‘ ‘31"?le 3 311111, 3333133 3 ~ 511). $4332.83 . 3313111 2 r 333 332g. .— ”,4 .3 O '9 . as“; .31 x". x . . .1 ‘1" Era-1’ '- '1 ‘. ,1 . '1 ' 3.3.33‘3'1'3312:333‘«~ '. 321333 :35: ' ~ . 1) l 3 I? . . 1 3 :7 33333‘h34 ‘3‘ 331:3”: , . s 1 - - 3333123 '3“ 313113.112515313111113 3 n a-, _. W 3 333333533 33133 3' n ' {33333.3 333.3‘3 (33111111131: ii“ ‘ 3 33 WM ' 2' 1'. ‘ ": '3 i; 333‘131 .1 (43;. ' I . 3 .1‘ {I 1 1 hi . "2.133333333533333 {x \ " .5. f ' I I I c ‘I‘ ‘7' ‘1 ‘ - . . ‘ _ . . ,sc. V 3 « 1 .‘u . ‘ " ‘ ' "flit 1'95”] ; ,m . ; " uv‘x’" J ' g' , , 0,. a "Ix M '1'. V ‘ ‘1' I.. . 1 I ' ' ' 1 "| 1- o . . . r. ‘ ‘ I ~ 3‘} «‘4 .. x u . ,> :r'" '. . ‘ ' ' A : I: 1‘ 3‘: aid-.5. v 't ' :h'u t o ' ‘l I ;‘ “V I..::. 0“.",0‘ L, ' '-Y-:x|v--- ... ., ‘ ' .:~S*2 "3:¢51:3:v“". L': '..-.' ' #5:?" "ii: ' ': ,.; v- . :'- . I U ' 52% . ~ . ,, ‘3 3%.”? 3 3xe 313 . 2 33:33.,5193 l 33‘ 131 11,1131” _ . 1 ' ‘33333333‘3333 ““3‘3‘33‘2’333 3313333323311 ., . . "13 3" 33333333333331333'313333333i 333‘ “3.433331333333533 3?)“ fit my; 15:33 {5:5 ‘ ‘3 33333333313333.2332W 13% «MM :wlfii MW 1.: 1:1 , . <‘ 3}: 3‘1'L,|;' 3313-1 “1331431111 31331;,“o'1b14’73 !. 1.' L; 1‘... 5111. 331343 11113313311 p3 13x3 {1133311 13‘. 1 333.33.2,3,31‘112113333 ‘33 11233.3. "2 «“33 {£333 ““3 ““3333: “11111.11. ‘5‘ ‘ '1 W“ 5 ‘z ”“3333" 33333, v.x x w ' z .23. n13" x 1'3 3.3:x1‘1153r. 13 I'm dz" wit??? _ 1.331. x Lax 13! 3313’} £3353: .21. 12:] :3231 . 1. 1 3:32. “537 3“ Kg ”ffitn 0' gull- 'v . '73. Zl'v3'~_" - ~. - 3 My: :‘SnJII'. " ‘ "'33‘ 32 {Quinn *2. - ; E "3""31 '1 ’31.” if“? 2 r. .1531... (3141‘1314‘53' 1111131" 1 333111339". 1 ~ 711111 #3333313 11 . .1 W. 33133 131.33 1 .31353“l“‘““3‘. I 3313:31‘12‘1‘3‘1333‘1id“. [313113 “3313123333 3113111111 ’33:“.113—11 “':13"“’.r‘ 11:.1: V ‘E‘ 333331233133113133331“ 31.13313. 13111.1. 11' 313:3‘33,‘.31111:1‘1'31x11 3' :3“ 31333333533313 31333333333 3313‘ 31x33 3.33133W13 “1‘311'1313133‘1'”? 33331'33‘33'1133‘13 f: - 11111113331331.11'11111111311111.331133333‘ 33 33‘11'33‘1 1 33,111 .1111133331 1133.333. 31 11} 42 .4—3 - ‘ "‘.'—-;:-‘_ 1 F.“ _ fl... 1-.‘9— _.... 30333333313111 13"1 . 3111 W33 3” «1111 13313‘1111“ 3331333331333 3333 3. 3333313133 3 1...; v 33333333331333.31333131113111313313333333'13 . L '3 3.33.: 23313.3 {33133333333333333313‘ . ms IIIIIIIIIIIIIIIIIII I III II IIIIII‘" I 32912516 “away Michigan. State University This is to certify that the thesis entitled A COMPARISON OF DIFFERENT METHODS USED TO MEASURE GAS TRANS— MISSION RATES THROUGH PLASTIC FILMS presented by LAURIE ANNE ROY has been accepted towards fulfillment of the requirements for IM:XW GH E - LOCKHART Major professor MAY 1 81 Date 5' 9 0-7639 I OVERDUE FINES: 25¢ per day per item RETURNI'K; LIBRARY MATERIALS: Place in book return to remove charge from circulation records A COMPARISON OF DIFFERENT METHODS USED TO MEASURE GAS TRANSMISSION RATES THROUGH PLASTIC FILMS BY Laurie Anne Roy A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE School of Packaging 1980 ‘/f I.“ G/I ABSTRACT A COMPARISON OF DIFFERENT METHODS USED TO MEASURE GAS TRANSMISSION RATES THROUGH PLASTIC FILMS BY Laurie Anne Roy With the increasing use of plastics as packaging materials, an accurate and rapid method of measuring gas permeabilities is needed. ASTM D 1434 is the standard procedure for the Manometric and Volumetric techniques of measuring gas transmis- sion rates, but because of the inherent problems with these methods, new sensing equipment has been developed for measuring gas transmission rates. These new methods are the oxygen specific Coulometric detector, a carbon dioxide specific pres- sure modulated infrared detector (PMIR) and an isostatic cell used in conjunction with a gas chromatograph as its detector. There is no published standard for the operation of these methods, therefore, information about the operational para- meters is not widely disseminated. The purpose of this study was to compare the instrumentation involved in the use of these sensing devices to see if a standard procedure could be written for them and to see if there is any correlation between permeability constants derived by each method. The effects of some experimental parameters were determined. Oxygen and carbon dioxide were used as the test gases. With many thanks to Dr. Hugh E. Lockhart, for his wisdom and guidance throughout this project. TABLE OF CONTENTS List of Tables ------------------------------------------- iii List of Figures ------------------------------------------- iv Appendix List ---------------------------------------------- v Introduction ----------------------------------------------- 1 Materials -------------------------------------------------- 7 Methods --------------------------------------------------- 10 Test Results ---------------------------------------------- 14 Discussion ------------------------------------------------ 22 Recommendations ------------------------------------------- 27 Bibliography ---------------------------------------------- 23 Appendix -------------------------------------------------- 30 ii LIST OF TABLES Experimental Design ------------------------------------- 9 Mocon Oxtran 100 Oxygen Transmission Rate --------------------------------------------------- 15 Mocon Permatran C Carbon Dioxide Transmission Rate --------------------------------------------------- 13 Isostatic Cell Oxygen Transmission Rate ---------------- 19 Isostatic Cell Carbon Dioxide Transmission Rate --------------------------------------------------- 20 Method Comparison, Mocon Equipment vs. Isostatic Cell ----------------------------------------- 21 Oxtran 100 vs. Isostatic Cell -------------------------- 23 iii LIST OF FIGURES The Mocon Oxtran 100 and Controls ----------------------- 3 The Mocon Permatran C and Controls ---------------------- 4 The Effect of Test Gas Flow Rates on the Mocon Oxtran 100 --------------------------------------- 25 iv APPENDIX LIST Standard Method of Test for Oxygen Gas Transmission Through Plastic Film and Sheeting Using a Coulemetric Sensor -------------------------------------------------- 31 Raw Data by Method and Material ------------------------- 52 Factors Affecting Calibration and Performance on the Oxtran 100 ---------------------------------------------- 68 Calibration Curve for Rotameter ------------------------- 7O INTRODUCTION With the increasing use of plastic films for packaging, their permeability with respect to gases becomes increasingly more important. The shelf life of packaged foods is often determined by the amount of gas entering the package. For this reason, an accurate method of measuring gas transmission rates is necessary and a rapid method is desirable for reasons of economy and timely availability (1). Presently, there are two standard methods, Monometric and .Volumetric, for determining the gas transmission rates of plastic films. These methods utilize a gas transmission cell with the sample mounted between two chambers forming a sealed semibarrier between them. With the Monometric method, gas transmission through the test specimen is indicated by a change in pressure. For the Volumetric method, the lower pressure chamber is maintained near atmospheric pressure and gas transmission is indicated by a change in volume (2). Both methods are difficult to operate and special laboratory conditions are required for their use. This equipment is dif- ficult to maintain and requires lengthy test times to obtain the permeability of high barrier materials. Due to these problems, new methods of determining gas trans- lmission rates have been developed. These methods utilize the concentration increase principle of gas permeation. Like the others, this method depends on the elimination of leakage, but its accuracy rests primarily on the method for analyzing the amount of gas transmission through the plastic film (3). An oxygen specific Coulometric detector and a