1:. my- w. an"... ‘WIIIN< u‘v-“P::> r. .r, .... _ ‘. -‘ _ .:...:.L- L; ”at. aha-a .. u.» u fl. '7‘1’“~«| a: . 221' - . ¢ , van-W id‘- 15; . afi- : .1 ;. V ’3’ “‘1 a“ fix ‘ :m‘ - “3h;- . ' y a .. ‘ § KL?" ‘ 3.}- ' nu.— v. z #4 “ ,p v...~ "3‘"??? = ; ~ «a ‘ , 4M .5353 " 3 .* ' ,. .53 Li" . gv‘...~u.~..r_. "WNW“... . . ”a“.-. .1“ w» arc-04f- v‘”~ . "- v-uo .. , THESlS MIC CHI GAN STATE UNIVER ll lllllll’lllllll iiIllllllilllllmllll 1293 01390 8920 This is to certify that the thesis entitled PERMEATION 0F Z-NONANONE VAPOR THROUGH LLDPE AFFINITY FILMS AS APPLY TO MODIFIED ATMOSPHERE PACKAGING presented by MOHD ARIFF WAHID has been accepted towards fulfillment of the requirements for MASTER degree in W Major professor Date FEBRUARY 5, 1996 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution LIBRARY Michigan 91319 Universlty PLACE ll RETURN BOXto monthl- chockout from your ncord. TO AVOID FINES Mom on or More data duo. _;-_ * _ H +_, E ‘1’ 0921"? f MSU Is An Nflmulvo Action/EM Opportunlty Intuition Warns-9.1 PERMEATION OF 2-NONANONE VAPOR THROUGH LLDPE AFFINITY FILMS AS APPLY TO MODIFIED ATMOSPHERE PACKAGING By Mohd Ariff Wahid A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Packaging 1 996 ABSTRACT PERMEATION OF 2-NONANONE VAPOR THROUGH LLDPE AFFINITY FILMS AS APPLY TO MODIFIED ATMOSPHERE PACKAGING By Mohd Ariff Wahid The sorption, diffusion and permeation of 2-nonanone in LLDPE was characterized at different concentrations, vapor activities and temperatures. The permeability was measured by quasi-isostatic method, and solubility and sorption profile data for 2-nonanone in LLDPE were determined by using microbalance technique. The results of this study showed concentration and temperature dependencies for the diffusion and permeability of 2-nonanone through LLDPE over the temperature range studied. Sorption behavior of 2-nonanone by LLDPE showed two mode sorption phenomena as a function of concentration and vapor activity at the temperature tested. A model for modified atmosphere packaging of minimally processed fresh produced can be developed in relation to the permeant transport process. The mass transfer model includes the permeability and equilibrium solubility of 2-nonanone in the polymer material and be used to calculate the amount of the package area needed to maintain target concentrations in the headspace of the package. TABLE OF CONTENTS LIST OF TABLES LIST OF FIGURES INTRODUCTION LITERATURE REVIEW Permeation of Organic Vapor Through Polymer Films The Nature of Perrneant The Mechanism of Diffusion and Sorption Variables Affecting Sorption and Diffusion The Nature of Polymer Polymer Material - Polyethylene Modified Atmosphere Packaging for Minimally Processed Fruits Properties of Organic Vapor - 2-Nonanone Theory of Permeation Factors Affecting the Diffusion and Solubility Coefficients Pressure Sorption Behavior Temperature iii vi vii 11 12 15 16 17 21 21 21 22 Measurements of Transport Properties Permeability Measurements Sorption Measurements MATERIALS AND METHODS Materials Polymer Film Penetrants Acrylonitrile Nitrogen Gas Experimental Procedures Permeability Measurement Sorption Measurement Gas Chromatographic Analysis RESULTS AND DISCUSSION Solubility of 2-Nonanone Vapor in LLDPE Affinity PL 1880 Film Equilibrium Solublity of 2-Nonanone in PL 1880 as a Function of Partial Pressure The Effect of Concentration on the Solubility Coefficient of 2-Nonanone iv 24 24 25 30 30 30 30 30 30 31 31 32 34 39 39 51 55 The Effect of Temperature on the Solubility and Permeability Coefficients of 2-Nonanone The Effect of the Temperature on the Diffusion Coefficient of 2-Nonanone The Effect of the 2-Nonanone Vapor Concentration on the Penetrant Permeability through Quasi-isostatic Technique Package Design for the Release of 2-Nonanone in the Headspace of the Package CONCLUSIONS RECOMMENDATIONS APPENDIX Appendix A Procedure of standard calibration curve construction Appendix B Calculation of the saturated vapor pressure of 2-Nonanone BIBLIOGRAPHY 59 60 64 74 84 86 87 89 90 LIST OF TABLES Table Title Page 1. Setting condition of gas chromatograph 35 2. Physical properties of LLDPE Affinity PL 1880 36 3. The partial pressure and concentration of 2-Nonanone 52 at 22 °c, 30 °c and 40 °c 4. The effect of vapor activity on the sorption of 2-Nonanone 57 at 22 °c 5. The effect of vapor activity on the sorption of 2-Nonanone 57 at 30 °C 6. The effect of vapor activity on the sorption of 2-Nonanone 58 at 40 °C 7. The effect of the 2-Nonanone concentration on 65 the permeation at 22 oC 8. The effect of the 2-Nonanone concentration on 65 the permeation at 30 °C 9. The effect of the 2-Nonanone concentration on 66 the permeation at 40 °C 10. Package design in relation with equilibrium solubility 76 of 2-Nonanone 11. Total quantity of liquid 2-nonanone to produce a headspace 79 concentration or vapor activity as a function of MI; vi LIST OF FIGURES Figure Title Page 1. Generalized transmission profile for the quasi-isostatic method 28 2. Generalized sorption profile curve for microbalance technique 29 3. Schematic diagram of sorption and permeability test apparatus 37 4. Standard calibration curve of 2-N0nanone at 23 °C 38 5. MM” vs time at 22 °C and Av = 0.13 41 5. WM... vs time at 22 °c and Av = 0.15 ‘ 41 7. WM... vs time at 22 °c and Av = 0.18 42 8. M.IM..o vs time at 22 °C and Av = 0.23 42 9. WM... vs time at 22 °c and Av = 0.38 43 10. WM.o vs time at 22 °c and Av = 0.48 43 11. WM... vs time at 22 °c and Av = 0.58 44 12. WM... vs time at 30 °c and Av = 0.13 44 13. wa vs time at 30 °c and Av = 0.18 45 14. WM.o vs time at 30 °c and Av = 0.24 45 15. WM.o vs time at 30 °c and Av = 0.33 46 vii 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. Mde vs time at 30 °c and Av = 0.49 WM.o vs time at 30 °c and Av = 0.66 WM.o vs time at 40 °c and Av = 0.18 Mthao vs time at 40 °C and Av = 0.21 Mums, vs time at 40 °c and Av = 0.28 Mthco vs time at 40 oC and Av = 0.31 Mthm vs time at 40 °c and Av = 0.46 WM.o vs time at 40°C and Av = 0.53 ‘ The effect of partial pressure and temperatures on the concentration of 2-N0nan0ne The effect of the vapor activitiy and temperature on the concentration of 2-Nonanone The effect of the concentration on solubility coefficient of 2-N0nanone at different temperatures Temperature dependence of the solubility coefficient for 2-Nonanone at Av = 0.18 Temperature dependence of the permeability coefficient for 2-Nonanone at vapor activity = 0.15, 0.3 and 0.5 Temperature dependence of the diffusion coefficient for 2-N0nanone at vapor activity = 0.15, 0.3 and 0.5 Transmission rate profile of 2-Nonanone at 22 °C and various vapor activities Transmission rate profile of 2—Nonanone at 30 °C and various vapor activities viii 46 47 47 48 48 49 49 50 53 56 61 62 63 67 68 32. 33. 34. 35. 36. 37. 38. 39 Transmission rate profile of 2-Nonanone at 40 °C and various vapor activities Transmission rate profile of 2-N0nanone at vapor activity of 0.23 The effect of 2-Nonanone vapor activity on Log P at different temperatures The effect of 2-Nonanone vapor activity on Log D at different temperatures The vapor activity of 2-N0nanone vs package area using desorption proses from the polymer film A typical package of slice apples to maintain 2-nonanone concentration in the headspace of a package through desorption process , ‘ A small pouch with 2-nonanone liquid in the package to maintain the concentration of 2-n0nanone in the headspace of the package The vapor activity of 2-Nonanone vs package area using small pouch in package system at different ratio of thickness, |1l|2, between small pouch and package ix 69 70 72 73 80 81 82 83 INTRODUCTION Efficient packaging is a necessity for almost every type of product. Therefore, packagingmaterials usually been selected on the basis of their performance. Barrier property is one of the most important parts of the spectrum of properties needed in a protective packaging material. Containers are usually designed to provide the most complete protection against penetration from oxygen and other gases, loss or gain of water vapor, loss of distinct aroma, and risk of absorptions of alien odors. Glass and metal easily handled this problem. Due to economic and technical reasons, plastic becomes an alternative to glass and metal packaging because of their processability, light weight and relatively low cost, but their limitation is in barrier performance. Packaging that serves a specific process function for the product which is an additional or different function from the common processing of product is sometimes called ‘functional packaging” or ‘active packaging’ (Ono ,1990). In some instances certain additives are incorporated into packaging film or within packaging containers to modify the headspace atmosphere and to extend product shelf life. Package design should incorporate mass transfer parameters to control product quality with the added active packaging function. Therefore, it is important to study the permeability, solubility and the diffusion factor of the packaging materials. 