner, middle, and outer layers of the lettuce head weighting between 40-50 g; and (2) 25 pieces of shredded lettuce (with initial weight comparable to the shredded sample prepared in section 6.2.2) that were subjected to a 5 min submersion in water and spin-dried using a kitchen type salad spinner (5 pumps). 152 6.2.3. Preparation of ClO2 solution The same methodology as described in chapter 3 was used to prepare the final ClO2 solution from a stock solution containing 1,000 mgClO2/L. 6.2.4. Concentration range to be used in absorption study Preliminary experiments were conducted to determine the concentration range of ClO2 gas to be used in absorption study. Twenty-five lettuce leaf pieces were placed upright in a glass container, along with ClO2 gas generated from ClO2 solution (1 L) as shown in Figure 6-1. The concentrations of gaseous ClO2 used were between 2.0 – 10.0 mg/L (with 2.0 mg/L increment) with exposure times of 15, 30, 60, 45, 90, and 120 min. Visual inspection of the exposed lettuce pieces showed that exposure to greater than 6.0 mgClO2/L, for less than 15 min, discolored the lettuce with the leaves changing from green to brownish white. Exposing shredded lettuce more than 60 min, even at the lowest ClO2 level, i.e. 2.0 mg/L also bleached the samples. Furthermore, the degree and area of discoloration increased as the concentration and/or time increased. Based on preliminary tests, selected ClO2 levels to be used in absorption work were 3.0 and 6.0 mg/L with predetermined exposure times of 7.5, 15, 30, 45, 60, and 90 min. A treatment time of 90 min was selected to observe the absorption behavior of the bleached lettuce pieces. 153 - 6.2.5. Quantification of residual ClO2 and ClO2 on shredded lettuce - Residual ClO2 and ClO2 on the lettuce leaf were quantified using a modification of the standard titration procedure for the examination of water and wastewater (Greenberg et al., 1992). Briefly, 25 pieces of shredded lettuce were exposed to ClO2 gas released from solution in the closed glass chamber (Figure 6-1) for a predetermined time period. - To measure residual ClO2 and ClO2 , the ClO2-exposed lettuce sample was washed with 300 ml of distilled and deionized water for 15 min (Han et al., 2004), after - which the washing solution was titrated for residual ClO2 and ClO2 , using an amperometric titration method (Greenberg et al., 1992). This specific amperometric titration (4500-ClO2 C) involves four titration steps with phenylarsine oxide (C6H5AsO), - in which free chlorine (Cl2), chloramines, ClO2 , and ClO2 can be determined separately. The titration procedure was described, in detail, in Appendix 4. For each experiment, the amount of ClO2 in solution was determined at the beginning and end of the treatment, using the thiosulfate titration method outlined in Appendix 2, to determine the recovery rate of the titration procedure for residual ClO2 - and ClO2 (see Appendix 5). 154 6.2.5.1. Evaluation of factors affecting chlorine dioxide absorption To determine the impact of cuts and excess water on the leaf, whole leaves and washed shredded lettuce samples prepared as described in section 6.2.2.1 was subjected to the same ClO2-treatments as described earlier, after which residual ClO2 and ClO2 - were determined. 6.2.6. Statistical analysis Each ClO2 treatment was repeated five times. The results obtained were statistically analyzed by analysis of variance (ANOVA) using the Statistical Analysis System (SAS) software (SAS Institute Inc., Cary, NC) at a confidence level of 95% ( = 0.05) with Tukey’s adjustment for comparison of means. 6.3. Results and discussion An experimental system that provided a constant ClO2 level of 3.0 and 6.0 mg/L in the gas phase over a certain period of time (≤ 90 min) was developed and used to study ClO2-absorption behavior by lettuce leaves. The absorption study was carried out by exposing different lettuce samples, i.e. shredded, whole, and washed shredded lettuces to ClO2 gas for different exposure times, after which the levels of residual ClO2 and ClO2 were quantified in the treated samples. 155 - 6.3.1. Absorption of ClO2 into shredded lettuce - The residual ClO2 and ClO2 values are showed in Figure 6-2. The levels of - ClO2 recovered indicate the amount of absorbed ClO2 gas that reacted with organic matter and/or normal flora in lettuce leaf (Han et al., 2004). At any given treatment time, - increasing the ClO2 level increased the amount of ClO2 recovered; and at any given ClO2 level, longer treatment times gave the same effect. Based on visual observation, bleached area was noticeable initially at treatment of 3.0 mg/L for 45 min and 6.0 mg/L for 30 min; and over half of lettuce samples were bleached after treatment of 3.0 mg/L for 90 min or 6.0 mg/L for 45 min. At equivalent treatments, such as 3.0 mg/L for 30 min or 6.0 mg/L for 15 min, the amounts of ClO2- recovered were comparable. - Differences in the amount of ClO2 recovered at different treatment conditions (Table 6-1) reflected different activity levels between ClO2 gas and organic matter and/or normal flora of the lettuce leaf. Microbial enumerations were not performed on any of these samples (samples were washed and discarded after recovering the residuals). However, in separate experiments using continuous exposure to ClO2 gas (details in Appendix 1, specifically in Table A-1) indicated that increasing the ClO2 levels and/or time of exposure, led to greater reductions of microbial background flora. Since the primary end-product from disinfecting mechanism of ClO2 is chlorite (USEPA, 1999), the differences in residual 156 - ClO2 recovered depended, at least in part, on the degree of disinfection occurring at each treatment condition. In regards to residual ClO2 recovered from lettuce, since only small amounts of ClO2 were absorbed without reacting (i.e. < 0.73 mgClO2/kg lettuce for every testing condition that did not cause noticeable discoloration), this portion of ClO2 could be reasonably neglected when calculating a gas treatment for lettuce, within the typical range used in food applications. Since the levels of residual ClO2 recovered from all lettuce samples were close to the detection limit of this particular titration procedure, which was calculated to be ˂ 0.32 mgClO2/kg lettuce, the variations of residual ClO2 determined were quite high, as can be observed from error bars in Figure 6-2. 157 2.5e+1 Recovered ClO2 and ClO2- (mg/kg lettuce) 2.0e+1 1.5e+1 ClO2 (mg/kg) at 3.0 mg/L ClO2 (mg/kg) at 6.0 mg/L ClO2- (mg/kg) at 3.0 mg/L 1.0e+1 ClO2- (mg/kg) at 6.0 mg/L 2.0e+0 1.0e+0 0.0 0 15 30 45 60 75 Treatment time (min) Figure 6-2. Absorption of ClO2 gas on shredded lettuce 158 90 105 - Table 6-1. ClO2 recovered from shredded lettuce after treatment with 3.0 and 6.0 mg/L ClO2 gas - ClO2 mgClO2-/kg lettuce 1 Treatment 3.0 mgClO2/L 7.5 min nd 15 min ≤ 7.17 ± 0.47 30 min ≤ 9.37 ± 1.71 45 min ≤ 10.82 ± 1.03 60 min 1 16.94 ± 1.19 90 min 2 6.0 mgClO2/L a,3 ≤ 6.62 ± 0.39 a b b b ≤ 9.40 ± 0.48 c 15.23 ± 1.06 d,4 ≤ 21.25 ± 1.12 c nd d,4 nd 20.36 ± 0.97 1 o For the solutions to release 3.0 and 6.0 mg/L of ClO2 gas into the chamber headspace, at 23 C, the actual amounts of ClO2 in 1 L of solution are approximately 87 and 175 mg. 2 Not determined. 3 4 Within columns, means sharing the same superscript letter are not significantly different (p > 0.05; n = 5). ≥ 50% of sample showed discoloration. 159 6.3.2. Factors affecting absorption behavior of ClO2 into lettuce Whole leaf and washed shredded lettuce samples were used to investigate if the presence of cuts or excess water could affect ClO2 absorption. The determined ClO2 - values for both sample types are listed in Table 6-2 and Table 6-3 and graphically shown, in comparison to those for shredded lettuce (original data reported in Table 6-1), in Figure 6-3. The levels of residual ClO2 recovered from whole leaf samples were comparable to those found in shredded samples i.e. < 0.51 mgClO2/kg lettuce, regardless of the treatment condition. - However, ClO2 levels were vastly different for whole leaf samples. The ClO2levels were only comparable to those found in shredded lettuce at 3.0 mg/L for 60 min and 6.0 mg/L for 30 min, where the leaf samples bleached severely (Figure 6-4). For ClO2-treatments that did not cause noticeable bleaching (i.e. 3.0 mg/L for ˂ 45 min and - 6.0 mg/L for 15 min), the ClO2 levels found were approximately 10 times lower than those for shredded lettuces. Increase in ClO2 uptake in shredded lettuce is likely due to the presence of cuts and/or bruises. Cutting damages cells and releases intracellular compounds which accelerate chemical and microbial degradations of lettuce by giving potential reactants and microorganism access to the plant tissue (Allende et al., 2006; Rico et al., 2007). Chlorine dioxide can also penetrate cuts or bruises and react more readily with organic matter in plant exudate. 160 Washing increased the initial weight of the shredded lettuce by 6.85 ± 0.71 %. Based on Figure 6-3, chlorite levels followed the same trend, and were comparable in the amount recovered. Levels of residual ClO2 recovered from washed shredded lettuce were also < 0.74 mgClO2/kg lettuce for all treatment that did not lead to discoloration. Thus, water uptake (at the level mentioned earlier) during washing did not significantly affect ClO2 absorption. However, visual observations (Figure 6-4) showed that discoloration first appeared on washed samples faster than those occurred on shredded samples. The increased moisture on the surface may also cause ClO2 gas to dissolve more easily (Ishi, 1958), with this reaction occurring more readily on specific areas where water droplet are located. Apart from excess moisture, washing and spin-drying introduced additional damage to shredded lettuce, which accelerated the biochemical reactions even further. - Due to concerns surrounding residual ClO2 and ClO2 on ClO2-treated food products (Couri et al., 1982; Qingdong et al., 2006), treated shredded lettuce (treated with 6.0 mgClO2, for 15 min) was placed on a clean surface at room temperature for 15 and - 60 min; and after which the amount of residual ClO2 and ClO2 were determined. These values were reported below (Table 6-4). 161 - Table 6-2. ClO2 recovered from whole leaf lettuce after treatment with 3.0 and 6.0 mg/L ClO2 gas - - ClO2 mgClO2 /kg lettuce Treatment 15 min 1 3.0 mgClO2/L a,2 ≤ 0.56 ± 0.31 a 30 min ≤ 0.92 ± 0.40 45 min a ≤ 1.25 ± 0.21 b,3 ≤ 1.80 ± 0.36 60 min 1 6.0 mgClO2/L ≤ 15.71 ± 0.86 90 min ≤ 15.25 ± 0.88 b 4 nd c,3 nd nd nd 1 o For the solutions to release 3.0 and 6.0 mg/L of ClO2 gas into the chamber headspace, at 23 C, the actual amounts of ClO2 in 1 L of solution are approximately 87 and 175 mg. 2 Within columns, means sharing the same superscript letter are not significantly different (p > 0.05; n = 5). 3 4 ≥ 50% of sample showed discoloration Not determined 162 - Table 6-3. ClO2 recovered from washed shredded lettuce after treatment with 3.0 and 6.0 mg/L ClO2 gas - - ClO2 mgClO2 /kg lettuce 1 Treatment 3.0 mgClO2/L 7.5 min nd 15 min ≤ 6.96 ± 0.27 30 min ≤ 9.36 ± 1.07 45 min ≤ 11.61 ± 0.92 60 min 1 ≤ 15.48 ± 1.60 90 min 2 6.0 mgClO2/L a,3 ≤ 7.64 ± 1.40 a b c a ≤ 9.75 ± 1.35 b ≤ 14.87 ± 1.24 c,4 ≤ 21.98 ± 1.17 d nd e,4 nd ≤ 21.62 ± 1.63 1 o For the solutions to release 3.0 and 6.0 mg/L of ClO2 gas into the chamber headspace, at 23 C, the actual amounts of ClO2 in 1 L of solution are approximately 87 and 175 mg. 2 Not determined 3 4 Within columns, means sharing the same superscript letter are not significantly different (p > 0.05; n = 5). ≥ 50% of sample showed discoloration 163 2.5e+1 Recovered ClO2 and ClO2- (mg/kg lettuce) 2.0e+1 1.5e+1 1.0e+1 Shredded lettuce at 3.0 mg/L Shredded lettuce at 6.0 mg/L Whole leaf at 3.0 mg/L Whole leaf at 6.0 mg/L Washed shredded lettuce at 3.0 mg/L Washed shredded lettuce at 6.0 mg/L 2.0e+0 1.0e+0 0.0 0 15 30 45 60 75 90 105 Treatment time (min) - Figure 6-3. Residual ClO2 recovered from whole leaf and washed shredded lettuce samples as compared to those of unwashed shredded lettuce 164 These findings indicate that; given enough time after ClO2 exposure, residual - ClO2 and ClO2 will decrease to undetectable levels. Thus, if packaging systems were designed to deliver ClO2 gas in a manner that releases the gas for a predetermined time and allows sufficient additional storage time (with the absence of ClO2 gas) for the residuals to degrade, any concerns regarding the presence of unwanted residuals could be - reasonably solved. However, degradation patterns for residual ClO2 and ClO2 are also - affected by commodity type. For example, about 6% of ClO2 still remained in o strawberry samples after 1 week of storage at 4 C (Han et al., 2004). - Table 6-4. Residual ClO2 and ClO2 in shredded lettuce after treatment with 6.0 mg/L ClO2 gas, for 15 min, and left at room temperature for 15 and 60 min Time left at room temperature (min) 15 2 ˂ 0.32 60 1 ClO2 mgClO2/kg lettuce ˂ 0.32 1 1 - - ClO2 mgClO2 /kg lettuce a,2 ≤ 4.90 ± 0.46 b ≤ 0.90 ± 0.05 Reached detection limit of the equipment. Within columns, means sharing the same superscript letter are not significantly different (p > 0.05; n = 5). 