ran ‘ "' 3...— - .-“.-.m’- ._ ‘9.— WA”.- v 9—- A 51qu OF THE EFFECT OF sremuzms ERRADIATION on THE PERMEABILITY OF CELLULosscs AND PETROLEUM wass Thesis for fho Dam“ of M. S. MICHIGAN STATE UNIVERSITY- Hugh ”Elvin Lockharf 19:60. fl T‘r,"l"‘!") ’ r'1 4L 1- nuCl The effect of two levels of sterilizing reflifltion tv of cellulOS'es and petroleum wax d i—h on the permeehil $7- 1‘.‘ ‘4. was stuflied. on-red eted samples were iroluded as a on levels unfier con— lJ. control. The particular rediat sideretion, one and six mefierods, fall into a borfler- 1ine erea between week and strong radiation. The findings of this study indicate that the effect on water vapor permeability is not] r“i‘ole. The errerimertsl work involved included selecti-n of an effective rethod for costind SO wil films of methyl cellulose and petroleum wax, perforwsrce and evaluation of water vapor transwission tests, and co struction of gas rerueebility apnaretus. A STUDY OF THE EFEECT OF STERILIZING IRRADIATION ON THE PERMEABILITY OF CELLULOSICS AND PETROLEUM WAXES B7 HUGH ELVIN LOCKHART A TLESIS Submitted to the College of Agriculture MicLican State University of Acriculture and Applied Science in partial fulfillment of the requirements for the degree of School of Peckafiing 1960 J ._‘\ ‘) A ‘ K1150 . I L JG}? IEZTT This study was pursued under the Sponsorship of the Quartermaster Food and Container Laboratory, Chicago, Illinois. Special thanks are extended to Dr. H. J. Raphael for his assistance and encourséenent during the course of this effort. Beverly Lockhart shouldered cheerfully the usual burden which hefslls the wife of a graduate student in undertaking the task of typinq. r1 '\ ‘1 “1 '1 I1"- - 1 1-7.1 :1 .Ligd-‘IJJ‘J O]? GUI: ‘1‘1‘Ji. .1. '~ 3 INTRODUCTION Purpose of Study __________________________________ _ 1 Scope of Study ------ ——-——— _________ _ ___________ ____ 2 Definitions and synlan tions ———————————————————— __- 3 BACKGROUND Irradiation --------- --- —————— —-—_- _____________ ____ o Permeability --------------------- - ______________ -__11 EI§TIIEIIU IIL-II P. KDC1I)U i5 Prel111ntry .ork _________________________________ __16 *jn-ter Vicki-JO? Tlti r631ir)sion Rate ————————————— ——————-——19 Oxygen Permeability ——————————————————————————— _—_—_22 TSTVL‘“ Data Irestvent-fieter Vaoor TIansnission Hate-------32 Juter Vapor lra Snission Rate Results —————————————— 33 Oxygen lermeability ------------------------- -------30 ONCLUSIOIS 1ND IJCQTTJUJLlJOIS Hater Vapor Lnsqsnlssxon Rate ---------------------- 37 Oxvf;on Iormenoility ——————————————————————————— —----37 411 1‘71:)12Jlig I [33019 83 Of fie C) --------- - ———————— -——-———--39 s" s l- 112".nslVS1:ables I—VIIf---~-~- ----------- ---———-—49 / BIBLIIJI'LL'XJLV iii. ———————— c— —————— —————— ——————————— I- —————— -—OO INTRODUCTION Purpose of Study This thesis covers approximately one half of a project undertaken by the writer for the United States Army Quartermaster Corps Food and Container Research Laboratories, Chicago, Illinois. It is anticipated that in the future most if not all of military food packaging will involve sterilization by irradiation as Opposed to such methods as sterilization by heating, or arrest of bacterial action by freezing. The reasons for this are readily apparent. Once irradiated, foodstuffs can be stored indefinitely under a wide variety of con- ditions, ranging from.below zero cold to trepical heat and humidity. The only requirement to be met is that of providing a barrier against contamina- tion during storage. For meats, one possible barrier (or package) is a dip coating of same kind. This kind of package is comparatively inexpensive, is easily and quickly applied, and is generally easy to open, since, ideally the dip coating material needs only to be slit once, then stripped off the product. In addition, a properly applied dip coating should provide a "per- feet" seal with no joints or breaks to give discon- tinnities in the package. Since irradiated foods need no special storage conditions, the ideal package for these foods would be one which can follow the product. That is, one which itself will withstand all conditions from arctic subzero cold to tropical heat and humidity. It is over this entire range that military food must be protected. Ideally, any given package should give protection over this range, thus elime inating many'meterials and many steps from.the handling of military foodstuffs. With.these considerations in.mind, this phase 'of the study concerns itself with the water vapor and oxygen permeability characteristics of dip coating materials before and after sterilizing irradiation doses when subjected to varying conditions of temperature and humidity. Scope of Study Basically, the Quartermaster Corps contract calls for investigation of two commercial cellulosic- materials and of waxes available from.one manufac- turer. The cellulosic materials are Ethocell, manu- factured by Dow Chemical Company and Wasc0pak manu- factured by Wasco Chemical Company. Ethocell is an ethyl cellulose and the Wascopak is believed to be the 88.1116. The difference between the two is most likely in the plasticizers used. These, of course, are not known. The wax selected is a blend of 70%9m1cro- crystalline, 20% paraffine and 10% butyl plastici- zer, manufactured by Kalamazoo Paraffine Company. In addition, it was requested that such.other materials as were discovered be investigated. In addition to the tests mentioned above, blocking tests and impact tests were to be conducted on all materials. In the interest of7brev1ty, coher- ence, and continuity, only the permeability tests on the cellulosics and the waxes have been selected for discussion in this report. Definitions and Explanations Dip coating is a definitive term. The product to be coated is dipped in the coating material. Normally this is a hot melt process, with cellulo- sics heated to 300°F - 325°F and waxes - those used in this study - heated to 200°F - 220°F. The coating material adheres to the