2 Understanding the permeation process of the polymer is critical in order to select suitable flexible packaging materials to protect packaged products. The measurement of the barrier properties has largely been restricted to gas and water vapor permeabilities. Recently it has been recognized that the barrier function of a package to various organic compounds is important, and has lead to the development of a method test to measure permeabilities of various organic vapors (Gillette, 1988; Hernandez, 1986; Zobel, 1988). Minimally processed fruits are products that can provide convenience and fresh-like quality. In 1995, 3% of the United States sales consisted of fresh-cut produce and by 2000, such ready-to-use produce is projected to fill 25% of shelf space (Leaversuch, 1995). A major factor contributing to the growth of this category is the refinement in the performance of plastic films. Films can be used to modify the package atmosphere and thereby substantially extend the shelf life of fresh produce. In this system, oxygen in air surrounding the product in the package is consumed by the product and carbon dioxide and water vapor are released, leading to a mixture of atmospheric gases different in proportion from that of air. Microbial growth is one of the major concern that can cause spoilage to minimally processed fruits in modified atmosphere packaging (MAP). Recently the application of antifungal natural volatiles to control decay in minimally processed fruits has been explored (Vaughn et al., 1993; Hamilton- Kemp, 1992; Ding et al., 1992; Davis et al., 1972). 2-Nonanone has been shown to have antifungal activity in packages of strawberry and raspberry fruits 3 (Vaughn et al., 1993). This natural volatile may function as an effective antifungal compound which might be used commercially to. prevent decay of stored fruits. According to (NIOSH, 1979), it has low mammalian toxicity (oral rat LD50_ 3200 mg/kg); a pleasant, fruity/floral odor, resistance to rapid decomposition, and adequately volatility (Vaughn et al., 1993). Therefore it has potential for commercial development as a slow-release fungistatic compound in prolong the shelf-life of fresh produce. Decay control using volatile hold some promise where it does not require direct application to the surface of the product, and are volatile in nature, so residue levels would tend to be low. According to Edney (1983), volatile materials can permeate the air spaces around the product more efficiently than liquids, leading to a greater likelihood of contact between the control agent and decay organism. Biologically active volatile materials are frequently encountered in nature and many are found in more than one source. For example, 2-nonanone can be found in attar of rose, clove oil, passion flowers, asparagus, tomato, corn and various fruits like raspberries and strawberries. The permeation rate and diffusivity through polymer by 2-nonanone is unknown. This information is needed to develop the relationship between film permeability characteristics and the volatility and biological activity of this material. Information on diffusivity, solubility and permeability of organic volatiles can help in developing a better understanding of the mechanism of the movement of organic volatiles through polymer structures. The diffusion and 4 solubility of organic volatile are important in the area of product quality when product quality is related to the effect of organic volatile in the closed environment and the transfer of the vapors out of the system. Knowledge of the organic vapor barrier properties of polymeric films can provide a means of designing a barrier structure for a specific application. The objectives of the study were as follows: 1. To determine the permeability and sorption of 2-nonanone for linear low density polyethylene (Affinity PL 1880). 2. Evaluation of the concentration and temperature dependencies on the transport process of natural organic penetrant. 3. To develop a better understanding of the mechanism and the variables that affect sorption and diffusivity of natural volatile compound in barrier polymer film. 4. To develop a package design which takes into account the permeant transport process. LITERATURE REVIEW Permeation of Organic Vapor Through Polymer Films The phenomena of mass transfer, includes permeation, absorption and desorption of gases or vapors into polymeric materials and is a very important aspect of packaging design. The permeability of packaging materials to organic vapors is of considerable interest, either to protect the packaged product from foreign odors or to retain its aroma and/or volatile compounds inside the package. According to Gillette (1988), the flux of a permeant through a film is a function of how much of the permeant is present in the film and the mobility of the permeant within the film. In other words, permeability is the product of the thermodynamic parameter of solubility and the kinetic variable of diffusion. For organic vapors, separating the permeability into its two components is important because the diffusion coefficients of these compounds are typically low enough that steady-state permeation is very slow to achieved . The significance of studying the mass transfer properties in polymer film have encouraged the development and application of various tests on the diffusion in polymer films. Solubility can become an important factor when low concentrations of a permeant must be contained in the product or the headspace of the package. Simply increasing the thickness of the packaging material may not prevent loss of the compounds if its solubility in the packaging material is sufficiently high. The compound will be effectively lost from the product by absorption into the 5 6 packaging material under these circumstances. The sorption of organic vapors in plastic films become quite complex as the penetrantlpolymer interaction increases. The sorbed vapors may swell and plasticize the polymer, resulting in increased mobilities of both polymer segments and penetrant molecules (Laine and Osbum, 1971). There is no standard test for the measurement of organic vapor permeability, although there are a number of methods have been described for vapor permeability measurements. An apparatus that could permit permeability measurements to be made at very low vapor levels was described by Zoebel (1982). An instrument for studying the transport of aromas in polymer films has been described (DeLaussus et al., 1988) which utilizes a mass spectrometer to detect the permeant. Modern Controls, Inc has recently introduced the Permatran-O to measure the permeation of organic vapors through polymeric materials. Various procedures for quantifying the rate of diffusion of organic penetrants through polymeric films have been described by Hernandez et al. (1986). These authors describe the isostatic and quasi-isostatic methods. Because thepermeation rate of organic compounds is not as fast as that of the permeant gases, the slope of the straight part of the permeation curve is very flat for organic substances. Therefore, in the permeation rate of organic vapors, the extrapolation to the time axis is very uncertain and the lag time can lead to negative diffusion constant (Robertson, 1972). In instances where swelling of the polymer occurs by organic vapors, both diffusion and solubility 7 constant become concentration and time dependent. Because these effects cannot be separated by combining time lag and steady state data, it is necessary to study the sorption behaviour as a function of time (Naylor, 1989). For this reason, it has been recommended (Kiszinowski, 1986) that the solubility constants of organic compounds should be determined by other methods such as sorption/desorption experiments. This is commonly done by suspending the polymer sample from a balance under a uniform concentration of vapor at a known temperature and pressure (Hernandez et al., 1986). The permeability and solubility of d-limonene vapor in package liners was studied by (Mohney et al.,1988) showed that the loss of aroma moieties can be result of sorption and permeation through the packaging material. ‘Holland and Santangelo (1988) introduced the ‘laminate film’ method for measuring vanilin transmission rates through food packaging films. The Plaque method (Kiszinowski, 1986) has been introduced when the organic compounds are sorbed from liquids. The Nature of Perrneant The Mechanism of Diffusion and Sorption The mechanism of permeation of a vapor through a polymer involves several steps (Van Amerongen, 1950). First, a diffusing molecule condenses on the high pressure side of the polymer. Then, this is followed by solution of the condensed vapor by the polymer. Next, the solution process is followed by 8 diffusion through the polymer, which takes place by the movement of a permeant molecule through a tangled mass of polymer chains and holes which are constantly forming and reforming as the result of thermal vibrations. Usually the holes are smaller than the permeant molecule, hence, several jumps in the same direction must be made before the molecule is displaced by a distance equal to its length. According to Robertson (1992), diffusion of a dissolved permeant in a polymer is viewed as a series of activated jumps from one vaguely defined cavity within the polymer matrix to another. Qualitatively, it is possible to increase the number or the size of cavities in a polymer or give chain segments more mobile increases the rate of diffusion. This can be achieved for instance, by increasing temperature or by the use of plasticizers. Following the diffusion process, there is desorption of the permeate on the low pressure side of the film by evaporation. Organic vapors diffuse by a more complicated mechanism which is dependent on the motions of both the polymer and diffusant molecule, which are comparable in size or larger than polymer chain segments. The total operational sorption process will likely involve one or more sorption modes. The nature of interactions between permeants and polymer is