IllllllllllllllllllllIll”\IHllllllllllllllllllllll 3 1293 017700 LIBRARY Michigan State Unlverslty This is to certify that the thesis entitled EFFECT OF Q-NONANONE VAPOR ON THE OXYGEN AND CARBON DIOXIDE PERMEABILITY OF TWO POLYMER FILMS presented by DeLynne Vail has been accepted towards fulfillment of the requirements for M . S . degree in Packagirg [MM L *— J Major professor Date J Iii-“l! 16L 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution PLACE IN REfURN Box to remove this checkout from your record. To AVOID FINES return on or before date due. MAY BE RECALLED with earlier due date if requested. DATE DUE I DATE DUE DATE DUE Filrioriiflmdi 1M GURU/WM“ EFFECT OF 2-NONANONE VAPOR ON THE OXYGEN AND CARBON DIOXIDE PERMEABILITY OF TWO POLYMER FILMS By DeLynne Vail A THESIS Submitted to Michigan State University In partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Packaging 1999 ABSTRACT EFFECT OF 2-NONAN ONE VAPOR ON THE OXYGEN AND CARBON DIOXIDE PERMEABILITY OF TWO POLYMER FILMS By DeLynne Vail In this study the oxygen and carbon dioxide permeability and diffusion coefficients of two polymer films were determined before and after exposure to 2-nonanone vapor. The two polymer films used were low density polyethylene, (LDPE), and a styrene — butadiene copolymer, called KRIO. The diffusion coefiicient of oxygen through LDPE film increased by an average of 3.2% when exposed to low concentrations, (197 — 256 ppm), of 2-nonanone and increased by an average of 19.4% when exposed to high concentrations, (346 — 443 ppm), of 2- nonanone. The steady state oxygen permeability of the LDPE film increased by an average of 2.4% and 2.9% after exposure to low and high concentrations of 2-nonanone respectively. For KRlO film, only the high concentration Of 2-nonanone had a significant effect on its diffusion coefficient for oxygen. Exposure to 2-nonanone increased KRl 0’s diffusion coefficient by an average of 11.8%. Low and high 2-nonanone concentrations increased the steady state oxygen permeability of KRIO 4.1% and 28.2% respectively. The LDPE films had an average of a 91.1% increase in carbon dioxide permeability. The KRIO films had an average increase in permeability of 52.2%. ACKNOWLEDGEMENTS I would like to thank the members Of my committee, my major professor Dr. Ruben Hernandez, Dr. Robert Tempelman, Dr. Jack Giacin, and Dr. Randy Beaudry for their help and guidance. In addition, I would like to thank my friends Laura Bix, Mark Newsham, and Donald Abbott for their encouragement and special knowledge they shared with me. Lastly, I would like to thank my best fi'iend Torben for his listening, understanding, kindness, and support along the way. iii TABLE OF CONTENTS List Of Tables ........................................................................ v List of Figures ........................................................................ vii Introduction ........................................................................... 1 Literature Review ..................................................................... 4 Materials and Methods .............................................................. 16 Results and Discussion ............................................................... 36 Conclusions ........................................................................... 57 Bibliography .......................................................................... 58 Appendices Appendix 1 .................................................................. 60 Appendix 2 .................................................................. 64 Appendix 3 .................................................................. 65 Appendix 4 ................................................................... 66 iv Table 5A: 5B: 5C: 6A: 6B: 6C: 7A: 73: 8A: 8B: LIST OF TABLES Summary Table: Oxygen Permeability of LDPE Films Summary Table: Oxygen Permeability of KRl 0 Films Summary Table: Carbon Dioxide Tests of LDPE Films Summary Table: Carbon Dioxide Tests of KRIO Films SAS Program Results for Low 2-nonanone Concentration Treated LDPE Films SAS Program Results for High 2-nonanone Concentration Treated LDPE Films Overall Comparison of D Values for LDPE SAS Program Results for Low 2-nonanone Concentration Treated KR10 Films SAS Program Results for High 2-nonanone Concentration Treated KRl 0 Films Overall Comparison Of D Values for KRIO Oxtran Steady State Oxygen Permeation Results for LDPE SAS Comparison p-Values for the Steady State Results in LDPE Oxtran Steady State Oxygen Permeation Results for KRl 0 SAS Comparison P-Values for the Steady State Results in KR] 0 Page 37 37 38 38 40 40 41 42 42 43 45 47 47 Table 9A: 9B: 10A: 10B: 11: 12: 13: 14: 15: Permatran Carbon Dioxide Steady State Permeation Results for LDPE Statistical Comparison of LDPE Permatran Values Permatran Carbon Dioxide Steady State Permeation Results for KRl 0 Statistical Comparison of KRIO Permatran Values Consistency Test Results for LDPE Tested on the Oxtran Consistency Test Results for KR] 0 Tested on the Oxtran Oxygen Test Results Summary Table Carbon Dioxide Test Results Summary Table Calibration Curve Values (Appendix 1) vi Page 50 50 52 52 54 55 56 56 61 LIST OF FIGURES Figure Page 1. Permeation Model 7 2. Diagram of an Isostatic Permeability Test Cell 13 3. Schematic Of the 2-nonanone Exposure System For Oxygen Permeability Testing 21 4. Schematic of the 2-nonanone Exposure System For Carbon Dioxide Permeability Testing 24 5. SAS Single Film Comparison Test 32 6. SAS Overall Comparison of Diffusion Coefficients 33 7. SAS Comparison of Steady State Oxygen Permeation Values 34 8. SAS Comparison of Steady State Carbon Dioxide Permeation Values 35 9. Relationship Between Concentration Of 2-Nonanone Exposure and Steady State Oxygen Flow in 1.25 mil LDPE 46 10. Relationship Between Concentration of 2-Nonanone Exposure and Steady State Oxygen Flow in 1.0 mil KRIO 48 11. Relationship Between Concentration of 2-Nonanone Exposure and Steady State Carbon Dioxide Flow in 1.25 mil LDPE 51 12. Relationship Between Concentration of 2-Nonanone Exposure and Steady State Carbon Dioxide Flow in 1.0 mil KRIO 53 13. (Appendix 1) 2-Nonanone Calibration Curve for the GC (1) 62 14. (Appendix 1) 2-Nonanone Calibration Curve for the GC (2) 63 vii INTRODUCTION In today’s time-crunched society consumers are looking for convenience in their food preparation. At the same time, they are also looking for the health benefits of what they eat. To this end, supermarkets have started selling minimally processed fruits and vegetables. Minimally processed is defined as fruits and vegetables that have been washed, peeled, cored and/or sliced. Some fruits and vegetables that have been researched for this type of application include apples, potatoes, carrots, onions, kiwi, and lettuce, (Gil et al., 1996). In order to sell minimally processed fruits or vegetables they have to be in some type Of package. Packaging protects the produce from mechanical damage. It also prevents contamination by insects, dirt and consumer handling and maintains quality of the produce by reducing the evaporation of moisture. The difference between fresh fruits and vegetables and other processed produce is that they remain living and respiring until they are cooked and/or consumed. Respiring fruits and vegetables consume oxygen and release carbon dioxide and heat energy. During this process they can also be subject to attack by microorganisms or ftmgi. Another hurdle in the packaging of fresh produce is condensation. Within a closed package of this type the relative humidity increases steadily as the fruit respires. This condensation makes a favorable environment for microorganism growth and decay if the temperature is warm enough. It has been shown that fungal spores that cause decay in fruit will germinate most rapidly when humidities reach 90% or higher and when the temperature is about 75° F (24° C) (Hardenburg, 1971). Some research has been done on spraying fruit with anti-ftmgal agents. Moyls et al. (1996) studied the effect of using acetic acid to prevent the growth of Botrytis cinerea on strawberries and grapes in MAP and found it to be successful. Leepipatanawit Observed that the substance 2-nonanone prevented growth of Penicillium expansum and Botrytis cinerea on apple slices, (Leepipattanawit et al., 1997). Vaughn et al. (1993) also reported that 2-nonanone inhibited the growth of the fungal species A. alternata, B. cinerea, and C. gloeosporioides on raspberries and strawberries. Since polymeric packaging materials are permeable they can interact with small molecules like gases, water and organic vapors. It has been shown that polymeric materials experience physio-chemical reactions when exposed to certain permeant molecules, (W ahid, 1996). The degree to which a polymer interacts with a permeant is dependent on the sorption or diffusion capability Of the permeant within the polymer and the resulting chemical reactions that may take place between the two. Sorption Of certain vapors and gases causes a swelling effect or morphological change in the polymer’s structure. This change can then have an effect on the polymer’s permeation properties. In modified atmosphere packaging it is important to know the permeation properties of the film being used. It is also important to know if the permeation properties will change over the shelf life Of the product. It is reasonable to ask that if an organic vapor such as 2-nonanone were being used on a product in MAP if it would interact with the polymer film being used. Studies have shown that the sorption of organic vapor can have a swelling effect on the polymer matrix. This can result in additional sorption of organic vapor molecules, (Hernandez and Giacin, 1998). If2- nonanone has a swelling effect on a polymer film this may alter the film’s permeability to oxygen and/or carbon dioxide. The packaging of fresh fruits and vegetables is determinate on specific amounts of oxygen and carbon dioxide getting through the polymer matrix. TOO little oxygen can send produce into anaerobic respiration which leads to fermentation and rot. A careful equilibrium Of oxygen and carbon dioxide needs to be kept in order for the fruit or vegetable to maintain freshness and withstand a decent shelf-life. There needs to be an accurate understanding of whether 2-nonanone affects certain MAP films. With this understanding it is possible to better predict the shelf life of MAP products. In this research project the following hypothesis is examined: The presence of 2- nonanone vapor will affect the oxygen and carbon dioxide permeability and diffusion values of two polymer films. The magnitude of the change will be quantified if the hypothesis is accepted. The goals Of this research were: 1. To develop a test apparatus that can expose polymer films to 2-nonanone vapor and subsequently test their oxygen and carbon dioxide permeability after this exposure. This will include modifications to the Oxtran 100 and Permatran CIV permeability testers. 