165 Whole lettuce leaf: 0 min Shredded lettuce: 15 min 30 min Bleached samples at 30 min Figure 6-4. Color profiles; Noted extensive discolorations on whole lettuce leaf and washed lettuce at 6.0 mgClO2/L treatment for 30 min 0 min Washed shredded lettuce: 15 min 30 min 0 min 15 min 30 min 166 Bleached samples at 30 min 6.4. Conclusion - Residual levels of ClO2 and ClO2 were successfully quantified on ClO2-treated lettuce leaves. The results showed that Romaine lettuce absorbed more ClO2 as the treatment time lengthened and/or the ClO2 level increased. Amount of ClO2 recovered from the lettuce were < 0.73 mgClO2/kg lettuce for every applicable treatment and could be dismissed when determining a ClO2 treatment for shredded lettuce. After treatment, amounts of both compounds decreased to undetectable levels if left to expose to air for sufficient time. The presence of cuts/bruises significantly increased ClO2 absorption in lettuce, while excess moisture did not significantly affect absorption. Excess moisture and more severe treatments increased changes in appearance, i.e. bleaching; of lettuce. 167 BIBLIOGRAPHY 168 Allende, A., Tomás-Barberán, F. A. and Gil, M. I. (2006). "Minimal Processing for Healthy Traditional Foods." Trends in Food Science & Technology 17(9): 513519. Anchor Rubber Products (2007) "Material Properties: Viton® O-ring." Retrieved August 27th, 2010, from http://www.anchorrubber.com/pdf/fda_viton.pdf. Couri, D., Abdel-Rahman, M. S. and Bull, R. J. (1982). "Toxicological effects of chlorine dioxide, chlorite and chlorate." Environmental Health Perspectives 46: 13-17. Greenberg, A. E., Clesceri, L. S. and Eaton, A. D. (1992). Standard methods for the examination of water and wastewater: 4500-ClO2 C. Amperometric method I. Standard methods for the examination of water and wastewater. Washington, D.C., American Public Health Association: 4-55 - 4-56. Han, Y., Selby, T. L., Schultze, K. K., Nelson, P. E. and Linton, R. H. (2004). "Decontamination of Strawberries Using Batch and Continuous Chlorine Dioxide Gas Treatments." Journal of Food Protection 67(11): 2450-2455. Ishi, G. (1958). "Solubility of chlorine dioxide." Chemical engineering (Japan) 22: 153 154. Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2009). "Mass Transfer Study of Chlorine Dioxide Gas through Polymeric Packaging Materials." Journal of Applied Polymer Science 114(5): 2929-2936. Qingdong, X., Guangming, Z. and Li, W. (2006). "Study on subchronic toxicity of chlorine dioxide and by-products in water." Journal of environmental science and health. Part A, Toxic/hazardous substances & environmental engineering 41(7): 1347-1353. Richardson, S. D., Thruston Jr., A. D., Caughran, T. V., Collette, T. W., Patterson, K. S. and Lykins Jr., B. W. (1998). "Chemical by-products of chlorine and alternative disinfectants." Food Technology 52(4): 58-61. Rico, D., Martín-Diana, A. B., Barat, J. M. and Barry-Ryana, C. (2007). "Extending and measuring the quality of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology 18(7): 373-386 USEPA (1999) "4. Chlorine Dioxide." Retrieved September 11th, 2007, from http://www.epa.gov/ogwdw/mdbp/pdf/alter/chapt_4.pdf Vaida, V. and Simon, J. D. (1995). "The Photoreactivity of Chlorine Dioxide." Science 268(5216): 1443-1448. 169 CHAPTER 7: EFFECTS OF PACKAGING DESIGN ON GAS DISTRIBUTION WITHIN THE PACKAGE 7.1. Introduction Given a proper packaging design, maximum surface exposure of food products to antimicrobial gases can be achieved. Several studies indicated that direct-contact between antimicrobial gas and the product surface is crucial for effective inactivation of microorganisms (Du et al., 2002; 2003; Ellis et al., 2006; Han et al., 2000; Lee et al., 2006; Lindsay et al., 2002; Shin, 2007). Packages contain many hard to reach areas where headspace gases have limited or no access to, including the center and the bottom of the package (Ellis et al., 2006; Gómez-López et al., 2009). In an earlier work by Shin (2007), ClO2 gas was used to sanitize chicken breasts packaged in sealed trays. When chlorine dioxide sachet was placed at one end of the tray and the meat at the other end, the meat’s surface near the sachet turned brownish black (Shin, 2007). In terms of antimicrobial efficacy of ClO2 gas, the results indicated that uniform gas distribution is required in order to sufficiently reduce the number of microorganisms, as microbial growth of the pathogen on the bottom of the meat was significantly higher than on the top (Shin, 2007). Ellis et al. (2006) investigated antimicrobial effect of ClO2 gas, generated from sachets containing ClO2 precursors, on packed chicken breasts. Higher microbial loads were found at the bottom of the meat where the area was less accessible to ClO2 gas generated by the sachet, as 170 compared to the top surface. Significant discoloration of the meat surface caused by the gas was observed at the area closer to ClO2 sachet, indicating that ClO2 was distributed unevenly within the package and react mainly with the surface adjacent to the sachet (Ellis et al., 2006). Furthermore, Du et al. (2002 and 2003) reported differences log10 CFU/site reduction for E. coli O157:H7 and Listeria monocytogenes, from different areas of apples, after ClO2 treatment (Du et al., 2002; 2003). This study explored the effect of packaging design on antimicrobial ability of ClO2 gas which should be considered when a packaging system is to be used as an additional microbial reduction step. This research was conducted in preparation for a collaborative project between Eastern Regional Research Center (ERRC), USDA, Rutgers University, and the School of Packaging at Michigan State University (MSU). The overall objective of the project was to evaluate the integration of 1) post-harvest strategies of produce sanitizing and 2) exposure to an antimicrobial through a packaging system; to improve microbial safety of fresh produce, by using packaging design to improve gas distribution in flexible and rigid container. The aim of this hurdle strategy is to achieve a microbial reduction of ≥ 5 log10 CFU/g for 2 target pathogenic microorganisms, i.e. Escherichia coli O157:H7 and Salmonella spp., inoculated on shredded Romaine lettuce and cherry tomatoes, respectively. The food product selected for this study portion was ready-to-eat (RTE) shredded lettuce which has been responsible for several foodborne outbreaks, in the past few years. RTE vegetables product is a minimally processed fresh produce, which is gaining popularity as it reduce the preparation time and waste generated (Zhou et al., 2004). 171 Shredded lettuce is one of the most important RTE products. As any other RTE leafy green product, the shelf life of shredded lettuce is limited due to discoloration (browning) and microbiological deterioration. Such processes are accelerated as the shredding process destroys the protective epidermal layer of the leaf and ruptures the cells, resulting in biochemical reactions. The cut surfaces also allow microorganisms the access to nutrients (Diaz and Hotchkiss, 1996; Zhou et al., 2004). Normally, minimally processed products are consumed without any further cooking, making contamination by pathogenic microorganism a very important issue. In this study, the selected pathogenic microorganism was Escherichia coli O157:H7 which has been reported as main microorganism among others, responsible of foodborne disease cases (Department of Health and Human Services, 2008; FSIS, 2005; Nowak et al., 2006; Phillips, 1996; Sy et al., 2005). This pathogen has been identified in food carrier such as raw or minimally processed foods, such as fresh spinach, shredded lettuce, and salads (Department of Health and Human Services, 2008; Lynch et al., 2006). This project focuses on the use of ClO2 gas in decontaminating shredded lettuce, by using a developed packaging system as an additional/extended step of food production. Once optimized, the new packaging system could effectively minimize the recurring problem of foodborne outbreaks related to retail and food services. Specific objectives of this section are to: 1. Identify ClO2 gas treatment conditions that can inactivate human pathogens on shredded lettuce without causing quality defects. 2. Identify a specific package design that ensures an effective gases distribution inside the package even in hard to reach areas. 172 3. Determine the efficacy of the packaging system in inactivating foodborne pathogens and prolonging the shelf life of shredded lettuce. 4. Evaluate a pilot scale treatment and scale up to demonstrate technical and economical feasibility. Objective 2 involved modifying the interior of a currently used flexible package for shredded lettuce. The approach is favorable since neither the packaging line nor the basic package will be changed significantly. The design development will aim to facilitate and maximize distribution of the gas, even in hard to reach areas by building a gas reservoir within the package, ensuring that ClO2 is available for an extended period. The designs will be optimized for maximum exposure between ClO2 gas and the product, as the effectiveness of any antimicrobial gas treatment also depends on its surface contact with the target product (Du et al., 2002). As part of objective 2, this preliminary study was also conducted to gather the necessary information to develop parameters for future packaging development work which will be performed by the research team, in the later parts of collaborative research project. Information regarding: 1) mass transfer of ClO2 as reported in Chapter 4, specifically, the amount of ClO2 gas that permeated through and was absorbed by the packaging material, 2) the exposure time of the materials to ClO2 gas that will not cause any significant alteration in the films’ properties (Chapter 5), and 3) ClO2-absorption behavior of lettuce leaf, will also be included as the parameters in the same future work on packaging design. 173 7.2. Materials and methods Two consecutive packaging studies were conducted using 3 different designs of flexible bags; and plain flexible bags as controls. These bags were used as package for shredded lettuce that was inoculated with E. coli O157:H7 and stored in refrigerator at 4 o ± 0.5 C. Microbial enumeration and visual inspection were performed at specific time intervals. 7.2.1. Preparation of inoculums Escherichia coli O157:H7 SEA13B88 (human feces, apple cider-associated disease outbreak), maintained at -80°C in trypticase soy broth (TSB; Becton Dickinson, Sparks, MD) and 10% (v/v) glycerol, was grown for 18-24 hours in TSB at 35 ± 0.5°C, transferred to a trypticase soy agar (TSA; Becton Dickinson, Sparks, MD) slant, and this working stock culture was stored at 4°C for no more than 21 days. Inoculum was prepared by transferring a loopful (1 μl) of the working stock to 10 ml TSB, which was incubated in a shaking incubator for 6 - 8 hours at 35 ± 0.5°C. Following incubation, 2.4 ml of the culture was transferred to 24 L of TSB, and then incubated at 35 ± 0.5°C in a shaking incubator for 18-24 hours. The culture was then centrifuged (6740 × g) at 4°C for 15 min. After decanting the supernatant, the resulting pellet was resuspended in sterile deionized water and centrifuged (6740 × g) for 15 min at 4°C. The supernatant was decanted and the pellet was resuspended in 3 L of sterile deionized water. The concentration of the inoculum was determined by serially diluting the inoculum in 0.1% peptone water (PW; Becton Dickinson, Sparks, MD) and plating on TSA. The 174 o resuspended culture was then stored at 4 C as a culture concentrate, for overnight. The culture was diluted to 24 L before use. 7.2.2. Inoculation of shredded lettuce Commercially available Romaine lettuce (Lactuca sativa L. var. longifolia) was purchased at a local supermarket in Wyndmoor, PA, US, and stored at 4 ± 2°C for a maximum of 24 hours before use in experiments. Damaged outer leaves were removed 2 from each head of lettuce, the lettuce was cut into pieces approximately 4 - 6 cm and immediately submerged into the E. coli O157:H7 inoculum suspension for 5 min. Excess liquid culture on the lettuce was removed using a salad spinner (OXO Good Grips Salad Spinner, OXO International, Ltd., New York, NY;) for 1 min (Figure 7-1) and then the lettuce was placed into an open container and allowed to dry for 2 hours at 22 ± 2°C in a biosafety cabinet. The lettuce was then placed into plastic bags and stored for 18-24 hours at 4 ± 2°C to allow bacteria attachment. 175 Figure 7-1. Inoculation of shredded lettuce; shredded lettuce were submerged in E. coli O157:H7 inoculums for 5 min and then spin-dried 176 7.2.3. Packaging design 7.2.3.1. Study 1 For the first study, two different bag designs (Figure 7-2) were used. These bags were modified from commercially available bag for leafy green products, PD-961EZ (Cryovac Inc., Duncan, SC) (see Appendix 6 for bag characterization) by placing gas reservoirs (GRs) in different locations. A thermal impulse heat sealer (Vertrod Crop., San Rafael, CA) was used to create the GRs by sealing the already existing bag. The seal, with small openings throughout, divided the bag into a large chamber for lettuce sample; and lengthwise ClO2 GRs, where ClO2 precursor was placed and humidity-activated to released ClO2 gas. The openings allowed gas to enter the bag chamber. Bags with different numbers of GRs and GRs with different numbers of openings were created, i.e. Design 1 with one-GR and 6 openings; and Design 2 with two-GRs, along each side of the bag, and 3 openings on each GR. 7.2.3.2. Study 2 In the second study, the GR was a low-density poly(ethylene) (LDPE) tube, with -17 a permeability coefficient (P) of 66.0 ± 1.09 × 10 -2 -1 kgClO2mm s Pa -1 (Netramai et al., 2009), that included ClO2 precursors. The tube was packed together with shredded lettuce and placed in the middle of the bag as shown in Figure 7-2. 