Pressure Modulated Infrared Detector were developed for analyzing gas transmission rates. Gas Chromatography is another method in common use. The effects of experimental parameters on these methods is unknown, that is why there is no standard method for their operation. The Mocon Oxtran 100 utilizes the Coulometric detector for measuring oxygen transmission rates of plastic films. The output of the Coulometric detector is consistent with the fun- damental principles of Faraday's Law (see Appendix 6). The instrument does not utilize vacuums or large pressure differ- entials. The Oxtran 100 and controls are illustrated in Figure 1 (4). The Mocon Permatran C, see Figure 2, utilizes the Pressure Modulated Infrared detector for measuring carbon dioxide trans- mission rate of plastic films. Unlike conventional Infrared techniques, this instrument does not use a mechanical chopper to modulate the light beam. Instead, the gas itself is pressure modulated at a fixed frequency by a mechanical pump. It detects the principle absorption band for carbon dioxide (4.3 hertz). A recently developed cell method uses the gas chromatograph as its detector. A wide range of gases can be analyzed using Ithis detector. Concentrations of different gases in mixture Temperature H Diffusion Control Cell Upper Chamber Flow Lower Bubbler Valve Meter Chamber Tubes Valve Oxtran 100 and Controls Diffusion Calibration Cell Device Temperature Chamber Flow Strip Control Valve Meters Chart Recorder Permatran C and Controls can be measured because the gas chromatograph can separate and analyze them individually. ASTM Committee F2.0 is currently repeating a study of these methods to see which one, if any, will lend itself to the development of a standard. The initial study was inconclusive due to the variability of data between laboratories and methods. Good repeatability of data was observed within laboratories but no correlation could be made between laboratories. (5) Some method dependence is expected but the variability between laboratories could have been due to poor operator technique and poor control of experimental parameters. The purpose of this study was to examine these methods, to determine the effects of some experimental parameters and to compare the Mocon equipment with the cell method to see if any correlation could be drawn between them for determining permeability. All equipment was operated by one person in one laboratory thus reducing the effects of such variables as operator tech- nique and variable laboratory conditions. Relative humidity, gas flow rates and film type were varied to determine if these parameters would have a significant effect on gas transmission rates. Relative humidity of 100% when tHe bubbler tubes on the Oxtran 100 are filled with distilled water and glass beads was proven using an electrical resistence hygrometer with dial readout. This was measured at flow rates between 10 and 30 cc/min. Operation of the instrument using dry bubbler tubes was assumed to be 0% relative humidity at any gas flow rate. The Permatran C could only be operated at 0% relative humidity because the detector is sensitive to moisture. The test gas was assumed to be at 0% relative humidity from the tank. Test gas from the tank was also assumed to be 0% relative humidity for the Isostatic Cell method. A bubbling system was used to humidify the test gases for the cell. This consisted of three plastic bottles half filled with distilled water. This system was connected by copper tubing to the tank and cell. a rotameter was used to control flow rate. 100% relative humidity was proven with this method at all flow rates using an electri- cal resistance hygrometer with dial readout. Gas flow rates were varied on all equipment. The Mocon equipment has built in gas flow meters. Nitrogen flow is con- trolled on the Oxtran 100 by the meter and oxygen flow is set visually by matching using the bubblers filled with distilled water. Gas flow meters are also used on the Permatran C to set the flow rate of carbon dioxide through the calibration device and the permeation cell. A rotameter was attached in the test gas flow line to the Isostatic Cell used with the gas chroma— tograph. A calibration graph for this meter is found in Appendix 4. Four different types of films commonly used as packaging materials were tested because of their different barrier properties. Their permeability to oxygen and carbon dioxide from these methods was calculated and compared. MATERIALS The test materials were supplied by ASTM Committee F2.0. These were samples of polyester, saran and FEP (flourinated ethylene propylene) which were used in their study. These films were chosen so that a comparison could be made between their data and the results of this study and because of their different barrier characteristics and their common use as packaging materials. Mylar, supplied by Modern Controls was also tested because it is used as the standardizing material for the Oxtran 100. The gas transmission rates for oxygen and carbon dioxide through five samples of each film (except FEP) were measured on different days as near 73°F as possible (i 2°F) using a driving force as near one atmosphere as possible. Each sample was run through a series of tests starting with the Oxtran 100 because its diffusion cell had the largest cir- cumference (100 cm2) thus giving the largest surface area. Oxygen transmission was measured at O and 100% relative humidity for each sample on the Oxtran 100. Next, the samples were tested at 0% relative humidity on the Permatran C with test gases of dif- ferent driving forces. A mixture of 5% COZ/Air balance was used for a driving force of .05 atmosphere and a tank of pure CO2 was used for a driving force of 1 atmosphere. Finally, the samples were tested by the Isostatic Cell method using both test gases at a driving force of 1 atmosphere and at 0 and 100% relative humidity, with analysis by gas chromato- graphy. FEP, because of its low barrier properties, was masked down using a light gauge aluminum foil to a surface area of 16 cm 2 and was tested using different driving forces depending on the instrument on which it was tested. A driving force of 1 atmosphere was used on the Oxtran 100, .05 on the Permatran C and .21 for oxygen and .05 for carbon dioxide for the iso- static cell. The gas flow rates were varied on all equipment to observe what effect this would have on transmission rates. See Table 1 for layout of the experimental procedure. TABLE 1 EXPERIMENTAL DESIGN Driving Force of Test Gas Percent Test Gas Flow Rate Relative Material Method (cc) Humidity (Atm.) SARAN Oxtran 100 10, 20 & 6O 0 & 100 ll Permatran C 20,50 & 100 0 .05 & l Isostatic Cell 20 & 80 0 & 100 1 FEP Oxtran 100 10, 20 & 60 O & 100 1 Permatran C 20,50 & 100 0 .05 & 1 Isostatic Cell 20 & 80 0 & 100 1 POLYESTER Oxtran 100 20 O & 100 l Permatran C 20,50 & 100 0 .05 & 1 Isostatic Cell 20 0 & 100 1 MYLAR Oxtran 100 20 0 & 100 1 Permatran C 20,50 & 100 0 .05 & 1 Isostatic Cell 20 0 & 100 l METHODS The Standard Method of Test for Oxygen Gas Transmission Rate Through Plastic Films and Sheeting Using a Coulometric Sensor developed by ASTM Committee D-20 was used as the basis for all test methods with the necessary alterations (see Appendix 1). This method was chosen because it was proven to give reproducible test results. Mocon Oxtran 100 Operation A 15 cm square test specimen is clamped into a cell having a 100 square cm diffusing surface. IBoth cell sides are purged with oxygen free nitrogen gas. The detector is inserted into the flow to establish a stable baseline reading. This repre- sents the amount of oxygen present in the carrier gas and that which has not or can not be purged from the film sample. When a stable zero reading has been established on the recorder, oxygen is introduced into the upper half of the diffusion cell. The nitrogen