2. To apply statistical programs to analyze the resultant permeability and diffusion values (after exposure to 2-nonanone vapor). LITERATURE REVIEW RESPIRATION OF FRESH PRODUCE Fruits and vegetables are different from other packaged food items in that they contain living tissue. And for this reason, they respire. The general equation of plant respiration is: GLUCOSE + 02 => H20 + C02 +ENERGY The plant takes in oxygen and together with its stores of glucose it creates water, carbon dioxide and energy in the form of heat. Normal air is a mixture of three main gases. There is approximately .03% carbon dioxide, 21% oxygen, and 78% nitrogen. Research has shown that produce exposed to reduced oxygen and elevated carbon dioxide levels has a delayed ripening period, reduced respiration rate, reduced ethylene production rate, delayed softening, and delayed compositional changes associated with ripening. (Kader, 1980) The speed at which a fruit ripens determines its shelf life. The faster it ripens the shorter its shelf life. If altering the gases surrounding the produce can slow the ripening process of produce it will thus extend its shelf life. MODIFIED ATMOSPHERE PACKAGING The most important methods to prolong shelf life of fruits and vegetables include refrigeration, maintenance Of high humidity, and harvesting at optimum maturity with minimal mechanical damage (Lee et al., 1995). Refrigeration slows down the respiration process of fruits and vegetables. High humidity keeps moisture in the cell tissue which keeps them alive. Harvesting at Optimum maturity means that the fruit is taken at a point where it will not ripen to the point Of decay before the consumer gets it but is ripe enough to have the taste qualities that the consumer prefers. Mechanical damage also speeds the process of fruit and vegetable ripening and decay. More recently there have been many developments in the use of modified atmosphere packaging to prolong the shelf life Of fi'uits and vegetables. Modified atmosphere packaging (MAP) is a passive system based on balancing produce respiration rate and package gas transmission rate, thus creating and maintaining the required CO2 and 02 levels under steady-state conditions in the package (Lee et al.,1996). A passive system does not introduce foreign gases to help prolong the shelf life of the commodity. Instead, this type of packaging relies on the interaction of respiration of the product and the permeability rates of the packaging film to produce the steady state levels Of CO2 and 02. The goal of MAP for fiesh produce is to prolong shelf life by reducing the rate of respiration Of the product. It has been determined that lowering oxygen concentrations below 8% will have a significant effect on slowing down fruit ripening, and the lower the oxygen concentration, the greater the effect. (Kader, et al., 1989). Each fruit has a minimum oxygen concentration tolerance and if it is exposed to less oxygen than this level it may go into anaerobic respiration. This will increase the accumulation of ethanol and acetaldehyde causing Off-flavors. Successful MAP will maintain near optimum O2 and CO2 levels to achieve the benefits of a modified atmosphere but will not exceed the limits of tolerance for minimum oxygen concentration which could lead to the above mentioned anaerobic respiration. MECHANISM OF MASS TRANSPORT Mass transfer of molecules into or out of a package can lead to further changes in the product, package, or both. Permeation through polymer membranes involves the transport of a gas or vapor through a homogeneous membrane. This membrane should be free of grSOss defects such as pores or cracks. There are three steps to permeation which are : 0 Absorption at the higher concentration surface of the polymer 0 Diffusion of the permeant through the polymer bulk phase 0 Desorption of the permeant on the low concentration side of the polymer Figure 1: The Permeation Model Absorption —-> O 00 O o (Orxygen O O 0 Carbon 0 O O Dioxide Molecules OO O O 50 O OO O O O Concentration Gradient “"‘Polymer —" Diffusion ——>~ l = polymer thickness »<———.—1—> Desorption O O O O O The process Of sorption is known to follow Henry’s law: C = 5* P (1) Where C is the concentration of the penetrant in the polymer, P = the partial pressure of the penetrant in the gas/vapor phase, and S = the solubility coeficient (Comyn, 1985) Films along with gases and vapors all have their own solubility coefficients. Research has shown that if a polymer and a gas or vapor have equal solubility coefficients then they will be mutually soluble (Hernandez and Giacin,l998). The diffusion Of the permeant through the polymer membrane can be described as a series of successive jumps where the permeant particle passes over the molecular barriers of the polymer matrix from one position to the next. This means that permeation is an “activated” process. The “jump” or particle diffusion depends on the rearrangement of the penetrant particles and the surrounding polymer. Molecular forces and movement activation energies must be overcome in order for the penchant particles to be able to move within the polymer molecular structure. Certain types of cohesive energies such as Van der Waal’s forces or hydrogen bonds must be overcome between molecules and chain segments in order to get them to break apart and let permeant particles pass. This process requires energy directed against the cohesive forces Of the polymer’s molecular structure. When enough energy is provided, the segments of the polymer structure begin to rearrange enough to allow for the passage of penetrant molecules. (Comyn, 1985). Difi'usion thus depends on the relative mobilities of the permeant particles to their surrounding polymer segments. Therefore it can be reasoned that anything which affects the arrangement of the polymer segments in their contact with permeant and the cohesive forces that hold the segments together will affect the diffusion of permeant particles. Factors that can influence the polymer/permeant relationship include: 1) 2) 3) Morphology of the polymer Studies have shown that the transport of gas or vapors through a polymer occurs only in the amorphous (non-crystalline) regions (Murray and Dorschner, 1983). The more regular the polymer chains the more easily they can pack close together. Close packing is ameasure of crystallinity. The more crystalline a polymer is, the more dificult it is for a permeant to pass through it. Polymer chain flexibility This is the ability of the polymer chains to move relative to one another. This is related to the glass transition temperature, (Tg), of the polymer above which chain mobility is high and below which chain mobility is extremely low. Below the T8 the size and frequency of voids between the polymer chains is fixed. Above the T8 the size and frequency of these voids is 3 functions of temperature. Usually the higher the temperature, the more mobility in the polymer chains. The more flexibility/mobility that the polymer has, the more free volume it has. Thus it is easier for permeant molecules to pass through it. Intermolecular forces Cohesive energy density, which produces strong intermolecular bonds and Van der Waal’s forces, is the measure of the strength of the bonds between molecules. The strength of the bonds between molecules in the polymer will affect the forces Of attraction between the polymer chains. This has a strong influence on the T8. The stronger the bonds the higher the T8 will be. 4) Concentration of permeant and permeant type Many sorbed penetrant molecules can act as plasticizing agents in the polymer structure, (Hernandez and Giacin, 1998). Plasticizing allows for the ease of movement between polymer chains and easier diffusion of penetrant molecules. Organic molecules are known to readily diffuse through polymer structures in which they solubilize easily, (Mohney et al., 1988). 5) Temperature Higher temperatures usually lead to higher molecular energies. Higher molecular energy increases the speed of movement of molecules. Higher temperature makes it much easier for penetrant molecules to diffuse through polymer molecules. F ick developed the first law of diffusion, (Equation 2). It states that the rate of transfer of a diffusing substance through a unit area is proportional to the concentration gradient measured normal the section, (Comyn, 1985). This law can only be applied to diffusion in the steady state. F = —D(C2 — C,)/1 (2) Where: F = rate of transfer of penetrant per tmit area at steady state D = difi‘usion coemcient (lengthz/time) I = thickness of the film Desorption takes place on the Opposite side of the polymer film as absorption. 