177 Design 1 – One-GR design Control Figure 7-2. Bag designs for study 1 and 2 178 Design 2 – Two-GR design Design 3 – Middle-GR design Figure 7-2. (cont’d) 179 7.2.4. Chlorine dioxide treatment For the first study, 8 g of ClO2 precursor (Special Mix with linear release, ICA TriNova LLC, Newnan, GA) was put in paper tubes which were inserted into the interior structure of flexible package. Precursor released two levels of ClO2; high level at 8 mgClO2/kg lettuce per day and low level at 4 mgClO2/kg lettuce per day (more details, see Appendix 7 and Appendix 8). The gas release profiles, determined using titration procedure outlined by ICA Trinova (ICA TriNova LLC, 2006), are shown in Figure 7-3. In the second study, ClO2, generated by precursor, was released by permeation through the LDPE reservoir (Appendix 8). The same two levels of ClO2 gas were also used in this experiment, i.e. 4 and 8 mgClO2/day (Figure 7-3). 7.2.5. Preparation of packaged shredded lettuce inoculated with E. coli O157:H7 Approximately 283 g of E. coli O157:H7-inoculated lettuce (the same weight as commercial products) were put in the bag along with paper tube that contained the mixed ClO2 precursors (Figure 7-4). The bags were sealed, photographed of Day 0, and stored, o standing upright, at 4 ± 0.5 C. Following 1, 4 and 7 days of storage, samples were collected and were photographed, visually inspected for changes in appearance and color, and evaluated for residual microbial populations. 180 60 Study 1 - Fast release Study 1 - Slow release Study 2 - Fast release Study 2 - Slow release Released ClO2 gas (mg) 50 40 30 20 10 0 0 1 2 3 4 5 6 7 8 Day Figure 7-3. Gas release profiles; 8 mgClO2/day type and 4 mgClO2/day type 181 Figure 7-4. Inoculated shredded lettuce in one-GR design and middle-GR design bags 182 7.2.6. Microbial enumeration Results from the preliminary experiments (Appendix 9) suggested that there was no significant difference in E. coli O157:H7 cell reductions on lettuce samples taken from the top and the bottom portions of both one-GR and two-GR bag types. Thus, the comparison only focused on samples taken from different locations that were parallel to the gas reservoirs. Twenty-five grams of shredded lettuce were taken from various points in each of the bag portion (Figure 7-5) (i.e. total of 75 g was taken from each bag in the first study and total of 50 g per bag in the second study) and homogenized with 75 mL Dey-Engley (DE) neutralizing buffer (Becton Dickinson, Sparks, MD) for 1 min, using stomacher blender. Undiluted DE buffer from the homogenate was serially diluted in sterile neutralizing buffer to desired dilutions, spread plated onto 2 Tryptic Soy Agar (TSA, o Difco, Becton Dickinson, Franklin Lakes, NJ) plates, incubated at 37 ± 0.5 C for 2 hours (for injured cells to recover), overlaid with Sorbitol MacConkey Agar (SMAC; Remel, Lenexa, KS) with cefixime, and potassium tellurite (CT; Invitrogen, Dynal AS, Oslo, o Norway), and then incubated at 37 ± 0.5 C for 22 hours and counted as E. coli O157:H7 (Keskinen et al., 2009). Due to certain restriction, all plates obtained from 1 day of o storage were incubated at 23 C for 72 hours. 183 Figure 7-5. Sample portions; For one-GR design, portion 1 = area next to the reservoir, portion 2 = middle area of the bag, and portion 3 = area most distant to the reservoir; For two-GR design, portion 1 and 3 = area next to the reservoirs and portion 2 = middle area of the bag; For middle-GR design, portion 1 = area next to the reservoirs and portion 2 = area further away from the reservoir from both sides of the bag 184 For control samples, inoculated shredded lettuce was packed in plain bags (details in section 7.2.5) without ClO2 gas. Log10 CFU/g sample of E. coli O157:H7 recovered from control samples represent populations of particular microorganism in lettuce with no ClO2 treatment. The values were reported in Table 7-2 and Table 7-3 and will be used later on for discussion purposes. 7.2.7. Statistical analysis The experimental design followed randomized complete block design (RCBD), as outlined in Table 7-1, in which the experiments were repeated twice (2 blocks). Each experiment involved three replicates. The results obtained were statistically analyzed by analysis of variance (ANOVA) using the Statistical Analysis System (SAS) software (SAS Institute Inc., Cary, NC) at the confidence level of 95% ( = 0.05) with Tukey’s adjustment for comparison of means. Table 7-1. Randomized complete block design information for study 1 Class Level Value Block 2 - Replicate 3 - Design 2 One-GR and Two-GR Concentration 2 4 and 8 mgClO2/kg lettuce per day Storage time 3 1, 4, and 7 days 185 7.3. Results and discussion To study how packaging design can affect efficacy of ClO2 gas, and lead to the optimization of a packaging system that, eventually, could improve safety and extend shelf-life of the products, two consecutive studies were conducted. Two levels of ClO2 gas were used in both experiments; 4.0 and 8.0 mgClO2/day. The high level of ClO2 gas used in the study was predetermined by exposing shredded Romaine lettuce to different levels of ClO2 gas in a closed chamber for 1 h. The level that was selected gave significant reduction in E. coli O157:H7 populations and did not cause a noticeable change in appearance (Appendix 7 and Appendix 8). The low level of 4 mgClO2/day was also selected to quantify the effect of lower ClO2 dose. 7.3.1. Study 1 The first experiment involved plain bags as control sample, and 2 types of modified flexible bags with different internal designs, as well as a plain bag as a control (Figure 7-2). Chlorine dioxide reservoir dimensions were determined by considering bag volume, ClO2 dose needed, ClO2 chemical characteristics, and lettuce weight per bag. The openings, placed on GR, were aimed to distribute equal amounts of ClO2 gas to the sample throughout the whole chamber. The total number of openings for both designs was the same, but the locations of the openings were varied. In Design 1 (one-GR), the openings were placed along one side of the bag; Design 2 (two-GR) had the same numbers of opening, but the openings were placed in different locations, as shown in Figure 7-2. As a result of the two designs, ClO2 gas entered the main chamber from 186 different directions, resulting in different gas distributions. The bags were packed with shredded lettuce inoculated with E. coli O157:H7, along with ClO2 as an antimicrobial o gas and stored at 4 C. After 1, 4, and 7 days of storage, the samples were subjected to microbiological analysis. Values of log10 CFU/g of E. coli O157:H7 for both packaging studies were reported in Table 7-2 and Table 7-3. The average log10 CFU/g reductions, shown in Figure 7-6, represent the differences of E. coli O157:H7 populations in lettuce packaged with ClO2 gas, for each bag design, at each ClO2 level; and without ClO2 gas. Due to variation of E. coli O157:H7 populations observed in control bag at different storage times, the discussions will focus on effects of different bag designs and different ClO2 levels (obtained from the same storage time). For one-GR design, increasing the amount of ClO2 gas generated per day (from 4 to 8 mgClO2/day) significantly increased the average log10 CFU/g reductions in the packaged sample after 1 and 4 days of storage, however, increasing the dose of ClO2 gas per day by 100% did not double the reduction in cell population of the pathogen. 187 1 Table 7-2. Log10 CFU/g of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in one-GR and two-GR design bags with 4 and 8 mgClO2/day Design / ClO2 level Storage time (day) E. coli O157:H7 (log10 CFU/g) Portion 1 Portion 2 Portion 3 a,2,3 1 Control 4 8.15 ± 0.08 - - One-GR / 4 mg Two-GR / 8 mg Two-GR / 4 mg 1 2 3 1 4 7 1 4 7 1 4 7 1 4 7 b - 8.48 ± 0.17 7.95 ± 0.07 7.94 ± 0.03 8.31 ± 0.04 8.01 ± 0.04 8.11 ± 0.04 8.28 ± 0.03 7.32 ± 0.06 7.32 ± 0.03 7.48 ± 0.05 7.41 ± 0.03 7.24 ± 0.02 7.51 ± 0.03 8.71 ± 0.07 7.62 ± 0.04 7.60 ± 0.05 7.95 ± 0.05 7.73 ± 0.04 7.82 ± 0.03 8.01 ± 0.04 7.55 ± 0.04 7.54 ± 0.04 7.74 ± 0.04 7.61 ± 0.03 7.54 ± 0.05 7.78 ± 0.02 b 7 One-GR / 8 mg Whole bag 6.99 ± 0.04 6.93 ± 0.06 7.25 ± 0.12 7.24 ± 0.07 7.27 ± 0.03 7.50 ± 0.10 7.30 ± 0.06 7.30 ± 0.08 7.46 ± 0.05 7.39 ± 0.03 7.31 ± 0.07 7.52 ± 0.05 7.92 ± 0.05 7.93 ± 0.08 8.30 ± 0.03 7.95 ± 0.03 8.08 ± 0.06 8.25 ± 0.02 8.04 ± 0.05 8.01 ± 0.04 8.28 ± 0.03 8.02 ± 0.04 8.06 ± 0.10 8.31 ± 0.05 GR = Gas reservoir Mean ± standard deviation Means sharing the same superscript letter are not significantly different (p > 0.05; block = 2, n = 3) 188 1 Table 7-3. Log10 CFU/g of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in middle-GR design bags with 4 and 8 mgClO2/day ClO2 level Storage time (day) E. coli O157:H7 (log10 CFU/g) Portion 1 Portion 2 a,2,3 1 0 mg 4 8.16 ± 0.04 - 8 mg 4 mg 1 2 3 b - 8.42 ± 0.05 7.48 ± 0.19 7.83 ± 0.01 8.36 ± 0.05 7.72 ± 0.02 7.81 ± 0.13 8.45 ± 0.10 8.70 ± 0.01 6.93 ± 0.01 7.19 ± 0.02 7.47 ± 0.11 7.63 ± 0.05 7.36 ± 0.08 7.76 ± 0.17 c 7 1 4 7 1 4 7 Whole bag 6.37 ± 0.06 6.55 ± 0.03 6.58 ± 0.22 7.54 ± 0.10 6.92 ± 0.07 7.07 ± 0.29 GR = Gas reservoir Mean ± standard deviation Means sharing the same superscript letter are not significantly different (p > 0.05; block = 2, n = 3) 189 Day 1 8 mg 0.53±0.03 4 mg a,a,1 Day 4 0.81±0.04 Day 7 0.76±0.05 a,b 0.59±0.02 a,b 4 mg 0.86±0.03 Day 7 0.97±0.04 b,b a,c (b) 0.60±0.03 Day 4 b,a 0.70±0.04 (a) Day 1 0.42±0.04 c,a 4 mg a,b 2 mg 0.55±0.02 0.87±0.05 b,c 0.93±0.02 (c) a,a 2 mg a,b b,b (d) Figure 7-6. Average log10 CFU/g reduction of E. coli O157:H7 of samples stored for 1, 4, and 7 days in one-GR design bag with 8 mgClO2/day (a), one-GR design bag with 4 mgClO2/day (b), two-GR design bag with 8 mgClO2/ day (c), and two-GR design bag with 4 mgClO2/day 1 First different script letters indicates statistically differences between means at  of 0.05 of different bag designs with the same storage time; Second different script letters indicates statistically differences between means at  of 0.05 of samples with different storage times within the same design 190 For bags with two-GR, increasing ClO2 gas from 4 to 8 mgClO2/day statistically increased reduction in cell population of E. coli O157:H7 only on the first day of storage, but, after day 4 and 7, the log10 CFU/g reductions of target microorganism were statistically the same, regardless of the level of ClO2 gas used. The locations of GRs (and the openings) affected antimicrobial efficacy of ClO2. For example, at 8 mgClO2/day, releasing antimicrobial gas from both sides of the bag (two-GR) reduced E. coli O157:H7 populations in shredded lettuce more than when the gas was released from only one side of the bag. This could be due to the shorter travel distance for ClO2 gas in two-GR bag, as compared to the one-GR bag that the gas has to travel to reach the target surface at the other end. The differences in log10 reductions from one-GR and two-GR bags were more pronounced at 4 mgClO2/day. From Figure 7-6, packing the samples in two-GR bags instead of one-GR bags, for 1 day, reduced E. coli O157:H7 by 0.07 log10 CFU/g and by 0.13 log10 CFU/g with 8 and 4 mgClO2/day, respectively. This finding suggests an interaction between bag design and ClO2 level, as supported by statistical outcomes in Appendix 11. Comparison between one-GR bag with 8 mgClO2/day (Figure 7-6(a)) and twoGR bag with 4 mgClO2/day (Figure 7-6(d)) showed that both had comparable reductions in cell population, after 1 and 4 days of storage, even though the amount of ClO2 gas in 191 the two-GR design was only half of what was released in the one-GR design (4 and 8 mgClO2/day, respectively). Furthermore, after 7 days of storage, the log10 CFU/g reduction of two-GR design bag with lower amount of ClO2 gas was higher than that of one-GR design with 8 mgClO2/day. Since the design had significant impact on improving antimicrobial efficacy of ClO2 gas, by reducing the traveling distance of the gas to its target surface, it should be possible to reduce the dose of ClO2, and still deliver the equivalent reduction in cell population of E. coli O157:H7. Utilization of lower ClO2 dose could also result in less appearance and texture changes (Ellis et al., 2006; Gómez-López et al., 2009). Apart from browning normally occurred at cut and bruised areas (Cantwell et al., 1998 ; Gómez-López et al., 2009; Lopez-Galvez et al., 1996; Peiser et al., 1998; Rico et al., 2007), some visual changes were also observed on the samples, i.e. softer, watery and/or bleached tissues, during exposure to ClO2 gas (Du et al., 2007; Han et al., 2000) (Figure 7-7 and Figure 7-8). These mostly occurred on lettuce pieces that were adjacent to the GRs. Samples from the bag with one-GR including the dose of 8 mgClO2/day showed more changes in appearance as compared to shredded lettuce in the bag with twoGR, but with 4 mgClO2/day dose. In day 4 and 7, the changes in appearance became more pronounced. 