an important factor in which the molecule of permeants are distributed in the polymer may results in cluster formation, or randomly dispersed permeant. The sorption of organic vapors in polymer films becomes more complex as the penetrant-polymer interaction increases. The sorbed vapor swells and plasticizes the polymer, resulting in increased mobilities 9 of both polymer segments and permeant molecules. The absorption of vapor is sometimes accompanied by time-dependent processes in the polymer which are slower than the micro-Brownian motion which promotes diffusion. These processes depend upon the nature of the polymer, the temperature, and the concentration of the permeant (Meares, 1965). Therefore polymer structure, free volume, chain stiffness of segmental mobility, and the availability of specific sites of interaction in the polymer, plus the physiochemical characteristics of the permeant determined the mode and mechanism of sorption and transport of the permeant within the polymer (Liu et al., 1991 ). Variables Affecting Sorption and Diffusion The size, shape and polarity of the permeate molecule, together with its facility of condensation, have important effects on permeability. Size and shape particularly affect diffusivity while solubility is influenced by polarity and ease of condensation of the molecule. The solubility of the permeate generally depends on its compatibility with the polymer (Pascat, 1986). According to Murray (1985), important permeant properties include volatility and the size and shape of the molecule. Volatility controls the maximum concentration; size and shape relate to the movement through temporary holes which develop in the polymer. Linear molecules will penetrate more readily than cyclic molecules. The diffusion of most organics and water is 10 complicated by concentration dependence, although gas transport at low pressure in polymer films is not concentration dependent (Hopfenberg, 1978). For the most systems, the diffusivity and permeability are generally higher when the polymer and permeant are similar. In polyethylene, which is a nonpolar material, the permeability is lowest for polar materials and highest for nonpolar materials such as hydrocarbons (Pinsky, 1957). According to Laine and Osbum (1971 ), the diffusion coefficient at zero activity, pressure and concentration, Do, generally decreases as the volume of the penetrant molecules increases, but branching has a greater effect than does molecular size. For example, addition of a methyl group on a given paraffin reduces the valve of D. more than does increasing the chain length by one carbon atom. This suggests that diffusion occurs preferentially along the direction of greatest length of the permeant molecule. The solubility coefficient, So, on the other hand, increases exponentially with the increase in molecular volume and crossectional area of the penetrant molecule. As a result of this compensating dependence of Do and So on permeant size and shape, the zero concentration permeability coefficient, Po, which equals the product of Do and So, is much less dependent on the size and shape of the penetrant than either term separately (Roger, 1965). 11 The Nature of Polymer The barrier properties of films depend on the specific molecular structures of the polymers involved (Pascat, 1986). A fairly systematic correlation has been established (Paine and Paine, 1983) between the activation energy for diffusion, Ed, the size of penetrant molecules and the T, of polymers for a wide range of gases and low molecular weight organic vapors. There is a greater dependence of both sorption and diffusion processes on the size and shape of the penetrant molecule in the glassy state than in the rubbery state. The low densities of plastic materials with their open structures allows small molecules pass through the structure without being affected by the diffusion process. In higher density polymers, the initialy sorbed vapor breaks up some of the more imperfectly ordered crystalline regions which would be proportional to the vapor concentration (Roger et al., 1956). Diffusion proceeds exclusively in the amorphus regions of the polymers which pass through the voids in these polymer regions. These voids are created by segmental motion of the polymer chains. Segmental mobility of the polymer chains increases with temperature (Roger et al., 1956). The diffusion process takes place in the amorphus domains of the polymer and is dependent upon the mechanical and thermal history of the polymer (Zobel,1985). Crystallites may reduce polymer chain segment mobility in the interstitial amorphus phase, thereby raising the energy barrier of diffusional transport of permeants. Futhermore, the degree of segment immobilization will become greater as the volume fraction of crystalline phase 12 increases and/or the average crystallite size decreases. The number and length of side chain branches in polymers are also important to the barrier properties. Polymer Material : Polyethylene The Ziegler-Natta or, often, Ziegler catalysts polymerize a wide variety of monomers to linear or stereoregular polymers. Ziegler (1955) introduced the polymerization of ethylene with aluminium alkyls and chlorides of transition metals of group IV and VI, which produce polyethylene with a catalyst system of anhydrous aluminium chloride and titanium tetrachloride under pressure and elevated temperature. According to Diedrich (1975), the original Ziegler type catalyst systems were of low activity, generating as little as 1500 g of polymer per gram of transition metal. Thus, Ziegler processes of the first generation required large quantities of catalyst and insufficient flexibility of the polymerization process to produce products for a wide range of application. Second generation Ziegler polyethylene processes has been developed by introducing high yield catalyst systems, which are capable of producing polyethylene without removal of catalyst and no need for additional processing steps. The catalyst can control the basic physical parameters of the polymer, for example, molecular weight, molecular weight distribution and the proportion of long and short side chains. 13 According to Billmeyer (1984), the physical properties of polyethylene are functions of three independent structural variables: molecular weight, molecular weight distribution or long chain branching and short chain branching. Properties dependent on crystallinity, such as stiffness, tear strength, hardness, chemical strength, softening temperature and yield point, increase with increasing density or decreasing amount of short chain branches in the polymer, whereas permeability to liquids and gases, toughness and flex life decrease under the same conditions. The effect of long chain branching on the properties of polyethylene is evaluated in terms of the breadth of the molecular weight distribution. A decrease in molecular weight distribution causes in a decrease in ease of processing but an increase in tensile strength, toughness and impact strength, and resistance to environmental stress cracking. Most of the differences in properties between branched and linear polyethylenes can be attributed to the high crystallinity of the latter polymers. Linear polyethylenes are stiffer and have a higher crystalline melting point and greater tensile strength and hardness then the branched material. Recently, a new polyolefin polymerization catalyst technology has been developed based on metallocene catalysts. This technology is based on a constrained geometry ligand attached to a group IV transition metal (69., Ti) catalyst center. The catalyst structure allows a significant increase in the flexibility for incorporating ethylene and other alpha olefin comonomers into the polymer structure. A new family includes homopolymers and copolymers of l4 olefin polymers has been produced. This new constrained geometry catalyst technology can produced an excellent physical and mechanical properties and excellent melt processability (Swogger, 1992). According to Leaversuch (1995) metallocene-based polyethylene will be central to minimally processed fruit and vegetable market growth. Metallocene polyethylene is suitable to maintain the optimal balance of oxygen, carbon dioxide and moisture necessary. It can be made at densities as low as 0.89 glml. Metallocene polyethylene delivers a higher oxygen and moisture vapor transmission rates than other polyethylene, thus highly efficient in allowing oxygen in and carbon dioxide out. Moisture control allows designers to minimize wilting. Metallocene polyethylene can provide clarity, toughness, extended heat sealing and FDA compliance for food use. Studies of organic vapor-polyethylene systems have been made by Rogers et al. (1960) and McCall and Slichter (1958) for the diffusion of saturated vapors in several polyethylenes at various temperatures. Michael and Parker (1959) studied the sorption of gases through polyethylene films. The authors found that the solubility depends primarily on the amount of crystallinity presents. The magnitude reflects both the volume fraction of amorphus polymer and the area fraction of amorphus material available for flow. Temperature dependence of branched polymer permeability is somewhat greater than linear polymer due to the increase in diffusional activation energy caused by the short chain branches (Michael et al., 1959). Polymers at temperatures below the 15 glass transition exhibit anomalous non-Fickian diffusion behaviour. The discussion of this phenomena can be found in