10 This is where the permeant moves Off of the polymer surface into the environment. The amount Of desorption that takes place is dependent on the concentration of permeant at or near this surface. When Henry’s Law is obeyed, the steady state rate of diffusion can be expressed as a combination of both Henry and Fick’s Laws: 92:1,.32 17:31): * 1 I (3) P = S* D (4) Ap = pressure difference between top and bottom faces Of the polymer film PERMEABILITY OF ORGANIC VAPORS THROUGH POLYMER FILMS There have been many studies done on the permeation of organic vapors through polymer films. Franz studied the permeation of d-limonene across a biaxially oriented polypropylene film (Franz, 1993). Theodorou and Paik (1992) studied the permeation of linaool, citral, ethyl butyrate, and d-limonene in low density polyethylene film. These studies were conducted to determine the barrier properties of these films to certain organic vapors commonly produced by food products. The concern was over the possibility that permeation of organic vapors may lead to loss of aroma or flavor compounds. In these studies it was the organic vapor that was the permeant measured. Studies involving organic vapors as permeants have shown that most organic vapors interact and swell the polymer (Theodorou and Paik, 1992) as a plasticizer would. At high vapor concentrations this swelling was significant enough to have an effect on the permeability coefficient of the vapor studied. Hernandez and Giacin (1998) found ethyl acetate caused PET film to swell. This penetrant-polymer interaction was attributed ll to the fact that ethyl acetate and PET have a similar polarity in structure and similar solubility coefficients. PERMEABILITY MEASUREMENTS The permeability of a plastic film (P) is a measure of the rate at which a permeant can pass through the film in a unit of time, dependent on permeant partial pressure, film thickness and film surface area. ((1)0) P = (ammo) q = quantity of permeant I = film thickness a = film area I = time One way to measure the permeability of a film to oxygen is the isostatic method. The isostatic method developed by MOCON, (Oxtran 100, Modern Controls Inc. Elk River, MN), involves a film sample that is mounted between two chambers (see Figure 2). The permeant gas, (oxygen), enters one chamber, creating a higher concentration of the permeant in this chamber. Because Of the pressure difi‘erential the permeant diffuses through the film and thus enters the second chamber of lower concentration. From this second chamber a carrier gas transports the permeant molecules (oxygen) to a sensor which quantifies the amount permeated per unit Of time and film area. It is called an isostatic method because the total pressure on both sides of the film is constant and generally is kept at atmospheric pressure. 12 Figure 2: Diagram of an Isostatic Permeability Test Cell in the Modified Oxtran 100 Test Film Out 02 In ; O2 O2 , —+[____ ,___I—> 2-Nonanone 2-Nonanone + N2+02 TO Sensor and N2 In To calculate the diffusion coefficient of a permeant through a film using the isostatic method the equation is: (Gavara and Hernandez, 1993). F‘ (4)”): (”’2”) <6) —= — ex Fm J}; 4Dt “1.3.5 p 4Dt This expression applies to the unsteady state portion of the curve only. F, is the flow rate Of the penetrant at time (t) and F... is the flow rate of the permeant at steady state. ORGANIC VAPORS USED ON FRUITS TO PREVENT FUN GAL GROWTH Recent studies have shown that some organic vapors have a fungistatic effect on selected fruits. Song showed that hexanal vapor was effective at the prevention of 13 growth Of Penicillium expansum and Botrytis cinerea on apple slices, (Song, et al., 1996). In other research, a modified atmosphere package was used in combination with fumigation with low concentrations of acetic acid on grapes and strawberries. This was shown to prevent storage rot and increase the shelf life of these two commodities. This method helped to prevent the growth of Botrytis cinerea, (Moyls, et al., 1996). Anderson studied the antifungal activity among volatile C6 and C9 aliphatic aldehydes, ketones, and alcohols. In his research he found that they were effective in the prevention of growth OfAlternaria alternata. He concluded that the C9 aldehydes and ketones, including 2-nonanone, were the most potent in their antifungal activity, (Anderson et al., 1994). Vaughn studied fifteen natural volatiles released by raspberries and strawberries during ripening. Of these fifteen volatiles he found that benzaldehyde, l-hexanol, E-2- hexenal, and 2-nonanone inhibited the growth of three fimgal species, Alternaria alternata, Botrytis cinerea, and Coletotrichum gIoeosporioides (Vaughn et al.,1993). Most recently, Leepipattanawit et al. (1997) used a vapor generating system to expose apple slices to 2-nonanone vapor. Through this research it was determined that 2- nonanone was efl'ective at the prevention of Penicillium expansum and B. Cinerea growth on apple slices and potato dextrose agar. STUDIES OF ORGANTC VAPOR INTERACTION WITH POLYMER FILMS Numerous studies have been conducted to research the effects of organic vapors on the permeability of polymer films. Most of these studies however, have concentrated on permeation rates of these organic vapors through selected polymer films and how to 14 measure them, (Franz, 1993). Some studies have paired two or more organic vapors to see how the presence of the other vapor(s) may affect the permeation of the first organic vapor (Nielsen and Giacin, 1994). There seems to be a lack of research in the area of how organic vapors influence the permeation rates of CO2 and 02 through polymer films. CO2 and 02 are of primary importance in modified atmosphere packaging of fruits and vegetables. Ampolsak (1992), studied the effect of ethanol vapor on the oxygen permeability Of selected films. This was done because organic compounds such as ethanol are generated during the anaerobic phase of fruit and vegetable respiration. 15 MATERIALS AND METHODS A. Materials 1. Polymer Test Films a. Low Density Polyethylene (LDPE) H-[ CH2 ]..-H Dow Chemical Company, (Midland, Michigan), supplied the LDPE film used. Its thickness was 1.25 mil (0.003175 cm). The LDPE film used had a crystallinity in the range of 40 — 50% and its density is 0.912 g / cm3. LDPE is produced by the polymerization of ethylene gas. It is made up Of both short and long chain branches. It contains no hydrogen bonding elements and has a non-polar structure. Its T8 is -120°C. This film has good Oil resistance but it is a poor barrier to most gases. It is highly permeable to oxygen and carbon dioxide. It also sorbs organic vapors easily. b. KRlI) The other film used was KRIO (Phillips Chemical Company, Bartlesville, Oklahoma). It is a styrene-butadiene amorphous block copolymer. The thickness was 1.0 mil (0.00254 cm). The density ofKRlO is 1.01 g/cm3. Its I, is 62 °c. KR10 meets FDA specifications (CFR 177.1640) for use with food. It is characterized as a film that is highly permeable to both oxygen and carbon dioxide gases. l6 Penetrant a. 2-Nonanone The penetrant was 2-nonanone, (Aldrich Chemical Company, Saint Louis, MO). Also called methyl-heptyl-ketone. H3C(CH2)5COCH3_ It was stored at 4°C until used. The density of 2-nonanone is 0.832 g/mL, molecular weight = 142.24 g/mol, and boiling point = 195°C. 2-nonanone is a colorless liquid found in the attar of rose, clove Oil, passion flowers, sorghum, asparagus, tomato, corn, cheese, and beer. It has some bactericidal activity. It is an alarm pheromone in ants, hornets, and honeybees. It is moderately toxic by ingestion. b. Acetonitrile CH3CN Acetonitrile,(EM Science, Gibbstown, NJ), was the solvent used to dilute 2-nonanone for the gas chromatograph tests. Gases a. Carrier Gas Nitrogen dry grade gas containing 2% hydrogen was used for the oxygen permeability tester (Oxtran 100). Pure nitrogen gas (100%) was used for the carbon dioxide permeability tester (Permatran CIV). Both gases were supplied by AGA Gas, Inc. (Cleveland, OH). 17 Permeant Gas Oxygen supplied in the form of compressed air (02 partial pressure of 0.21 atm) (AGA Gas, Inc., Cleveland, OH), was used as the oxygen source for the Oxtran 100 tests. Carbon dioxide gas, (AGA Gas, Inc., Cleveland, OH), was used as the carbon dioxide source for the Permatran CIV tests. Equipment a. Oxtran 100 The Oxtran 100, (Modern Controls Incorporated, Minneapolis, Minnesota), was the instrument used to measure the oxygen permeability Of the test films. It uses the isostatic method to test oxygen permeance Of films or packages. It has a single film testing station. b. Oxygen Transmission Rate Datalogger Model DL200 Oxygen transmission was monitored using this datalogger, supplied by Modern Controls Incorporated, Minneapolis, Minnesota. Data for oxygen permeability was collected every two minutes. c. Permatran CIV Supplied by Modern Controls Incorporated, Minneapolis, Minnesota. It uses the Dynamic Accumulation method to test carbon dioxide permeance of films or packages. It has three film testing stations. (I. Chart Recorder Model L6512 The chart recorder, (Linseis), recorded the steady state permeation values of each film to carbon dioxide in ten-minute increments. 