192 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was more visible Day 4: Browning occurred more extensive and darker in color, especially at the cut area of midribs Day 7: Browning found on most of the lettuce pieces; the texture became watery and soft in many pieces, especially the ones contained midrib parts Figure 7-7(a). E coli O157:H7 inoculated shredded lettuce packed in plain bags with 0.0 mgClO2/kg (Control for study 1), stored at o 4 C for 0, 1, 4, and 7 days 193 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was visible Day 4: Browning was darker in color, especially at the cut area of midribs; bleaching was noticeable on many lettuce pieces at the area next to GRs; some pieces also became watery and soft Day 7: Browning was more extensive; the texture became watery and soft in many pieces, especially the ones located next to GRs; more bleaching also observed o Figure 7-7(b). E coli O157:H7 inoculated shredded lettuce packed in one-GR bags with 8.0 mgClO2/kg, stored at 4 C for 0, 1, 4, and 7 days 194 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was visible on more lettuce pieces Day 4: Browning was visible; some shredded pieces at the area next to GRs became watery and soft Day 7: Browning was more extensive; the texture became watery and soft in many pieces, especially the ones located next to GRs; bleaching was visible in some lettuce pieces o Figure 7-7(c). E coli O157:H7 inoculated shredded lettuce packed in one-GR bags with 4.0 mgClO2/kg, stored at 4 C for 0, 1, 4, and 7 days 195 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was darker in color Day 4: Browning was visible; some shredded pieces at the area next to GRs became watery and soft Day 7: Browning was more extensive; the texture became watery and soft in many pieces, especially the ones located next to GRs; bleaching was visible in many lettuce pieces o Figure 7-7(d). E coli O157:H7 inoculated shredded lettuce packed in two-GR bags with 8.0 mgClO2/kg, stored at 4 C for 0, 1, 4, and 7 days 196 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was darker in color Day 4: Browning was more extensive and darker in color Day 7: Browning was more extensive; the texture became watery and soft in some pieces, especially the ones located next to GRs; bleaching was visible in some lettuce pieces o Figure 7-7(e). E coli O157:H7 inoculated shredded lettuce packed in one-GR bags with 4.0 mgClO2/kg, stored at 4 C for 0, 1, 4, and 7 days 197   Figure 7-8. Changes observed in samples packaged with ClO2 gas, at 8 mgClO2/day level; bleached surfaces and watery and softer tissues. 198 1.16 a,a,1 ±0.04 0.24 a,b ±0.04 0.20 a,b ±0.07 Day 4 1.48 a,a ±0.05 0.48 a,b ±0.08 Day 7 1.46 a,a ±0.12 0.41 a,b ±0.03 Day 1 8 mg 4 mg 0.92 b,a ±0.07 0.20 a,b ±0.02 0.14 a,b ±0.04 0.46 a,b ±0.02 1.14 b,a ±0.02 0.33 b,b ±0.06 0.30 b,b ±0.04 0.40 a,b ±0.04 1.22 b,a ±0.10 0.46 a,b ±0.02 0.43 a,b ±0.03 (a) (b) Figure 7-9. Log10 CFU/g reduction of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in one-GR design bag with 8 mgClO2/day (a), one-GR design bag with 4 mgClO2/day (b), two-GR design bag with 8 mgClO2/day (c), and two-GR design bag with 4 mgClO2/day 1 First different script letters indicates statistically differences between means at  of 0.05 of different sample bags with the same portion; Second different script letters indicates statistically differences between means at  of 0.05 of samples with different portions within the same bag 199 0.85 b,a ±0.05 0.11 b,b ±0.04 0.84 b,a ±0.05 Day 4 1.11 b,a ±0.08 0.39 b,b ±0.04 Day 7 1.25 b,a ±0.05 0.43 a,b ±0.03 Day 1 4 mg 4 mg 2 mg 0.76 c,a ±0.03 0.13 b,b ±0.03 0.68 c,a ±0.10 1.09 c,a ±0.03 1.10 b,a ±0.06 0.34 b,b ±0.09 1.17 d,a ±0.02 1.23 b,a ±0.05 1.19 b,a ±0.05 0.40 a,b ±0.05 1.21 b,a ±0.03 (c) (d) Figure 7-9. (cont’d) 200 2 mg The effects of proximity were further explored to evaluate how antimicrobial gas impacts the product throughout the bag, by considering log10 CFU/g reductions of E. coli O157:H7 in different areas of the bag, as shown in Figure 7-9. After day 1, for one-GR and two-GR designs, regardless of the level of ClO2 gas used, reductions in cell population of E. coli O157:H7 in the samples, adjacent to the GRs, were significantly higher than those samples further away from the reservoirs. In samples from one-GR bags, reductions in E. coli O157:H7 were statistically equal between samples taken from the middle area and the area most distant from the reservoir. Similar trends were observed in day 4 and 7 of storage. As expected, these results confirmed that ClO2 gas was most effective in the area close to the GRs, thus, proximity of antimicrobial gas releasing location to the target surface is crucial to maximize the antimicrobial effect (Ellis et al., 2006). The softer and watery shredded leaves could partially block the gas flow to other areas of the bag, and further reduced accessibility to the target surfaces, especially for those areas further away from the GRs, resulting in low kills in the area distant from the releasing locations (Du et al., 2002; Ellis et al., 2006). 7.3.2. Study 2 To minimize the distance between ClO2 gas releasing location and the shredded lettuce, the second experiment was conducted by modifying the interior of the bag and placing the reservoir in the middle of a package chamber (Figure 7-2, Figure 7-5, and Figure 7-10). Chlorine dioxide precursors were put in a LDPE tube, which has very low 201 barrier to ClO2 gas. Gas delivery was regulated by permeability of the selected film, so that the gas would be readily released and available for gas-phase decontamination within the package chamber (Appendix 8). For this study, ClO2 was released through the entire surface of LDPE tube, while, in the previous study, ClO2 gas was mainly delivered through the 6 openings located vertically along the edges of the chamber. The differences between microbial populations in control bag and those in middle-GR bag (Table 7-3) were reported as log10 CFU/g reductions of E. coli O157:H7 as shown in Figure 7-11. The calculated average log10 CFU/g reductions for the whole bag were shown in Figure 7-12. The results followed the same trend as those found in the first experiment; increasing ClO2 gas in the headspace significantly increased the reduction in cell population of E. coli O157:H7 (Han et al., 2000; Lee et al., 2004). Also, appearance changes, including bleached and watery tissues, more noticeable in lettuce from the 8 mgClO2/day bag, with those changes most evident in lettuce pieces next to the GRs (Ellis et al., 2006; Gómez-López et al., 2009). For lettuce bags containing a dose of 4 mgClO2/day dose, no improved microbial inactivation was observed between the two-GR (Figure 7-6(d)) and middle-GR bags (Figure 7-12(b)). Thus, using either two-GR design bag or a middle-GR design bag, with 4 mgClO2/day, yield similar microbial inactivation, even though the releasing areas of both designs were not equal. This could be attributed to the reactive nature of ClO2 gas. 202 More extensive understanding of ClO2 gas inside the package is needed to further understand this finding. On the other hand, comparison of log10 CFU/g reductions between samples taken from two-GR design bags (Figure 7-6(c)) and middle-GR design bags (Figure 7-12(a)), with 8 mgClO2/day, showed that moving the reservoir from the edges to the center of the chamber and increasing the gas releasing surface area significantly improved the E. coli O157:H7 killing efficiency of ClO2 gas, on day 1, 4, and 7, by 0.61, 0.37, and 0.27 log10 CFU/g, respectively, but, at this ClO2 dose, the appearance of the lettuce was compromised (Figure not shown). Given that two-GR design bags, with 4 and 8 mgClO2/day, both showed comparable average log10 CFU/g reductions in the first experiment, it was likely that the soft and watery lettuce pieces, adjacent to the reservoir opening, partly blocked the flow of ClO2 gas, preventing the gas distribution, especially, those in 8 mgClO2/day bags. Increasing the release area from 6 small openings to the entire surface of a LDPE tube could be the main factor in improving the gas distribution in the middle-GR design (with 8 mgClO2/day), in a more uniform fashion, resulting in the significantly higher average log10 CFU/g reduction of E. coli O157:H7. 203 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was more visible Day 4: Browning occurred more extensive and darker in color, especially at the cut area of midribs Day7: Browning found on most of the lettuce pieces; the texture became watery and soft in many pieces, especially the ones contained midrib parts Figure 7-10(a). E coli O157:H7 inoculated shredded lettuce packed in plain bags with 0.0 mgClO2/kg (Control for o study 2), stored at 4 C for 0, 1, 4, and 7 days 204 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was visible Day 4: Browning was darker in color, especially at the cut area of midribs; bleaching was noticeable on many lettuce pieces around GRs; lettuce pieces near GRs also became watery and soft Day 7: Browning and bleaching was more extensive; the texture became watery and soft in many pieces, especially the ones located next to GRs Figure 7-10(b). E coli O157:H7 inoculated shredded lettuce packed in middle-GR bags with 8.0 mgClO2/kg, stored o at 4 C for 0, 1, 4, and 7 days 205 Storage: Day 0 Day 1 Day 4 Day 7 Day 0: Browning was visible at the cut of some lettuce pieces; bruises also present (darker green color), especially on the leaf area Day 1: Browning was darker in color Day 4: Browning was more extensive; bleaching was noticeable on some lettuce pieces around GRs; some lettuce pieces near GRs also became watery and soft Day 7: Browning was more extensive; the texture became watery and soft in some pieces, especially the ones located next to GRs; bleaching was visible in many lettuce pieces Figure 7-10(c). E coli O157:H7 inoculated shredded lettuce packed in middle-GR bags with 4.0 mgClO2/kg, stored o at 4 C for 0, 1, 4, and 7 days 206 Day 1 1.76 a,a,1 ±0.06 0.66 a,b ±0.19 0.61 b,a ±0.10 0.43 b,b ±0.02 Day 4 1.87 a,a ±0.04 0.59 a,b ±0.01 1.51 b,a ±0.07 0.61 a,b ±0.13 Day 7 2.13 a,a ±0.22 0.34 a,b ±0.05 1.63 b,a ±0.29 0.26 a,b ±0.10 (a) (b) Figure 7-11. Log10 CFU/g reduction of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in middle-GR design bag with 8 mgClO2/day (a) and middle-GR design bag with 4 mgClO2/day (b) 1 First different script letters indicates statistically differences between means at  of 0.05 of different sample bags with the same portion; Second different script letters indicates statistically differences between means at  of 0.05 of samples with different portions within the same bag 207 Day 1 1.21±0.11 a,a,1 Day 4 1.23±0.02 Day 7 1.24±0.11 0.52±0.05 a,a 1.06±0.08 a,a b,b 0.94±0.17 (a) b,a a,b (b) Figure 7-12. Average log10 CFU/g reduction of E. coli O157:H7 of samples stored for 1, 4, and 7 days in middle-GR design bag with 8 mgClO2/day (a) and middle-GR design bag with 4 mgClO2/day (b) 1 First different script letters indicates statistically differences between means at  of 0.05 of different bag designs with the same storage time; Second different script letters indicates statistically differences between means at  of 0.05 of samples with different storage times within the same design 208 Even though, at the testing conditions, utilization of ClO2 as an antimicrobial gas in the package gave only up to 1.24 ± 0.09 log10 CFU/g reduction of E. coli O157:H7 (for middle-GR design, with 8 mgClO2/day; Figure 7-12(a)), it is beneficial to note that the average values represent the average kills for a particular design and ClO2 level, but the closer look specifically at each bag portion indicated that, at the area next to gas releasing locations, as high as 2 log10 CFU/g reduction could be achieved. 7.4. Conclusion Three types of flexible bags with different modified interior structures were used to package Escherichia coli O157:H7 inoculated shredded lettuce, along with 2 levels of ClO2 gas, i.e. 4 and 8 mgClO2/day which were released from gas reservoirs located in different areas of the bag interior, i.e. from one side of the bag, from both sides of the bag, and from center of the bag in one-GR, two-GR, and middle-GR, respectively. The o samples were kept at 4 C and stored for 1, 4, and 7 days. Log10 CFU/g reduction of E. coli O157:H7, from different sections of the bag, were determined. The results showed that increasing the level of ClO2 gas per day increased the reduction in cell population of E coli O157:H7 with the highest log10 CFU/g reduction found in samples taken from the area next to the gas reservoir. The calculated average log10 CFU/g reductions for the whole bag showed that ClO2 gas gave up to 1.24 ± 0.09 log10 CFU/g reduction of E. coli O157:H7, depending on the modified interior design, and two-GR and middle-GR bag 209 designs gave higher log10 CFU/g reduction than those from one-GR design, for both ClO2 gas levels. The bag design that minimized the distance between the gas reservoir and the sample showed higher log10 CFU/g reduction. Changes in texture and appearance of the sample were visually observed, especially in the samples taken from the area exposed to higher level of ClO2 gas and considered to be one of the factors that affects the gas distribution inside the package. This study also indicated that; design optimization can result in applying smaller dose of ClO2 gas (4.0 mgClO2/day) to achieve the same level of log10 CFU/g reduction of E. coli O157:H7 with one-GR at a higher dose. Delivery of ClO2 gas within the packaging system could also be adjusted and regulated to suit specific conditions of particular product requirements by selecting polymer films with different barrier properties; or by changing locations or numbers of openings of GRs. 210 BIBLIOGRAPHY 211 Cantwell, M., Rovelo, J., Nie, X. and Rubatzky, V. (1998 ). Specialty salad greens: postharvest physiology and shelf-life. III International Symposium Diversification of Vegetable Crops, Beijing, China Department of Health and Human Services (2008) "Summary Statistics for Foodborne Outbreaks, 2006." Retrieved November 1st, 2008, from http://www.cdc.gov/foodborneoutbreaks/documents/2006_line_list/2006_line_list .pdf. Diaz, C. and Hotchkiss, J. H. (1996). "Comparative Growth of Escherichia coli O157 : H7, Spoilage Organisms and Shelf-Life of Shredded Iceberg Lettuce Stored under Modified Atmospheres." Journal of the Science of Food and Agriculture 70: 433438. Du, J., Fu, M.