product at a thickness deter- mined by viscosity, melt temperature, temperature of the dipped product and time of immersion. Generally, coating thickness increases with higher viscosity, lower temperature of melt and lower temperature of product. Longer immersion time also results in thicker coating. To say that the coating adheres to the product is not entirely true, at least not in high degree. Actu- ally, an ideal dip coating would not adhere to the product at all, but would solidify around it and by cohesive force alone seal it off from the environment. The better dip coating materials approach this ideal closely, hence their ease of removal. It has been noted that in cases where spots of good adhesion have occurred, small pieces of meat have been torn out of the block of meat upon removal of the coating. This effect was ordinarily associated with.localized thawing of frozen:meat before and during dipping. This ine dicates the importance of dipping frozen meats. Water vapor transmission rate or water vapor permeability is defined as the weight of water transmitted per unit time per unit area of test specimen. This is easily determined as the slope of the plot of weight of water gained by a dessi- cant versus time. The transmission of water vapor is usually not constant over an entire determination, but displays considerable variability during the first few readings and again at the end of the test when the dessicant approaches saturation. For most purposes the slope of the curve is computed from.a curve drawn on coordinate paper. However, for this study the slopes were determined statistically in order to obtain a quantitative es- timate of the variation occurring. water vapor permeability data are, of course,a direct measure of the rate at which a material transmits water vapor to or from.the item it en- closes. These data are adequate for the compar- ison of different materials, but for a good est- imate of package performance more information.must be known. Of prime importance is the amount of moisture gain or loss the product can tolerate. Also of prime importance is information regarding such discontinuities in the materials as seals and seams. In the particular case of dip coatings, discontinuities may occur at the point of suspenp sion of the product or at points of contact if the package should be handled prior to complete setting of the film. .The cellulosic materials studied are subject to the accumulation of air bubbles within the melt during melting and dipping. These, if included in a coating, would contribute discontinuities to the coating. Because of the unpredictability of all of these sources of discontinuity, every effort_was made to avoid their inclusion in the study. The results reported are for clear films without de- tectable discontinuity. Gas permeability is a concept similar to water vapor permeability. By definition, gas permeability is volume of gas transmitted per unit area per unit, time at standard temperature and pressure. In this study, the test gas is oxygen. The interest in oxygen arises from the fact that this gas has a deleterious effect on meat, affecting color and palatability. The same arguments and restrictions apply to use of this data as were stated in the para- graphs referring to water vapor permeability. It has been determined that a proper sterilizing radiation dose for foods is on the order of six megarads radiation. 1 Dunn, Campbell, Pram and Hutchins (11.)2 report that spore forming micro- organisms are difficult to eliminate and that these are completely destroyed by 2 mega roentgens ra- diation. They report also that all of the micro- organiSms in.raw:milk and soil have been destroyed by’l mega roentgen. The mechanism.of radiation sterilization has been stated nicely by Trump and van de Graff(13): "In.the case of living tissue, injury - and in the limdt death - of a cell is the result of physicochemical changes brought about by direct absorption of ionizing energy or indirectly by the reaction of surrounding irradiated tissue." -—TTT-?;T;Ete conversation with Professor Lyman J. Bratzler, Department of Animal Husbandry, Michigan State University. (2) Numbers in parenthesis refer to references appended. We will see later that this mechanism.is really a modification of the general effect of radiation upon materials in general. Radiation dosage is computed on the basis of radiation energy absorbed. Thus, any source of energy may be used to produce a given dose. The amount of energy required to produce a particular dose will depend on the dosage unit used, the material being irradiated and the particular radiation source involved. Three units of dosage are in common use, the rad, the roentgen and the rep. The rad is defined as the absorption of 100 ergs radiation energy per gram of material irradiated. The roentgen (r) is defined as 83.8 ergs per gram of air. The amount of radiation required for this will produce 93 ergs per gram of water or tissue. The rep (roentgen equivalent physical) springs frmm the difference noted above for the roentgen. This unit is equivalent to 83.8 ergs radiation energy absorbed per gram.of tissue or water(12). For purposes of comparison, the following radiations will produce a one roentgen dose of radiation absorbed (12): 1.1 x 10 2 mevr rays / cm2 7.2 x 10 thermal neutrons / 'm 7.2 x 107 2 mev neutrons / cm ll..3 x 107 l mevfl particles / cm 2 2 . 