the literature (Roger et al., 1956). Affinity PL 1880 polyolefin plastomer is the first family of homogenous ethylene alpha-olefin copolymer with 12% octene comonomer developed by the Dow Chemical Company using in-site technology, for use in a variety of demanding packaging applications, including high speed, form-fill seal products. It has narrower molecular weight distribution compare to linear low density polyethylene produced by Ziegler-Natta process. It has excellent ultimate hot tack strength and low temperature seal initiation, outstanding optics and abuse resistance. This film has excellent compatibility with other polyolefins, allowing efficient blending and coextrusion. Modified Atmosphere Packaging For Minimally Processed Fruits Modified atmosphere packaging involves modification of atmospheric conditions, via the commodity’s respiration as it responds to the surrounding physical environment. Initial modification can be made by the replacement of the air in the headspace of the package with a mixture of atmospheric gases different in proportion from that of air or passively modified by fresh produce inside the package. The recommended concentration of oxygen and carbon dioxide for modified atmosphere storage of fruits and vegetables can be found in the published literature (Kader, 1985). Minimally processed fruits and 16 vegetables contribute 3% to the United States fresh produce sales in 1995 and by 2000 such ready-to-use produce is projected to fill 25% of shelf space (Leaversuch, 1995). These products have the attributes of convenience and fresh—like quality. The key importance of minimally processed fruits and vegetables is the control of enzymes from the produce itself or from the invading microorganisms (Robertson, 1993), to maintain the firm, crisp texture, bright and light color. According to King and Bolin (1989), the primary spoilage mechanisms are the metabolism of the tissue and microbial growth; both will cause deterioration of the tissue and must be controlled to maintain tissue viability. Decay control can be achieved through the application of chemical food additives such as 802 or the use of heat or irradiation. The use of natural preservative compound has a greater appeal and increase in interest in the use of organic volatiles to control decay. In a modified atmosphere packaging system, natural organic compounds can be used to control decay by adding the compound in the packaging film of by flushing into the headspace of the package. 2-Nonanone is a promising organic volatile that effectively controls fungal growth on minimally processed fruits (Vaughn et al., 1993). Properties of Organic Vapor : 2-Nonanone 2-Nonanone is a natural volatile compound found in various fruits like raspberries and strawberries (Vaughn et al., 1993). The formula of l7 2-nonanone is CH3(CH2)6COCH3, with a -C=O functional group, known as the carbonyl group in the chemical structure. It is a colorless liquid with a melting point of -9 °C and boiling point of 194 °C. The density of 2-Nonanone is 0.832 glml. This natural volatile may function as an effective antifungal agent if sufficient concentration could be maintained in the gas space surrounding the fruit. According to (NIOSH, 1979), it has low mammalian toxicity (oral rat LD50, 3200 mglkg); a pleasant, fruity/floral odor; resistance to rapid decomposition and adequate volatility (Vaughn et al., 1993). Theory of Permeation. According to Jost (1960), diffusion through polymer materials at constant pressure and temperature can be expressed by Fick’s Law. The first Fick’s Law described the rate (J) of transfer of diffusing substance through unit area of a section as being proportional to the concentration gradient aC/ax. The equilibrium uinidirectional diffusion along the x axis can be expressed as, J = -D ac/ax. l l) where J represent the mass of the substance which passed through a unit area perpendicular direction during a unit of time, D is the diffusion coefficient and c is the concentration of the permeant substance. Diffusion as a function of time 18 on a reference volume is given by Fick’s second Law. (D (2) fig 61 9211’ Q 6x Equation (2) is also valid if D is a function of penetrant concentration and/or time. When D varies with time, t , the diffusion is often called non-Fickian. Integration of Fick’s first law gives an expression relating flux (J) to the diffusion coefficient (D), and gas concentration Cg and c1, the steady state concentration of the permeant at the two surfaces of a gas barrier to thickness, L, by, c2 J: (x')jD.ac = Did-Clix] , (3) cl Similarly,tfor a given plastic barrier at equilibrium, the steady state due to permeation is given by, 92 J = lxlllPfip = Plp2-pilL" (4) pl where p, and p; are the partial pressure of the permeating species on either side of the barrier, and P is the permeability coefficient. From these expressions, the 19 total amount of permeating substance (0) passed through the polymer area (A) during time (t) can be evaluated thus, Q = DAt (c2-cr) = PM (92-91) (5) L L To relate the concentration of the substance (c) within the polymer barrier to the concentration of substance in the gas or vapor phase in equilibrium with the polymer, Henry’s Law is assumed, c=Sp (a where S is the solubility constant of the gas or vapor in the polymer. Therefore, the gas or organic vapor barrier characteristics of plastics may be expressed by reference to three coefficients as mention by (Gillette, 1988), permeability is the product of the thermodynamic parameter of solubility and the kinetic variable of diffusion. P=DS (N where, the diffusion coefficient, D, the permeability coefficient, P, and the solubility constant characteristize the process of gas transfer through a polymer. The diffusion coefficient is a kinetic term that describes how fast a permeant moves in a polymer and it also measure of how much time is required to reach steady state. The value of the diffusion coefficient for a selected 20 penetrant/polymer, is determined by the size of the permeant and the size and frequency of fluctuations of polymer molecules. It is a combination of geometry and thermal effects. N9 (1985) summarized the dependence of diffusion on concentration in polymers as follows; D = K exp (ABC) fo‘2 (8) where, K = a jumping frequency factor A = a factor related to the minimum hole size for a jump to occur F0 = the polymer free volume fractioning in the absence of the vapor B = a measure of the ability of the migrating compound to increase the free volume of the polymer c = the concentration of the compound The solubility coefficient is a thermodynamic term that describes how many permeant molecules move in a polymer. It can be determined by temperature, chemical activities, and intermolecular interactions plus the state of the polymer relative to its glass transition temperature (De Lassus and Strandburg, 1991). 21 Factors Affecting the Diffusion and Solubility Coefficients Pressure The permeability coefficient is independent of the pressure of the diffusing gases and vapors provided there is no interaction between the polymer and diffusing material. However, where there is strong interaction, the permeability is found to be pressure dependent and, in general, it increases as the pressure increases. This is due to an increase in diffusion coefficient, D, promoted by the plasticizing effect of the sorbed permeant, and an increase in the solubility coefficient, S, due to the shape of the sorption of permeant (Robertson, 1992). Sorption behavior The term sorption is used to describe the initial penetration and dispersal of permeant molecules into the polymer matrix and includes both adsorption and absorption as well as cluster formation (Naylor, 1989). Sorption behaviour has been classified on the basis of the relative strengths of the interactions between the permeant molecules and the polymer, or between the permeant molecules themselves within the polymer. Three major types of sorption behavior are Henry’s Law sorption, Langmuir-Type sorption and Dual-mode sorption. 22 Temperature At low pressures or at enhanced temperature, the solubility of gases in a polymer is dependent of the pressure, and the Equation (6) is analogous to Henry’s Law (Stepek et al., 1987). An increase in temperature contribute energy to increase the segmental motion of the polymer chains, which increases the permeability of permeants through the polymeric films. The temperature dependence of the solubility coefficient over relatively small ranges of temperature can be represented by an Arrhenius-type relationship: 3 = So exp(-AHs)/RT (9) where 8,, is a constant, AHs is the heat of solution of the gas in the polymer. For condensable vapors AHS is negative due to the contribution of the heat of condensation, and thus S decrease with increasing temperature. R and T stand for the universal gas constant and absolute temperature, respectively. The temperature dependence of the diffusion coefficient can also be represented by an Arrhenius-type relationship: D = Doexp(-Ed)/RT (IO) 23 where Do is a constant, and Ed is the activation energy of diffusion which is usually a linear function of temperature. Ed is associated with the energy required for hole formation against the the cohesive forces of the polymer plus the energy necessary to force the molecule through the surrounding structure. For organic liquids in polymer, this value is generally between 10 to 40 kcallmole (Laine and Osbum, 1971 ). 