-R., Li, M.-M. and Xia, W. (2007). "Effects of Chlorine Dioxide Gas on Postharvest Physiology and Storage Quality of Green Bell Pepper (Capsicum frutescens L. var. Longrum) " Agricultural Sciences in China 6(2): 214-219. Du, J., Han, Y. and Linton, R. H. (2002). "Inactivation by chlorine dioxide gas (ClO2) of Listeria monocytogenes spotted onto different apple surfaces." Food Microbiology 19: 481-490. Du, J., Han, Y. and Linton, R. H. (2003). "Efficacy of chlorine dioxide gas in reducing Escherichia coli O157:H7 on apple surfaces." Food Microbiology 20: 583-591. Ellis, M., Cooksey, K., Dawson, P., Han, I. and Vergano, P. (2006). "Quality of Fresh Chicken Breasts Using a Combination of Modified Atmosphere Packaging and Chlorine Dioxide Sachets." Journal of Food Protection 69(8): 1991 - 1996. FSIS (2005) "Poultry preparation." Retrieved November 22nd, 2009, from http://www.fsis.usda.gov/factsheets/Poultry_Preparation_Fact_Sheets/index.asp. Gómez-López, V. M., Rajkovic, A., Ragaert, P., Smigic, N. and Devlieghere, F. (2009). "Chlorine Dioxide for Minimally Processed Produce Preservation: A Review " Trends in Food Science & Technology 20(1): 17-26 Han, Y., Linton, R. H., Nielsen, S. S. and Nelson, P. E. (2000). "Inactivation of Escherichia coli O157:H7 on surface-uninjured and -injured green pepper (Capsicum annuum L.) by chlorine dioxide gas as demonstrated by confocal laser scanning microscopy " Food Microbiology 17: 643-655. ICA TriNova LLC (2006). Profiling gas release from ClO2 sachets. Standard Operation Procedure: Profiling gas release from ClO2 sachets, ICA Trinova. Keskinen, L. A., Burke, A. and Annous, B. A. (2009). "Efficacy of chlorine, acidic electrolyzed water and aqueous chlorine dioxide solutions to decontaminate 212 Escherichia coli O157:H7 from lettuce leaves." International Journal of Food Microbiology 132: 134-140. Lee, S.-Y., Costello, M. and Kang, D.-H. (2004). "Efficacy of chlorine dioxide gas as a sanitizer of lettuce leaves." Journal of Food Protection 67(7): 1371-1376. Lee, S.-Y., Dancer, G. I., Chang, S., Rhee, M.-S. and Kang, D.-H. (2006). "Efficacy of Chlorine Dioxide Gas Against Alicyclobacillus acidoterrestris Spores on Apple Surfaces." International Journal of Food Microbiology 108: 364-368. Lindsay, D., Brozel, V. S., Mostert, J. F. and Holy, A. v. (2002). "Differential efficacy of a chlorine dioxide-containing sanitizer against single species and binary biofilms of a diary-associated Bacillus cereus and a Pseudomonas fluorescene isolate." Journal of Applied Microbiology 92: 352-361. Lopez-Galvez, G., Saltveit, M. and Cantwell, M. (1996). "Wound-induced Phenylanine ammonia lyase activity: factors affecting its induction and correlation with the quality of minimally processed lettuces " Postharvest biology and technology 9: 223-233. Lynch, M., Painter, J., Woodruff, R. and Braden, C. (2006) "Surveillance for FoodborneDisease Outbreaks --- United States, 1998-2002." Retrieved November 1st, 2008, from http://www.cdc.gov/mmwr/preview/mmwrhtml/ss5510a1.htm?s_cid=ss5510a1_e. Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2009). "Mass Transfer Study of Chlorine Dioxide Gas through Polymeric Packaging Materials." Journal of Applied Polymer Science 114(5): 2929-2936. Nowak, B., Sammet, K., Klein, G. and Mueffling, T. v. (2006). "Trands in the production and storage of fresh meat - the holistic approach to bacteriological meat quality." International Journal of Food Science and Technology 41: 303-310. Peiser, G., López-Gálvez, G., Cantwell, M. and Saltveit, M. E. (1998). "Phenylalanine ammonia lyase inhibitors control browning of cut lettuce " Postharvest Biology and Technology 14(2): 171-177 Phillips, C. (1996). "Review: Modified atmosphere packaging and its effects on the microbiological quality and safety of produce." International Journal of Food Science and Technology 31: 463-479. Rico, D., Martín-Diana, A. B., Barat, J. M. and Barry-Ryana, C. (2007). "Extending and measuring the quality of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology 18(7): 373-386 213 Shin, J. (2007). Evaluation of the Effect of Chlorine Dioxide and Allyl-isothiocyanate on the Growth of Salmonella Typhimurium and Listeria Monocytogenes on Fresh Chicken Breast and Effect of Chlorine Dioxide Exposure on the Physical Properties of Plastic Films. School of Packaging. East Lansing, Michigan State University. Ph.D. Sy, K. V., Murray, M. B., Harrison, M. D. and Beuchat, L. R. (2005). "Evaluation of gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes, and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection 68(6): 1176 - 1187. Zhou, T., Harrison, A. D., McKellar, R., Young, J. C., Odumeru, J., Piyasena, P., Lu, X., Mercer, D. G. and Karr, S. (2004). "Determination of acceptability and shelf life of ready-to-use lettuce by digital image analysis." Food Research International 37: 875-881. 214 CHAPTER 8: OVERALL CONCLUSIONS AND FUTURE WORKS 8.1. Overall conclusions This work presents a new approach for the implementation of chlorine dioxide (ClO2) as an antimicrobial gas in food packaging applications, particularly, for fresh and fresh cut produce. Such antimicrobial packaging systems could also be regarded as part of a hurdle technology where the packaging system will provide the last stage of sanitation to assure a safer food product. To be able to consider ClO2 gas as an antimicrobial agent for vapor-phase decontamination in food packaging systems, the knowledge of its compatibilities with packaging materials, especially with polymeric films which are a major component of typical food packaging materials, is essential. The presence of ClO2 gas in the packaging system should be a compliment to the active-protective function of the package and does not compromise the system’s performance, e.g. barrier properties as well as mechanical properties. The understanding on how packaging could be modified to impact antimicrobial effects of ClO2 gas is also critical in order to maximize efficiency of the gas. Therefore, packaging design could turn into an important tool to improve performance of 215 disinfecting gases. Although ClO2 is being considered in this specific work as the antimicrobial gas, this approach could implemented into other gases or vapors. This work was aimed to study the potential use of ClO2 gas in food packaging applications. Specific objectives, along with the corresponding hypotheses, were stated earlier in Chapter 1. A summary of the main findings are outlined here: 8.1.1. Mass transfer of ClO2 gas -19 According to the P values for selected films which ranged between ≤ 7.0 × 10 -16 – 9.68 × 10 2 kgClO2·m·m ·s·Pa, BOPP, PET, PLA, nylon, and multilayer EVA/EVOH /EVA were categorized as high barrier to ClO2, while PEs, PVC, and PS were regarded as having low barrier properties to ClO2 gas. The activation energy of permeation (EP) of PET and PLA films were also determined (Netramai et al., 2009). 8.1.2. Impact of ClO2 gas on chemical, physical, mechanical, and barrier properties of packaging materials After 14 days of treatment, ClO2 exposure caused decreases in tensile and barrier properties in some films, for example, PEs and PET, while improved barrier properties to O2 were observed in treated nylon film. Changes in permselectivity (PCO2/PO2) were found in many polymer types, with HDPE showing the most change at 46.8%. 216 8.1.3. Antimicrobial activity of ClO2 gas on shredded lettuce Antimicrobial effects of ClO2 gas on microorganisms residing on lettuce have been reported many times, during the course of this entire study and found to increase as the applied concentration and/or exposure time increased. For example, ClO2 used in gasphase decontamination inside the package of shredded lettuce inoculated with E. coli O157:H7, ClO2 at a level of 8 mgClO2/kg lettuce gave higher log10 CFU reduction of E. coli O157:H7 as compared to 4 mgClO2/kg lettuce. But, higher ClO2 level used also increase changes in appearance, i.e. watery and soften tissue, of shredded lettuce. Regardless of the general trend, i.e. increasing the ClO2 level increased average log10 CFU reduction, the degree of disinfection depended largely on different packaging designs, which will be concluded in the next section. - Residual ClO2 and ClO2 absorbed by lettuce was quantified to study the lettuce/ClO2 absorption pattern and determine if the amount of ClO2 absorbed by lettuce is significant enough to take into account when selecting ClO2 dose for vapor-phase decontamination. According to the findings, increasing concentration and/or time of exposure increased absorption of ClO2 by lettuce. The more severe treatment conditions also increased appearance changes. 217 8.1.4. Effects of packaging design on antimicrobial effectiveness To study the influences of packaging design on antimicrobial effects of ClO2 packaging system for shredded lettuce, 3 types of flexible bag with different modified interior structures were used. Each design released ClO2 gas from different locations. The studies indicated that the bag design allowing the shortest travel distance of ClO2 gas to the sample helped improved reduction of E. coli O157:H7 population. Through optimization, lower levels of ClO2 could be used to achieve the comparable level of log10 CFU reduction and still maintain the produce’s appearance and texture. 8.2. Applications and future works 8.2.1. Mass transfer of ClO2 gas Information on compatibilities of ClO2 gas and polymeric materials, especially regarding mass transfer characteristics and the possible effects of ClO2 exposure on different types of polymeric materials from this study is the beginning step of effectively including antimicrobial gas in the packaging system; or designing the packaging system, by selecting suitable material, for the system that included ClO2 gas. Polymeric materials show wide range of barriers to ClO2 gas (Netramai et al., 2009). With this basic knowledge, one can narrow down the material choice to best fit required applications, for example, if the packaging system needs to maintain ClO2 gas 218 within the package for long-term exposure, then materials with high barrier to ClO2 gas are recommended. However, if the package needs to deliver certain amount of ClO2 for particular time frame, the information on permeability coefficient of many polymer films could serve as initial guideline for material selection, as in the case of LDPE selection for ClO2 gas reservoir in packaging study (Chapter 7). Mass transfer study of ClO2 gas should be explored further by varying the testing conditions, especially expanding the temperature range to cover typical storage o temperature for fresh produce, i.e. 4 C, since the permeation behavior of the polymer is temperature dependent (Selke et al., 2004; Van Krevelen, 1997). The study on EP of PET and PLA also clearly indicated this fact. The calculation on ClO2 level for packaging study 2 (Appendix 8) also demonstrated the disadvantage of the lack of said information. A system with a continuous supply of ClO2 gas at a constant concentration is also desirable. The developed system described in Chapter 3 was stable for the testing o durations used, i.e. ≤ 24 hours for permeation study at 23 C, but, if the longer testing time is necessary, for example, at the lower temperature where permeation process happen in slower rate, ClO2 solution might not be a suitable gas source (even though, at lower temperature, ClO2 gas should also degrade in a slower rate). 219 8.2.2. Impact of ClO2 gas on properties and performance of packaging materials Regarding chemical degradations of polymeric films caused by exposure to ClO2 gas, since statistical differences exist between untreated and ClO2-treated film after 14 days of treatment and the results were obtained at 10 mg/L of ClO2 gas which is at the upper end of its range in food applications (refer to Table 2-2, Chapter 2), this study was conducted under an extreme scenario used in food applications. Therefore, if any further study is needed, it should be done at specific exposure condition that represents the use of ClO2 gas in particular produce application. On the other hand, for other applications, such as in medical products’ packaging systems which might require higher level of ClO2 for the applications, the influences of ClO2 gas on the package performance could be of greater concern and additional tests should be performed, before finalizing material selection. 