100 A one rad dose would then be produced by 83.8 times any of the above source radiations (e.g. 1.3 x 109 2 mev X rays / c1112). The general effects of radiation on plastics are quite extensive when they occur. Among these effects can be listed(12): Breaking of chemical bonds and for- mation of free radicals Dissociation, or degradation of the molecule, and Rearrangement of the molecule. These effects may occur at different levels for different molecules; or, at low radiation levels, they may not occur at all. BACKGROUND Irradiation The effect under investigation is that of irradiation upon three specific materials, two cellulosics and a microcrystalline wax formula- tion. The literature reveals some disagreement among researchers as to the extent of the effects of radiation on cellulosics as used in packaging. One thing is clear, however. The one megarad dosage included in this study is well below the dose level Which causes change in cellulosics, while the six megarad dosage falls in a rather ill- defined area where, apparently, changes may or may not occur. Saeman, Lawton and Millett(ll) report the depolymerization of cellulose at 107 roentgens, but little effect at lOSroentgens. Karel and Proc- tor(6) report a deleterious effect on cellophane at doses above 3.5 megareps, but attribute this to the effect of radiation on coatings and plasticizers. For example, they found the WVTR of coated cello- phane to be as much as 2 - 2.5 times greater for dosages of 7 megareps than for non-irradiated film. The essence of the theory of the damaging effects of irradiation has been stated by Slater(lO); "Radiation damage is the result of displacement of atoms in a solid by particles passing through it, of the particles which.remain embedded in the solid as impurities, and of the ionization produced by the particles." He modifies this statement with the explanation that the effect of radiation on covalent compounds is one of electronic excitation rather than atomic. In the covalent bond, electrons are displaced from.one configuration to another and are there trapped for a relatively long time. This effect can occur at relatively low energy levels. It is not necessarily damaging, since the shift in configura- tion could bring about an improvement in properties as is the case with crosslinking(12). Charlesby(3) points out that in order to be greatly affected, a plastic must consist of linear or branched chains. Here, then, is an indication of what might be ex- pected. The cellulose molecule is of course not linear or branched, its configuration being a three dimensional ring structure. This would indicate that a cellulosic material ought not to be greatly affected. However, the plasticizers used in cellophane and other cellulosics may be of a linear or branched chain arrangement,(6) and here is where the effect may occur, as was found experimentally by Karel and Proctor as mentioned above. With reference to the waxes, the wax itself is a linear molecule(5) so it seems reasonable to ex- pect that there may be some effect on the material as a result of irradiation. Permeability The mechanism of permeation through a film for water vapor and the permanent gases (02, N2, H2) is essentially the same. Obviously, in the presence of holes or cracks, the movement of gas or vapor is a mass movement, or diffusion governed by laws of flow through tubes and orifices (1,8). This, how- ever, is considered to be outside the area of in- terest in the investigation of gas barriers for packaging. In the absence of these faults in the barrier, the mechanism is one of activated diff- usion(9). That is, the gas or vapor dissolves in one surface of the film, diffuses through the film under a concentration gradient and evaporates from the film.at the low concentration, or low-pressure side. Under steady state (equilibrium) conditions, Fick's law has been found to apply and the trans- mission proceeds according to: do q== ~D d3: where q is the amount of gas diffusing through unit area in unit time, D is the diffusion con- stant and dc/dx is the concentration gradient across thicknes 3 (1X0 Assuming D to be independant of concentration, this expression can be integrated across the total thickness of the rilmf to obtain: <1: D (0% -02) where c1 and c2 are the concentrations of the gas in the two surfaces. Normally gas concentrations are measured in terms of the pressure, p, of the gas which is at equilibrium with the film, and c can thus be exp pressed by means of Henry's law c:= S”. p where S is the solubility coefficient of the gas in the film. Thus, q= D-S (P1 -P2) .1 The product D-S is referred to as the permeability constant P, and we have P: D°S 1'9 (P1 “ P2) This treatment assumes that D and S are independent of concentration. This is generally true for per- manent gases, but deviations occur for water and organic vapors. P is always found to be independent of thickness.(9)3 Rogers and associates(9) and Ridde11(8) have shown that the usual xpression for P is: -T§7_Thi§_development of permeability theory has been taken from.Rogers et a1 "Studies in the Gas and Vapor Perm ability of Plastic Films and floated Papers,’ Reference 9. where Q is the total volume of gas passing through the film, I is the thickness of the film, A is the area under test, t is the time, and dp is the pressure differential between the two surfaces of the film, The units for P are usually expressed as cm} of gas at S.T.P. per second per cm.Hg per mm thickness per square centimeter surface area. .tPhysically, diffusion can.be thought of as occurring through inter atomic "holes" which in- evitably occur in.the tangle of polymer chains in the essentially amorphous film. Then, to an extent, the diffusion is dependent on the looseness of the polymer structure. Anything which affects the mo- bility of the structure affects the diffusion. Mdbility refers to the freedom of the atomic structure to move; lengthen, contract, entangle or disentangle. During all of this, a greater or lesser number of holes is appearing, disappearing and reappearing. In a sense, greater mobility imr plies greater effective looseness. Thus, addition of plasticizers, or increase in temperature will increase the mobility of the polymer structure and so increase the diffusion through the film. Diffusion necessarily depends on the size relation- ship between the interatomic spaces and the size of the diffusing molecule. Thus, Rogers and associates report a regular decrease in diffusion for the series oxygen, nitrogen, carbon dioxide. Solubility is highly dependent on the principle of "like dissolves like." In