continues to flow through the lower cell (see flow diagram, Appendix 1). Moisture was added to the gases using bubbler tubes con- taining distilled water and glass beads. As oxygen diffuses through the barrier, it is entrained by the nitrogen and carried to the detector. The detector current rise is recorded on a strip chart. When the oxygen 10 11 transmission rate is at equilibrium, the recorder value levels off. Detector current is directly related to oxygen trans- mission rate (see Coulometric Detector, Appendix 3). (6) Mocon Permatran C Operation A film sample of 6 cm square is clamped in a diffusion cell. One side of the film sample is exposed to a carbon dioxide concentration, the other side is exposed to air which has been passed through a bed of ascarite. This condition is maintained until equilibrium conditions are reached which may take a few minutes or several hours. At this point the stream of air is deflected past the ascarite and is circulated in a closed loop past one side of the sample, into the sensing chamber and then back to the sample again. As carbon dioxide diffuses through the sample it accumulates in the capture volume and this amount is recorded on a strip chart recorder. The slope of the recorder trace over time (A_V/At) is proportional to the carbon dioxide diffusion rate through the sample. In order to determine what a recorder deflection (AV) means in terms of pure cubic centimeters of carbon dioxide, the instrument must be calibrated. This is done by clamping an impermeable material in an open diffusion cell and closing the cell which contains the film sample. The special cali— brating device which is a part of the instrument has a known volume which is filled with the test gas and then entered in the circulation 100p. The deflection on the strip chart recorder is used as the basis for determining rate of trans- mission through the film sample. The recorder trace, from the film sample, is timed (At) to the deflection point of the calibration procedure. Isostatic Cell - Gas Chromatograph Method A 6 cm square film sample is clamped between two compart- ments of the cell. Both cell surfaces are accurately machined and polished so that when a sample is clamped between them, the force from the clamping device is applied to a narrow circular area on both sides of the sample to give an effective seal. Each compartment is fitted with a gas sample port and inlet and outlet valves. The lower chamber has a support for the film sample, which is a circular ring that fits into a slot in the bottom of the chamber (7). Calculation of the permeability of a sample of film from measurements made with the Isostatic Cell requires a knowledge of the internal volume of the measuring compartment. The volume of each compartment was determined by sealing aluminum foil in the cell and flushing the compartments with nitrogen. The inlet and outlet valves of the cell were then closed, and a known volume of nitrogen was withdrawn from each compartment and replaced with the same volume of oxygen. After an equili- bration period of one hour, the oxygen contents of the cell were redetermined. The volume (V) of each compartment was calculated from the relation: V = (Va - 100) / (Cl - C0) 13 where Va is the volume of oxygen added and C0 and C1 are respectively the percentages of oxygen before and after the addition. With the film sample to be tested clamped between the compartments the lower (measuring) compartment is flushed with nitrogen. The test gas is flowed through the upper compartment across the film sample. A test gas stream of 100% relative humidity was obtained by bubbling the gas through three bottles half filled with distilled water before it flowed through the Isostatic Cell. Flow rate of the test gas was controlled by a rotameter. At periodic intervals a sample was withdrawn from the measuring compartment and replaced with the same amount of nitrogen. The permeability of the film was calculated from the increase in the concentration of the test gas in the measuring compartment. TEST RESULTS The standard definition of gas transmission rate (GTR) is the volume of gas flowing normal to two parallel surfaces, at steady-state conditions through unit area, under unit pres- sure differential and the conditions of test (temperature, specimen, thickness). An accepted unit of GTR is 1 cm3 (at standard conditions) / 24 hours meter2 atmosphere. The test specimen thickness and test temperature must also be stated. The gas permeability coefficient (P) is the volume of a gas flowing normal to two parallel surfaces at unit distance apart (thickness), under steady-state conditions, through unit area under unit pressure differential at a stated test temperature. An accepted unit of P is 1 cubic centimeter / square meter 24 hours atmosphere at the stated temperature of the test (generally 23°C). P=g(STP)-t/A-T-p where g is the volume of gas (cc) t is the thickness (mil) A is the area of the test specimen (m2) l4 15 T is the time_(24.hours) p is the pressure differential (atm.) The Gas Permeability Coefficient is a basic property of a material independent of specimen geometry. It is related to the diffusion rate and the solubility of a gas. Steady state is attained when the volume of test gas transmitted as measured at STP (standard conditions of temper- ature and pressure) becomes linear with time. P was calculated from GTR for the film samples from all measuring techniques as near 23°C as possible. Table 2 is a summary of the data for the samples tested on the Oxtran 100. See Appendix 2 for the permeabilities of each sample. The Permeability Coefficient (P) was calculated from the voltage deflection of the strip chart recorder using the following formula: 5 = (v0 - v1) - RL where V0 and V1 are respectively the final voltage deflection and the baseline deflection. RL is the value of the load re- sistor which converts millivolts to cc/m2/24 hours. There are two resistors—-1 mv = 10 cc/m2/24 hours and 1 mv - 100 cc/m2/24 hours. The smaller factor is used for high barrier materials. 16 TABLE 2 MOCON OXTRAN 100 OXYGEN TRANSMISSION RATE P (:2) Percent cc . mil Relative Temperature 2 Material Humidity (°C) m ~24 hroatm 0 SARAN 0 22.5 61.5 17.4 100 22.5 66.9 15.9 FEP 0 22.5 19,100 916 100 22.5 19,000 3,560 MYLAR 0 22.5 44.1 4.39 100 22.5 44.2 2.25 POLYESTER 0 22.5 42.2 6.46 100 22.5 41.6 4.84 17 In order to calculate the Permeability Coefficient of a film sample on the Permatran C, transmission rate must be expressed in terms of equivalent volume at 0°C, 760 mmHg, it is necessary to correct the volume of the calibrating valve (Va) to obtain the volume (Vo) that would be occupied by the same mass at 0°C, 760 mmHg. _ Pa . 273 V0 ' V°" 60 Ta - 273 Volume of the calibrating valve stem hole (0.0209 cc) where: Va Ta = Laboratory ambient temperature (23°C) Pa = Atmospheric Pressure (742.5) Vo = Mass-equivalent volume at 0°C, 760 mmHg The carbon dioxide transmission rate throughthe film is calculated using the following relationship: _ Vo (cc) . 1440 minutes COZTR - t (minutes) day See Table 3 for the data from the Permatran C. The Permeability calculation of the Isostatic Cell for a homogenous polymer film to a gas may be defined by the relation: P = q . 1 - A ' t - AP where: 1 (cc) is the quantity of gas permeating through a film of thickness 1 (mil) and area A (m2) in time t (hours) with a partial pressure difference AP (atm.) across the film. Table 4 and Table 5 give the P values for oxygen and carbon dioxide transmission respectively (see Appendix 2 for data). Table 6 is a comparison of the Isostatic Cell and the two Modern Control methods of measuring gas transmission rates - Oxtran 100 and the Permatran C. This data is based on the mean of the permeabilities of the samples at 0% and 100% relative humidity. 