18 B. Methods 1. Setup Schematic for 2-nonanone Vapor Generation and Subsequent Oxygen Permeability Testing The apparatus used consisted Of an Oxtran 100 and a piping system to generate and control the vapor stream of 2-nonanone, and a datalogger, (Figure 3). Normally the Oxtran is connected to a test gas, (oxygen or air), and a carrier gas, (98% nitrogen and 2% oxygen). The system for this research project was modified in that the carrier gas line is split into three lines. One of these lines goes into the 2-nonanone washing bottle which contains pure (99%+) 2-nonanone liquid. The bubbler in this bottle creates 2- nonanone vapor Of a concentration Of approximately 2000 ppm. The required concentrations Of 2-nonanone were reached by blending it with a stream of carrier gas coming from the “mixing line” shown in Figure 3. The flow of gas and vapor streams was controlled by a series of needle valves. The flow of gas and vapor streams was monitored using flow meters. Exiting Ofl of the “mixing line” is another line of pure carrier gas. This line is used when running a regular oxygen permeability test, thus not exposing the film to any 2-nonanone. OXYGEN MEASUREMENTS The oxygen permeability of the film samples was determined in accordance with the ASTM Standard D3985-81, “Oxygen Gas Transmission Rate Through Plastic Film and Sheeting Using a Coulometric Sensor”. The Oxtran 100 Permeability Tester (Modern Controls, Inc., Elk River, MN), employs an isostatic method. The gas that permeates the film is conveyed to the sensor by a carrier gas. Each film sample was 19 tested three times. The first and second measurements were for oxygen permeability control, where the film sample was not exposed to 2-nonanone vapor. These tests were done in duplicate to ensure stability of the baseline before the film was exposed to the 2- nonanone. After each measurement was finished, the film was allowed to equilibrate in the test cell for approximately four hours before the next test. When the two non-exposed oxygen tests were completed the valves were adjusted on the system of piping to allow for the carrier gas to generate a 2-nonanone vapor stream and for the vapor to enter into the Oxtran carrier gas stream. The Oxtran was left in the “carrier purge” mode in order to expose both sides Of the film to the 2-nonanone/carrier gas stream. The LDPE films were exposed in this fashion for approximately 5 to 6 hours and the KRl 0 film were exposed for approximately 3 1/2 days. After the designated exposure time was reached, an oxygen permeability test was run on the film, while continuing exposure to 2-nonanone vapor on one side only. The output for these tests was collected by the DL 200 Datalogger. The datalogger converted the voltage response on the coulometric detector to an oxygen permeance response. Readings were recorded every 2 minutes and were reported in units of cc/m2*day. The results of this third test, after 2-nonanone exposure, were then compared statistically to the average of the results of the first two regular oxygen tests to determine if there was a significant difference in the “before 2-nonanone exposure” oxygen permeation values to the “afier 2-nonanone exposure” oxygen permeation values. We 20 were interested in the effect Of the first exposure therefore only one measurement was carried out on each film. Figure 3: Schematic of the 2-Nonanone Exposure System for Oxygen Permeability Testing W i Carrier ———-> 98% 2% H 2 -Nonanone Oxygen Washing Bottle Source Waste to Hood AIR Smplins [‘5 Port Oxtran 21 °/. N o. D I 3-Way Flow Connector N Valve B Flowmeter Datalogger Both of the test films, LDPE and KRIO, are known to be highly permeable to oxygen and thus an aluminum mask was used to reduce the exposed surface area of the film by a factor often. Compressed air was used as the oxygen test gas source to reduce the amount Of permeated oxygen conveyed to the sensor, as it contains only 21% oxygen. 21 CARBON DIOXIDE MEASURMENTS The setup for the carbon dioxide testing was very similar to that for the oxygen testing. (Figure 4). There were, however, two differences compared to the Oxtran 2- nonanone setup (Figure 3). The first was that the 2-nonanone vapor was generated using the test gas stream, (carbon dioxide), and not the carrier gas stream. The reason for this was that the carbon dioxide stream is in constant contact with the film samples whereas the carrier gas stream is only in contact with the films periodically during the actual testing period. The other difference was that the 2-nonanone vapor generating line was physically taken Off -line when not exposing the films to the vapor. The pure carbon dioxide line was hooked up straight from the source tank when nmning the regular carbon dioxide tests. The Permatran CIV system is similar to the Oxtran 100 in that as the molecules Of carbon dioxide permeate through the test film they are transported to a sensor by the carrier gas. The sensor on the Permatran is different fi'om the Oxtran in that it is an infrared sensor. The continuous flow method was employed to analyze the test films. This method is used for the evaluation of moderate to high transmitting films such as LDPE and KRIO. When the continuous flow method is used, the Permatran is acting as a comparitor in which the test films are being compared to a given reference film value. In order to do this the value of the reference film has to be determined. Running a dynamic accumulation test on the reference film provides a method for determining the permeability of the test film under control conditions. Once the reference film value has been determined, then the reference film is run along side the test films in the 22 continuous flow method and through a series Of calculations the values of the test films can be determined using the value of the reference film. The continuous flow test method was run on the films six times before 2-nonanone exposure and then six times after 2-nonanone exposure. Unlike the single station Oxtran 100, the Permatran had three stations for testing films. Once the 2-nonanone vapor generating system was interfaced to the Permatran CIV system the three test stations were being exposed simultaneously to the 2-nonanone vapor. After the designated exposure time the vapor was allowed to continue flowing while the last set of six tests were run. The output from the Permatran CIV system was monitored on a strip chart recorder, with lines denoting a certain amount of voltage created by the test samples. The voltage response was compared to that of the reference film. Through a series Of calculations involving the value of the reference film, (see Appendices 4 and 5), the CO2 permeation values of the test films were determined. The results were analyzed statistically to determine if the CO2 permeation values before 2-nonanone exposure were significantly different from the CO2 permeation values after 2-nonanone exposure. 23 Figure 4: Schematic of the 2-Nonanone Exposure System for Carbon Dioxide Permeability Testing CARBON H DIOXIDE p GAS H Sampling Exhaust Port 2 - Nonanone Washing Bottle H Permatran NITROGEN GAS Data Recorder 2-NONAN ONE CONCENTRATION DETERMINATION Two different concentrations of 2-nonanone were used during the oxygen permeability testing and one concentration was used for the carbon dioxide testing. Quantification of 2-nonanone concentration in the test chamber(s) was determined by a gas‘ chromatographic analysis using a flame ionization detector. The settings for the gas chromatograph are shown in Appendix 1. 24 CALIBRATION CURVE Standard concentrations were prepared by diluting a certain amount of known concentration stock solution of 2-nonanone and acetonitrile. In this way, several concentrations Of 2-nonanone in acetonitrile were prepared. A calibration curve was generated using a gas chromatograph. The quantity injected versus area unit response was plotted, (Appendix 1). The equation of the line describing the relationship between the quantity injected and detector response is y=20,644,615x, where y equals area unit response (AU), and x equals quantity Of 2-nonanone injected (x 10'6 grams). R2 = 1.00. At various times during the exposure period of the film to the 2-nonanone a sample of the vapor stream was taken at the sampling port using a 500 pl syringe. As shown in figures 3 and 4, the sampling port was located just before the entrance of the carrier gas line on the Oxtran and just before the carbon dioxide entrance on the Permatran. The contents of the syringe were then injected into the gas chromatograph and analyzed for area unit response. This response was then converted into a concentration amount using the calibration curve. DETERMINATION OF FILM EXPOSURE TME TOEONAN ONE VAPOR Each film needed to be exposed to the 2-nonanone vapor long enough for the 2- nonanone vapor to reach steady state permeation through the film before exposing it to oxygen. The equation used to estimate the time to reach steady state of a permeant through a film is: (Hernandez, 1996) 25 T=— (7) T = lagtime l = thickness of the film D = diffusion coefficient of the permeant through the film The diffusion coefficient of 2-nonanone through LDPE is known to be D= 3.1 x 10'14 mz/sec (Wahid, 1996). The thickness of LDPE film used in this study was 31.75 x 106 meter, (1.25 mil). Solving for T here we get 4.5 hours of exposure time to reach steady state permeation of 2-nonanone vapor through LDPE film. The diffusion coefficient of 2-nonanone though KR10 film is not known and therefore had to be estimated. The diffusion coefficients Of d-limonene and ethyl acetate through KR10 film (McDowell, 1997). 2-nonanone has a molar volume of 171 , d- limonene of 162, and ethyl acetate Of 98. Therefore 2-nonanone is likely to behave more like d-limonene than ethyl acetate when diffusing through the KR10 film and therefore have a similar diflirsion coemcient. When substituting the difi‘usion coefficient of ethyl acetate (3.3 x 10'14 mz/sec) (McDowell, 1997) into the above equation, the exposure time to reach steady state is about 3 hours. Ifone substitutes the diffusion coefiicient of d- limonene through KR10 film (3.0 x 10'15 m2/sec) (McDowell, 1997) into the above equation the exposure time to reach steady state is 30 hours. (The thickness of KR10 film used was 2.54 x 10'° meter or 1 mil). Because the molar volume of 2-nonanone is larger than that of d—limonene it is estimated that the time to reach steady state permeation will be longer. Therefore a safe estimate was determined to be 80 hours. 