8.2.3. Improving antimicrobial capacity of ClO2 through packaging design 8.2.3.1. Important factors in packaging design In the packaging study, the effects of packaging design on antimicrobial activities of ClO2 gas were quantified by determining log10 CFU reduction of E. coli O157:H7 in treated shredded lettuce located in different areas of the bag. Even though, at the testing conditions, utilization of ClO2 as an antimicrobial gas in the package gave only up to 220 1.24 ± 0.09 log10 CFU/g reduction of E. coli O157:H7 (for middle-GR design, with 8 mgClO2/day), it is beneficial to note that the average values represent the average kills for particular design and ClO2 level, but the closer look specifically on each bag portion indicated that, at the area next to gas releasing locations, as high as 2 log10 CFU/g reduction could be achieved. Thus, interior characteristics of the package, i.e. location and size of the releasing area for ClO2 gas, influenced its antimicrobial effects on E. coli O157:H7 inoculated on shredded lettuce; minimizing the travel distance between ClO2 gas and the target surface, as well as maximizing the releasing surface of antimicrobial gas, proved to increase the log10 CFU/g reduction of particular pathogen. These findings could be utilized in modifying the interior of packages, for food products, that will be used in conjunction with any antimicrobial gas included in the packaging system, without changing the packaging exterior. Many other important factors are also known to influence sanitizing effects of antimicrobial gas. Some studies showed that it was more difficult to inactivate microorganisms on injured or cut surfaces, as compared to those on smooth surfaces (Han et al., 2000; Mahovic et al., 2009). Therefore, shape and size of the products, as well as the surface characters, including damages and bruises, are also an important parameter to be considered (Du et al., 2002; 2003; Gómez-López et al., 2009; Han et al., 2000; Mahovic et al., 2009; Yuk et al., 2006). Du et al. reported the different in log10 CFU/site reduction of E. coli O157:H7 and Listeria monocytogenes, obtained from different areas of the fruit, after ClO2 treatment (Du et al., 2002; 2003). Therefore, additional studies 221 should be extended to the applications of ClO2 gas as a headspace gas for other types of packaged produce, especially one that has different shape, size, and surface characters, such as cherry tomatoes, baby carrots, etc (Du et al., 2002; Han et al., 2000; Mahovic et al., 2009; Yuk et al., 2006), for better understanding on the factors affecting the antimicrobial gases’ efficiency in vapor phase decontamination for food packaging application. Microbial enumeration should also be included in any future study on absorption pattern of ClO2, as it could help explaining ClO2-consumption behavior of produce at different treatment conditions, besides, it is important to determine if consumption of ClO2 gas could also be influenced by microbial load on the food products, as mentioned earlier. Future experiments should also included testing temperature at 4oC, for both - absorption test and experiment on degradation of ClO2 and ClO2 on exposed produce, since the absorption behavior might depend on 1) solubility of ClO2 in solution and 2) the influences of temperature of its disinfecting capacity (USEPA, 1999). To expand the scope of ClO2-absorption study further, other commodities should be included in future works. Different types of produce provide entirely different shapes, textures, and surface characteristics, as well as different skin pigments and normal flora resided on its surface, thus, different ClO2-absorption behavior, which could partially be indicated by different sanitizing results when the same treatment were applied to various food products, could be expected (Du et al., 2003; Han et al., 2000; Han et al., 2004; Rodgers et al., 2004; Sy et al., 2005). 222 8.2.3.2. Computational modeling of antimicrobial packaging system Conventional way of studying the antimicrobial effect of ClO2 gas in a particular packaging system would be as what has been done in this study (chapter 7), by conducting the actual experiment. Gathering enough information through repeated experiments could be time consuming and might not be applicable for certain testing conditions. In many cases, computational modeling based on principle of finite element analysis (FEA) could be a very useful tool in packaging development to help predict and, to a certain extent, to quantify the outcome of the study. Computational fluid dynamic (CFD) is the science of solving fluid flow, heat and mass transfer, chemical reactions and related phenomena. FEA is a numerical technique applied to solve complex partial differential or integral equations encountered in many continuous engineering domain, e.g. elasticity, aeronautical problem, etc. by discretize the domain into a set of ‘finite’ sub-domains, or ‘element’ (or volume or cell). Considering each finite element would simplify or even eliminate the need to solve complicated domain, giving a reasonable estimation depending on the required precision and accuracy. The overview of CFD is shown in Figure 8-1 CFD process starts with defining the model’s goals, then identifies the model domain, and designs and creates the grids (e.g. shape and size of the package and releasing location of ClO2 gas). To set up an appropriate numerical model to solve for specific system, for example product/package system, first, one needs to select appropriate physical model, e.g. turbulence for gas mixing and monitoring surface contact within the package. Material properties (e.g. chemical and physical properties of ClO2, mass transfer behavior of the package, and absorption behavior of ClO2 on lettuce) 223 and boundary conditions, are necessary in model building. After computing the solution, the results will be verified. And one might consider revise the model, if the outcome did not reflect real circumstance. Actual experiment might also be conducted to confirm the prediction of the model, but the number of real tests performed could be significantly smaller. Gathering parameters necessary to build an appropriate CFD model could be difficult and time-consuming, especially when dealing with new type of system where many data has never been generated before, as in the case of food product/packaging/antimicrobial gas system. The information obtained from this work, i.e. compatibilities of ClO2 gas with polymeric materials, the effects of packaging design, especially on gas releasing locations on antimicrobial activities of ClO2, as well as, the ClO2-absorption behavior of lettuce are among a few parameters that could be beneficial in CFD modeling for packaging study. Many additional data type are needed to develop the CFD model for food packaging study in general, for example, the effects of shape and texture of food products. 224 Figure 8-1. CFD modeling overview. Adapted from (Fluent Inc., 2005). 225 To be able to effectively monitor and quantify the behavior of ClO2 gas inside the package, in real-time, small, precise, and accurate ClO2-sensor is needed. Unfortunately, most sensors for ClO2 are suitable for atmosphere measurement which requires large volume of gas sample or suitable for determining ClO2 in solution from (USEPA, 1999). Some chemical sensors are developed for detecting trace amount of ClO2 gas, but the size of the equipment is still too large to incorporate into the food packages, as in the case of detector used in mass transfer study (Chapter 3 and 4). Using CFD in predicting packaging performance under various circumstances has become a useful tool for many studies in the past few years (Abdul Ghani, 2003). However, there are many challenges in applying it in antimicrobial packaging study. One of a major drawback, for example, is to correlate the outcome of the model, in terms of gas circulation, with the sanitizing efficiency of selected antimicrobial gas incorporated into the packaging system. If such approach could successfully be developed, it could be applied to any potential antimicrobial gas/packaging system. Since the use of ClO2 gas in antimicrobial packaging application could also be introduced as one of preservation methods in hurdle strategy, before the products are being delivered to the consumers (Allende et al., 2006; Gómez-López et al., 2009; Huang et al., 2006; Rico et al., 2007). By applying several technologies to assure safety and improve shelf-life of food products, milder treatment for each technique could be used to achieve significant results. Future studies should explore the effects of the combination of antimicrobial packaging with other mild, yet effective, technologies, such as sequential 226 cleaning and ultrasound (Allende et al., 2006; Huang et al., 2006; Pao et al., 2009; Singh et al., 2002), as well as the use of other packaging techniques, like modified atmosphere packaging (MAP) (Jin and Lee, 2007). Many processes can be applied throughout the food production lines to give synergistically disinfecting effect to improve safety of food products, while maintain their near-fresh quality. 227 BIBLIOGRAPHY 228 Abdul Ghani, A. G., Farid, M. M., and Chen, X. D. (2003). "A computational and experimental study of heating and cooling cycles during thermal sterilization of liquid foods in pouches using CFD." Journal of Process Mechanical Engineering 217: 1-9. Allende, A., Tomás-Barberán, F. A. and Gil, M. I. (2006). "Minimal Processing for Healthy Traditional Foods." Trends in Food Science & Technology 17(9): 513519. Du, J., Han, Y. and Linton, R. H. (2002). "Inactivation by chlorine dioxide gas (ClO2) of Listeria monocytogenes spotted onto different apple surfaces." Food Microbiology 19: 481-490. Du, J., Han, Y. and Linton, R. H. (2003). "Efficacy of chlorine dioxide gas in reducing Escherichia coli O157:H7 on apple surfaces." Food Microbiology 20: 583-591. Fluent Inc. (2005). Fluent 6.2: Flow modeling software training notes. Ann Arbor, MI, Fluent Inc. Gómez-López, V. M., Rajkovic, A., Ragaert, P., Smigic, N. and Devlieghere, F. (2009). "Chlorine Dioxide for Minimally Processed Produce Preservation: A Review " Trends in Food Science & Technology 20(1): 17-26 Han, Y., Linton, R. H., Nielsen, S. S. and Nelson, P. E. (2000). "Inactivation of Escherichia coli O157:H7 on surface-uninjured and -injured green pepper (Capsicum annuum L.) by chlorine dioxide gas as demonstrated by confocal laser scanning microscopy " Food Microbiology 17: 643-655. Han, Y., Selby, T. L., Schultze, K. K., Nelson, P. E. and Linton, R. H. (2004). "Decontamination of Strawberries Using Batch and Continuous Chlorine Dioxide Gas Treatments." Journal of Food Protection 67(11): 2450-2455. Huang, T.-S., Xu, C., Walker, K., West, P., Zhang, S. and Weese, J. (2006). "Decontamination Efficacy of combined Chlorine Dioxide with Ultrasonication on Apples and Lettuce." Journal of Food Science 71(4): M134-M139. Jin, H. H. and Lee, S. Y. (2007). "Combined effect of aqueous chlorine dioxide and modified atmosphere packaging on inhibiting Salmonella Typhimurium and Listeria monocytogenes in mungbean sprouts." Journal of Food Science 72(9): M441-445. Mahovic, M., Bartz, J. A., Schneider, K. R. and Tenney, J. D. (2009). "Chlorine dioxide gas from an aqueous solution: reduction of Salmonella in wounds on tomato fruit and movement to sinks in a treatment chamber." Journal of Food Protection 72(5): 952-958. 229 Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2009). "Mass Transfer Study of Chlorine Dioxide Gas through Polymeric Packaging Materials." Journal of Applied Polymer Science 114(5): 2929-2936. Pao, S., Kelsey, D. F. and Long III, W. (2009). "Spray washing of tomatoes with chlorine dioxide to minimize Salmonella on inoculated fruit surfaces and crosscontamination from revolving brushes." journal of Food Protection 72(12): 24482452. Rico, D., Martín-Diana, A. B., Barat, J. M. and Barry-Ryana, C. (2007). "Extending and measuring the quality of fresh-cut fruit and vegetables: a review." Trends in Food Science & Technology 18(7): 373-386 Rodgers, S. L., Cash, J. N., Siddiq, M. and Ryser, E. T. (2004). "A comparison of different chemical sanitizers for inactivating Escherichia coli O157:H7 and Listeria monocytogenes in solution and on apples, lettuce, strawberries and cantaloupe." Journal of Food Protection 67(4): 721-731. Selke, S. E. M., Culter, J. D. and Hernandez, R. J. (2004). Plastics Packaging: Properties, Processing, Applications, and Regulations. Cincinnati, OH, Hanser Gardner Publications, Inc. Singh, N., Singh, R. K., Bhunia, A. K. and Stroshine, R. L. (2002). "Efficacy of Chlorine Dioxide, Ozone, and Thyme Essential Oil or a Sequential Washing in Killing Escherichia coli O157:H7 on Lettuce and Baby Carrots " LebensmittelWissenschaft und-Technologie 35(8): 720-729 Sy, K. V., Murray, M. B., Harrison, M. D. and Beuchat, L. R. (2005). "Evaluation of gaseous Chlorine Dioxide as a Sanitizer for Killing Salmonella, Escherichia coli O157:H7, Listeria monocytogenes, and Yeasts and Molds on Fresh and Fresh-Cut Produce." Journal of Food Protection 68(6): 1176 - 1187. USEPA (1999) "4. Chlorine Dioxide." Retrieved September 11th, 2007, from http://www.epa.gov/ogwdw/mdbp/pdf/alter/chapt_4.pdf Van Krevelen, D. W. (1997). Properties of Polymers. New York, Elsevier Science B.V. Yuk, H.-G., Bartz, J. A. and Schneider, K. R. (2006). "The Effectiveness of Sanitizer Treatments in Inactivation of Salmonella spp. from Bell Pepper, Cucumber, and Strawberry." Journal of Food Science 71(3): M95-M99. 