the case of gases, condensation is of even greater importance, the more easily condensed gas being the more soluble. Barrer(1) reports a linear relationship between the logarithm.of solubility and the boiling point of a gas dissolved in rubber. V Permeation by water vapor follows the pattern outlined above, but certain special effects must be considered. Many films (e.g. cellulosics, nylon) are hydrophilic substances. These films are plasticized by the water vapor, and diffusion increases with increased plasticizing. This explains the observed phenomenon of increased water vapor permeability with increased relative humidity. The behavior of waxes when exposed to gas or vapor conditions conducive to permeation is characteristic of crystalline substances. In a crystalline substance, "holes" do not exist within the crystal, that is, there is no free space for the unhindered passage of a molecule of vapor or gas. In addition, the solubility of gas or vapor is de- creased by increasing crystallinity. Fox(S) des- cribes waxes as platy, crystalline substances. He reports the failure of waxes to follow the permea- bility statement of Fick's law, and attributes this to the platy structure. He proposes that the only practical path for gases and vapors to follow through a wax film is one which requires the molecule to seek a route through the microscopic cracks which exist between the crystalline plates existing in waxes. This argument seems to explain the extremely low permeabilities reported for waxes. Brabender(2) reports that a 100%:micro- crystalline wax passed only 0.01 grams of water vapor through ten square inches surface area in twenty-seven days. EXPERIMENTAL PROCEDURE Preliminary‘Work The first step in the investigation was that of determining the thickness of coat of each.material when dipped at the manufacturer's recommended temp perature. Here a direct approach was used. Frozen meat was cut into blocks approximately 2" x 3" x l", the size being dictated by the size of the melting pot available. The meat blocks were then dipped into the molten material and air cooled. After cooling, one-half inch strips were cut out of the coating, the length of the strip running from tap to bottom.of the meat chunk. The thickness of the strips was then determined by means of a T M I deadweight micrometer. It was found that there was a regular increase in thickness from.top to bottom.for all samples. This is attributed to the considerable run down of'molten material which was noticed during the cooling period. For the purposes of this study an average was taken of all readings top to bottom. Here the tacit assumption was made that the variation in permeabilities would be linear with variation in thickness, thus allowing the taking of averages on a simple numerical basis. The averages determined were #8 mils for Ethocell; Liz mils for Wascopak; and 65 mils for microcrystalline wax. Having obtained an average coating thickness, it was necessary to cast sheets of the materials at the average thickness determined for each. It was desirable to cast fairly large sheets, since one of the tests to be used requires a sample of five inches diameter. After some experimentation it was found that the most effective casting plate for the cellulosics was one of aluminum,-% inch.thick. This thickness was found necessary to support the weight of the material without bending. The large mass of metal also speeded cooling of the cast material. The sides of the plate were built up to the required thickness with.masking tape, this being rapidly applied and easily varied as to height. The most easily handled sheet size was found to be 6" x 12", two sheets being cast on a 13%" square plate. Once the material was poured on the plate, it was necessary to remove the excess with a doctor blade. 18 To avoid too rapid surface cooling and consequent scuffing of the cast sheet, the doctor blade was heated in a 100°C oven for several hours and then returned to the oven immediately after each use. While it was essential not to chill the air-sheet interface, it was found just as important to pro- vide a quick chill at the plate-sheet interface. With.anything less than a quick chill, adhesive forces came into play making removal of the sheet difficult or impossible. Proper use of the doctor blade proved to be an art. Variations in pressure and speed of travel across the sheet surface could cause variations in sheet thickness of as much as 20-30 mils. Practice, however, yielded results, and samples were obtained with extreme variation of S mils on either side of the desired thickness. Casting of the wax involved yet another problem. The adherent properties of the wax are marked. A l" aluminum.plate was used, but this did not pro- vide a sufficiently rapid chill to prevent adher- ence. It was decided that a safe release agent would be petrolatum. This proved to be effective. Petrolatum was selected because of its similarity to the wax itself. Excess petrolatum.was removed from the sample by wiping first with a dry cloth and then with a cloth.saturated in acetone. No special procedure was used to obtain re- presentative samples, since in.most cases only small lots were received from.the manufacturer. In addition, it was assumed that the melt process of manufacture would provide uniformity within each batch, at least. water Vapor Transmission Rate The test used for water vapor transmission was the Technical Association of the Pulp and Paper Industry (TAPPI) Standard T 14118m-Ii9, water Vapor Permeability of Paper and Paperboard. This test was specified by the Quartermaster Corps for water vapor permeability tests. Briefly, the method specifies that a membrane of the material to be tested shall separate an atmos- phere below 5% relative humidity at 73°F from.one of 50% relative humidity and 73°F. The low R.H. atmos- phere is attained by means of a dessicant contained in a test cell, the mouth of which is covered with the test material. The edges of the Specimen are