26 CONSISTENCY TEST The data from the Oxtran permeability tests was subjected to a consistency test developed by Gavara and Hernandez, (1993), specifically for continuous flow experimental data. When nmning an oxygen permeability test, the need arises to detect variations in the system’s parameters such as temperature and concentration variations and to determine if they have affected the consistency Of the data. The correctness of the permeability data will affect all future calculations regarding diffusion coefficients and/or steady state values. The consistency test involves determining the t 1/4, t 1/2, and t 3,4 values. These values correspond to the ‘/4 time it takes to reach steady state, '/2 the time it takes to reach steady state, and 3/4 the time to reach steady state, respectively. These values are used to determine values for K1 and K2. K; = (t 1/4 )/ (t 3/4). K2 = (t 1,4 )/ (t 1/2 ). The range for the accepted consistent experimental values of K are 0.42 5 K1 5 0.46 and .65 5 K2 5 0.69. Tables 11 and 12 show the results of the consistency tests for the Oxtran tests of LDPE and KR10. STATTSTTCAL ANALYSIS The statistical methods employed in this study were used to determine whether to accept or reject our hypothesis. Specifically, we wanted to know if there were significant differences between: 27 1) The oxygen diffusion coefficients (D) corresponding to zero 2-nonanone exposure, low concentration 2-nonanone exposure, and high concentration 2-nonanone exposure. 2) The steady state rate Of oxygen permeation values at zero 2-nonanone concentration versus low 2-nonanone concentration and high 2-nonanone concentration exposed films. 3) The carbon dioxide permeation coefficient values at zero 2-nonanone concentration versus the ones exposed to 2-nonanone vapor. All of the statistical analyses were carried out using the SAS computer software program using an input of all the data for the LDPE and KR10 films. (SAS/STAT User’s Guide. Version 6, 4th Edition, 1990, SAS Institute, Cary, NC). This program was not used to compare any data between LDPE and KR10. 1) Singular Film Comparison of Diffusion Coefficients Each film had a series of three oxygen tests run on it. The first two tests were control tests, where the film was not exposed to 2-nonanone vapor. The third test was nm on the film after it was exposed to the 2-nonanone vapor. The results of the two control tests were compared to the result of the test after 2-nonanone exposure. There were five replicates of LDPE films tested at low 2-nonanone concentration exposure, five replicates Of LDPE films tested at high 2-nonanone concentration exposure, five replicates Of KR10 films tested at low 2-nonanone exposure and five replicates Of KR10 films tested at high 2-nonanone concentration exposure. We refer to a “run” as an individual permeation experiment. Using the SAS program PROC NLH‘I (SAS, 1990) 28 the data, for one film at a time, was then fitted to a non-linear regression curve using the equation: (Gavara and Hernandez, 1993) HT '2 l” [—W) 8 F,” J; 4(D+6)t ”2,,“1’ 4(D+6)t H where: 6 = 0, if treatment = control 5 at 0, if treatment = 2-nonanone The parameter (D + 6 ) was estimated for each film. The diffusion coefficient vale, D, was the control, (at zero 2-nonanone concentration). The (6 ) was the difference between the control and the treated diffusion coefficient value. (D + 8) is the value of the diffusion Of oxygen in the presence of 2-nonanone. A 95% confidence interval was provided for both parameters. If the confidence interval for (5 ) did not contain zero then the estimated difference between the control D-value and the treated D-value was statistically significant, at a Type I error rate of 5%. 2) Comparison of All Diffusion Coefficients Each polymer film, LDPE and KR10, had ten sections of data. Five sections were used to compare low concentration of 2-nonanone vapor to control and five were used to compare high concentration of 2-nonanone vapor to control. This design further allowed an indirect comparison of low to high 2-nonanone concentration. HYPOTI-IESIS The data was labeled according to three treatments. Treatment , (trtl ), corresponded to no 2-nonanone exposure (regular oxygen test). Treatment 2, (trt 2), 29 corresponded to low concentration 2-nonanone exposure. Treatment 3, (trt 3), corresponded to high concentration 2-nonanone exposure. Using SAS PROC NLIN on all sets of data a non-linear regression analysis was used to estimate 3 D-values: those for 1) zero 2-nonanone exposure, 2) low 2-nonanone concentration exposure, and 3) high 2-nonanone concentration exposure. The effect of films was also accounted for in the analysis. The effect of films was also accounted for in the analysis. The 95% confidence intervals for each D-value are also computed. If any two confidence intervals do not overlap then the corresponding D-values are surely known to be significantly different (P<0.05). Nevertheless, two D-values may be statistically different even if their confidence intervals overlap. 3) Comparison of Steady State Values of Oxygen Permeation Data was entered and labeled according to one of the three treatments, film number and the corresponding steady state oxygen permeation value Observed. To determine if the steady state values of permeation differed between treatments an AN OVA test was run on SAS. Pair-wise t-tests were run to compare each of the three treatments’ steady state values to the others. The analyses used blocked on film. Ifthe p-value between a comparison of two treatments is less than 0.05 this means that the films were significantly difl‘erent at steady state. 4) Comparison of Steady State Values of Carbon Dioxide Permeation With the Permatran tests there were only two treatments. Treatment 1 was no 2- nonanone exposure and treatment 2 was high concentration 2-nonanone exposure. 30 First, all the data for one type of film (LDPE or KR10) is entered and labeled according to treatment 1 or treatment 2 and film section. An analysis of variance (ANOVA) based on the use of SAS PROC GLM was used, (SAS, 1990). Treatment differences were assessed by an analysis of variance by blocking on film section. Figures 5 - 8 are flowcharts Of how the above-mentioned statistical programs ran. 31 Figure 5: SAS Single Film Comparison Test Film 1 Data Film 2 Data Film 3 Data (non-exposed) (non-exposed) (Exposed to 2-Nonanone) t(t1me, seconds) vs. t(tirne, seconds) vs. t(time,seconds) vs IMO: permeation y.(O2 permeation yt(02 permeation at time t) at time t) at time t) G“ 0} Flux Value Get 02 Flux Value Get 02 Flux Value ”Yr/y“ =y‘/y“ =yI/yll Combine All Data control values versus 2-nonanone treated values i Non-Linear Regzession (Stochastic M_Odel) Fits data to the following equation and solves for D __Y_t.. _( 4 )[ [2 )1” [—7122] ________] Y." 11k ’ J? 4(D +5)“ “:3 Exp 4(D +5)! + em 7 0 ar 1 on-e Dv 2 ex sed 21m Least Squares Method D is the D-value for neatment I; (D+ 6) is the D-value for treatment 2 Ho: 5=0 ( mlvs.tl't2) OUTPUT D estimates, D confidence intervals, and sums of squares Ifthe confidence interval for 6 does not contain 0 the the D values are Statistically different between the two treatments ( at a = 0.05) Note: D is the diffusion coefficient of the non-exposed (control), f] is the effect of the jth film, eijk is the error pertaining to the kth time period on the jth film, and $in ~ N(0,02). 32 Figure 6: SAS Overall Comparison of Diffusion Coefficients Di is the diffusion coefficient for the ith treatment, 1] is the effect of the jth film and cm, is the error term pertaining to the ith treatment, jth film and kth time period. eijk ~ NllD (0,02). Trt 1 = no 2-nonanone exposure D = diffusion coefficient for tn 1 Trt 2 = low concentration 2-nonanone exposure D1 = diffusion coefficient for tn 2 Trt 3 = high concentration 2-nonanone exposure D2 = diffusion coefficient for trt 3 En_ter Data for Each Film Test x = time (seconds) vs. y = Flux calculate flux = permeation value / steady state value 1 Combine all Data Labeled by treatment 1, 2, or 3 Blocking on film i Non-Linear Regression Fits data to the following equation and solves for D L--_ 4 12]” [—n__l_22+] Y”yk-(‘/;t_)[4Dltl ”IZMCXP 4D ti +8,” 1 Compare Treatment 1, Treatment 2, and Treatment 3 D values Nonlinear Least Squares Method H01: D=D1,H 02: D=D2,H o3: D1=D2 l OUTPUT D estimates, D confidence intervals, degrees of freedom,and mean square error. Iftwo of the confidence intervals do not overlap then the difi'usion coeficients are surely known to be significantly different. 33 Figure 7: SAS Comparison of Steady State Oxygen Permeation Values Enter all data treatment film steady state value 1,2, or3 1-10 YSS & General Linear Models Procedge Analysis of Variance (AN OVA) Fischer’s Least Significant Difference Test Blocking on Film H01: ”11%, H02: 1111:1113, 1103;012:0113 OUTPUT F-values, means, and standard error, mean differences, P-values, and Coefficient of Determination (R2) 34 Figure 8: SAS Comparison of Steady State CO2 Permeation Values Enter all data Treatment 1 versus Treatment 2 Control 2-Nonanone treated General Linear Models Procedure Analysis of Variance (ANOVA) Ho: trt1=trt2 OUTPUT F-values, Mean values, and letter groupings 35 RESULTS AND DISCUSSION The Standard Calibration Curve. The relationship between the GC response and 2-nonanone concentration was found to be linear (See Appendix 1 ). The GC-response was linear with 2-nonanone concentration according to the following equation 1’ = 20644615 X where X is the quantity of 2-nonanone injected ( ug ) and Y is the GC response in area units. The coefficient of determination (r2) was 0.995. The original hypothesis of this research project was whether the presence of 2-nonanone vapor afiected the oxygen and carbon dioxide permeability and diffusion values of the two films studied. The statistical analysis of the data was done using four different SAS programs. The first SAS program, called the Single Film Comparison Test, was used to compare the diffusion coemcients (D) of oxygen an a film by film basis. This test takes the data through non-linear regression and a least squares method to determine the D values and whether they are statistically different between the non-exposed and exposed film. Each film sample had three tests run on it that included two oxygen control tests and one test after the exposure to 2-nonanone vapor. The two control tests were denoted as treatment 1 while the third test was denoted treatment 2. Tables 5A, SB, 6A, and 6B show the results of this analysis. 