1 From the above two equations it follows that: .0 l - Po exp(-Ep )/RT (1 l) 13 II ' (DoSo) exp [-( Ed+Hs)/ RT] ' (l 2) where Ep (= E, + H.) is the apparent activation energy for permeation. Generally, the solubility coefficient decreases for vapors, and the diffusion coefficient increases with temperature for both gases and vapors (Brown, 1981 ). Therefore, permeability coefficients of different polymers for one type of permeant determined at one temperature may not be in the same relative order at other temperatures. 24 Measurements of Transport Properties Permeability Measurements A quasi-isostatic test method or accumulation technique has been developed (Hilton and Nee, 1978) for determining the permeability of organic vapors through barrier films. The polymer film is mounted in a permeability cell and the vapor permeating through the film is monitored such that the total quantity permeated can be plotted as a function of time. Baner et al. (1986) described a method based on a quasi-isostatic procedure for determining the diffusion of organic penetrants through polymeric films. Baner et al. (1986) employed a continous flow of organic vapor through the high concentration cell chamber to assure a constant vapor gradient. A gas chromatographic method was developed for the permeation rate measurements. Barrer (1939) presented a solution of Equation (2) which allowed an approximative determination of D, the diffusion coefficient: D = L2/69 (13) where 9 is the intersection of the projection of the steady-state portion of the transmission curve and is called the lag-time (Figure 1). The steady state permeability coefficient (P) can be determined from the quasi-isostatic method by using the Equation (14), 25 P = yL/Ab (14) where y is the slope of the straight portion of the transmission rate curve (mass/time), L is a thickness of the film, A is an area of the film exposed to the permeant in the permeability cell and b is driving force given by the concentration or partial pressure gradient. By plotting log[t1,2( dM/dt)] as a function of 1lt, it is possible to obtain information about the concentration dependency of the diffusion coefficient, D (Baner, 1987). Sorption Measurements Sorption measurements have been used to select potential barrier polymer films for specific applications. F ujita (1961) described the general behavior of sorption and permeability of organic vapors. Baner (1987) applied the equilibrium vapor pressure and microbalance gravimetric technique to study the sorption and diffusion of toluene vapor through polymeric films as a function of penetrant concentration. Solubility of vapors is usually determined in sorption apparatus by measuring gravimetrically the equilibrium amount of the vapor absorbed by a known volume or weight of polymer. A gravimetric technique carried out at equilibrium vapor pressure can continually measure the weight gain and loss by the sample film as a function of time. The solubility coefficient 26 and diffusivity coefficient are thus determined by sorption measurement (Hernandez et al., 1986). Crank (1975) described the diffusion equation appropriate for the sorption of penetrant by a polymer film as: Mt/Mw = I - (8/1t2) [exp(-D1t2’1)/L2+l/9 exp(-9Dit21)/L2] (15) where M and M... are the amount of penetrant sorbed or desorbed from the polymer film sample at any time (t) and the equilibrium sorption level after infinite time, respectively, t is the time to attain M, and L is the thickness of the film sample. The sorption diffusion coefficient (D.) and diffusion coefficient for desorption (D.) can be calculated from Equation (16), where WM..= 0.5 (Figure 2). D. = D. = 0.049L2lt1)2 (16) Where t1); is equal to the time required to reach a sorption level equal to half of the equilibrium value, M... Solubility coefficient (S) values can be calculated from sorption experiments by yhe following equation. 5 = m (17) 27 Where S is the solubility coefficient expressed as mass of vapor sorbed at equilibrium per mass of polymer per driving force . M.. is the total amount (mass) of vapor absorbed by the polymer at equilibrium for a given temperature, w is the weight of the polymer sample under test and b is a value of the permeant driving force. Total Quantity Permeated (Q) 28 49+" Time Figure 1. Transmission rate profile of the quasi-isostatic method 29 MJMm t1/2 _ Time Figure 2. Generalized sorption profile curve MATERIALS AND METHODS Materials Polymer Film Linear low density polyethylene, polyolefin plastomer called Affinity PL 1880, 1.9 mil thick from Dow Chemical Company with 12% octene comonomer group was used in the experiments. The polymer had a density of 0.902 glcc. The physical properties was illustrated in Table 2. This material was selected based on the structural properties, highly permeable to oxygen which is suitable for fresh produce, and relatively inexpensive. Penetrants Research grade 2-nonanone (CH3(CH2)eCOCH3) from Adrich Chemical Company (purity greater than 99 %, boiling point of 192 °CI743 mm, molecular weight 142.24 glmol and density 0.832 g/cc) was employed through out, as the permeant molecule. Acrylonitrile Used as a solvent for constructing a 2-nonanone calibration curve. Nitrogen Gas Used as a carrier gas. High purity dry nitrogen 99.98%. 30 31 Experimental Procedures Permeability Measurement The permeability test based on the quasi-isostatic method (Figure 3) was used to collect the permeation data at different concentrations of the volatile compound through a sample film at 22, 30 and 40 °C (Hernandez et al., 1986). The film to be tested has been mounted in the permeability cell which was comprised of two aluminium cell chambers and placed in constant temperature chamber. A constant concentration of permeant vapor was passed through the center cell chamber at a constant rate. The center cell was separated from the two edge chambers by the test film. This cell design allows the permeability of two film samples to be determined concurrently, under identical conditions. A constant concentration of a permeant vapor was generated by bubbling nitrogen gas through liquid permeant maintained at constant temperature. The vapor permeating through the film into the lower chambers was then quantified by sampling and analysis by gas chromatography, until the permeation rate attained steady state. At predetermined time intervals an aliquot (500 pl) of headspace was withdrawn from the low concentration cell chambers with a gas tight high performance syringe (Hamilton, N0. 1750) and injected directly into the gas chromatograph with flame ionization detection. An equal volume of nitrogen was replaced into the cell in order to maintain a constant total pressure (1 atm). 32 To determine the diffusivity and permeability values, the increase in penetrant quantity in the lower concentration cell chamber was plotted as a function of time and the resultant transmission rate profile was related to the permeability of the film sample. Sorption Measurement The sorption measurement system was obtained by using an equilibrium vapor pressure and microbalance gravimetrictechnique (Figure 3). A Cahn Electrobalance (Cahn Instruments, Inc., Cerritos, CA) will be employed for the gravimetric technique (Hernandez et al., 1986). The electrobalance and sample tube were placed in the temperature controlled chamber, which maintained at 22 °C, 30 °C and 40 °C. A film sample was cut into small pieces and a weight between 30 to 50 mg was used since the sensitivity of the system is 5 ug for a sample mass of approximately 30 mg. The nichrome wire was hooked to one end of the sample, and tube assembly with the sample hanging freely in hangdown tube. The wire hooked to the sample was also hooked to the balance arm. This system allows continous collection of sorption data until equilibrium is reached as a function of penetrant concentration and temperature. A constant concentration of permeant vapor was flowed continously through the hangdown sample tube. A constant concentration permeant vapor can be produced by using a vapor generator system. Before actual testing was conducted, rotameter 33 settings were determined to provide a range of vapor activities. Vapor activity was calculated by dividing the experimentally determined vapor pressure by the saturated vapor pressure (Appendix B). Rotameters were used to provide an indication of the settings required for the desired vapor activities. The gas flows to the rotameters were regulated by Nupro ‘M’ series needle valves. For the calculation of vapor activity, a standard calibration curve for 2-nonanone was prepared. A detailed procedure to measure the calibration factor is presented in Appendix A and Figure 4 shows the standard calibration curve of 2-nonanone. Sorption profiles and solubilities were obtained by a continuous recording of weight-gain measurement at 22 °C, 30 °C and 40 °C until the system reached steady state. After this stage was reached, the desorption profiles was measured by closing the valve that attached to the vapor generator system that carried a constant concentration of permeant vapor through the tube. This would allow the permeant to desorbed out of the sample. To determine whether the initial portion of the sorption of the sorption curve followed Fickian behavior, and to estimate the diffusion coefficient, the following method was applied with the aid of a computer assisted fitting point. The values of M.. were selected by refering to a plot of the experimental M vs time, t, data. The M.. value for a Fickian diffusion process was assigned, the t1); value can be obtained from a plot of MIM. versus time value and the diffusion coefficient (D.) calculated by refering to Equation (16). By substituting the D. value into Equation (15), calculated values of M were obtained. By using minimum sum of square 34 between the calculated and experimental values, the calculated values were selected to plot MIM. versus time and compared with the experimental values. Gas Chromatographic Analysis Analysis of permeant concentration was carried out by a gas chromatographic procedure with flame ionization detection. A Hewlett-Packard Model 5890A gas chromatograph equipped with flame ionization detector interfaced to a Hewlett-Packard Model 3392A integrator was employed for quantification. The gas chromatographic conditions are presented in Table 1. A standard curve of detector response versus quantity has been constructed from standard solutions of known concentration. Figure 4 shows the standard curve, where response is plotted as a function of 2-nonanone quantity. Elution time for 2-nonanone was at 11.7 min. 35 Injection temperature 220 °C Max. oven temperature 250 °C Final temperature 200 °C Detect temperature 250 °C Rate 7.5 °C/min He carrier gas 45 l/min Initial time 1 min Final time 30 min Column SPB-5 Fused silica capillary column 30 m, 0.32 mm ID 0.25 )1 film thickness Table 1. Setting condition of gas chromatograph 36 Physical Properties Values Percent comonomer, octene 12.0 Melt index, dg/min 1.0 Density, gmlcc 0.902 DSC Melting Point, °C 100 Film Properties, 2.0 mil Puncture Resistance, chm3 26.0 Dart Impact, 9 >830 Elmendorf Tear Strength, 9 MD 355 CD 500 Ultimate Tensile, psi MD 7170 CD 3800 Ultimate Elongation, % MD 570 CD 560 Seal Initiation Temperature, °C 85 Table 2. Physical properties of linear low density polyethylene Affinity PL 1880 Ca 0 - . Cu N P 4 '- c1. 8 N N . :- S R N Re a g P: >Tv N _ . R I 1! Waste 1 Z ‘3. U". Figure 3. B - Organic Vapor Bubbler Ch - Control Temperature Chamber Cn - Cahn Electrobalance Ct - Computer Terminal Cu - Control Unit N - Needle Valve P - Plotter Pc - Permeability Cell R - Rotameter Re - Regulator S - Sampling Port Sf - Sample Film T - Nitrogen Tank Tv - Three Way Valve Schematic Diagram of Sorption and Permeability Test Apparatus Area Response (AU) 500000 400000 300000 200000 100000 38 I I I I 2O 40 60 80 Quantity Injected (ng) Figure 4. Standard Calibration Curve of 2-Nonanone at 23°C 100 RESULTS AND DISCUSSION Solubility of 2-Nonanone Vapor in LLDPE Affinity PL 1880 Film. Plots of M/M.. vs time for sorption of 2-nonanone in Affinity PL 1880 film at 22 °C are shown in Figures 5 to 11. The experiments were conducted at the I vapor activities of the permeant ranging from 0.13 to 0.58. Superimposed on the experimental data is the calculated curve according to Equation (15) with the diffusion coefficient, US, as calculated from Equation (16). Similar plots were prepared for solubility of 2-nonanone at 30 °C and 40 °C and are shown in Figures 12 to 17 and Figures 18 to 23, respectively. The vapor activities ranged from 0.13 to 0.66 at 30 °C and 0.18 to 0.53. at 40 °C. For these cases, the calculated curves were also compared to the F ickian model. At 22 °C, at a concentration of 2-n0nanone less than 2 ppm (w/v) or vapor activity less than 0.38, equilibrium times of less than 10 hours were observed for the F ickian diffusion process. As shown, the theoretical curve fits the experimental data very well and the initial portion of the curve is approximated by a straight line. The Fickian diffusion pathway may be assumed to be completed before applicable stress-relaxation has occured. It can be seen that in the early stage of sorption, the 2-nonanone vapor gradient provides the major driving force and the transport process is dominated by Fickian diffusion. At 22 °C, as the concentration of 2-nonanone increase above 2 ppm or vapor activity above 0.38, the agreement between the experimental and calculated results, following the 39 40 same procedure as above, is quite poor as shown in Figures 10 and 11. This trend occurred at vapor activities above 0.33 as shown in Figures 15 to 17, at 30 °C, and vapor activities above 0.28, as shown in Figures 20 to 23, at 40 oC. The sorption process is sufficient to develop a significantswelling stress which produces gradual swelling of the polymer structure and allows additional sorption. Therefore the level of vapor concentration determines sorption behavior. Berens (1977) and Mohney et al. (1988) also observed this behavior and suggested that this behavior involved a ‘two-stage’ sorption process with a non-Fickian, relaxation-controlled mode of sorption superimposed on the Fickian diffusion. 41 120 100 - MJMm x 100 so - 60 ’- 500 1000 1500 Time (min) Figure 5. Mm... vs Time at 22 °C and Av = 0.13 120 MJM... x 100 0 experiment — calculation 500 1000 1500 Time (min) Figure 6. MM» vs Time at 22 °C and Av = 0.15 III-.i'n 42 Mdex 100 120 .00 1 80 . 60 . 0 experiment 40 t . . calculation 20 . 0L 1 l 1 l l 0 100 200 300 400 500 600 Time (min) Figure 7. MJvas Time at 22°C and Av = 0.18 MJNLx 100 O 09 ——celculation 500 1000 1500 Time (min) Figure 8. st Time at 22 °c and Av=o.23 43 120 i l 100 - + fl— L _‘ ) 0 3° * i 2 .3 X 60 I g 0 experiment 40 ,0 -—calculation 20 .- oe + ' 0 200 400 600 800 1000 1200 Time (min) Figure 9. M. INL vs Time at 22°C and Av = 0.38 120 100 - 1 ° ° 80 t. 8 ‘- x 50 . . o emnment S. e —calculatlon 4o . 20 ,. at L 0 500 1000 1500 Time (min) Figure 10. MJMIyS Time at 22°C and Av = 0.48 120 l I I 1W> l. 40 ~ I experiment calculation Tine (min) Figure 11. MJNLvs Time at 22 °C and Av I 0.58 “x1” 120 1m .. ' A A_ A e w h 60 . 40 .. A 0 experiment 20 .. calculation 0 1 A 1 a 1 0 200 400 600 800 1000 1200 1400 Tine(mln) Figure 12. MJNLvs Time at 30 °C and Av - 0.13 45 120 Mlex 10 o A A A A A A 4A 0 200 400 600 800 1000 1200 1400 1600 Time (min) Figure 13. MJMavs Time at 30°C and Av = 0.18 120 100 .. ; L 4x F... 80 , O r- ;a 60 ‘ , S 40 0 experiment I. —calculation 20 . o A A A A A A A_ 02004006008001000120014001600 Time(min) Figure 14. MJvas Time at 30°C and Av '-' 0.24 46 120 100 - “L L0 6 8° ' O 1- ? 6°“ ‘3. 0 experiment 5 40 ‘- —calculation 0 20 l- 06 0 200 400 600 800 1000 1200 1400 Time (min) Figure 15. MJNLvs Time at 30 °C and Av =0.33 120 e 100 l. 3 0 80 ,3. g 60 0 experiment 5 —celculation 40 20 o l J l l 0 500 1000 1500 2000 2500 Time (min) Figure 16. MJMovs Time at 30 °C and Av = 0.49 47 120 . . O 100 —%4 80 l- 60 .- 1° 40 b 0 experiment —calculation 20 A ole 0 500 1000 1500 2000 Time (min) Figure17. MJNLvs Time at 30 °C and Av = 0.66 MJMJr 100 120 100- -——celculation Oexpen'ment 0 200 400 600 Time (min) 800 1000 1200 Figure 18. MJNLvs Time at 40 °C and Av = 0.18 48 —calculation 3 F £605 Oexpen'ment S 0 e - J 0 500 1000 1500 2000 Time (min) Figure 19. MJNLvs Time at 40 °C and Av = 0.21 120 0 0 CI 0 a P 0 experiment calculation m I 0‘ 1 A 1 l l J L 02004006008001000120014001600 Time(min) Figure 20. MJMDVS Time at 40°C and Av = 0.28 49 WM”): 100 120 . O 100 4 . 80 .- 0 experiment —calculation 20 0 0 500 1000 1500 2000 Time (min) Figure 21. MJNLvs Time at 40 °C and Av = 0.31 120 e 9 ° 7 100 ~ A L 60 60 0 experiment 40 -—calculatlon 20 o l l l l 0 500 1000 1500 2000 2500 3000 Time (min) Figure 22. MJMcvs Time at 40 °C and Av = 0.46 50 MtIMoo x 100 120 100- l 500 1000 1500 2000 2500 3000 3500 Time (min) 0 experiment ——celculation Figure 23. MJMovs Time at 40 °C and Av = 0.53 51 Equilibrium Solubility of 2-Nonanone in Affinity PL 1880 as a Function of Partial Pressure. The equilibrium solubility of 2-nonanone vapor diluted in nitrogen gas in polyethylene film sample was determined by measuring the uptake of 2-nonanone at different values of partial pressure and temperature. Figure 24 presents the plot of equilibrium solubility of 2-nonanone in the films expressed as gram of 2-nonanone per gram of polymer at 22 °C, 30 °C and 40 °C. From the plot, it can be seen that there is a linear relationship between the 2-nonanone equilibrium solubility in the polymer and 2-nonanone partial pressure. Table 3 presents values of 2-nonanone partial pressure and concentration. The range of concentration studied was between 0.46 to 4.21 mgIL. At this short range of concentration, the linear behavior indicates that the equilibrium solubility of this compound obey Henry’s law (Equation 6). This agreement also observed at 30 °C and 40 °C, as shown in Figure 24. At 30 °C and 40 °C, as the partial pressure increases, so does the equilibrium solubility, resulting in a higher concentration of 2-nonanone in the polymer. The total penetrant sorption process will likely involve one or more sorption modes as the penetrant molecules are sorbed within the polymer in different modes (Rogers, 1965). Futhermore, the specific sorption process may be described as a function of penetrant concentration and temperature. The relationship between concentration and temperature on the solubility of 2-nonanone can be described as a function vapor activity of the permeant. Figure 25 shows a linear relation 52 between the vapor activity and the equilibrium solubility of 2-nonanone. As the vapor activity increase, the equilibrium solubility of 2-nonanone increase at 22 °C, 30 °C and 40 °C. Temperature has a profound impact on equilibrium solubility of 2-nonanone, as shown in Figure 25. As the temperature increase, the equilibrium solubility decrease at one level of vapor activity, due to an increase in saturated partial pressure. The curves at 22 0C, 30 °C and 40 0C were closed each other due to short range of temperature studied. 