sealed into the cell by means of a wax mixture of 60%:microcrystalline wax and h0% crystalline par- affine wax. The test area is defined by a template placed on the specimen around which the molten wax. is poured. Shrinkage of the wax during cooling is negligable, so the test area is defined as being equal to the area of the template in contact with the Specimen. This area must be accurately known for computation of the water vapor transmission rate. The area was found to be 53.28 cm? and Sh.20 cm? for the two temp plates used. The test cell is a small dish about 3 inches in diameter, with a grooved lip to accept the edge sealing wax. (See Figure I) The dessi- cant used was silica gel in a quantity of about 30 grams occupying approximately ho cubic centi- meters volume. The same method was used for evaluation of water vapor transmission rate at lOOOF and 90% relative humidity, which conditions were additionally specified by the Quartermaster Corps. Preliminary tests showed that for the Wascopak at 50% and 73°F an inflection point was reached at about the 130th hour. From 0-60 hours a straight line was obtained, and again from.l70-280 hours a straight line was obtained. It should be noted that this second straight line had approximately the same slope as the first segment. For this reason, and because the first 60 hours is probably the most significant in applications to food, it was decided i .‘ImjjTl 1‘5“.) 0 _~«.. 0 1' , “(J ('2 ( ry- r. m . ~-- ‘1 to continue the tests for sixty hours then stop. Five specimens of each radiation level (0,1,6 megarads) were tested for each sample of dip coating material. In most cases nine readings of weight were obtained on each specimen before the sixty hour period elapsed. Oxygen Permeability The Quartermaster Corps specification for this test was only that a currently approved standahd test be used. Investigation revealed that a test develop- ed at Dow Chemical Company was winning wide favor. In fact, a standard test recently published by the American Society for Testing Materials (ASTM Designation Dui3h-56T Tentative Method of Test f9; Gas Transmission Rate of Plastic Sheeting) is based on this test. The ASTM test was used for this deter- mination. Basically, the method of conducting the test is as follows: A sheet of the material under test is exposed to pure oxygen at a pressure differential through the sheet of one atmosphere. The flow of oxygen through the sheet is measured by means of its effect on a cap- illary mercury column, the rate of depression of the column being proportional to the rate of flow through the sheet. The apparatus used consists of a vacuum.tight cell and manometer (See Figure 2) in which the spec- imen is tightly clamped. The gas is introduced at the tOp of the cell by means of appropriate tubing from a pressurized cylinder of gas. (See Figure 3) The system is closed at a vacuum.pump which.serves the purpose of evacuating the system prior to in- troduction of the test gas (See Figure 3). The en- tire system is evacuated and flushed with the test gas several times to insure that the permeability being measured is that for the test gas only. In the final step of preparation, the system is evacuated, the cell valve closed, and the mercury dumped into the cell manometer from the storage bulb. Then the gas is introduced into the system.to a pressure differential of one atmosphere as measured on the U-tube manometer (See Figure 3). One of the cells used in the apparatus was pur- chased from Custom Scientific Instrument Company Model CS—89. The other three were manufactured locally. The cell manometers for the locally manufactured cells were purchased from.H. S. Martin and Company, Chicago and were calibrated in the School of Packaging laboratory. 0. US“?- vs Y» 1/‘7' ,aL.‘ The first step of the calibration required the deter- mination of the volume of the capillary in cubic millimeters to three significant figures. This was done by mercury emplacement. A mercury pressure system.was devised using a Yale hypodermic syringe of 50 cc capacity. This was mounted in a locally manufactured screw operated closing device (See Figure 5 ) . The entire system.was filled with.mercury and attached with a spring clamp to the inverted cell manometer (See Figure A). A glass tubing connector was flanged and ground in the packaging laboratory. The face of each flange was lightly greased to aid in sealing. Between the two flanges was placed a three component slide system. The slide was copper shimtstock 0.012" thick, about two inches long by three eighths of an inch wide with a one eighth inch hole drilled in.the center (See Figure h). This was placed across the middle of the lower flange area. Two smaller pieces of the same material were placed on each side of the slide to provide support. The manometer was then lowered into contact with the slide assembly and the whole assembly clamped to- gether. By means of the screw operated hypodermic plunger, the mercury was forced into the capillary. Figure h Thrcury 'huttcr, Hanomctcr Calibration liéuro S Care was taken to obtain as flat a meniscus as pos- sible.' This meniscus was then aligned with one of two marks scribed on the tube. The slide was then pushed through, sealing the mercury into the capillary. Then by weight difference the amount of mercury and consequently the volume of the capillary were de- termined. This volume was determined for two sec- tions of the capillary - one from the opening to the first scribed line, the other from.the first line to the second, a distance of 65 mm.1 0.2mm. These volumes were determined by the above method with a variation.between highest and lowest values which was of the order of 1% of the average volume. At least three determinations were made for each vol- ume of each tube. From the volume and the distance between the scribed lines, the area of the cap- illary was determined. These values will all be used directly or indirectly in later calculations. The remaining volume - in the body