36 Tables 1 and 2 summarize the concentration of 2-nonanone at which each film sample was exposed to before testing for oxygen permeability and the effect on oxygen permeability that was measured after 2-nonanone exposure. ‘ Table 1: Oxygen Permeability of LDPE Films Concentration of Run Number 2-nonanone Exposure Change in Permeability (ppm) ( ”/01 l 240 0.8 2 256 4.2 3 197 2.4 4 209 4.1 5 220 0.7 6 443 4.1 7 381 0.6 8 386 -0.3 9 346 6.5 10 358 3.5 Table 2: Oxygen Permeability of KRlO Films Concentration of Run Number 2-nonanone Exposure Change in Permeability (Ppm) ( %) l 283 5.0 2 317 5.6 3 279 3.9 4 335 1.1 5 282 5.1 6 322 28.3 7 337 29.2 8 300 27.5 9 433 22.7 10 518 33.1 37 Tables 3 and 4 summarize the concentration of 2-nonanone at which each film sample was exposed to before testing for carbon dioxide permeability and the effect permeability that was measured after 2-nonanone exposure. Table 3: Carbon Dioxide Permeability of LDPE Films Concentration of Run # 2-nonanone Exposure Change in Permeability (ppm) ( °/o ) 1 351 78.7 2 351 110.8 3 351 93.5 4 333 59.0 5 333 82.2 6 333 100.0 7 454 77.2 8 454 73.8 9 454 144.4 Table 4: Carbon Dioxide Permeability of KR10 Films Concentration of Run # 2-nonanone Exposure Change in Permeability (ppm) L%) 1 370 51.0 2 370 44.3 3 370 34.7 4 349 28.8 5 349 29.4 6 349 31.1 7 446 78.8 8 446 63.3 9 446 108.0 38 Tables 5A and 6A show the results for the low concentration 2-nonanone exposure to the LDPE and KR10 films respectively. Listed as D in these tables are the difiusion coefficient values for the control oxygen tests that were conducted. The D + 8 values represent the diffusion coefiicients of the same films after exposure to 2-nonanone. The last column denotes whether or not H0 was rejected. If the answer is YES this means that we reject the null hypothesis (P<0.05) and therefore the diffusion coefficient values were significantly different. 8 is the incremental difference in diffusion. The null hypothesis is Ho: 8 = 0 meaning that the incremental difference in diffusion is zero after exposure to 2-nonanone. The confidence intervals given in this program are for 5. Therefore if the confidence interval does not include zero then 5 does not equal zero and therefore Ho must be rejected. The tables 5A and 5B show the original and after exposure D-values and the difl‘erence between the two for low and high concentration 2-nonanone exposure respectively. The values are shown for each run. Table 5C gives an overall comparison, combining all the data to show the difference between the D-values of low and high concentration exposure. All values are for LDPE film samples. 39 Table 5A: Nonlinear Regression Estimates of D-Coefficients for Low 2NN Concentration (A) Exposure of LDPE Films All values are in units of mZ/sec RUN D n+6 5 H. W. 13 4.91 E-13 5.21 E13 2.96 E-l4 YES 23 5.10 E-13 5.28 E-13 1.79 E-l4 YES 33 5.10 E-13 5.22 E—l3 1.19 E-14 YES 4a 4.97 E-l3 5.02 E-l3 5.16 E-15 NO 5a 5.11 E-l3 5.28 E-l3 1.73 E-14 YES Table 5B: Nonlinear Regression Estimates of D-Coefficients for High 2NN Concentration (B) Exposure of LDPE Films All values are in units of m2/sec RUN D D + 5 0 Required? lb 5.52 E-13 5.70 E-13 1.77 E-14 YES 2b 4.95 E-13 5.91 E-13 9.63 E-l4 YES 3b 4.94 E-l3 5.05 E-l3 1.12 E-14 YES 4b 4.53 E-13 4.64 E-l3 1.10 E—l4 YES 5b 4.61 E-l3 4.74 E-l3 1.30 E-l4 YES The second analysis, “Overall Comparison of Diffusion Coefficients”, was used to average all the diffusion coefiicients at a given concentration of 2-nonanone exposure. The D value for the control tests was called D1. The D value for the low 2-nonanone concentration tests was D2, and the D value for the high concentration tests was D3. Nonlinear regression analyses were used to estimate D1, D2, and D3. Table 5C for LDPE shows that none of the confidence intervals for D1, D2, or D3 overlaps meaning that all difl‘usion coemcients differed from each other. In other words, both the low concentration 2-Nonaone and the high concentration 2-nonanone exposure of the films had an effect on the D value of LDPE, the latter having the stronger effect. Table 6C for KR10 shows that the D1 and D2 confidence intervals overlap with each other but neither D1 nor D2’s confidence intervals overlap with that of D3. This means that the low 40 concentration 2-nonanone exposed KR10 film did not appear to have a significantly difierent D value than the control and the low concentration 2-nonanone treated film. Table 5C: Overall Comparison of D Values for LDPE All values are in writs of mz/sec Parameter Estimate 95% Confidence Interval Lower Upper D 4.96 E-13 4.94 E-13 4.98 E-13 D1 5.12 E-l3 5.07 E-13 5.16 E-l3 D2 5.92 E-13 5.86 E-13 5.98 E-l3 D = No 2-NN exposure D1 = Low Cone. 2-NN exposure D2 = High Cone. 2-NN exposure The tables 6A and 6B Show the original and after exposure D-values and the difference between the two for low and high concentration 2-nonanone exposure respectively. The values are shown for each run. Table 6C gives an overall comparison, combining all the data to show the difi‘erence between the D-values of low and high concentration exposure. All values are for KR10 film samples. 41 Table 6A: Nonlinear Regression Estimates of D—Coefficients for Low 2NN Concentration (A) Exposure of KR10 Films All values are in units of m2/sec RUN D D + 8 8 H. Rejected? 13 2.84 E-13 3.19 E-l3 3.45 E-14 YES 2a 3.08 E-13 2.90 E-l3 -l.77 E-l4 YES 33 2.37 E-13 3.08 E—13 7.14 E-l4 NO 4a 2.34 E-l3 2.38 E-13 4.09 E-15 NO 53 2.40 E-l3 1.70 E-13 -7.00 E-14 NO Table 6B: Nonlinear Regression Estimates of D-Coefficients for High 2NN Concentration (B) Exposure of KR10 Films All values are in units of mZ/sec RUN D D + 5 5 Rejeliied? 1b 2.36 E-13 2.55 E-l3 1.91 E-l4 YES 21> 2.65 E-13 2.91 E-13 2.62 E-14 YES 3b 2.50 E13 3.31 E-13 8.05 E-14 YES 4b 3.12 E-13 3.11 E13 -8.76 E-l6 N0 51: 2.76 E-l3 3.08 E13 3.17 E-14 YES As seen in tables 5A, for LDPE, four out of five film runs, (runs 1a,2a,3a, and 58), at low 2-nonanone concentrations had different D values after exposure. Table 6A, KR10, shows that three out of five film runs, (3b, 4b, and 5b), showed no significant difference in their before and after exposure D values. Tables SB and 6B are set up in a similar manner to that of 5A and 6A. The difference is that the 2-nonanone concentration that the films were exposed to was higher. Again, D] was the D value of the control films and D1+ 8 was the D value of the films after exposure to 2-nonanone. Tables 5B and 6B show that both the LDPE and KR10 films had all of their D values increase significantly as a consequence of the exposure to the higher concentration of 2-nonanone vapor. 42 Table 6C: Overall Comparison of D Values for KR10 All values are in units of m2/sec Parameter Estimate 95% Confidence Interval Lower Upper D1 2.62 E-13 2.60 E-13 2.65 E-13 D2 2.66 E-l3 2.60 E-13 2.71 E-13 D3 2.93 E13 2.88 E—13 2.99 E-13 D = No 2-NN exposure D1 = Low Cone. 2-NN exposure D2 = High Conc. 2-NN exposure The third analysis, “Comparison of Oxtran Steady State Values”, was an analysis of variance (ANOVA) used to determine if the steady state oxygen permeation values of the films tested changed significantly after exposure to 2-nonanone. The control, (non- exposed), steady state values were compared to the 2-nonanone exposed steady state values of each film tested. The values are all shown in tables 7A and 8A with their corresponding percent changes in permeance, along with the concentration of 2-nonanone that each film was exposed to. Table 7A shows that the LDPE film samples did not show as large of a change in permeance compared to the KR10 film results in table 8A. The percent change in permeance for the LDPE films ranged from —0.3% up to 6.5% whereas the percent change in permeance for the KR10 films ran from 1.1% up to 33.1%. Treatment 1 corresponds to the steady state values for the non-exposed film tests, treatment 2 corresponds to the steady state values after exposure to low concentration 2- nonanone vapor, and treatment 3 corresponds to the steady state values of the films treated with high concentration 2-nonanone vapor. Tables 7B and 88 use a grid design to compare treatments with their p-values calculated using the SAS program. As seen in Table 7B, treatments 1 and 2 are significantly difl‘erent from one another as denoted by a P-value less than 0.05. A p—value of less than 0.05 means that you can say with 95% confidence that the two treatments compared are difl‘erent. 