22 °C 30 °C 40 °C Partial Concentration Partial Concentration Partial Concentration Pressure 2-Nonanone Pressure 2-Nonanone Pressure 2-Nonanone “’3’ (mg/L) (Pa) (mg/L) (Pa) (mg/L) 7.97 0.46 14.55 0.82 26.13 1.43 9.19 0.53 20.14 1.14 30.49 1.67 11.03 0.64 26.85 1.52 40.65 2.22 14.09 0.82 36.92 2.08 45.01 2.46 23.28 1.35 54.83 3.09 66.79 3.65 29.41 1.71 73.85 4.17 76.95 4.21 Table 3. The partial pressure and concentration of 2-nonanone at 22 °C, 30 °C and 40 °C Equilibrium Solubility (g/g) (1012 (101 ~ (1008 - (1006 - (1002 ~ 53 10 20 30 40 50 60 70 80 Partial Pressure (Pa) Figure 24. The effect of partial pressures and temperatures on the concentration of Z-Nonanone 54 0.012 I 0.01 _ E’ 3’ 0.008 - E 3 2 o 0.006 - U) E .2 5 0.004 . mac "5" I30C g -220 0.002 » 0 0 0.2 0.4 0.6 0.8 Vapor Activity Figure 25. The effect of vapor activity and temperature on the concentration of 2-nonanone 55 The Effect of Concentration on the Solubility Coefficient of 2-Nonanone The effect of concentration on the solubility coefficient of 2-nonanone in the test polymer sample at 22 °C, 30 °C and 40 °C is shown graphically in Figure 26. It was found that the sorption data could be represented by a linear function of the logarithm of the soubility coefficient versus the concentration of 2- nonanone. At 22 °C, the solubility coefficient is constant with the increase in vapor concentration from 0.4 to 2.1 ppm (wlv). There is no significant increases in solubility coefficient as the concentration increase. The same trends were observed at 30 °C and 40 °C, with the concentration ranging from 0.8 to 4.2 ppm (wlv) and 1.4 to 4.2 ppm (wlv) respectively. Table 4, 5 and 6 summarized the effect of 2-nonanone concentration on permeability coefficient, diffusion coefficient and solubility coefficient of Affinity PL 1880 at 22 °C, 30 °c and 40 °c using gravimetric technique. Figure 26 also ilustrates the effect of temperature on the solubility coefficient of 2-nonanone. As the temperature increased, the solubility coefficient of 2-nonanone decreased probably due to an increase in saturated partial pressure and relatively large heat of condensation. Solubility Coefficient (g/g.Pa) 56 50 e40C 40 ~ I30C A 22 C 30 ~ A A _A A 20 . A5..-.-.-.A_.-- I ..... . .-...._. .._._ I 1 e 10 L 3 I 4...... ..,... 0 0 0.5 1 1.5 2 2.5 3 3.5 4 Concentration, ppm (wlv) Figure 26. The effect of the concentration on solubility coefficient of 2-Nonanone at different temperatures 57 Vapor Concentration Equilibrium Solubility Diffusion Activity 2-Nonanone Solubility, C* Coefficient, S Coefficient, D (ppm) lx 10") (x 10..) (x 10°“) (WM (010) (ole-Pa) (mzlsec) 0.13 0.4 1 0.03 1.7 1 0.5 21.4 1 0.5 3.1 1 0.3 0.15 0.5 1 0.04 1.8 1 0.4 20.7 1 0.4 3.3 1 0.5 0.18 0.6 1 0.02 2.5 1 0.5 22.6 1 0.5 3.8 1 0.2 0.23 0.8 1 0.05 2.9 1 0.6 20.9 1 0.6 4.0 1 0.5 0.38 1.4 1 0.04 5.1 1 0.3 21.7 1 0.3 4.2 1 0.5 0.48 1.7 1 0.03 7.1 1 0.6 24.2 1 0.6 4.4 1 0.4 0.58 2.1 1 0.06 8.9 1 0.4 25.0 1 0.7 4.8 1 0.3 Table 4. The effect of vapor activity on the sorption of 2-nonanone at 22 °C Vapor Concentration Equilibrium Solubility Diffusion Activity 2-Nonanone Solubility, C* Coefficient, S Coefficient, D (ppm) (X 10-3) ' (x105) (X 10'") (wlv) (gig) (glgPa) (mzlsec) 0.13 0.8 1 0.02 1.8 1 0.4 12.71 0.4 4.11 0.3 0.18 1.1 1 0.05 2.2 1 0.4 11.01 0.4 4.21 0.2 0.24 1.5 1 0.04 3.3 1 0.2 12.1 1 0.2 4.51 0.4 0.33 2.1 1 0.03 3.9 1 0.3 10.41 0.3 4.91 0.6 0.49 3.1 1 0.05 8.2 1 0.3 15.01 0.3 5.11 0.5 0.66 4.2 1 0.04 11.8 1 0.5 15.91 0.5 5.61 0.4 Table 5. The effect of vapor activity on the sorption of 2-nonanone at 30 °C 58 Vapor Concentration Equilibrium Solubility Diffusion Activity 2-Nonanone Solubility, C* Coefficient, S Coefficient, D (ppm) (9/9) (X 10-5) (X 10'“) (wlv) (glg.Pa) (mzlsec) 0.18 1.4 1 0.2 2.2 1 0.3 8.5 1 0.3 4.4 1 0.4 0.21 1.7 1 0.4 2.6 1 0.3 8.6 1 0.3 4.5 1 0.3 0.28 2.2 1 0.5 3.5 1 0.4 8.7 1 0.4 4.8 1 0.1 0.31 2.5 1 0.3 3.8 1 0.3 8.4 1 0.3 4.9 1 0.4 0.46 3.7 1 0.4 7.5 1 0.1 11.2 1 0.1 5.4 1 0.5 0.53 4.2 1 0.3 8.9 1 0.5 11.6 1 0.5 6.3 1 0.5 Table 6. The effect of vapor activity on the sorption of 2-nonanone at 40 °C 59 Effect of the Temperature on the Solubility and Permeability Coefficients of 2-Nonanone Temperature has a profound impact on both the solubility and permeability of the organic volatile compound, as indicated in Equations (9) and (11). Figure 27 shows the Arrhenius plot of the solubility coefficient of 2-nonanone at vapor activity, Av = 0.1810005. The heat of solution of 2-nonanone was determined to be 32 kcallmole. An increase in temperature contributes energy resulting in an increase in segmental mobility of the polymer chains, thus a decrease in the solubility coefficient of 2-nonanone. An Arrhenius plot of the permeability constant, P, of 2-nonanone vapor through Affinity PL 1880 film vs. 1/T (°K) is shown in Figure 28. From this figure, the temperature dependency of the permeability constant can be represented by Equation (11). From the slope of Figure 28, the activation energy of the permeation process, Ep, was determined to be 37 kcallmole. When compared with the activation energy of oxygen through LDPE above Tg (E. = 10.18 kcallmole) (Mannapperuma et al., 1989), the activation energy of 2-nonanone vapor is high. ' A possible explanation for these results is that the molecular size of 2-nonanone is relatively large as compared to oxygen and requires a higher activation energy for diffusion. The strong temperature dependence on permeability of 2- nonanone which enhances polymer chain conformational mobility may affect the swelling of the polymer. Although this study was conducted at 20 °C, 30 °C and 60 40 °C respectively, the permeability coefficient at lower temperature can be extrapolated from Figure 28. Effect of the Temperature on the Diffusion Coefficient of 2-Nonanone An Arrhenius plot of the diffusion coefficient (D) of 2-nonanone vapor through polyethylene film vs 1IT (°K) is shown in Figure 29. From the slope of this figure, the activation energy of the diffusion process (Ed) was determined to ' be 14 kcallmole. The activation energy appeared to be higher than the activation energy of a permeant gas, such as oxygen through LDPE above Tg (E. = 9.58 kcallmole) (Mannapperuma et al.,‘1989). This result may be explained considering that the molecular size of 2-nonanone is larger than oxygen molecule. Michael et al., (1959) indicated that the side chain branches in the polymer chain may required a higher activation energy for diffusion. The activation energy value determined for the 2-nonanone/polyethylene system was similar in order of magnitude to the activation energy of diffusion determined for other data of organic vapor/LDPE systems. According to Laine and Osbum (1971), the activation energy of 15 organic vapors tested ranging from 6 to 30 kcallmole. In addition to molecular size, a further explanation for the strong temperature dependence of 2-nonanone diffusion through the polyethylene sample is the swelling effect by the sorbed permeant. Solubility Coefficient (glg.Pa x 10‘) 61 e e 0 e 3.15 3.2 3.25 3.3 3.35 3.4 111' (K) Figure 27. Temperature dependence of the solubility coefficient for 2-Nonanone at Av = 0.18 62 Permeability Coefficient (kg mrm2 sec Pa x 10“) 1 A A L 0.00314 0.00319 0.00324 1 0.00329 0.00334 0.00339 1rr (K) Figure 28. Temperature dependence of the permeability coefficient for 2- nonanone at vapor activity = 0.15, 0.3 and 0.5 Diffusion Coefficient 63 100 A :0 A F x o 8 10 - O "‘ n E to e 0 e015 no.3 A05 1 A 3.18 3.28 3.38 l/T (K) ' Figure 29. Temperature dependence of the diffusion coefficient for 2-Nonanone at vapor activity = 0.15, 0.3 and 0.5 64 The Effect of the 2-Nonanone Vapor Concentration on the Penetrant Permeability with a Quasi-lsostatic Technique Representative transmission rate profile curves for Affinity PL 1880 film samples at 22, 30 and 40 °C are presented in Figure 30, 31 and 32, respectively, where the total quantity of 2-nonanone permeated (Q) in micrograms is plotted as a function of time over the vapor activity range studied. The transmission rate profile curves as shown are illustrative of the effect of concentration on the permeability and lag time values. The reported data were the average of duplicate studies, since good agreement was shown by the replicate experiments, with an estimate error value of 10%. As shown, an initial induction time, a non-steady state or a time lag occured during the initial stage of the experiment as the permeant passed (through the film and established a steady-state profiles. The mass transport parameters at 22, 30 and 40 °C, that includes lag time, diffusion coefficient and permeability coefficient, calculated from these data, are summarized in Tables 7, 8 and 9, respectively. Figure 33 clearly illustrates the temperature dependence of the permeability coefficient, P, and the lag time diffusion coefficient, Dlag, with P increasing and D... decreasing with an increase in temperature. The observed temperature dependent permeability constants suggest penetrant/polymer interaction, that is swelling of the polymer matrix, resulting in configurational changes and alteration of polymer chain conformational mobility and thus of penetrant diffusivity. 