of the cell - is determinable from the direct measurement of the openings in question. These measurements were made with a Starrett No. 122 Vernier Caliper graduated in 0.001" on the vernier. The problem of constructing an adequate vacuum system consumed a large portion of the research time available. The method used specifies that a vacuum of 0.2 mm mercury or better shall be attained. This is not considered to be a high vacuum.and so this range is not discussed in high vacuum.literature. Experience proved, however, that this degree of vacuum.is neither easily attained nor easily held once reached. Since the vacuum was originally felt to be rather easily attained, a plastic tubing vacuum system.was tried, using polyethylene tubing with brass compression fittings manufactured by Dekoron Products Company. The highest vacuum.ob- tained with this system was 5 mm, probably as a result of the multiplicity of threaded fittings in the system. The system was then converted to copper tubing, using solder joints and flare fittings treated with glyptal. The flare fittings were in- cluded for easy take-down of the system when nec- essary for cleaning purposes. This system.attains a vacuum.of 0.8 mm.mercury using a Cenco Hyvac 2 vacuum.pump and cold trap to remove oil and water vapors. Cooling is accomplished with acetone and dry ice. For the sake of efficiency, and in order to have a continuous record, electrical recording of the mercury depression in the cell manometer was selec- ted. A platinum wire was incorporated in the cap- illary leg of the manometer by the manufacturer. This is connected in parallel with a 50 ohm variable resistor to a Minneapolis-Honeywell Model Y1§3 x 6h - W8 Brown 8 point recorder with range 0-2 ohms plus an adjustable padding resistor for range 0-8.7 ohms. This arrangement provides for range spreading as well as for zero adjustment. The changing resistance as the mercury column rises or falls is in this way recorded automatically for each cell. This system, of course, required calibration and this was done as follows: mercury was admitted to the manometer and the level adjusted to zero reading, the level reading being made with a cathetometer. The mercury column was then lowered by steps until several different levels had been read with the cathetometer. For each.mercury level reading, the recorder position was recorded, and a graph prepared of mercury level versus recorder pen position (See Figure 6). 30 T“ (.3 .22mm0uLoQ Audueo: a w a wand u one.“ fie; gfm.~ SEHULQQQU LQVLQOUQ Q euhmlk 4Q~ 2 Unity Recorder The slope of the line so determined is taken as the recorder travel constant to be used in later cal- culations as specified in ASTM test le3h-56T. RESULTS Data Treatment - Water Vapor Transmission Rate Raw data (cumulative weights) were converted to weight gained, and this data was treated statistically. The regression of weight gain on time was calculated using a standard regression analysis (see sample calculation, Appendix II, Table VI). Where lack of significance of the re- gression was indicated, the regression coefficient was not calculated. The coefficient has the usual statistical notation, "b". For the Wasco material, two cases were found, by means of the gap test or the test for stragglers, in which the values were "wild". In order to come plete the variance table, these were treated as missing values and an appropriate value was found by using a series of approximations. In determining the effect of radiation levels, a two way analysis of variance was employed using the "F" statistic. The variation,/(_Illrs g/m / hhrs UaSCOpak O .OOBh 11.9 , 1 .00263 12.6 26.0 6 .00256 12.3 Dow C .00115 5.52 Ethocell 1 00107 5.1h .h32 6 .00109 5.23 hicro- 0 3.66xlO_§ .176(53pec.) Crystalline 1.ibx10 , .117(ns;ec. ).0816 max 6 -7.1 x10 5 -.3h1(1spec.) Table 3 Thickness of Ester Vapor Transmission Specimens (All Tests) Specimens tested at lOOOF, gap R.H. Dose Material Level Specimen l 2 Ethocell O 65 b2 1 37 63 6 __ __ fiascopak 0 L2 5 1 DA. 0 6 hl bl Nicrocrystelline O 65 oo wax 1 66 62 6 7o 65 .1 . V, f 1 T bpec1mens tested at 73oF, SON R.d. Ethocell O 68 Eh l 51 SO 6 U7 he Wascopak 0 ha MO l "C) 3L) 6 £2 to Kicrocrystslline 0 OS 66 flax l 3 TO 6 71 66 0 Co Cot-O b I—J-F‘bu UL C" C.\ Q\\,_) _ {T 4:151 \\ :1- '2" U) lbLF‘ 0+4 PFJFJI~QULFTJ C\O\ Mfg .._.. Mr‘wh—fl " ‘.| .__- ,— tJUUUtJbinGfio4d h\ ‘ a \ r \.I \4 fl , E 53 H. H ‘4 ~ i\) \FLUI 03 31“" *“ I O\O‘\O\T—“\2Jw I 4:“ka I—‘xO CD I U1 r32" (ix-d C xii-“‘L-J {T‘U‘L‘xng \O O-§:’\»\o H l-—' |--J CO Hon irradiated Elapso Tifle, L.08 1.88 12.16 1H.1o 27.; £32230 d hrs. , - (‘3 Table L deimh 1t Gein of Ester Vapor Tron msn sion Tsmvles ‘9seon9‘ 9-H 19 002 90 532. . height Gain rrfims 1 2 3 L- 5 0.0597 0.1986 0. 0665 9.0786 0.0687 0.118L 0.19L8 0. 123 9 0.1L73 0.1315 0.1832 0.2803 0.1F55 0.2198 0.1985 0.2150 0.2207 0.2111 0.2L91 0.2276 0.313L 0.L78L 0. 3086 0.3585 1.3336 0.Ll83 0.6127 0.L137 0.L706 0.L501 9.5LL8 0.7626 .E363 0.5863 0.5763 0.6936 0.9329 0.6790 0.7161 0.72L3 diation Dose Iei3ht Gain, Trams 1 2 3 L 0.0795 0.0695 0.0737 0. 0736 0.0601 0.1503 0.13 73 0.1L76 0.1503 0.2102 0.2L96 0.2L23 0.2539 .2623 0.2621 0.3603 0.3563 0.3713 0. 3652 0.3693 0.L757 0.L8L1 0.9891 0.L793 0.L805 0.601L 0.6193 0.6L20 0.6051 0. 6058 0.7L50 0.7761 0.6078 0. 7L72 0. 7L75 0.9783 1.031L 1.0721 0.9753 0.9780 adiation Dos e Jeight Gain, grams 1 2 3 5 0.0899 0.0736 0.0581 0. 069 5 0. 0663 0..2 ’6 0.2135 0.1852 0.2120 0.1971 0.L397 0.3722 0.2938 0. 3773 0.L632 0. 693 0.L86L 0. 3896 0.L831 0.L799 0. 7L20 0.6322 0.E;7L 0.6.69 0.:<999 C. 0570 O.C282 0.66;); 6217 C. 70 '9 1.0705 0.9297 0.7L63 0. 9215 C.“rl l..2276 1.0733 0.8602 1.0639 0 .97L5 L3 I 'll-lllll‘ll 'l'llu'llul I] n s a . a s . a n O A n n s A x g \ a a a V I o I - . a v . s 9 1 s A 6 . u o 4 ~ Q A \ I . 1 5 6 J a A a I u o. a O . 4 a ~ n Tab Height Gain of U 10 5 ater Vapor Transmission Samnles {wagon k 30- (LA 7301], 502913.11. Non irradiated Elapsed T1016, hrs. 1 L.08 0.017 8.L2 0. 030 12. 08 0. 030 21. 08 0. 