18 TABLE 3 MOCON PERMATRAN C CARBON DIOXIDE TRANSMISSION RATE Cell _ Temperature P(§) a Material [C02] (°C) SARAN 5 23 577 131 100 23 536 119 FEP 5 23 27,500 5,090 100 23 17,900 7,650 MYLAR 5 23 361 17.7 100 23 337 13.4 POLYESTER 5 23 379 17.6 100 23 349 7.49 19 TABLE 4 ISOSTATIC CELL OXYGEN TRANSMISSION RATE Percent Relative P Material Humidity (x) n 0 SARAN 0 120.7 5 9.69 100 118 4 '888 FEP 0 19,700 5 2,650 100 18,400 4 8,040 MYLAR 0 85.2 5 5.27 100 78.5 4 9.98 POLYESTER 0 74.9 5 7.75 100 76.2 4 4.23 20 TABLE 5 ISOSTATIC CELL CARBON DIOXIDE TRANSMISSION RATE Percent Relative fi _ Material Humidity (x) n O SARAN 0 463 5 112 100 496 5 77.3 FEP 0 15,600 5 1,910 100 14,100 5 1'03° MYLAR O 278 5 22 100 254 5 27.5 POLYESTER 0 288 5 18.5 100 273 5 35.4 21 TABLE 6 METHOD COMPARISON MOCON EQUIPMENT vs. ISOSTATIC CELL fi Test Isostatic Oxtran 100 Permatran Material Gas Cell SARAN O2 119 64.2 C02 479 536 FEP 02 19,600 19,000 CO2 14,800 17,900 MYLAR 02 75.5 41.9 C02 280 337 POLYESTER 02 82.2 44.2 262 348 DISCUSSION Comparison of the permeability coefficients between different methods yielded consistent differences between the methods. The Oxtran 100 was consistently lower than the Isostatic Cell and the Permatran C was consistently higher than the Isostatic Cell. Similar trends in data were observed from method to method such as no difference in permeation values for Mylar and Polyester within the accuracy of the test method (5%), permeation values for Saran are on the order of 30% higher than Mylar and Polyester and similar permeation values were observed from all methods for FEP. Also, the standard devia- tions were on the same order of magnitude for all samples from method to method. Closer analysis showed that a numerical relationship could be derived from the mean differences and these are listed in Table 7 and Table 8. Dividing the mean P value from samples tested on the Isostatic Cell into the mean P value from samples tested on the Oxtran 100 gives a ratio difference of .54 for Saran, Mylar and Polyester. According to this relationship, the value determined for P for a certain film sample on the Isostatic Cell can be multiplied by the ratio of the 22 ‘7 23 TABLE 7 OXTRAN 100 VS. ISOSTATIC CELL Material Saran FEP Mylar Polyester Ratio of Difference P (Oxtran 100) / P (Isostatic Cell) .53 .96 .54 .55 TABLE 8 PERMATRAN C VS. ISOSTATIC CELL Material Saran FEP Mylar Polyester Ratio of Difference P (Permatran C) / f (Isostatic Cell) 1.1 1.3 1.2 1.2 24 difference between methods, in this case .54, to give the P value for the Oxtran 100 for that sample. The reverse is also true, meaning that the P value from the Oxtran 100 can be divided by the ratio difference to obtain the P value from the Isostatic Cell method. A relationship was also found between the Permatran C and the Isostatic Cell which held for all samples tested. P derived on the Permatran C is approximately 1.2 times greater than P derived on the Isostatic Cell. Comparing P values of individual samples between methods also yielded the same ratios. This information can be used as the groundwork for deter- mining a standard method. The fact that numerical relationships could be derived to compare data from different methods when the work was done in one laboratory by one person indicates that if experimental parameters are tightly controlled and the same operational techniques are used, the differences due to detection methods and instrumentation can be compensated for. The permeability coefficients measured by ASTM Committee F2.0 showed similar relationships for the ratio difference between methods for Polyester when comparing the Permatran C to the Isostatic Cell. Because of the small amount of data from ASTM F2.0 for the Isostatic Cell no valid conclusion could be drawn from this but this indicates that further work should be done so that if these ratios derived from this study are found in work from other laboratories they can be incorporated into a standard procedure. 25 There were no similarities in the permeability coef- ficients measured in this study and those of ASTM Committee F2.0. The data from this study may have been different due to the conditioning of all the samples 48 hours before testing at either 0.1% relative humidity or 100% relative humidity. This can not be proved without further laboratory testing which could not be done due to the time limitations on this study. Relative humidity had no effect on the gas transmission rates on the film samples on any equipment used. A 15% de- crease in permeability was expected for Polyester at 100% relative humidity but was not observed (8). This may have been due to the conditioning of the samples before testing. Gas flow rate had a significant effect on the Oxtran 100 with higher flow rates giving a decrease in permeability (See Figure 3). This result differs from increased flow rate data observed by the Ontario Research Foundation (9) where their results showed no effect on permeation values at increased flow rates. This may have been due to the fact that they controlled the oxygen flow rate during testing which was not done for this study. Flow rate had no effect on the other methods. Using test gases of different concentrations on the Permatran C had no significant effect on the gas transmission rate for the sample being tested other than increasing test time. This proves that different concentrations of test gas , can be used to increase test time for poor barrier materials without having any significant effect on the permeability coefficient. 26 Figure 3 The Effect of Test Gas Flow Rate on the Mocon Extran 100 Flow Rate versus Permeability 2000 __ 23 1500 _. L: c ‘0' N N E \ o 3 um 1000 __ 500__ 1 1 4 1 J .10 20 30 40 50 Flow Rate (cc) RECOMMENDATIONS Test the largest number of film samples that would be economically feasible. Gas flow rates between 10 and 30 cc/min. should be used on the Oxtran 100. Any flow rate higher than this will give a decrease in the permeability of the sample. Laboratory temperature must be tightly controlled when using the Mocon equipment because changes in laboratory con— ditions will effect the cell temperature. Decreased concentrations of test gas should be used for poor barrier materials on the Permatran C to obtain more accurate data because this will cause a increase in test time. The thickness of all samples tested should be accurately measured and used as a multiplication factor when determining permeability. BIBLIOGRAPHY BIBLIOGRAPHY Linowitzki, V., "Method of Measurement of the Permeability of Plastic Films," Kunstoffe. American Society for Testing and Materials, Method of Test for Gas Transmission Rate of Plastic Film and Sheeting, D1434-75. Smyser, H.D., "Now: Quick Test Times Check Films Gas Transmission," Package Engineering, May, 1970, pps. 71-730 Mocon Oxtran 100, Operator's Manual. ASTM Committee F-2 on Flexible Barrier Materials. ASTM Committee D-20 on Plastics, Standard Method X-70-038-3. Davis, E.G., Huntington, J.N., "New Cell for Measuring Permeabilities of Film Materials," Australian Packaging, June, 1978, pps. 35-37. Demorest, R.L., "Testing for Gas and Water Vapor Transmission Rates," Food and Drug Packaging, May 4, 1976, pps. 10, 18, 22. Soroka, W., Castelletti, L.T., "Effect of Mocon Oxtran 100 Operating Variables on Polyester Film Oxygen Transmission Rate,: Report No. ENG. R-79-32, Ontario Research Foundation, Mississauga, Ontario, Canada. (a paper) 29 APPENDIX APPENDIX 1 STANDARD METHOD OF TEST FOR OXYGEN GAS TRANSMISSION RATE THROUGH PLASTIC FILM AND SHEETING USING A COULOMETRIC SENSORl 1. Scope 1.1 This method covers a procedure for determination of the steady—state rate of transmission of oxygen gas through plastics in the form of film, sheeting, laminates, coextru- sions, or plastic coated papers or fabrics. It provides for the determination of (1) oxygen gas transmission rate (OZGTR), (2) the permeance of the film to oxygen gas (P02), and (3) oxygen permeability coefficient (POZ) in the case of homo- geneous materials. 