65 Vapor Partial Thickness Lag Diffusion Permeability P Activity Pressure (mil) Time Coeff. Coeff. (fs) (Pa) (min) (x 10“) (x 10‘6) (mzlsec) (kg.m/m2.sec.Pa) 0.08 4.9 1.9 118 5.5 10.3 1.9 1 0.3 19 0.13 8.0 1.9 100 6.5 10.3 2.2 1 0.2 22 0.23 14.1 1.9 96 6.7 10.5 2.3 1 0.5 23 0.33 20.2 1.9 90 7.2 10.3 2.7 1 0.3 27 0.54 33.1 1.9 50 12.9106 4.8 1 0.6 48 Table 7. The effect of the 2-nonanone concentration on the permeation at 22 °C Vapor Partial Thickness Lag ‘ Diffusion Permeability P Activity Pressure (mil) Time Coeff. Coeff. (fs) (Pa) (min) (x 10“) (x 10“) (mzlsec) (kg.mlm2.sec.Pa) 0.23 25.3 1.9 59 10.9102 2.9 1 0.4 29 0.30 33.0 1.9 57 11.4104 3.9 1 0.2 39 0.42 46.2 1.9 45 14.4102 5.3 1 0.3 53 0.54 59.4 1.9 38 17.0105 5.4 1 0.3 54 Table 8. The effect of the 2-nonanone concentration on the permeation at 30 °C 66 Vapor Partial Thickness Lag Diffusion Permeability P Activity Pressure (mil) Time Coeff. Coeff. (fs) (Pa) (min) (x 10“) (x 10‘“) (mzlsec) (kg.m/m2.sec.Pa) 0.13 18.9 1.9 53 12.21 0.3 3.5 1 0.4 35 0.23 30.5 1.9 40 16.21 0.3 3.8 1 0.2 38 0.29 42.1 1.9 35 18.41 0.4 4.3 1 0.4 43 0.44 63.9 1.9 30 21.51 0.4 5.5 1 0.3 55 0.56 81.3 1.9 20 32.21 0.3 7.0 1 0.3 70 Table 9. The effect of the 2-nonanone concentration on the permeation at 40 °C Quantity permeated (ug) 100 90 80 70 60 50 40 30 20 10 67 O 0 e 0.54 I 0.33 - 0.23 A 0.12 e 0.08 I ._ -__.__——-— uni-fl " 50 100 150 200 250 300 350 400 Time (min) Figure 30. Transmission rate profile of 2-nonanone at 22 °C and various vapor activities Quantity Permeated (ugl 68 160 140 ) ° ”-54 no.42 -0.3 12° ' A023 e012 100 » 80 » . - 60 - ' ' ll 40 - 20 t 0 réf-L-‘EF' - 0 50 100. 150 200 250 300 350 400 Time (min) Figure 31. Transmission rate profile of 2-nonanone at 30 °C and various vapor activities Quantity Permeated (ug) 180 160 140 120 100 80 60 40 20 69 O 0 I ' e 0.56 l(144 -(129 —(123 , e 0.13 50 100 150 200 250 300 350 400 Time (min) Figure 32. Transmission rate profile of 2-nonanone at 40°C and various vapor activities Quantity Permeated lug) 70 140 ~ 022C 12° ' A30c 0400 100 ~ 80 . 60 - ll 40 . o O 20 - o _. ’4 _-’-" A . . 4 0 100 200 300 400 Time (min) Figure 33. Transmission rate profile of 2-nonanone at vapor activity = 0.23 71 Figure 34 demostrates that the permeability coefficient (P) is exponentially dependent on the penetrant concentration as found in the literature (Zobel, 1982,1985; Baner et al., 1986; Hernandez et al., 1986). The observed concentration dependency of the permeability coefficient shows penetrant/polymer interaction occurred during the experiment resulting in swelling of the polymer matrix by 2-nonanone vapor and corresponding changes in polymer chain conformations, leading to an increase in penetrant diffusivity, and therefore permeability. The diffusion coefficient values of 2-nonanone were determined from the permeability data based on Equation (13). As shown in Figure 35, there is an increase in diffusion coefficient with an increase in vapor activity of 2-nonanone over the vapor activity range studied. The diffusion coefficient values are calculated from the transient state region of the transmission rate profile curve because at the steady state penetrant/polymer interaction may lead in a gradual relaxation of the polymer structure, resulting in a change in the free volume of the polymer. In addition to molecular size, another possible explanation for the strong dependence of 2-nonanone diffusion through polyethylene film is the swelling effect by the sorbed permeant. In Figure 35, as the temperature increases, the diffusion coefficient increases. The polymer segmental mobility is greater and it is easier to increase the free volume. The magnitude of the dependence of the diffusion coefficient on the concentration is influenced by the temperature prevailing during the transport process and the molecular size of the penetrant. 72 10 x 10“) Perrneabllity Coefficient (kg.mlm’.sec.Pa 1 I L I I l 0 0.1 0.2 0.3 0.4 0.5 0.6 Vapor Activity Figure 34. The effect of 2-nonanone vapor activity on Log P at different temperatures Diffusion Coefficient (mzlsec x 10“) 73 100 D/ 10 ~ 0 O 0 e 22 C l: 30 C -4oc 1 I L I I I 0 0.1 0.2 0.3 0.4 0.5 0.6 Vapor Activity Figure 35. The effect of 2-nonanone vapor activity on Log D at different temperatures 74 Package Design for the Release of 2-Nonanone in the Headspace of the Package 2-Nonanone is one of several promising antifungal natural volatiles that has been used to prevent decay in packaged fruit (Vaughn et al., 1993). To develop a model that can predict the package headspace concentration after flushing the package with antifungal volatile or by adding it into the polymer material itself, requires knowledge of fungi response, product’s toxicity, mass transport and solubility parameters of the antifungal volatile compounds in the polymer material. The optimum concentration level of 2-nonanone in the headspace of the package that can prevent decay in fruit should be investigate in order to develop the actual product/package system (Leepipattanawit, 1996). The permeability and solubility characteristics of 2-nonanone have been studied in this work and for relatively high concentration it produces some swelling of the polyethylene. We assume the 2—nonanone desorbed into the headspace of the package is half of the 2-nonanone initial content in the polymeric film. The concentration level of 2-nonanone in the headspace can be expressed as, Czu = M2NNHS 118) 75 where C2" is the concentration of 2-nonanone in the headspace, M2" is the amount of 2-nonanone desorb into the headspace of the package and VHS is the volume of the headspace. VHS is related to the total package volume by, VHS = VPKG - VPRT (19) where me is the total package volume and VP." is the product volume. According to Leepipattanawit (1996), 300 x 10‘ liter of liquid 2- nonanone per liter of air in the headspace is needed in the headspace of the packaging system to control the fungi growth of sliced apple at room temperature. The first method that can be applied to maintain the concentration of 2-nonanone in the headspace of the package is by assuming that 2-nonanone will be initially sorbed in the film of area, A, and it will be desorbed into the headspace after the package is formed. Table 10 summarizes the package area needed to control the microbial growth in the package at vapor activity ranging from 0.13 to 0.58 by assuming that 50% of the organic vapor desorbed into the headspace from the film which is given by the following equation, Y = 55.695X"-‘°18 (20) where, Y = package area (cmz) X = vapor activity of 2-nonanone 76 Figure 36 illustrates the relationship between the vapor activity and the package area. As the vapor activity increased, the equilibrium solubility, C*, in the polymer increased, which will reduce the area needed to develop the packaging system for sliced apple. A typical package can be a pouch or a covered tray as shown in Figure 37. Vapor Partial Equilibrium Weight Area Activity Pressure Solubility Polymer Package (Pa) (mom/1000090 (m9) (000’) . (t=1 .9mil) 0.13 7.96 ‘ 0.0017 2231 512 0.15 9.2 0.0019 A 2002 460 0.18 11.03 0.0025 1527 351 0.23 14.09 0.0029 1291 297 0.38 23.28 0.0051 751 172 0.48 29.41 0.0071 535 122 0.58 35.54 0.0089 427 98 Table 10. Package design in relation with equilibrium solubility of 2-nonanone at 22 °C 77 Another method that can be applied to maintain the concentration of 2-nonanone in the headspace of the package is by including a small pouch with 2-nonanone liquid in the package as illustrated in Figure 38. By assuming that the transmission rate of 2-nonanone from the small pouch is equal to the transmission rate of 2-nonanone through the package, we can get the ratio between the area of the small pouch and the package at various material thicknesses and vapor activities. In this method, we want to maintain in the headspace a target concentration of 300 ulll during 3 weeks. We also need to calculate the thicknesses of the small pouch, A, that will give the same rate of permeation through package, B. Transmission rate across A, Q1It1= P1A (p. -p1) I I, (21) Transmission rate across B, 02/12 = 1323 (p1- p2) I I2 (22) where, A and B = area of the small pouch and package respectively P1 and P2 = permeability coefficient Or and Q; = quantity permeated t1 and t2 = time p. = saturated partial pressure p, and p2 = partial pressure l1 and I2 = thickness 78 01,11 = 02/12 = P1A (ps’p1) / I1 = P23 ([31 - ()2) I 12 (23) Assuming that P1 = P2 A (P: " 91112 = B (91111 A/ B = (911 l1/(Ps-Pt) I2 (24) Assuming that the area of the small pouch is 8 cm2, Figure 39 gives the area of the package 8 as a function of 2-nonanone activity in the headspace, for three values of 11,12. Table 11 summarizes the total quantity of 2-nonanone to be placed into the small pouch to produce a headspace concentration of 2-nonanone as a function of lrllz. Calculation of the amount of 2-nonanone needed in the small. pouch to keep the 2-nonanone concentration in the headspace of the package B during 21 days from Equafion(211. For a specific activity value, at vapor activity = 0.33 partial pressure, p, = 20.2 Pa. partial presure, p2 = 0 saturated partial pressure, p. = 61.3 Pa permeability coefficient = 2.7 x 10'"3 kg.m/m2.sec.Pa thickness = 2 mil pouch area = 8 cm2 0 = (2.7 x 10"°)(8 x 10")(61.3-20.2)(21 x 86400) I (2 x 25.4 x 10‘“) 3.17x10'7 kg 0.317 mg 79 Ratio of 0.25 0.30 0.50 Thickness (11/ 12) Vapor Concentration Amount of 2-Nonanone Activiy (mg/L) (mg) 0.08 0.27 0.30 0.18 0.15 0.23 0.8 0.31 0.19 0.15 0.33 1.15 0.32 0.19 0.16 0.54 1.88 0.38 0.23 0.19 Table 11. Total quantity of liquid 2-nonanone to be place into the small pouch to produce a headspace concentration of vapor activity as a function of I ll; 80 600 a" s 400 9. I! 0 h < l' 0 a) . N .1 '5 : g 200 " y = 55.695x01.1018 o t 4 4 A 1 0 0.1 0.2 ‘ 0.3 0.4 0.5 0.6 Vapor Activity of 2-Nonanone Figure 36. The vapor activity of 2-nonanone vs package area using desorption process from the polymer film 81 50%of2-naxrn’iedescrbwldlhepodmm 50%01‘ 2—nonmone dsabintolhe pod