061 28.08 O. 070 €3.08 0. 006 7L. 8 0.11L $3.021éb 58. 92 0.14 1 Kegarad Radiation Elepsed Time, hrS. 1 ’|.08 0.017 8. L2 0.052 12.08 0001.3 21.08 06 28.50 0.090 S. 08 0.108 dLoSB 0.130 530'-2 00153 58.92 0.107 6 Megarads Radiation Elapsed Time hrs. 1 L.0 0.015 8.L2 0. 029 12.08 0. 039 21.08 0. 005 28.58 0. 080 35. 08 0.105 llbw. 0.126 53. L2 0. 1L9 SBQQ2 0.103 Weight Gain, grams 2 0.018 0.031 0.010 0.063 0.082 0.098 0.117 0.137 0.1L9 3 0.020 0.036 0. 075 0. 097 0.11L 0.136 0.159 0.173 0.016 0.028 0. 038 0. 058 0. 076 0. 092 0.110 0.129 0.1L0 0. 818 0. 0 3 0. 617.3 0. 070 0.091 0.109 0.131 0.153 0.166 Weight Gain, grams 2 0. 016 8383.8 0. .080 0.10L 0.126 0.1L8 0.162 3 0.017 0.0 2 O. 012 0. 038 0. 090 0.108 0.130 0.152 0.166 0.016 0.029 0.038 0.063 0.082 0. 099 0.119 0.1L0 O 153 5 . 0.015 0.0211, 0- 039 0. 00L_ 0. 08L 0.101 0.122 O o l‘llL 0.158 Weight Gain, grams 2 0.019 0.033 0.055 0.075 0.091 0.110 0.130 3 0. 017 0. 033 0. 01!.) 0. 073 0.00 0.11 0.100 0.165 L 0.01L 0.028 0.036 0.060 0.080 0.097 0.117 0.139 0.152 5 0.012 0.025 0.03L 0.050 0.075 0.091 0.110 0.129 0.101 94 li'lalll' a o a c a a a . a. a I o r s n o n x a .. a 7 n u a a w a 1 1 . 1 _ a n a a . . q a . n 1 1 .. ll I! l I II II I A x: - 1. I n r o 5 0 X r u G 1 N .t o a u n a 0 q o n Table 6 Gain of Transmission Ethogell - 100 F, 90» R.H. Non irradiated Elapsed Time, 11.33 12.75 22.25 28.08 36.50 L7.16 53.00 61.58 hrs. 1 0.0235 0.0762 0.1509 0.2072 0.2592 0.3393 0.3821 2 0. 0202 0. 086 0.189 0 .2032 2668 0. L212 3Q10'21‘ 0.1L5L11 1 Negarad Radiation Dose Elapsed Time, hrs. L.25 12.58 22.08 28. 00 €6.L_2 L6. 02 52. 75 61 061.2 1 0.0107 0.0760 0.1590 0.2081 0.2 22 0.3 38 0 061-123 0.L809 2 0.0298 0. 08’2 0.1L30 0.1olg 0.:L 0. 3196 0. 3596 0. L162 6 Megarads Radiation Dose Elapsed Time, .58 5-33 16.91 22 01 28 01 L0.25 L6.11 .50 62.50 hrs. 1 0.0392 0.0725 0 1379 0.1905 0.2396 320C) OO_) 3 131le 0.1127 2 0. 0329 0.0601 0.1170 0.1522 0.1981 0. 7’3 0.3209 0 .36L5 0.L321 Hater Vapor Samples ieiaht Gain, grams 3 0.0229 0. 0698 0:12)“) 0.16 0 0. 21-0 0.3081 0.3180 0.3691 weight 3 0. 031Q 0. OQU_O 0.1661 0. 2162 C). 21%37 0. 3695 0. L160 0. 62L1 Jeight 3, 0.0367 0.0665 0. 21 0.1?819 0. 228L 0.3159 0. 693 0.5051 L 0. 0280 0. 083 0 0.1626 0.1936 0.2513 0. 0.3360 07 0,131.1 aApparent failure after 47th hour Gain, L 0.0L12 0.11L5 0.1990 0.2581 0. 33115 0:11-387 OoU—(S7 0. 5711.1 5 Q. 0100 0. 0650 0.1175 OolFO 0. 203 0.2700 0.3050 003529 grams 5 0. 0512 0. OQL3 0.163 0.2126 0 2 a 2366 OQU-OS3 0. L603 Gain, grams L 0.0778% 0. 0010 0.120? 0.17 0. /200-0 00318Q 0.371LO 0.U27U 0. “L12 %Liquid water on surface of sample 5 0. 03L7 0. 065L 0.132.0 0.1030 23OL 0:3105 0. 702 0.1-260 0.5927 lllllll III! D a Tab Weight Gain of Transmissio Ethoc 7302, 50 Non irradiated Elapsed Time, hrs. 1 L.75 0.006 8.08 .010 21.25 .02 28033 003 U0900 .OBU El 92 .062 60. 08 .071 l Megarads Radiation Elapsed Time, hrs. 1 U075 00005 0.08 .010 10.75 .01L 21.25 .026 28.33 .03L L6. 08 .050 51. 92 .05 60. 08 .06 6 Megarads Radiation Elapsed Time, hrs. 1 L.75 0.00L 8.08 .010 10.75 .014 £1.25 .027 8033 .030 16. 08 .055 9:1. 92 .062 60 .08 .072 le 7 Hater Vapor n Samples e11 A R.H. height Gain, grams 2 3 L 5 0.005 0.007 0.006 0.005 .910 .012 .011 .011 .013 .017 .016 .015 .025 .030 .028 .027 .0 3 .0L0 .038 .037 .0L8 .058 .056 .05L .055 .066 .063 .061 .063 .076 .072 .071 Weight Gain, grams 2 3 L 5 0.00L 0.005 0.005 0.00L .009 .010 .010 .009 .013 .01L .015 .013 0025 .025 .02 .025 .03L .030 .03 .03L .050 .050 .051 .050 .057 .056 .057 .057 .065 .065 .066 .066 Reight Gain, grams 2 3 L 5 0.00L 0.30L 0.003 0.00L .010 .000 .007 ".005 .01L .012 .011 -.00L 027 .22 .021 .008 .036 .031 .029 .017 .055 .0L6 .0L3 .03 L .062 .053 .0L 9 .0L1 .072 .060 .057 0L8 - E“ b\ ll'rt-lll‘l 'Ill! . .I I I'll! \\ neicht Gain of T“"nsnis ion Iicroorystalline Lax Table 8 1000 F, 903 2.1. Non irradiated Llapsed Time, 1 L.'2 10.58 21.90 27.25 0.08 1-5.2g f2 25 1 Hogarad Radi Elapsed Time, hrs. L.33 10.E3Q 21.L2 2? l? 30. 00 6 Kegarads Elapsed Time, L-33 10.50 21.L2 27.17 5 0.00 )2:17 ‘ A’_ Il- ‘S o ‘1 O h16. 1 2 -0.0107 -0. 0065 " .0095 "' . H.109 ‘ 00105 ‘ .0009 ” .0061 “ .0003 — .0081 - .0058 - .0106 — 0089 - .0071 - .0063 ation l 2 -0. 0077 -0.0018 - .0000 - .0020 — .0090 - .0012 - 00061 + .0056 - .0076 + .0001 — .0097 — .00L3 - .0075 — .0012 Radiation 1 2 -0. 003 3 -0.0023 _ .0,62 - .0010 "' .OOIB + .0102 - .00L9 + .0100 - .0050 + .0103 _ 0601 + @020 - .0033 + .0032 "ater Vapor .amples height Gain, 3r ms 3 L -0.0178 -0.0065 -0 .0056 - .0180 - .0061 - .006L ‘ 00177 “ .0052 - .OOD1_L "' .0070 " .005 "" .0052 - .016i - .001L- - .00L6 - .019 - .0068 - .0066 — .0157 - .00 39 - .0036 -eignt Gain, gers 3 L 5 -0.00L3 -0.00L0 -0.00L1 ” 00050 - .0019 ‘ .0037 - .0056 - .001L - .0012 — .0065 - .0017 - .0021 - .0061 - .000 - .2013 — .0082 + .000 - 003.1.1 - .0055 + .0030 - .1005 Ueimht Gain, grams -0.0(_.7‘q - .00 8 ‘ 00008 - 0007 +-+I 0.. '7DCDf 6 1 ~45 HLQ L -0. 03 6 _ 9‘20 - .0020 - IQOUl + .0070 + .0050 - .0006 + 5 +OoOOQl .0026 00089 .0081 .0090 .008L .0073 I 'l i l J I II I I Q r t l \ n. I i v. p 1 I n. ‘ Q 5 cs 4. n. . m A a h a 0 s i x u .. h 1 u 5 w I o a q A u _ .. o 1 n, x . . . . w n. a n 1 v \ \ 'lll Table 9 weight Gain of Transmission 3 Iioroorystalline “ax 73oF, 500 1. Non irradiated Elapsed Time, hrs. 1 2 5.33 0.0025 0.0022 12.00 .0032 .0033 2.33 38 .00 7 29.67 .0018 L5 58 .00L2 .00.Lio 5L. L2 .00L0 .00L8 80 .L2 .00L L .00L_3 l Hegarad Radiation Elapsed Time, hrs. 1 2 5.33 0.002; 0.0023 32 .00 .0020 .0020 2:2.33 .0028 .0028 20 .87 .0038 .00L2 JS:E 8 .0091 .OOBA EL. ]2 .00L2 .0050 6 Megarads Radiation Elapsed Time, hrs. 1 2 5.33 0.0008 0.0018 12.00 .001L .0028 22.33 .0023 .0027 29.87 .0015 .0037 LE. .58 .0015 .0027 L2 .0005 .0038 80. L .0007 .0031 H Weight 3 0.0018 .0019 .0021 .0023 .0018 .0018 We 181,11; 3 0.0003 .0002 -.0003 -.0011 -.0023 —.0032 “.0035 Later Vapor amples Gain grams L 5 0.0023 0. 