43 Treatments 1 and 3 are also significantly different from one another because their comparison P-value is 0.015. In contrast, treatments 2 and 3 have a corresponding p- value of 0.8008 that denotes a non-significant difference between treatments. This basically means that the steady state oxygen permeation values did change significantly when comparing non-exposure to either low or high concentration exposure Of 2- nonanone. But the difference between the low concentration exposure effect and the high concentration exposure effect was not statistically significant. In the table 8B, for KR10, it is shown that all the comparisons Of treatments give a significant p—value meaning that they are all significantly different from one another. The KR10 film steady state oxygen permeation was affected significantly by the exposure to 2-nonanone vapor. This is demonstrated by the fact that the p-values for treatment 1 versus 2 is below 0.05 and the p-value of the comparison between treatment 1 and 3 was also below 0.05. In addition, the p-value of the comparison between low concentration exposure (trt2) and high concentration exposure (trt 3) is significant. This means that the change in concentration of exposure had a significant efi'ect on the change in oxygen permeation. 44 Table 7A: Oxtran Steady State Oxygen Permeation Results for LDPE *Appendix 4 shows the calculations used to determine the quantity of 2-nonanone injected Steady State Steady State Percent 2-nonanone Quantity 01' Run Flow Flow Change in Concentration isngiiijfn“; Number N on-Egrposed Exppsed Flow (ppm) (p g) cc/m Oday cc/m Oday (%) (vol vapor-l vol air) 1 1029 1037 0.8 240 100 2 1118 1165 4.2 256 106 3 1134 1161 2.4 197 82 4 1096 1141 4.1 209 87 5 1038 105 0.7 220 92 6 1085 112 4.1 443 184 7 1105 1112 0.6 381 158 8 1098 1094 -0.3 386 161 9 1060 1129 6.5 346 144 10 1100 1139 3.5 358 149 Table 7B is a grid which compares the p-values of ANOVA tests run on the steady state oxygen flow rates between treatments. Table 7B: ANOVA p-Values for Steady State Results in LDPE Treatment 1 2 3 1 -- 0.0318" 0.0150" 2 0.0318" -- 0.8008 3 0.0150" 0.8008 -- "Values below 0.05 indicate a statistically significant difl‘erence between treatments 45 Figure 9: Relationship Between Concentration of 2-Nonanone Exposure and Steady State Oxygen Flowin 1.25 mil LDPE 1180 - Steady State Oxygen Flow (cc/Iii day) A —l O O 1020 t l l i T 1 150 200 250 300 350 400 Concentration of 2-Nonanone (vol/vol) 46 Table 8A shows steady state oxygen flow rates for KR10 before and after exposure to varied concentrations of 2-nonanone and the percent change in steady state oxygen flow after exposure. Table 8A: Oxtran Steady State Oxygen Permeation Results for KR10 Steady State Steady State Percent 2-nonanone Quantity 0f Run Flow Flow Change in Concentration 0:“3‘Izg’3ct“; Number N on-Exposed Exppsed Flow (PPm) (p g) cc/m Oday cc/m Oday (%) (vol vapor/ vol air) 1 1347 1414 5.0 283 118 2 2662 2811 5.6 317 132 3 1478 1535 3.9 279 116 4 1416 1432 1.1 335 139 5 1454 1528 5.1 282 117 6 1907 2447 28.3 322 134 7 2144 2771 29.2 337 140 8 1391 1774 27.5 300 125 9 1560 1914 22.7 433 180 10 1910 2542 33.1 518 215 Table 8B is a grid which compares the p-values of ANOVA tests run on the steady state oxygen flow rates between treatments. Table 8B: SAS Comparison P-Values for Steady State Results in KR10 Treatment 1 2 3 1 -- 0.0410* 0.0001* 2 0.0410* -- 0.0001 "‘ 3 0.0001* 0.0001“ -- *Values below 0.05 indicate a statistically significant difi‘erence between treatments 47 Figure 10: Relationship Between Concentration of 2-Nonanone Exposure and Steady State Oxygen Flowin 1.0 mil KR10 3000 3 L N 01 O O o 2000 1500 ” , 1000 Steady State Oxygen Flow (cc/Iii day) 500 0 T: I 1 150 250 350 450 Concentration of 2-Nonanone (vol/vol) 48 The fourth analysis, “SAS Comparison Of Steady State Permeation Values”, was used to compare the steady state permeation of carbon dioxide values of non-exposed films to the permeation values after these films were exposed to a given concentration of 2-nonanone vapor. The difference between this SAS program and the program for oxygen steady state values is that this program is only comparing two treatments, non- exposed versus exposed. Tables 9A and 10A show the actual steady state values before and after exposure to 2-nonanone for LDPE and KR10, respectively, along with their corresponding percent changes in permeance. These tables also Show the concentration at which these tests were run. Table 9A and 10A Show that both the LDPE and KR10 films had a relatively large percent change in carbon dioxide permeance after exposure to 2-nonanone. Tables 9B and 10B show the results Of the SAS program analysis for the Permatran results. This was again an analysis of variance. The program just denotes by a letter grouping next to each mean value whether the means are statistically considered the same or difl‘erent. As shown in both tables 9B and 10B the letter grouping for both means, non-exposed and 2- nonanone exposed, are different for both films. This means that the mean carbon dioxide permeation values were significantly affected by 2-nonanone exposure for both the LDPE and KR10 films studied. 49 Table 9A: Permatran Carbon Dioxide Steady State Results for LDPE Steady State Steady State Percent 2-nonanone Quantity of Flow Rate Flow Rate Change in Concentration 2-nonanone in Run N on-Exposed Exposed Flow (ppm) 0'5 ml Injected Number cc lmz-day cc lmz-day (%) (1‘ g) 1 5020 8969 78.7 351 146 2 5217 10998 110.8 351 146 3 5217 10096 93.5 351 146 4 5231 8316 59.0 333 139 5 4829 8799 82.2 333 139 6 4628 9255 100.0 333 139 7 5043 8936 77.2 454 189 8 4906 8528 73.8 454 189 9 4770 1 1662 144.4 454 189 Table 9B: Statistical Comparison of LDPE Permatran Values Mean values with a different letter grouping are si 'ficantly different Mean Steady State Flow Rate Treatment Letter Grouping (cc I “‘2 ' d”) A 4985 Non-Exposed B 9506 Exposed 50 Figure 1]: Relations hip Between Concentration of 2-Nonanone Exposure and Steady State Carbon Dioxide Howin 1.25 mil LDPE 12000 - 11500 0 A A o _u 01 O O O O O L 7?? ~31 8 E E :1 o a; 10000 § 9500 g ' . “E 9000 i x o g 0 8500 g *5 8000 w t t t t 250 300 350 400 450 500 Concentration of 2-Nonanone (vol/vol) 51 Table 10A: Permatran Carbon Dioxide Steady State Permeation Results for KR10 Steady State Steady State Flow Rate Flow Rate Percent Quantity 01' Run Non-Exposed Exposed Change in 2-nonanone 2'9““0“ 1“ Number cc lm2-day cc /m2-day Flow Concentration 0'5 ml Injected (%) (ppm) (“8’ 1 3823 5774 51.0 370 154 2 4084 5895 44.3 370 154 3 3883 3923 34.7 370 154 4 3360 4326 28.8 349 145 5 3219 4165 29.4 349 145 6 3622 4748 31.1 349 145 7 4752 8496 78.8 446 186 8 4320 7056 63 .3 446 186 9 3600 7488 108.0 446 186 Table 10B: Statistical Comparison of KR10 Permatran Values Means with a different letter grouping are significantly different Letter Grouping Mean Steady State Flow Rate Treatment (cc I In2 - day) A 3 85 l Non-Exposed B 5909 Exposed 52 Figure 12: Relations hip Between Concentration of 2-Nonanone Exposure and Steady State Carbon Dioxide Flowin 1.0 mil KR10 8500 ‘ o 8000 7500 0 7000 . ’ 6500 6000 00 5500 5000 4500 Steady State Oxygen Flow (cc/niday) .0 4000 I o 3500 l l T i 250 300 350 400 450 Concentration of 2-Nonanone (vol/vol) 53 500 Tables 11 and 12 Show the results of the consistency tests run on the data from the Oxtran tests. The study by Gavara and Hernandez (1993) states that the resulting K values have to be within certain limits in order to call the data consistent. The values for K] must be within the range of 0.42 and 0.46 and the values for K2 must be within the range of 0.65 and 0.69. The K] and K2 values reflect that the Oxtran tests were within range for consistency. Table 11: Consistency Test Results for LDPE Tested on the Oxtran Test Number K1 K2 54 \ Table 12: Consistency Test Results for KR10 Tested on the Oxtran Test Number K1 K2 1 0.50 0.70 2 0.49 0.70 3 0.47 0.69 4 0.44 0.67 5 0.46 0.68 6 0.46 0.68 7 0.60 0.85 8 0.47 0.69 9 0.47 0.69 10 0.46 0.68 l 1 0.46 0.68 12 0.43 0.66 13 0.42 0.65 14 0.43 0.65 15 0.43 0.65 16 0.44 0.66 17 0.44 0.66 18 0.43 0.65 19 0.43 0.66 20 0.44 0.66 21 0.47 0.68 22 0.47 0.69 23 0.44 0.67 24 0.48 0.77 25 0.45 0.68 26 0.45 0.67 27 0.44 0.69 28 0.44 0.67 29 0.45 0.67 30 0.41 0.76 55 SUMMARY TABLES Tables 13 and 14 are to be read in columns. The first column describes which type of exposure to 2-nonanone. The following colurrms list, by letter designation, if the diffusion, permeation, or flow coefficients were significantly different from one another. A different letter designation means statistically significant difference. Percent change is that between the given value and the non-exposed value. Table 13: Oxygen Test Results Summary Table LDPE KR10 TREATMENT Diffusion Permeation Diffusion Permeation Coefficient Coefficient Coefficient Coefficient l A A A A (non-exposed) 2 B B A B (low concentration 3.2% 2.4%, 1.50/0 4.1%, exposure) . 3 C B B C (lush “mama“ 19.4% 2.9% 11.8% 28.2% exposure) Table 14: Carbon Dioxide Test Results Summary Table TREATMENT FILM LDPE KR10 Non-Exposed A A Exposed B (91.1%) B (52.2%) 56 1) 2) 3) 4) CONCLUSIONS Two testing systems were successfully developed to expose polymer films to 2- nonanone vapor and subsequently test their oxygen and carbon dioxide permeability. Four different analyses were designed in SAS. One can statistically analyze the diffusion coefficients of oxygen through single films before and after exposure to 2- nonanone. Another analyzes the diffusion coefficients of a group of films with three different treatments. The last two compare the steady state permeation rates of control versus treated films for both oxygen and carbon dioxide. Table 13 shows that the low concentration exposure of 2-nonanone to LDPE film increased the LDPE’S diffusion coefficient by 3.2% and also increased it steady state oxygen permeation by 2.4%. Exposure of LDPE film to high concentration 2- nonanone vapor increased its diffusion coefficient by 19.4%. It also increased its steady state oxygen permeation by 2.9%. In addition, Table 13 shows the results of KRl 0’s exposure to 2-nonanone. Low concentration exposure caused a 1.5% increase in its diffusion coefiicient and a 4.1% increase in its steady state oxygen permeation. Exposure to high concentration 2—nonanone caused an 11.8% increase in the diffusion coefficient and a 28.2% increase in its steady state oxygen permeation. Table 14 shows that after exposing the LDPE film to 2-nonanone its steady state carbon dioxide difi‘usion increased by 91.1%. Table 14 also shows that after exposing KR10 to 2-nonanone vapor the steady state rate of carbon dioxide increased by 52.2%. 