2. Applicable Documents 2.1 ASTM Standards D 374 Thickness of Solid Electrical Insuraltion, Methods C and D2 1This method is under the jurisdiction of Committee D-20 on Plastics and is the direct responsibility of Subcommittee 020.70 on Analytical Methods. 2Annual Book of ASTM Standards, Part 35. 31 32 D 618 Conditioning Plastics and Electrical Insu- lating Materials for Testing2 D 883 Definition of Terms Relating to Plastics2 D 1434 Gas Transmission Rate of Plastic Film and Sheeting2 D 1898 Sampling of Plastics2 E 96 Water Vapor Transmission Rate of Material in Sheet Form2 E 177 Use of the Terms Precision and Accuracy as Applied to Measurements of a Property of a Material3 E 380 Standard for Metric Practice3 F 372 Water Vapor Transmission Rate of Flexible Barrier Materials using an Infrared Detec- tion Method4 3. Summary of Method 3.1 The oxygen gas transmission rate is determined after the sample has equilibrated in a dry test environment. In this context, a "dry" environment is considered to be the one in which the relative humidity is less than one percent. 3.2 The specimen is mounted as a sealed semi-barrier be- tween two chambers at ambient atmospheric pressure. One chamber is slowly purged by a stream of nitrogen and the other chamber contains oxygen. As oxygen gas permeates 3Annual ASTM Book of Standards, Part 41. 4Annual Book of ASTM Standards, Part 21. 33 through the film into the nitrogen carrier gas, it is trans- ported to the coulometric detector where it produces an electrical current, the magnitude of which is proportional to the amount of oxygen flowing into the detector per unit time. 4. Significance and Use 4.1 OZGTR is an important determinant of the packaging protection afforded by barrier materials. It is not, however, the sole determinant, and additional tests, based on experi- ence must be used to correlate packaging performance with OZGTR. This method is convenient and lends itself well to quality control measurements. It is suitable as a referee method of testing, provided that buyer and seller have agreed on sampling procedures, standardization procedures, test conditions, and acceptance criteria. 5. Definitions 5.1 Oxygen Transmission Rate (OZGTR) - the quantity of oxygen gas passing through a unit area of the parallel sur- faces of a plastic film per unit time under the conditions of test. The SI unit of Transmission Rate is 1 mol/(mZ-s). The test conditions, including temperature and oxygen partial pressure on both sides of the film, must be stated. 5.2 Oxygen Permeance - is defined as the ratio of the O GTR to the difference between the partial pressure of O2 2 .on the two sides of the film. The SI unit of permeance is 34 1 mol/(m2-5~Pa). The test conditions (see para 5.1) must be stated. 5.3 Oxygen Permeability(POZ) - is the product of the permeance and the thickness of a film. The permeability is meaningful only for homogeneous materials, in which case, it is a property characteristic of the bulk material. This quantity should not be used unless the relationship between thickness and permeance has been verified on tests using several different thicknesses of the material. The SI unit of oxygen permeability is 1 mol/(m‘s-Pa). The test conditions (see para 5.1) must be stated. 5.4 Customary units of OZGTR - a commonly used unit of OZGTR is 1 cm3(STP)/m2-d) at l atmospheric pressure difference where 1 cm3(STP) is 44.58 micromoles, 1 atmosphere is 0.1013 MPa, and one day is 86.4 x 103 s. OZGTR in SI units is ob- tained by multiplying the value in customary units by 5.160 x 10’10. 6. Interferences 6.1 The presence of certain interfering substances in the carrier gas stream may give rise to unwanted electrical out- puts and error factors. Interfering substances include free clorine and some strong oxidizing agents. Exposure to carbon dioxide should also be minimized to avoid damage to the sensor through reaction with the KOH electrolyte. .7. Apparatus 7.1 Oxygen Gas Transmission Apparatus ~ as diagramed in 35 in Figure 15 along with the following: 7.1.1 Diffusion Cell - Shall consist of two metal halves, which, when closed upon the test specimen, will accurately define a circular area. Typical acceptable diffusion cell areas are 100 cm2 and 50cm2. The volume enclosed by each cell half when clamped is not critical; it should be small enough to allow for rapid gas exchange, but not so small that an un- supported film which happens to sag or bulge will contact the top or bottom of the cell. Each half of the diffusion cell shall be provided with a thermometer well for measuring temperature. 7.1.1.1 O-Ring - an appropriately sized groove machined into the oxygen (or test gas) side of the diffusion cell retains a neoprene O-ring. The test area is considered to be that established by the inside contact diameter of the compressed O-ring when the diffusion cell is clamped shut against the test specimen. The area, A, can be obtained by measuring the inside diameter of the imprint left by the O-ring on the specimen after it has been removed from the diffusion cell. 7.1.1.2 The nitrogen (or carrier gas) side of the diffusion cell shall have a flat, raised rim. Since this rim is a criti- cal sealing surface against which the test specimen is pressed, it shall be smooth and flat, without radial scratches. 7.1.1.3 Diffusion Cell Pneumatic Fittings - each half of 5Suitable apparatus, identified as Oxtran 100 Model 100 and Oxtran Model 10-50, can be obtained from Modern Controls Inc., 19228 Industrial Blvd., Elk River, MN 55330. 36 the diffusion cell shall incorporate suitable fittings for the introduction and exhaust of gases without significant loss or leakage. 7.1.1.4 It is desirable to thermostatically control the diffusion cell. A simple resistive heater attached to the carrier gas side of the cell in such a manner as to ensure good thermal contact is adequate for this purpose. A therm- istor sensor and an appropriate control circuit will serve to regulate the cell temperature unless measurements are being made close to ambient temperature. In this case, it is desirable to provide cooling coils to remove some of the heat. 7.1.1.5 Experience has shown that arrangements using multiple diffusion cells are a practical way to increase the number of measurements which can be obtained from a coulo- metric sensor. A valving manifold connects the carrier gas side of each individual diffusion cell to the sensor in a predetermined pattern. Carrier gas is continually purging the carrier gas sides of those cells that are not connected to the sensor. Either test gas or carrier gas, as is approp- riate, purges the test gas chamber of any individual cell. 7.1.2 Catalyst Bed - a small metal tube with fittings for attachment to the inlet on the nitrogen side of the diffusion cell shall contain 3 to 5 grams of 0.5% platinum or palladium catalyst on alumina6 to provide an essentially oxygen free carrier gas. 6A suitable catalyst can be obtained from Englehard Industries Division, Chemical Dept., 429 Delancey St., Newark, NJ 07105. 37 7.1.3 Flowmeter - a metal/glass, ball-in-tube flowmeter having an operating range of 5-100 ml/min is required to monitor the flow rate of the nitrogen carrier gas. 7.1.4 Flow Switching Valves - two four-port ball valves for the switching of the nitrogen and test gas flow streams. 7.1.5 Coulometric Sensor - an oxygen-sensitive coulometric sensor5 operating at an essentially constant efficiency shall be used be used to monitor the quantity Of oxygen transmitted. 