00 2L .0027 .00 l .0031 .00 S .00L2 .0056 .0034 .0051 .00LL .0055 .0039 .0053 G ain, grams u .0022 .0027 .0030 .0037 .0033 .0037 0.0001 -0000h '00012 -.0011 ’00028 “00031 -.0038 .J b .0028 .0035 oOObl .00L7 .OOILB oOOb-7 Gain, grams 5 0.0088 .0073 .0075 .0077 .0075 .0060 .0088 All-I. I I'll ' I‘ll I'll-I‘ .l‘ v I _ : a a 1 v A t 4 \ q n 1 A . ) a a \ ! )x m I r a . n a a C A. A .l C . ll '1‘" II III III‘ pl? . APPEITDIX II Table I Analysis of Variance 'jater Vapor Transmission Rate UaSCOpak 1000p, 90% R.H. Specimen 0 158* 203 150 150 180 823 Radiation 1 162 173 180 101 157 833 Levels 0 201 177 181 175 15o 850 T 517 553 071 092 M73 2500 These values are b x 10h Source of Sum of Kean Variance Squares df Square V” F Total 401,000 10 Replications 7,000 A 1,750 Radiation Levels 157,000 2 78,500 2.25 n.s. Error 277,000 8 30,000 180 True V’e = 0.0186 ngO/hr Radiation Levels HOW-’0 Source of Variance Total Replications Radiation Levels Error Table II Analysis of Variance Eater Vapor Transmission Rate UaSCOpak 73oF, 50% R.H. 1 232-2:- 270 208 770 Sum.of Square S7th103 1h, 2815x103 M Specimen 2 3 a 237 27S 225 208 270 208 232 295 250 733 880, 723 alues are b X 105 Inf-e a n 3 df Square 708x103 289x103 578x103 2 2351x103 8 208x103 5.82332102 True V—e = .005’1312 gHZO/hr 5 208 201 233 702 T 1237 1313 1278 3828 .98 n08. R.) Table III Analysis of Variance Water Vapor Transmission Rate Dow Ethocell 100Or, 905 R.H. Specimen l 2 3 8 5 T, 0 788a 873 028 727 588 3550 Radiation 1 820 085 930 080 787 813 Levels 0 880 091 800 839 900 807 T 2810 2289 2309 2500 2235 11709 g 4 These values are b x 10’ Source of Sum of Kean Variance Squares df Square V“ F Total 100,820 18 Replications 17,217 8 Radiation Levels 80,779 2 20,389 1.50 n.s. Error 108,830 8 13,003 110.8 OH True Ve = .00117 9' O/‘nr 2 Specimen 1 2 3 8 0 117* 102 122 118 Radiation 1 107 108 100 107 Levels 0 120 120 110 95 T 388 330 338 320 These values are b X 105 Source of Sum of Kean Variance Squares df Souare Total 1035 18 Replications 169 8 b2.25 Radiation , Levels 100 2 80.00 Error 700 8 88.25 True V’e : .00009 gHOQ/hr Table IV Analysis of Variance Hater Vapor Transmission Rate Dow Ethocell 73oF, 50$ R.H. S 115 109 95 31 9-8 ‘Kj—‘k w -:':‘!—3 0V6". U1 U1 U1»;- H O \1 N n.s. 58 Table VA Analysis of variance Water Vapor Transmission Rate Microcrystalline flax 730p, 50% RoHo Specimens 1 2 3, 8 J5 T O 305' ’302 3.0 3.0 5.0 1803 Radiation 1 3.0 1.0 0 2.8 3.2 13.2 Levels 0 0+ 0 -7.1 0 0 -7.1 T 6.5 7.8 -3.5 S.}_l_ 8.2 214.01% These values are b x 105 + Where regression proved not significant, the value is entered as 0. Source of Sum of Mean Variance Squares df Square V‘ B1 Total 120.8 18 Replications 31.0 8 Radiation Levels 72.3 2 36.1 , 12.8%* Error 23.1 8 2.9 1.7x10‘5 V2 = .000017 g/nr 55 Table VB Analysis of Variance Hater Vapor Transmission Rate Kicrocrystalline flax 730R, 50; R.R. Specimens 1 2 3 8 5 T 0 3.5% 3.2 3.0 3.0 5.0 18.3 Radiation 1 3.0 8.6 -1.0 2.8 3.2 12.2 Levels 0 —1.1 2.0 ~7.1 —8.7 -0.1 —11.0 T 50).;— 008 ~L!_.5 .7 8.1 19.5 r These values are b X 109 Source of Sum of Kean Variance Squares df Square F V— Total 109.00 18 Replications 85.15 8 11.29 Radiation Error 28.25 8 3.53 1.88 True V’e'=1.88 x 10"5 : 0.000023H90/hr ‘2‘ ab] 8 VI Regression Analysis Wample Calculation ’fe need to obtain the sums of squares of deviations. These are obtained from: (EXY Q .._.._.._.. 1 9 _7 {KM 31A - TH P 9 f, g3» =zy- - and. 5 xy :tXY - 11 3nd b =ZXC Using H - O -.l as an example: ix :200211 N: 9 21 = .716 ZXY = 28.516 £3181? t 21.181 11 €3Qf:7.335 (€15)?- 3.023932 E y’é.rf’>1703f‘a a *2 : 11,037.0qu A, .4 '. o 7;”3‘20970 , arr A. (U , 1T 5X1 3,1Ulo‘CU (5x? 2 - zLx’ In the analysis table, Source of Variance SS df 33 *IJ Total .017038 8 1/ Regression .017016 1 17,016x10"Q Error .000022 7 3x10"0 and finally, .E;EK - b 3 51:“ 3101 " .00232 Table VII Height Gain Licrocrystalline naX flax placed on F1388 plate, eXposed for two days to 73oF, £03 3.1. then eXposed at 1000p and 905 R.H. Petrolatum removed with cloth and acetone 1t. Gain, grams Elapsed time, hrs 1. 2 3 11 2.33 -.0010 -.0006 -.0005 -.0005 1.19 +.0002 +.0006 +.0001 +.0001 7.12 +.0008 4.0013 +.0012 +.0012 11.12 +.0010 +.0015 +.0011 +.0010 23.67 +.0006 +.0018 +.0009 ¢.0003 27.17 +.0010 -.0070 +.0018 +.0016 29.33 +.0012 -.0066 +.0022 +.0011 16.08 4.0001 -.0071 +.0016 +.0009 51.50 +.0015 —.0008 +.0015 +.0011 Table VIII dater Vapor Transmission Test Kicrocrystalline hax release agent was used with this wax. Elapsed Height Gain, grams Time, hrs. 1 3.12 .0039 9.16 .0030 19.12 .0028 21.12 .0072 23.12 .0065 25.67 .0066 27.12 .0060 h2.92 .0038 50.12 .0063 2 .0067 .0018 .0001 .0021 .0023 .0031 .0027 .0008 .0018 l. 3. 1. 5. 9. 10. 60 Bibliography Barrer, uichard H., DifoSion In and ThroUZQ Colids, Cambridge University Press London,l)11 Brabender, George J., Hater Vapor Perreabili y; of Moisture Sensitive Katerials Taper Trade J Oct. 19, 1911 Charlesby, Arthur, Beneficial Effects of Radiation on Plastics and Elastomers, Nucleonics Vol I1, Ho. 9, 1950 Dunn, Cecil G.; Campbell, Gilliam L.; Fram, H arvey; Hutchins, Ardelia: Biological and Photo-Chemical Effects of High unorgy, Elec- trostatically Produced Roentgen Rays and Cathode Rays, J. Applied Physics 19 ppbO5—olo Fox, R.C., flax Crystal Structure vs UVTR Hodern Packaging October , 1938 Karel h. and Proctor B. E. Cathode Raw 9 : Irradiation of Celloohanes Modern Packaginv g ’ ._. Q January 1957 McDermott, P. F. Noisture : wigration Refrig Eng #2 (Aug. lghl) Riddell, G.L., Science and Packaging, Chemistry and Industry 66 Jan 18, 1517 Rogers, 0.; Meyer, J.A.; Stannett, V.; Szwarc, M. Studies in the Gas and Vapor Permeability_ of Plastic Films and Coated Papers, Parts I & II Tappi, Vol 39, No. 11 (1958? Slater, J.C. Effects of Radiation on faterials J. Applied Physics 22,237 LarIBI ,9 L... 13. 17. Saeman, Jerome F.; L9.Wton, alliot J.; Killet, Terri 11 A. Effect of :i:h ”ncrsy Cathode Rays on Cell_ulos e Ind. And hng Chemistry 11 (12) p} 231 -2e51 Sun, K.H. Effects of Atomic Radiations on Hirh Polymers, Kode1 n 1lasti lCS (327(1) September 1951 Trump, J.G. and Van de Graff, R. J. Irra d:iation of Biological Fatcrials by H1:h 'n