230 APPENDICES 231 APPENDIX 1: Antimicrobial efficacy of ClO2 gas against E. coli O157:H7 on lettuce This study focus on studying the topographic characters of E. coli O157:H7 biofilm on Romaine lettuce leaf surfaces, before and after being exposed to ClO2 gas by using Scanning Electron Microscope (SEM) (Netramai et al., 2010). Research plan:  Lettuce sample preparation: Shredded Romaine lettuce pieces were inoculated o with inoculum of E. coli O157:H7 SEA and kept at 4 C, for overnight, to allow bacteria attachment. The 48-hour inoculated samples had E. coli O157:H7 population of 8.33 ± 0.05 log10 CFU/g sample.  Chlorine dioxide treatment of lettuce sample: Inoculated lettuce samples were treated with 0.1 and 0.2 mg/L of ClO2 gas (continuous exposure) for 0 - 60 min in an air-tight chamber as shown in Figure A-1. After predetermined time intervals, the samples were taken out; 25 g of treated sample was sampling for E. coli O157:H7 enumeration on Tryptic Soy Agar (TSA) overlaid with Sorbital MacConkey Agar with Cefixime, and Tellurite (CT-SMAC) (Han et al., 2002).  SEM procedure: Both untreated and ClO2-treated- inoculated lettuce pieces were cut into specific size (circle with diameter of 1 cm), underwent fixation and dehydration processes using the Microscopic Imaging facility at ERRC, USDA. Samples were fixed by immersion using 2.5% Glutaraldehyde for 1 hour 232 followed by a 0.1M Imidazole buffer rinse (2X). Tissue was dehydrated through an Ethyl Alcohol series of 50, 80, 90 and 100% (3X), for 1 hour each. The samples were then critical point dried with Liquid CO2 (Denton Vacuum, CP-1) process treatment, and sputter coated (Edwards Scan Coat Six) with gold (20 mA for 30 sec, repeated 2 times). SEM microscopic images were obtained using Quanta 200 FEG Environmental Scanning Electron Microscope using a voltage of 10.0 kV at 3.0 nm spot size. Figure A-1. Chlorine dioxide gas treatment of shredded lettuce in an air-tight chamber Results: The untreated surfaces of inoculated lettuce leaf (48 hrs after inoculation) were covered with continuous layer of E. coli O157:H7 biofilm as shown in Figure 2-3. Bacteria cells were visible in some area and distributed unevenly, accumulated more in the wrinkle surfaces. 233 After the inoculated surfaces were exposed to 0.2 mg/L of ClO2 gas for up to 60 min, microbial enumeration showed E. coli O157:H7 reduction of ≤ 1.57 ± 0.05 log10 CFU/g sample, as listed in Table A-1. The ClO2-treated surfaces showed the absence of biofilm and the biofilm being destroyed (Figure 2-4, Chapter 2), in many area. By visual observation of the surfaces, using SEM, the number of E. coli O157:H7 cells visibly on the surfaces also noticeably decreased. It should be noted that the observed results from SEM images showed possibility of biofilm being partially removed, after ClO2 treatment, but the microbial population was only reduced by less than 2 log10 CFU/g sample. Furthermore, another research conducted by Ölmez and Temur on sanitizing effects of ozone (which is considered to be stronger oxidizing agent than ClO2) and chlorine solutions, on biofilm of E. coli O157:H7 on lettuce leaf, indicated that there was no removal of biofilm even though the population of target microorganism was reduced up to 1.5 log10 CFU/g sample (Ölmez and Temur, 2010). Further works are recommended, to confirm effects of ClO2 gas on biofilm of E. coli O157:H7 on lettuce leaf. 234 Table A-1. Log10 CFU reduction of E. coli O157:H7 and total aerobic plate count (TAPC) of ClO2 treated lettuce sample Treatment\Media E. coli O157:H7 (log10 CFU/g) 0.1 mgClO2/L 30 min 60 min 1 2 a1,2 0.69 ± 0.04 b 1.05 ± 0.06 0.2 mgClO2/L a 1.14 ± 0.02 b 1.57 ± 0.05 Mean ± standard deviation Different superscript letters indicates statistically differences between means, in the same column, at  of 0.05 235 APPENDIX 2: Thiosulfate titration for determination of ClO2 in solution The analytical procedure for determining ClO2 level in solution was described below. This method was modified from titration method outlined by ICA Trinova (ICA TriNova LLC, 2006; Post and Moore, 1959). Apparatus:  50 ml Burette and Stand  50 ml Erlenmeyer flask  pH meter  Automatic titrator equipped with pH meter  10% (wt/wt) Potassium Iodide (KI) solution  0.1 N, 0.01 N, and/or 0.001 N Sodium Thiosulfate (Na2S2O3) (Certified or Reagents: solution)  2 N Sulfuric acid (H2SO4)  Starch indicator 236 Procedure: The procedure outlined below is for manual titration. 1. Sampling known amount of ClO2 solution (Vs mL) and put into Erlenmeyer flask. Add KI solution if the volume is too low. 2. (Add starch indicator, if required.) 3. Titrate the solution to a colorless end-point with Na2S2O3. Record result as Vn. 4. Adjust pH by adding H2SO4 to a pH of ≤ 2. 5. Allow the solution to stand in the dark for 10 min. 6. Titrate the solution to a colorless end-point with Na2S2O3. Record result as Va. 7. Discard the solution. Calculation: To obtain results in mg ClO2/L: V  N  67.5 103 ClO2 (mg / L)  a 4  Vs where N = Normality of Na2S2O3. 237 APPENDIX 3: Results for consistency test on isostatic method In continuous flow permeation, the value of the permeant flow at time t, Ft, is described below: Ft F where     n 2 2   4   2    exp       4 Dt     4 Dt n 1,3,5     (1) Ft is the permeant flow ratio in the 0 partial pressure side and t is the F permeation time. For short time, equation (1) explains permeation process from time t = 0 up to 95% of steady state and can be simplified as:  4  1/ 2   exp x  X   Where (2) 2 Ft is and X is . 4 Dt F To be able to assume that polymeric material characteristics, permeant properties, temperature, and concentration gradient remain unchanged or constant, during permeation process, the consistency test must be performed. The test consists of 1) 2 determining R of a straight line plot between 1 xt versus time to reach at   t , with intercept pass the origin (0,0); and 2) determining K1 and K2 where 238 K1  2 t1 / 4 t and K 2  1 / 4 . Table A-2 reported values of K1, K2, and R obtained t1 / 2 t3 / 4 from consistency test for permeation experiments conducted and reported in Chapter 3. All values indicated that there is no unacceptable variation of system parameters, such as permeant concentration and temperature fluctuation, and isostatic permeability experimental conditions can be assumed (Gavara and Hernandez, 1993).  2 Table A-2. K1, K2, and R for consistency test 2 0.42 ˂ K1 ˂ 0.46 0.65 ˂ K2 ˂ 0.69 R (set to pass the origin) HDPE 0.45 ± 0.01 0.68 ± 0.02 0.9837 LDPE 0.44 ± 0.02 0.66 ± 0.01 0.9703 LLDPE 0.43 ± 0.01 0.67 ± 0.02 0.9793 BOPP 0.46 ± 0.00 0.67 ± 0.01 0.9512 PS 0.43 ± 0.01 0.67 ± 0.02 0.9534 PVC 0.43 ± 0.01 0.66 ± 0.01 0.9841 PET 0.45 ± 0.01 0.67 ± 0.01 0.9703 PLA 0.44 ± 0.02 0.67 ± 0.01 0.9782 Nylon 0.44 ± 0.02 0.68 ± 0.01 0.9703 Polymer 1 EVA/EVOH/EVA 1 nd Not determined 239 APPENDIX 4: Amperometric titration for determination of ClO2 in solution - The analytical procedure for determining ClO2 and ClO2 level in solution was described below. This method was modified from standard amperometric titration method for the examination of water and wastewater (4500-ClO2 C) (Greenberg et al., 1992; Haller and Listek, 1948). Apparatus:  50 ml Erlenmeyer flask or specific sample cup for titrator  pH meter  Titrator equipped with potentiometer  5% (wt/wt) Potassium Iodide (KI) solution  0.00564 N and/or 0.000564 N Phenylarsine oxide, PAO ((C6H5)AsO) Reagents: (Certified solution)  Phosphate buffer, pH 7.0  6 N Sulfuric acid (H2SO4)  6 N Sodium hydroxide (NaOH) 240 Procedure: 1. Divide sample solution into 3 portions, 200 mL each. 2. Portion 1: Adjust pH by adding NaOH to a pH of ≥ 12. After 10 min, adjust pH to 7 by adding H2SO4. Add 1 mL of KI solution. Titrate the solution to an end-point with PAO. Record result as A. 3. Portion 2: Adjust pH by phosphate buffer to a pH of 7. Add 1 mL of KI solution. Titrate the solution to an end-point with PAO. Record result as B. 4. Portion 3: Add 1 mL of KI solution. Adjust pH by adding H2SO4 to a pH of ≤ 2. After 10 min, adjust pH to 7 by adding NaOH. Titrate the solution to an end-point with PAO. Record result as C. 5. Discard the solutions. Calculation: To calculate ClO2 in mg ClO2/L: ClO2 (mg / L)  1.9B  A - To calculate ClO2 in mg Cl2/L:  ClO 2 ( mgCl 2 / L )  4 A  5 B  C 241 APPENDIX 5: ClO2 degradation profile and recovery rate of the experimental setup for absorption study The degradation profile of ClO2 in the solution and the recovery rate of titration - procedure for residual ClO2 and ClO2 were discussed below. Degradation profile of ClO2 in the solutions: The degradation rate of ClO2 in the solutions (which give out ClO2 gas level at 3.0 and 6.0 mg/L) was calculated from the different of ClO2 in solution at particular time and the initial amount at the beginning of the experiment (0 min), as followed: Degradatio n Rate (%)  mgClO2,T 0  mgClO2,Tt 100 mgClO2,T 0 where mgClO2,T 0 and mgClO 2,Tt = mg ClO2 in 1 L of solution at the exposure time of 0 and t min, respectively. The degradation profiles are reported in Table A-3. As expected, the amount of ClO2 in the solution gradually decreased with time, with slower rate for solution at higher level of ClO2. 242 Table A-3. Degradation profile (%) of ClO2 in solution according to time (min) Degradation profile (%) Time (min) 1 1 3.0 mg/L 15 3.96 ± 0.56 0.53 ± 0.20 30 5.70 ± 0.30 1.16 ± 0.51 45 5.94 ± 0.68 2.02 ± 0.18 60 8.110.96 2.62 ± 0.60 90 1 6.0 mg/L 14.18 ± 1.19 4.00 ± 0.22 For the solutions to release 3.0 and 6.0 mg/L of ClO2 gas into the chamber headspace, o at 23 C, the actual amount of ClO2 in 1 L of solution are approximately 87 and 175 mg. - Recovery rate of titration procedure for residual ClO2 and ClO2 : Recovery rates of titration procedure are calculated according to equation described below, and are listed in Table A-4. Re cov ery Rate (%)  mgClO2, R  mgClO2,Chlorite  100 mgClO2,T 0  mgClO2,Tt  where mgClO2, R = mg ClO2 recovered from the procedure; and mgClO2, Chlorite = - mg ClO2 as converted from mg ClO2 recovered. 243 Table A-4. Recovery rate (%) of titration procedure for residual ClO2 and ClO2 - Recovery rate (%) 1 1 3.0 mg/L 9.79 ± 1.22 1 6.0 mg/L 42.99 ± 21.00 For the solutions to release 3.0 and 6.0 mg/L of ClO2 gas into the chamber headspace, o at 23 C, the actual amount of ClO2 in 1 L of solution are approximately 87 and 175 mg. - Many literatures reported that, generally, ClO2 made up to 70% of end-products from reactions of ClO2 with organic matters found in waste water (USEPA, 1999). Thus, the recovery rates at both ClO2 levels may not represent the actual efficiency of the titration method. Also, noted that the amount of ClO2 disappeared from the solution (through auto-degradation or other means, Table A-3) could also contributed to the reduction of ClO2 in the solution over time. Table A-5 shows amount of ClO2 disappeared during the experiment (Decrease, mg), which was calculated from the difference between ClO2 at the beginning of the experiment (ClO2@0, mg) and remaining ClO2 at the end of the experiment (ClO2@t, mg). Amount of ClO2 lost through auto-degradation is listed in column ‘Auto-Degradation, mg’, while amount of ClO2 recovered in the sample could be seen by combining residual ClO2 and ClO2 (column ‘ClO2-, mg’ and ‘ClO2, mg’). 244 - Table A-5. Sample of raw data (shredded lettuce) for absorption study Level, Time, mg/L min Wt sample, g ClO2@0, mg ClO2@t, mg Decrease, mg AutoDegradation, mg ClO2-, mg ClO2, mg Recovery rate, % 6.0 30 48.1711 46.9669 49.7478 164.9194 179.8453 167.2313 162.9450 177.7275 165.2400 1.9744 2.1178 1.9912 1.9102 2.0830 1.9369 0.7464 1.43E-02 0.6483 4.28E-02 0.7196 1.43E-02 38.53 32.68 36.85 3.0 30 44.5234 43.4751 44.9116 83.7169 86.2228 83.6494 78.9244 81.2869 83.6494 4.7925 4.9359 4.7925 4.7696 4.9124 4.7657 0.4625 7.13E-03 0.5139 4.28E-02 0.3879 5.63E-02 9.80 9.54 9.27   245 APPENDIX 6: Characterization of PD-961EZ bag used in packaging study 2 PD-961EZ bags (26.67 × 43.18 cm ) were supplied from Cryovac (Cryovac Inc., Duncan, SC) in March, 2009. The bags were characterized to obtain chemical profile (as indicated by FT-IR spectrum), as well as important thermal and barrier properties (Table A-6), using procedures as previously outlined in Chapter 4 and 5. The determined characteristics were reported in Figure A-2 and thermal properties showed that the film had 2 Tm which coincided with those of LDPE and LLDPE, indicating that the film could be LDPE and LLDPE blend. 