7.1.6 Load Resistor - the current generated by the coulo- metric cell shall pass through a resistive load across which the output voltage is measured. Typical values for the load resistor are 5.3 ohms or 53 ohms. These values yield a con— venient relationship between the output voltage and the oxygen transmission rate in customary units (cm3(STP)/m2-d. 7.1.7 Voltage Recorder - a multi-range, potentiometric strip chart recorder shall be used for measuring the voltage developed across the load resistor. The recorder should be capable of measuring a full scale voltage of 50 millivolts. It should be capable of measuring voltages as low as 0.100 millivolts and have a resolution of at least 10 microvolts. An input impedance of 5000 ohms or higher is acceptable. 8. Reagents and Materials 8.1 Nitrogen Carrier Gas - shall consist of a nitrogen and hydrogen mixture in which the percentage of hydrogen shall fall between 0.5 and 3.2 percent by volume. The carrier gas .shall be dry and contain not more than 100 ppm of oxygen. A 38 commercially available mixture known as "forming gas" is suitable. 8.2 Oxygen Test Gas - shall be dry and contain not less than 99.5 percent oxygen (except as provided in para 14.11). 8.3 Sealing Grease - a high-viscosity silicone stopcock grease or a high-vacuum grease is required for sealing the specimen film in the diffusion cell. 9. Precautions 9.1 Extended use of the test unit with no moisture in the gas stream may result in a noticeable decrease in output and response times from the sensor (equivalent to an increase in the calibration factor, O). This condition is due to drying of the sensor and can be corrected by injecting approximately 2 ml of water into the sensor at its inlet connection. 9.2 Temperature - is a critical parameter affecting the measurement of OZGTR. Careful temperature control can help to minimize variations due to temperature fluctuations. During testing, the temperature shall be monitored periodically to the nearest 0.5 K. The average temperature and the range of temperatures found during a test shall both be reported. 9.3 The sensor will require a relatively long time to stabilize to a low reading characteristic of a good barrier after it has been used to test a barrier such as low density polyethylene. For this reason, materials of comparable gas transmission qualities should be tested together. 9.4 Back Diffusion of Air - into the unit is undesirable. 39 Care should therefore be taken to ensure that there is a flow of nitrogen through the system at all times. This flow can be low when the instrument is not being used. 9.5 Elevated Temperatures - can be used to hasten specimen outgassing, provided that the treatment does not alter the basic structure of the specimen (crystallinity, density, etc.) This can be accomplished by the use of the heaters in the diffusion cells. 10. Sampling 10.1 The sampling units used for the determination of O GTR 2 shall be representative of the quality of product for which the data are required, in accordance with Recommended Practice D1898. Care shall be taken to ensure that samples are repre- sentative of conditions across the width and along the length of a roll of film. 11. Test Specimens 11.1 Test Specimens - shall be representative of the material being tested and shall be free of defects, including wrinkles, creases, and pinholes, unless these are a characteristic of the material being tested. 11.2 Average Thickness - shall be determined to the nearest 2.5 micrometers (.0001 inches), using a calibrated dial gage (or equivalent) at a minimum of five points distributed over the entire test area. Maximum, minimum and average values shall be recorded. 40 11.3 If the test specimen is of an assymetrical construc- tion, the two surfaces shall be marked by appropriate dis- tinguishing marks and the orientation of the test specimen in the diffusion cell shall be reported (e.g. "side II was mounted facing the oxygen side of the diffusion cell"). 12. Calibration 12.1 Insert a sheet of Standard Reference Material 14707 in the diffusion cell and measure E0 and Be as described in Section 15, Procedure. Determine the specimen area, A. Convert the permeance value given on the SRM certificate to an OZGTR in the units in which results are to be reported. Use this value of OZGTR to solve the formula given in para. 15.1 for Q, the calibration constant. Repeat the above process using additional sheets of the SRM until the confidence interval for Q defined by the measurements is within acceptable limits. When operating an instrument with multiple diffusion cells, it is desirable to keep a sheet of the SRM in one of the diffusion cells at all times to assure the reliability of the rates being measured. CAUTION: The O GTR of the SRM is adversely affected by water 2 vapor in the test gas or the carrier gas. 12.2 In principle, four electrons are produced by the sensor for each molecule of oxygen gas which passes into it. Experi- ence indicates that production models of the sensors achieve 7Standard Reference Material 1470 is a polyester film whose oxygen permeance has been certified. It is available, in packages of 15 sheets of film, from: Office of Standard Reference Materials, National Bureau of Standards, Washington, D.C. 20234. 41 efficiencies of 95 to 98 percent. Any significant drift in the calibration factor Q should, therefore, be investigated as to its cause and corrective action undertaken. Specimen to specimen variability of SRM 1470 is such that a.Change should never be made in the calibration factor, Q, as the result of a measurement using a single sheet of the SRM. 12.3 The value of Q will be a function of the units in which'the results are to be expressed. If it is desired to change units, Q can be transformed to its proper value in the new set of units simply by using the appropriate relationships between base units (quantity of matter, length, and time) in the new and the old sets of units. 13. Conditioning 13.1 The test specimen shall be trimmed to a size appropri- ate for the diffusion cell in which it will be mounted. In general, this means that the seal around the edge of the diffusion cell should not be impaired if the specimen bulges or sags slightly. After trimming, the specimen shall be conditioned in a desiccator (0.1% R.H.) for a minimum of 48 hours. This does not imply that 48 hours will be sufficient to bring all materials to a condition where their measured OZGTR'S will be reproducible. Previous experience should serve as the primary guide to the suitability of a given conditioning regimen. If a material is being tested with which the user has no previous experience, the effect of .conditioning time should be investigated and a regimen 42 selected such that there is no significant effect due to conditioning time. 13.2 Elevated Temperatures - may be used to accelerate outgassing of films, provided that the temperatures are not so high as to cause alterations of the physical and chemical state of the film (see para 9.5). 13.3 Testing for OZGTR - shall be done in a temperature- controlled environment with the apparatus placed in a draft free location. 14. Procedure for Measuring OZGTR 14.1 Preparation of Apparatus - If preceding tests have exposed the apparatus to high moisture levels, it will be necessary to outgas the system, particularly the catalyst bed, to desorb residual moisture. Water shall be removed from the nitrogen and test gas bubblers. The system can then be dried by slowly purging overnight using dry carrier gas and with the sensor bypassed. Heating the apparatus will speed the drying and outgassing process. 14.2 Inserting the Specimen - Place valve VI (Figure 1) in the sensor bypass position to avoid swamping the detector with air. Unclamp the diffusion cell and open it. Apply a thin layer of sealing grease (see para 8.3) around the raised rim of the lower half of the diffusion cell. Remove the test specimen from the desiccator and place it upon the greased surface, taking care to—avoid wrinkles or creases. ‘Lower the .upper half of the diffusion cell into place and clamp both halves tightly together. 