57 BIBLIOGRAPHY Ampolsak, S. 1992. Effect of Ethanol Vapor on the Oxygen Permeability of Packaging Polymer Films. MS. Thesis. Michigan State University, E. Lansing, MI. Anderson,R.A., TR. Hamilton-Kemp, D.F. Hildenbrand, C.T. McCracken, R.W. Collins, and PD. Fleming. 1994. Structure-Antifungal Activity Relationships Among Volatile C5 and C9 Aliphatic Aldehydes, Ketones, and Alcohols. J. Agric. Food Chem. 42: 1 563- 1 568. Comyn, J. 1985. Introduction to Polymer Permeability and the Mathematics of Diffusion. Polymer Permeability. Elsevier Applied Science Publishers, London and New York. pp 1 — 6. Franz, R. 1993. Permeation Of Volatile Organic Compounds across Polymer Films- Partl: Development of a Sensitive Test Method Suitable for High-Barrier Packaging Films at Very Low Permeant Vapor Pressures. Packaging Tech. and Science. 6: 91-102. Gavara, R. and Hernandez, R]. 1993. Consistency Test for Continuous Flow Permeability Experimental Data. J. Plastic Film and Sheeting. 9: 126-138. Gil, M.I., F. Artes, and F .A. Tomas-Barberan. 1996. Minimal Processing and Modified Atmosphere Packaging Effects on Pigmentation of Pomegranate Seeds. J. Food Science. 61: 161. Handbook of Plastics, Elastomers, and Composites. Harper-Editor. Chapter 8. Plastics in Packaging. Hernandez, R]. 1996. Hardenburg, RE. 1971. Efi‘ect of In-Package Environment on Keeping Quality of Fruits and Vegetables. HortScience. 6: 198 - 199. Hernandez, RI. and JR. Giacin. 1998. Factors Afiecting Permeation, Sorption, and Migration Processes in Package-Product Systems. Food Storage Stability. CRC Press LLC. pp 290-300. Kader, AA. 1980. Prevention of Ripening in Fruits by Use of Controlled Atmospheres. Food Technology. March 1980: 51-53. Kader, A., D. Zagory, and EL. Kerbel. 1989. Modified Atmosphere Packaging of Fruits and Vegetables. Critical Reviews in Food Science and Nutrition. 28: 1 - 22. Lee, L., J. Arul, R. Lencki, and F. Castaigne. 1995. A Review on Modified Atmosphere Packaging and Preservation of Fresh Fruits and Vegetables: Physiological Basis and Practical Aspects — Part 1. Packaging Technology and Science. 8: 315 — 325. 58 Lee, L., J. Arul, R. Lencki, and F. Castaigne. 1996. A Review on Modified Atmosphere Packaging and Preservation of Fresh Fruits and Vegetables: Physiological Basis and Practical Aspects — Part H. Packaging Technology and Science. 9: 1 — 12. Leepipattanawit, R., R.M. Beaudry, and R]. Hernandez. 1997. Control Of Decay in Modified-atmosphere Packages of Sliced Apples Using 2-nonanone Vapor. J. Food Science. 62: 1043 — 1046. McDowell, D. 1997. Characterization of the Barrier Properties of Styrene Butadiene Copolymer (KR10) Film. MS. Thesis. Michigan State University. E. Lansing, MI. Mohney, S.M., Hernandez, R.J., Giacin, J .R., Harte, BR, and Miltz, J. 1988. Permeability and Solubility Od d-Limonene Vapor in Cereal Package Liners. J. Food Science. 53: 253-257. Moyls, A.L., P.L. Sholberg, and AP. Gaunce. 1996. Modified-atmosphere Packaging of Grapes and Strawberries F umigated with Acetic Acid. HortScience. 31: 414 — 415. Murray, L]. and R.W. Dorschner. 1983. Permeation Speeds Tests, Aids Choice of Exact Material. Package Engineering. 28 (3): 76-84. Nielson, T.J. and J .R. Giacin. 1994. The Sorption of Limonene/Ethyl Acetate Binary Vapour Mixtures by a Biaxially Oriented Polypropylene Film. Packaging Tech. And Science. 7: 247-258. Song, J ., R. Leepipattanawit, W. Deng, and RM. Beaudry. 1996. Hexanal Vapor Is a Natural, Metabolizable F ungicide: Inhibition of Fungal Activity and Enhancement of Aroma Biosynthesis in Apple Slices. J. Amer.Soc. Hort. Sci. 121(5):937-942. Theodorou, E. and J .S. Paik. 1991. Effect of Organic Vapor Interaction on permeation rate Through Polymer Films. Packaging Tech. and Science. 5: 21-25. Vaughn, S.F., G.F. Spencer, and BS. Shasha. 1993. Volatile Compounds from Raspberry and Strawberry Fruit Inhibit Postharvest Decay Fungi. J. Food Science. 58: 793 - 795. Wahid. M.1996. Permeation of 2-nonanone Vapor Through LLDPE Affinity Films as Apply to Modified Atmosphere Packaging. MS. Thesis. Michigan State University, E. Lansing, MI. Wiley Encyclopedia of Packaging Technology, Edited by M. Bakker and D. Eckroth, Copyright 1986, John Wiley and Sons, Inc. 1997. 59 Appendix 1 2 - Nonanone Calibration Curve for Gas Chromatography Instrument: Hewlett Packard 5890 Gas Chromatograph (GC) Column - SPB — 5 (non-polar column) Conditions: Oven Temperature : 100 °C Initial Temperature: 60 °C Initial Time: 1 minute Rate: 7.5 degrees / minute Final Time: 30 minutes Final Temperature: 200 °C Injection Temperature: 220 °C Range: 2 Attenuation: 0 Reagents: 1. Acetonitrile - HPLC Grade CH3CN EM Science, Gibbstown, NJ. 2. 2 - Nonanone - 99+ % CH3(CH2)5COCH3 Aldrich Chemical Company, Milwaukee, WI. Density 0.832 gm/ml Units: 1 uL of 2-nonanone / Liter Of Acetonitrile corresponds to 1 ppm (v/v) Example: 1300 uL of 2-nonanone / L = 1300 ppm Procedure: 1. A 1300 ppm stock solution of 2-nonanone in acetonitrile was made. A 100 ml flask was used. The amount of 2-nonanone weoghed out into the flask to achieve 1300 ppm was figured by: 0.13m] x 0.832 g/ml = 0.1082g Density of 2-nonanone = 0.832 g/ml 0.1082 g of 2-nonanone was placed into the 100 ml flask and was then diluted with acetonitrile to the line to achieve the 1300 ppm stock solution. 1. Three different dilutions of the stock solution were made in 25 ml flasks to achieve 200 ppm, 400 ppm, and 800 ppm concentrations. The equation used for this was: VoCo : VnCn 60 Where Vo = volume of original solution (stock solution) to add Co = concentration of original solution (1300 ppm) Vn = volume Of new solution (25 ml) Cn — concentration Of new solution (200, 400, or 800 ppm) For 200 ppm: V0 (1300 ppm) = (25 ml)(200 ppm) Vo = 3.85 ml 3.85 ml of stock solution was pipeted into a 25 ml flask and diluted to the line with acetonitrile. This procedure was repeated for the 400 and 800 ppm solutions. Volume conversions of the four different concentrations (200,400,800, and 1300 ppm) in there injected form are as follows: 200 ppm : 200x 10'6 ml/ml x 0.832 g/ml = 166.4 x 106 g 2-nonanone /ml acetonitrile 166.4 x 10'° g/ml x 0.001 ml = 1.082 pg Of2-nonanone (quantity injected) The above calculation was repeated for 400, 800, and 1300 ppm. Table 15: Calibration Curve Values Concentration of Quantity of 2-nonanone Average Area Unit 2-nonanone (ppm) Injected ( * 10'° g) Response (AU) 0 0 0 200 0.166 2,880,563 400 0.333 5,829,077 800 0.666 13,699,297 1300 1.082 22,773,886 Seven injections at each concentration were run on the GC to determine the average area unit response for each of the four different concentrations. The resulting calibration curves are shown in figures 13 and 14. 61 ArealklitlbsponeMAU) Figure 13: 2-Nonanone Calibration Curve for the GC (1) 0 t a: 0 500 1000 Wmon—Nonarmlrjectedmpm) 62 1500 rea Unit Response (AU) Figure 14: 2-Nonanone Calibration Curve for the GC (2)! 25000000 “.7 20000000 ~ 1 5000000 1 0000000 5000000 I I y = 20644614.84x R2 = 1.00 l —— +__ O 0 0.2 0.4 0.6 0.8 1 1.2 Quantity of 2-Nonanone Injected ( x 10" grams ) 63 Appendix 2 Determination of Reference Film Carbon Dioxide Permeation Value Using the Dynamic Method Given: Volume of CO2 detected, cc (V d) 0.0248 Area of sample film, cm2 (As) 50 Pressure differential, atm 1 Recorder velocity, mm/sec (Rv) 0.167 (1 cm/min) Recorder paper advance, measured, mm (Rp) 57** F ilrn Thickness, measured, mil (I) 2'” ** example munbers 1. Calculate Time of Permeation 0 : 0 = Rp / RV = 57 mm = 341 seconds = 5.69 minutes 0.167 mm/sec 3. Calculate Gas Transmission Rate (GTR) : GTR = vd / ( 0 * A.) = 0.0248 cc = 8.72 x 10'5 _e_e 5.69 min * 50 cm2 min * cm2 4. Calculate Permeability Constant (P): P= GTR*I Ap P = 8.72510-5cc x 2mil x 1 x 10020m2 x 1440minutes minutes * cm2 1 1 atm 1 m2 1 day "U II 2511 cc *mil m2*day*atm 64 Appendix 3 Determination of Test Film Carbon Dioxide Permeation Values Given: Pre-determined permeation value Of reference film to be 2511 cc * mil m2 * day * atm (See Appendix 2) The Permatran CIV Continuous Method was used running three test films against this reference film. The voltage values were found as follows: Reference film: 0.56 volt Test Film A : 0.64 volt Test Film B : 0.62 volt Test Film C : 0.54 volt 1. Determine Ratios: Film A / Reference Film = 0.64 volt / 0.56 volt = 1.14 Film B / Reference Film = 0.62 volt / 0.56 volt = 1.11 0.96 Film C / Reference Film = 0.54 volt / 0.56 volt 2. Determine Transmission Rates: Rate of Test Film = Rate of Reference Film x Ratio FilmA = 2511 cc x1.14 = 2863 cc mZ-day m2°day FilmB = 2511 cc x1.]1= 2787 cc mz-day mZ-day FilmC = 2511 cc x 0.96 = 2411 cc mZ-day mzoday 65 1) 2) 1) Appendix 4 Calculation for the Conversion of ppm to pig For 2-Nonanone Concentration A 500 uL sample was drawn during testing of the 2-nonanone vapor stream and injected into the gas chromatograph. The gas chromatograph gave out readings in area units (AU). The AU value was then put into the equation for a line from the calibration curve seen in Figure 13 as y and the equation was solved for x which was the concentration of 2-nonanone in ppm (vol/vol). ppm was then converted to 11g by the following example: 240 ppm (vol/vol) = 240 * 10°6 ml/ml (240 r 10'6 mel) * (0.832 g/ml) = 199.68 * 10'6 g/ml density of 2-nonanone (200.88 * 10'6 gml) * 0.500 ml = 99.84 * 10'6 g = 99.84 ug sample size measured 66 "I1111111111111111111“