2.5 Absorbance, O.D. 2.0 1.5 1.0 0.5 0.0 4000 3500 3000 2500 2000 1500 Wavenumber, cm-1 Figure A-2. FT-IR spectrum of PD-961EZ bag 246 1000 FT-IR spectrum of the bag indicated that the film has chemical profile similar to -1 those of PE materials, with the exception of the peak at around 1750 cm region which coincided with that of carbonyl group in EVA material (Figure 5-6, Chapter 5). The film could be PE material coated with thin layer of EVA. Thermal properties showed that the film had 2 Tm which coincided with those of LDPE and LLDPE (Table 5-2, Chapter 5), indicating that the film could be LDPE and LLDPE blend. Table A-6. Important properties of PD-961EZ bag Properties Values (unit) o Tm ~107 and 118 C -19 PH2O 1.21 ± 0.10 × 10 PCO2 12.1 × 10 PO2 1.03 ± 0.12 × 10 PClO2 1 1 42.8 ± 1.07 × 10 -17 -2 -1 kg·m·m ·s ·Pa -17 -17 Mean ± standard deviation 247 -2 -1 kg·m·m ·s ·Pa -1 -1 -2 -1 -1 -2 -1 -1 kg·m·m ·s ·Pa kg·m·m ·s ·Pa APPENDIX 7: Identification of ClO2 levels for packaging study This study observed effects of different ClO2 treatments on population of microorganism on shredded lettuce. Research plan:  Chlorine dioxide treatment of lettuce sample: Air-tight chamber (265 L) was filled with 0.1 mg/L ClO2 gas before starting the batch treatment, and then 1.7 kg of un-inoculated shredded lettuce was laid on the bottom of the chamber. Samples were taken out at 15 and 30 min of exposure for microbial enumeration on Tryptic Soy Agar (TSA) (ClO2 level was monitor during the experiment and the concentration read 0.0 mg/L at approximately 30 min of exposure). Results: Log10 CFU reduction of total aerobic plate count (TAPC) of ClO2 treated lettuce samples were shown in Table A-7. Since ClO2 level was down to 0.0 mg/L at the end of 30 min, it is assumed that the total of 0.1 mg/L × 265 L = 26.5 mg of ClO2 was used up by 1.7 kg of shredded lettuce; and, assumed that ClO2 level decreased linearly, at 15 min of exposure, 26.5/2 = 13.25 mg of ClO2 was consumed by the sample. 248 Based on visual observation, ClO2 level at 8 (~7.79) mg/kg per day (or ~2.264 mg/283 g lettuce per bag per day) was selected as a high level ClO2 treatment. The level of 4 mg/kg per day (or ~1.132 mg/283 g lettuce per bag per day) was then chosen as a logical choice to study the effect of concentration on antimicrobial effects of ClO2, in the packaging study. Table A-7. Log10 CFU reduction of total plate count (TAPC) of ClO2 treated lettuce sample Treatment time 15 min 30 min 1 2 TAPC (log10 CFU/g) a,1,2 0.87 ± 0.08 b 1.37 ± 0.10 Visual observation Many lettuce pieces bleached; mostly at the top layer of lettuce bed. Mean ± Standard Deviation Different superscript letters indicates statistically differences between means, in the same column, at  of 0.05 249 APPENDIX 8: Amount of ClO2 precursors to be used in packaging study This section shows calculation of amount of ClO2 precursors to be used in packaging study. Materials and chemicals:  Fast release ClO2 precursor (0.5 mgClO2/g precursor per day)  Slow release ClO2 precursors (0.05 mgClO2/g precursor per day)  LDPE tube (0.875 × 10.5 in ) with P, at 23 C, = 66.0 ± 1.09 × 10 2 -2 -1 o -17 -1 kgClO2mm s Pa (Netramai et al., 2009) Calculations for packaging study on one-GR and two-GR designs: To fix the weight of precursors used in the study to 8 g, the calculations of amount of each precursor to be used in the study are as shown in Table A-8. The actual ClO2 gas release profiles were shown in Figure 7-3, Chapter 7. In Table A-8, amount of fast and slow release compounds are equal to x and y g, respectively. To avoid variation of using different weights of precursors to generate 2 levels of ClO2, the weight was fixed at 8 g, i.e. x + y = 8 g. 250 Table A-8. Calculation of amount of precursors to be used in packaging study 1 Fast release compound: Slow release compound: GR: Total precursor wt = Total released ClO2 wt = 0.5 mg/g/d chemical 0.05 mg/g/d chemical 8 g z mg = = 0.4 0.4 will be used for: will be used for: x+y 0.5x + 0.05y x y >>> >>> z= x= 0.45x + (z- for 1 day, z = Use: 2.264 mg ClO2 / 283 g lettuce Total mg ClO2 = x= 4.14 g y= 3.86 g 2.264 for 1 day, z = Use: 1.132 mg ClO2 / 283 g lettuce Total mg ClO2 = x= 1.63 g y= 6.37 g 1.132 )/0.45 wt-based: 251 g g y= z= 8-x 0.4+0.45x Calculations for packaging study on middle-GR design: Amount of ClO2 gas permeated through LDPE tube was calculated based on the following equation (Rodriguez et al., 2003; Selke et al., 2004): P q A  t  p (1) q or   P  A  t  p  (2) LDPE film was selected to be used as a GR tube in study 2 due to its low barrier to ClO2 gas (Netramai et al., 2009). Giving that 1) P is not constant, for this particular case, it varies according to the amount of ClO2 gas constantly released (which should make up to 2.26 mgClO2/day, for high level ClO2 treatment) and those permeated o o through LDPE tube, at 4 C; and 2) actual value of P at 4 C (which should be lower than o that at 23 C) was not available, amount of ClO2 permeated per sec was estimated using equation (2), assuming 1) there was no permeation in the first second of the process; and 2) at any given second, partial pressure of ClO2 in the tube is constant. t 1    x  t   q   i  i 1   3.64 10 6 P  A   tube volume        qt =  252 (2) where qt = amount of ClO2 permeated through LDPE in 1 sec, at time t (mg); P = 66.0 × -17 10 -2 -1 -1 kgClO2mm s Pa o -2 2 at 23 C; A = 1.18 × 10 m ; -5 linearly released from precursor per sec, i.e. 2.62 × 10 is mg ClO2 assumed to be mg for high level release; t = -2 time into permeation process (sec); tube volume = 4.19 × 10 -5 L; ℓ = 2.79 × 10 m; and 6 3.64 × 10 is the conversion factor from mgClO2/L to Pa. As partial pressure of ClO2 gas inside LDPE tube increase, amount of permeated ClO2 will continuously increase, due to partial pressure built up. When qt = permeated amount of ClO2 becomes constant. Time to reach qt = , the  depends on P of LDPE which is temperature dependent (Rodriguez et al., 2003; Selke et al., 2004). Theoretically, P of the same polymer should be smaller as temperature decreases, resulting in longer time taking to reach qt = . Figure A-3 demonstrates qt at different P. In this study, for LDPE GR to be able to deliver constant amount of ClO2 per day in this particular packaging system, sealed LDPE tubes containing ClO2 precursor should be left in chemical fume hood for predetermined time to achieve linear release at qt = , o and then packed along with shredded lettuce. Since P of LDPE at 4 C is not known and due to time constraint, the sealed tubes were left in chemical fume hood for approximately 5 hours. The actual ClO2 gas release profiles were shown in Figure 7-3, Chapter 7. 253 3.0e-5 Released ClO2 gas (mg) 2.5e-5 2.0e-5 at 11.0 x 10-17 kgClO2-m-m-2-s-1-Pa-1 at 33.0 x 10-17 kgClO2-m-m-2-s-1-Pa-1 1.5e-5 at 66.0 x 10-17 kgClO2-m-m-2-s-1-Pa-1 1.0e-5 5.0e-6 0.0 0 200 300 400 600 800 1000 1200 1400 1600 1800 Time (sec) Figure A-3. Calculated amount of permeated ClO2 (mg) from LDPE tube with different permeability coefficients 254 APPENDIX 9: Preliminary study on sample portions for packaging study Results of preliminary study to observe if the effects of gas distribution to the top and bottom portions of the bag are significantly different, was conducted. The procedure followed what has been outlined in Section 7.2.5 and 7.2.6 (except lettuce samples were taken from 6 different areas instead of 3 areas). The precursor used was of different type, as shown below. 22 Released ClO2 gas (mg) 20 18 16 14 12 10 8 6 0 1 2 3 4 5 6 7 8 Day Figure A-4. Gas release profile The results were provided in Figure A-5. Since there was no different between top and bottom portions (except in two-GR at Day 4 of storage), in the same lengthwise portion, the sampling portions for packaging study were finalized to 3 lengthwise portions as applied in Section 7.2.6. 255 Day 1 8 mg Day 4 Top 0.72 ±0.11 ±0.04 ±0.06 Bottom ±0.07 ±0.04 ±0.08 ±0.03 0.73 a ±0.06 ±0.03 a ±0.10 a 0.15 a 0.26 a a 0.22 a 0.27 a ±0.09 0.20 a 0.21 a 0.69 ±0.08 ±0.02 0.13 a 0.25 a 1.08 Day 1 Day 7 ±0.03 0.31 a,1 0.70 Day 7 Day 4 1.02 ±0.05 a 0.21 a 0.23 a ±0.07 ±0.2 a 0.23 a ±0.05 Figure A-5(a). Log10 CFU/g reduction of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in one-GR design bag with 8 mgClO2/day 1 Different script letters indicates statistically differences between means at  of 0.05 of different portion (top vs bottom areas) located horizontally in the same bag 256 Day 1 4 mg Day 4 Top ±0.08 ±0.04 0.64 ±0.03 a ±0.02 ±0.02 ±0.02 ±0.01 0.59 ±0.00 ±0.03 0.62 a ±0.04 ±0.07 a ±0.03 a 0.80 a 0.16 b 4 mg 0.58 a 0.12 a ±0.06 a 0.65 a 0.14 a 0.75 Bottom 0.79 a 0.16 ±0.03 Day 1 Day 7 0.15 a 0.59 Day 7 Day 4 0.81 ±0.04 a 0.62 a 0.12 a ±0.04 ±0.04 a 0.65 a ±0.06 Figure A-5(b). Log10 CFU/g reduction of E. coli O157:H7 of sample portions stored for 1, 4, and 7 days in two-GR design bag with 8 mgClO2/day 1 Different script letters indicates statistically differences between means at  of 0.05 of different portion (top vs bottom areas) located horizontally in the same bag 257 APPENDIX 10: Results on moisture content of shredded lettuce for packaging study The analytical procedure for determining moisture content (%) in E coli O157:H7 inoculated shredded lettuce was described below. The results were reported in Table A-9 and Table A-10. Apparatus:  Aluminum dish  Analytical balance  Vacuum chamber or Hot-air oven  Dessicator Procedure: 1. Weight around 5 g of chopped shredded lettuce (WF) into aluminum dish with known weight (WA). 2. Let dry in vacuum chamber (or hot-air oven). 3. Put sample in dessicator, until the temperature cool down to room temperature. 4. Weight dried sample and repeat step 2-4 until sample weight does not change more than ± 0.0005 g; record as WD. 5. Discard the sample. 258 Calculation: To obtain moisture content, MC (%): MC %   WF  WD  W Al  100 WF 259 1 Table A-9. Moisture content (%) of shredded lettuce samples, for packaging study on one-GR and two-GR designs, after storage for 0, 1, 4, and 7 days Treatment Moisture content (%) Day 0 Day 1 Day 4 a Control 95.16 ± 0.08 One-GR, 8.0 mgClO2/kg 95.67 ± 0.17 One-GR, 4.0 mgClO2/kg a,2 95.12 ± 0.06 95.77 ± 0.08 Two-GR, 8.0 mgClO2/kg 95.68 ± 0.19 Two-GR, 4.0 mgClO2/kg 95.60 ± 0.34 1 2 b b b b Day 7 b 96.09 ± 0.13 b 95.92 ± 0.20 b 95.93 ± 0.14 b,c 95.89 ± 0.25 b 95.41 ± 0.51 b 95.77 ± 0.15 b 95.79 ± 0.08 b 95.90 ± 0.26 c 96.00 ± 0.26 b 95.71 ± 0.60 GR = Gas reservoir Different superscript letters indicates statistically differences between means, between different storage days, at  of 0.05 260 1 Table A-10. Moisture content (%) of shredded lettuce samples, for packaging study on middle-GR , after storage for 0, 1, 4, and 7 days Treatment Moisture content (%) Day 0 Day 1 a 95.01 ± 0.06 Control middle-GR, 8.0 mgClO2/kg middle-GR, 4.0 mgClO2/kg 1 2 Day 4 a,2 94.99 ± 0.11 94.98 ± 0.21 95.08 ± 0.28 a a Day 7 a 95.35 ± 0.23 a 95.13 ± 0.32 a 95.07 ± 0.61 a 95.36 ± 0.91 a 95.35 ± 0.19 b 95.19 ± 0.13 GR = Gas reservoir Different superscript letters indicates statistically differences between means, between different storage days, at  of 0.05 261 APPENDIX 11: Statistical analysis result for packaging study Table A-11. Statistical analysis, for Study 1, on interaction between different bag design, ClO2 gas level, and storage time (at  of 0.05), as indicated by Log10 CFU/g reduction of E. coli O157:H7 The SAS System The GLM Procedure Class design conc block Class Level Information Levels Values 2 1 2 2 4 8 2 1 2 Source Model Error Corrected Total DF 7 64 71 Sum of Squares Mean Square 1.88267292 0.26895327 0.27941458 0.00436585 2.16208750 R-Square 0.870766 Coeff Var Root MSE 9.225075 0.066075 logred Mean 0.716250 Source design conc time block design*conc time*design time*conc time*design*conc DF 1 1 1 1 1 1 1 2 Mean Square 0.03076705 0.05326705 1.20650208 0.00006806 0.04351250 0.04625208 0.00460208 0.001542 Type III SS 0.03076705 0.05326705 1.20650208 0.00006806 0.04351250 0.04625208 0.00460208 0.003083 262 F Value 61.60 Pr > F <.0001 F Value 7.05 12.20 276.35 0.02 9.97 10.59 1.05 0.35 Pr > F 0.0100 0.0009 <.0001 0.9010 0.0024 0.0018 0.3084 0.7056 BIBLIOGRAPHY 263 Gavara, R. and Hernandez, R. J. (1993). "Consistency Test for Continuous Flow Permeability Experimental Data " Journal of Plastic Film and Sheeting 9(2): 126138. Greenberg, A. E., Clesceri, L. S. and Eaton, A. D. (1992). Standard methods for the examination of water and wastewater: 4500-ClO2 C. Amperometric method I. Standard methods for the examination of water and wastewater. Washington, D.C., American Public Health Association: 4-55 - 4-56. Haller, J. F. and Listek, S. S. (1948). "Determination of Chlorine Dioxide and Other Active Chlorine Compounds in Water." Analytical Chemistry 20(7): 639–642. Han, Y., Linton, R. H., Nielsen, S. S. and Nelson, P. E. (2002). "A comparison of methods for recovery of chlorine dioxide-injured Escherichia coli O157:H7 and Listeria monocytogenes." Food Microbiology 19(2-3): 201-210. ICA TriNova LLC (2006). Profiling gas release from ClO2 sachets. Standard Operation Procedure: Profiling gas release from ClO2 sachets, ICA Trinova. Netramai, S., Rubino, M., Auras, R. and Annous, B. A. (2009). "Mass Transfer Study of Chlorine Dioxide Gas through Polymeric Packaging Materials." Journal of Applied Polymer Science 114(5): 2929-2936. Netramai, S., Rubino, M., Auras, R., Annous, B. A. and Matthews, K. (2010). Surface study on antimicrobial effects of ClO2 gas on E. coli O157:H7 resided on lettuce leaf. IFT10 - IFT Annual Meeting & Food Expo, Chicago, IL, Institute of Food Technologists. Ölmez, H. and Temur, S. D. (2010). "Effects of different sanitizing treatments on biofilms and attachment of Escherichia coli and Listeria monocytogenes on green leaf lettuce " LWT - Food Science and Technology 43(6): 964-970. Post, M. A. and Moore, W. A. (1959). "The Determination of Chlorine Dioxide in Treated surface Waters." Analytical Chemistry 31(11): 1872-1874. Rodriguez, F., Cohen, C., Ober, C. K. and Archer, L. A. (2003). Principles of Polymer Systems. New York, Taylor & Francis. Selke, S. E. M., Culter, J. D. and Hernandez, R. J. (2004). Plastics Packaging: Properties, Processing, Applications, and Regulations. Cincinnati, OH, Hanser Gardner Publications, Inc. USEPA (1999) "4. Chlorine Dioxide." Retrieved September 11th, 2007, from http://www.epa.gov/ogwdw/mdbp/pdf/alter/chapt_4.pdf 264