43 14.3 Purging the System - Place valve V2 (Figure 1) in the "carrier position". Start the nitrogen carrier flow and purge air from the upper and lower diffusion cell chambers, using a flow rate of 50 to 60 ml/min (as indicated by the flowmeter). After three or four minutes, reduce the flow rate to the de- sired value between 5 and 15 ml/min. Maintain this configura- tion for 30 minutes. 14.4 Establish Eo - After the system has been flushed with nitrogen for 30 minutes, the valve VI is moved to the "Insert sensor" position. This diverts the carrier gas which has passed through both sides of the diffusion cell into the sensor. At this time, the sensor output, as displayed on the voltage recorder, will usually increase abruptly, indicating that oxygen is entering the sensor with the carrier gas. The most likely sources of this oxygen are (1) outgassing of the sample, (2) leaks in the system, or (3) a combination of (1) and (2). The operator shall observe the recorder trace until the sensor output current stabilizes at a constant value with no signifi- cant trend in either direction. Thick samples may require a purge of several hours, or even overnight, before a steady low value of sensor current is obtained. During this time, valve VI should be in the sensor bypass position except for brief periods when the zero level is being checked. Once a steady low value of sensor current has been obtained, valve VI should be switched to the "insert sensor" position and left there. The observed deflection of the strip chart re- , corder at this time is recorded and labelled Eo. It has 44 been found helpful to periodically test the OZGTR of a piece of brass shim stock in order to ascertain that no leaks or contamination of the carrier gas have developed. 14.5 Once Eo has been established, switCh V2 into.the "oxygen purge" position. This cuts off the flow of nitrogen into the test gas side of the diffusion cell and substitutes a flow of test gas into the diffusion cell. 14.6 The sensor output current, as indicated by the strip chart recorder, should increase gradually, ultimately stabil- izing at a constant value. While some thin films with high diffusion coefficients may reach equilibrium in 30 to 60 minutes, thicker, or more complex structures may require a number of hours to reach a steady rate of gas transmission. The constant value of the voltage on the strip chart recorder should be recorded and labelled as Be. 14.7 Temperature - shall be obtained by monitoring the temperature in the thermometer wells on both sides of the specimen. The specimen temperature may be assumed to be midway between the two values. 14.8 Standby and Shutoff Procedure - At the conclusion of a test, but at a time when it is expected that other tests will be performed soon, the instrument should be placed in a standby condition by taking the following steps: (1) turn VI to "sensor bypass", (2) turn V2 to "carrier purge", (3) turn off the oxygen supply, and (4) reduce the nitrogen flow rate to less than 5 ml/min. These steps will economize on carrier and test gases and will minimize the danger of ruining 45 the sensor because of a film failure while the instrument is not being used for testing. It is desirable to maintain a slow flow of nitrogen through the instrument when it is not being used in order to reduce the back diffusion of air into the sensor. When the instrument is not being used for a long period of time, the electrical power may be turned off. 14.9 Tests In a Moist Environment - Although the coulo- metric method may prove to be suitable for testing barrier materials in a moist environment, test data on which to base suitable precision and accuracy statements are not available. Specific procedures for testing in a moist en- vironment are being developed as a separate ASTM Recommended Practice. 14.10 9 GTR at Temperatures Other than Ambient - may be 2 determined by thermostatically controlling the diffusion cell or by using the heater referred to in para 7.1.1.4, provided that the temperature of the carrier gas does not adversely affect the operation of the sensor. 14.11 Testing Poor Barriers - Films having transmission rates in excess of 200 cm3(STP)/(m2-d) when tested with an oxygen partial pressure difference of one atmosphere are de- fined as poor barriers. Examples of such materials, depend- ing on thickness, include polyethylene, polycarbonate, and polystyrene. High oxygen concentrations in the carrier gas from the testing of poor barriers will tend to produce de- tector saturation. One way to avoid this problem is to use a a test gas which is a mixture with a known concentration of 46 oxygen in nitrogen. The permeance of the film should be calculated for the appropriate partial pressure difference from the permeance and the desired partial pressure difference. Another way to reduce the oxygen concentration in the carrier gas when testing poor barriers is to mask off most of the area of the test specimen using a mask of thin metal or alumi- num foil on both sides of the test specimen by use of a suit- able adhesive such as contact cement or epoxy. The specimen area then becomes equal to the open area of the mask. The effect of varying the area of the open hole in the mask should be tested to ensure that the mask is performing properly. 15. Calculations 15.1 Determine the OZGTR — from the following formula: 2 A-RL Ee = steady state voltage level (see para 14.6). E0 = zero voltage level (see para 14.4). A = specimen area (see para 7.1.1.1). Q = calibration constant (see sec. 12). R + value of load resistance (see para 7.1.6). L 15.2 The Permeance, PO - of the specimen can be determined 2 by means of the following formula: PO2 = P where p = partial pressure of oxygen which is the mol fraction of oxygen times the total pressure (normally, one atmosphere) in the test gas side of the diffusion cell. The partial pressure of 02 on the carrier gas side is considered to be zero. 47 15.3 The Oxygen Permeability Coefficient, 302 - can be determined by means of the following formula: go =PO -t 2 2 where t = average thickness of the specimen (see para 11.2). Results should be expressed as permeabilities only in cases where materials have been determined to be homogeneous by investigation of the relationship between specimen thickness and permeance. 16. Report 16.1 The report shall include the following: 16.1.1 A description of the test specimen, including an identification of the two sides of the material if they are different, a statement as to which side was facing the test gas, the location of the specimen in the lot of material of which it is representative, and the dimensions of the test specimens. 16.1.2 The average thickness of the test specimen as de- termined in para 11.2 and the standard deviation of the thick- ness values. 16.1.3 The barometric pressure at the time of the test. This information is not required if the pressure of the gas on the test gas side of the diffusion cell is maintained by an accurate pressure regulating device. 16.1.4 The partial pressure of the oxygen gas on the test gas side of the diffusion cell and a statement as to how it . was determined. 48 16.1.5 The rate of flow of the nitrogen carrier gas during the test. 16.1.6 The temperature of the test specimen (to the nearest 0.5 K) and the method used to determine the temperature. 16.1.7 The values of OZGTR, Permeance (if desired), and the permeability (if desired). 16.1.8 A description of the apparatus used including, if applicable, the manufacturer's model number and serial number. 16.1.9 A statement of the means used to Obtain the calibra- tion factor, Q. 16.1.10 The effective area for permeation, A, and a descrip- tion of how it was obtained. 17. Precision and Accuracy 17.1 The data given below were obtained from the statistical analysis of an inter-laboratory test program in which sheets of a polyester film nominally 25 micrometers thick were distri- buted to seventeen participants3. The participants used two different models of a commercially available coulometric gas transmission measuring instrument, corrected the data for atmospheric pressure fluctuations, and reported permeance values. No significant difference was found between results obtained from the two models, and the results from the two models were pooled. 17.2 Variability - The statistical analysis of the data 8Supporting data for this method can be obtained from ASTM Headquarters, 1918 Race Street, Philadelphia, PA 29203 by requesting RR:D20-XXXX. 49 took the unequal number of replicates from each respondent into account. The between laboratory standard deviation, sigb, is 5.0 cm3(STP)/m2-d-atm, and within laboratory standard devia- tion, sigw, is 1.2 cm3(STP)/m2-d-atm. 17.3 A Weighted Average Material Value, Xbar, and its standard error were calculated from the data, taking the with- in and between lab variability into account. Xbar was found to be 59.36 cm3(STP)/m2-d~atm and the standard error of 1.21 cm3(STP)/m2-d-atm. 17.4 The material used in the inter-laboratory test was from the same manufacturing lot as NBS Standard Reference Material 1470, which NBS has found to possess an OZGTR equal, at a pressure differential of one atmosphere (.10133 MPa), to 63.8 cm3(STP)/m2-d-atm with a standard error of 0.4 cm3 (STP)/m2-d-atm. This discrepancy may be attributable to differences in sampling, test methodology, or analysis of the data. ‘— T I DIFFUSION SPECIMEN CELL L WWY\ 33;: I J REA fl ‘ I OXYGEN CARRIER ATER PURGE C(’—N\1)PURGE CATALYST m gggggg BYPASS V2‘ 0 SENSOR v1 ’/\ I 'CONTROLLER I I I I I SENSOR I I FLOWMETER Figure l A PRACTICAL ARRANGEMENT OF COMPONENTS FOR THE MEASUREMENT OF OXYGEN TRANSMISSION RATE USING THE COULOMETRIC METHOD ALTERATIONS Section 8.2 - Test gases of 0% and 100% relative humidity were used on the Oxtran 100 and Cell Method. A test gas of 0% relative humidity was used on the Permatran C because tector is sensitive to moisture. Different concentrations of test gases were used when FEP because of its low barrier characteristics. A 5% Balance mixture was used on the Permatran C. Air was the oxygen supply and 5% COZ/Air Balance mixture were the Isostatic Cell. Section 14.1 - The apparatus was conditioned at least previous to testing at the required conditions. 51 the de- testing COZ/Air used as used on 12 hours APPENDIX 2 RAW DATA BY METHOD AND MATERIAL Oxtran 100 SARAN Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 44 100 63 2 0 87 100 43 3 0 66 100 72.5 4 0 64 100 86 5 0 46.5 100 70 52 53 Oxtran 100 FEP Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 20,000 100 24,400 2 0 19,000 100 21,800 3 0 19,000 100. 15,900 4 0 17,500 100 18,100 5 0 20,000 100 15,000 54 Oxtran 100 MYLAR Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 38 100 I 43 2 0 41 100 44.5 3 0 46 100 41 4' 0 48 100 46 5 0 47.5 100 46.5 55 Oxtran 100 POLYESTER Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 42.5 100 ‘ 37.5 2 o 32.5 100 42.5 3 0 40 100 37 4 0 47.5 100 49 5 0 48.5 100 42 56 Permatran C SARAN Gas Flow Rate 2 P Sample # [C02] (cc) (cc/m /24 hrs.) 1 5 20 355 5 50 353 5 100 349. 100 20 352 100 50 353 100 100 347 2 5 20 613 5 50 '598 5 100 605 100 20 590 100 50 586 100 100 595 3 5 20 672 5 50 676 5 100 675 100 20 665 100 50 663 100 100 658 4 5 20 699 5 50 670 5 100 666 100 20 493 100 50 491 100 100 461 5 5 20 610 5 50 590 5 100 578 100 20 591 100 50 591 100 100 590 57 Permatran C F132 P Gas Flow Rate 2 Sample # [C02] (cc) (cc/m [24 hrs.) 1 5 20 23,200 5 50 34,100 5 100 45,500 100 20 13,800 100 50 16,800 100 100 25,200 2 5 20 18,400 5 50 29,900 5 100 37,900 100 20 10,200 100 50 13,100 100 100 14,600 3 5 20 12,200 5 50 20,500 5 .100 26,500 100 20 12,000 100 50 14,100 100 100 13,500- 4 5 20 20,200 5 50 27,800 5 100 31,000 100 20 27,700 100 50 32,900 100 100 35,600 5 5 20 13,700 5 50 25,100 5 100 - 37,700 100 20 12,000 100 50 12,500 100 100 20,000 58 Permatran C MYLAR P Gas Flow Rate 2 Sample # [C02] (cc) (cc/m /24 hrs.) 1 5 20 376 5 50 376 5 100 370 100 20 355 100 50 352 100 100 355 2 5 20 373 5 50 380 5 100 372 100 20, 357 100 50 339 100 100 340 3 5 20 340 5 50 353 5 100 341 100 20 343 100 50 339 100 100 354 4 5 20 365 5 50 361 5 100 366 100 20 334 100 50 333 100 100 332 5 5 ' 20 345 5 50 347 5 100 350 100 20 317 100 50 316 100 100 318 59 Permatran C POLYESTER P Gas Flow Rate 2 Sample # [C02] (cc) (cc/m /24 hrs.) 1 5 20 343 5 50 - 351 5 100 322 100 20 343 100 50 348 100 100 343 2 5 20 372 5 50 382 5 100 373 100 20 348 100 50 349 100 100 348 3 5 20 ' 380 5 . 50 384 5 100 373 100 20 347 100 50 357 100 100 354 4:? 5 20 378 5 50 400 5 100 395 100 20 353 100 50 360 100 100 380 5 5 20 379 5 50 378 5 100 378 100 20 338 100 50 340 100 100 339 60 Isostatic Cell - O2 SARAN Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 117 100 118 2 0 118 100 119 3 0 118 100 118 4 0 143 ' 100 --- 5 0 118 100 107 61 Isostatic Cell - O2 FEP Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 22,800 100 30,400 2 0 20,700 100 13,000 3 0 20,800 100 15,400 4 0 22,600 100 ' ------ 5 0 16,200 100 15,000 62 Isostatic Cell - O2 MYLAR Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 81.4 100 89.0 2 0 81.7 100 66.2 3 -0 84.2 100 83.6 4 0 84.3 100 -—-- 5 0 94.3 100 75.2 63 Isostatic Cell - O2 POLYESTER Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 0 62.1 100 80.1 2 0 73.6 100 75.7 3 0 79.6 100 77.3 4 0 81.9 ' 100 78.7 5 0 77.1 100 69.2 64 Isostatic Cell — CO2 SARAN Percent 2 P Sample # ' Relative Humidity (cc/m /24 hrs.) 1 0 293 100 379 2 0 412 100 514 3 0 529 100 480 4 0 573 100 593 5 0 509 100 513 65 Isostatic Cell - CO2 FEP. Percent 2 P Sample # Relative Humidity (cc/m [24 hrs.) 1 0 . 13,500 100 14,800 2 0 15,600 100 15,500 3 0 14,000 100 13,300 4 0 17,600 - 100 14,000 5 0 17,600 100 13,000 66 Isostatic Cell - CO 2 13 MYLAR Percent 2 Sample # Relative Humidity (cc/m /24 hrs.) 1 0 259 100 252 2 0 290 100 257 3 0 290 100 258 4 0 250 100 260 5 0 301 100 242 67 Isostatic Cell - CO2 POLYESTER Percent 2 P Sample # Relative Humidity (cc/m /24 hrs.) 1 o 299 100 263 2 0 298 100 262 3 o 286 100 246 4 o 303 100 260 5 0 257 100 335 APPENDIX 6 FACTORS AFFECTING CALIBRATION AND PERFORMANCE OF THE OXTRAN 100 Fundamental Relationships The Oxtran 100 utilizes a coulometric cell that operates in accordance with well-known relationships established by Faraday. Unlike other coulometric devices, this cell does not require a bias voltage; it is a constant—current generator, the out— put of which is a linear function of the mass flow rate of oxygen into the cell. On entering the cell, each oxygen molecule reacts at the surface of a graphite cathode to capture 4 electrons: O + 2H 0 + 4e 40H- 2 2 The hydroxide ions ultimately react at the porous cadmium anode to release electrons and form cadmium hydroxide: 2Cd + 40H” - 4e Cd(OH)2 Since each oxygen molecule causes the transfer of 4 electrons, one mole (22.4 liter @ 0°C, 760 mm Hg) of oxygen is equivalent to 4 "Faraday's". 68 69 Noting that one Faraday = 96,500 ampere—seconds, each mole of oxygen will produce 4 x 96,500 - 3.86 x 105 ampere-seconds. Re-stated another way: One cc 02/24 hrs. = 1.99 x 10.4 amperes. In actual practice, the cell current is displayed in terms of the d—c voltage developed across a fixed loan resistance in the cell circuit. Since the detector is a constant-current source, it is possible to choose a value of loan resistance that will yield a convenient relationship between the observed voltage and the desired units in which oxygen transmission rate is to be expressed. At the present time, most laboratories in the United States express oxygen transmission rate in terms of cc (stP)/100 in2/ 24 hours. 70 ZHS\QE MMB<3 #QBm Aflflm m0 mmazmo Bfl UZHDflHM mqflow OO OH OO Om OO om ON OH O HO. H mO. m H.O OH ~.O ON om OO m.O om OO Om O.H OOH ONH m.. OOH OOH O.H OOH m.m OOH m .m ooh 02¢ .BMDU ZOHBflMmHflfio ¢ XHQmemfl ZHS\HS MH< *DBW