Available online at www.sciencedirect.comComposites: Part B 39 (2008) 782– permeability of various graphite/epoxy composite laminatesfor cryogenic storage systemsSukjoo Choi 1, Bhavani V. Sankar *Department of Mechanical and Aerospace Engineering, P.O. Box 116250, University of Florida Gainesville, FL 32611, USAReceived 2 July 2007; accepted 27 October 2007Available online 24 November 2007AbstractExperiments were performed to investigate the effect of cryogenic cycling on the gas permeability of various composite laminates forcryogenic storage systems. Textile composites have lower permeability than laminated composites even with increasing number of cryogenic cycles. Nano-particles dispersed in one of the ply-interfaces in tape laminates do not show improvement in permeability. Micrographs of sections of various specimens provide some insight into formation of microcracks, and damage before and after cryogeniccycling. In laminated tape composites microcracks in various layers connect and form an easy path for gas leakage. Composites whereinplies of different orientations are dispersed rather than grouped show excellent performance even after cryogenic cycling. In textile composites the damage is restricted to regions contained by the weave yarns and hence the permeability does not increase significantly withcryo-cycling.Ó 2007 Elsevier Ltd. All rights reserved.Keywords: A. Laminates; C. Fractography; E. Prepreg; Gas permeability1. IntroductionIn order for the next generation of space vehicles to beaffordable, it is critically important to achieve a significantreduction in their structural weight thus reducing the costof launching payloads into space. Typically various gasstorage tanks account for about 50% of the dry weight ofa space vehicle. Fiber reinforced composite materials canoffer significant weight reduction and they are candidatematerials for various cryogenic storage systems e.g., theliquid hydrogen (LH2) storage tanks, in the space vehicles.Fiber reinforced composite materials offer many advantages in the design of cryogenic storage tanks such as highspecific stiffness and specific strength, and low coefficientof thermal expansion in the fiber direction.*Corresponding author. Tel.: 1 352 392 6749; fax: 1 352 392 7303.E-mail address: [email protected] (B.V. Sankar).1Present address: Department of Aerospace Engineering, Texas A&MUniversity, College Station, TX, USA.1359-8368/ - see front matter Ó 2007 Elsevier Ltd. All rights reserved.doi:10.1016/j.compositesb.2007.10.010The LH2 storage tanks experience extreme temperaturesduring refueling operation and also during atmosphericreentry of the space vehicle. The external structural loadscombined with the thermal stresses cause microcrack initiation and propagation which could lead to delamination incomposite tanks [1]. The cryogenic storage tank requiresthat the permeability is extremely low, in fact almostimpermeable, so that the cryogenic fuel will not leakthrough the walls of the storage after cryogenic cycling.Therefore, the gas permeability of the material is a criticalfactor for effective and reliable performance.The phenomenon of cryogenic-cycling or simply cryocycling in which the composite structure is subjected toroom temperature and cryogenic temperature alternately,can lead to progressive damage. In fiber reinforced composites, thermal stresses develop at cryogenic temperature,which causes microcrack initiation and propagation. Thermal stresses develop because of the difference in thermalexpansion of the fiber and matrix materials at microscale,and also due to the difference in thermal expansion of adjacent layers of the laminate at macro-scale [2]. When the

S. Choi, B.V. Sankar / Composites: Part B 39 (2008) 782–791sum of thermal stresses and stresses due to external loadsexceeds a critical value for the material system and layup,microcracks develop. When microcracks in the polymermatrix grow, they become transverse cracks, and when atransverse crack reaches the interface between two layers,the crack deflects through the interface and delaminationinitiates [3]. Combination of microcracks and delaminations provide a pathway for the cryogenic fuel and othergasses to leak through the walls of the storage system.The purpose of the present study is to investigate thefundamental issue of gas permeability in composite materials. Gas permeability has been predicted using analyticalmethods and measured experimentally by many researchers. Roy and Benjamin [3] developed an analytical modelto estimate the permeability based on the crack openingdisplacement with a given delamination length, crack density and loading conditions. Stokes [4] performed an experimental investigation to evaluate the permeability of IM7/BMI laminated composites under bi-axial strains. The permeability experimental facility was constructed followingthe ASTM Standard D-1434 [5] in which the permeabilityis determined by the pressure difference across the specimenas a function of time for a specific duration. As the strainwas gradually applied, the permeability increased initiallyand then reached a steady-state until the specimen failed.The crack densities in each layer were measured using optical microscopic inspection. The advantage of this test isthat it minimizes the error due to ambient pressure differences during the test. However, this experimental methodrequires a sophisticate pressure transducer with high sensitivity which is capable of detecting infinitesimal pressurechange across the specimen accurately.Kumazawa et al. [6] performed an experimental investigation to measure the gas leakage of fiber reinforced composites under bi-axial strain and thermal load. Through acombination of experimental testing and finite elementanalysis, the leakage rate as a function of temperaturechange was determined as the crack density is assumed tobe constant.Grimsley et al. [7] constructed an experimental facilitybased on ASTM Standard D-1434 [5] to measure permeability of hybrid composites and related films. The volume-flow rate was estimated by measuring the rate ofmoving distance of a liquid indicator in a glass capillarytube. And, then the gas transmission rate is converted tovolume-flow rate using the ideal gas law. The permeanceis calculated by the gas transmission rate per upstreampressure. Herring [8] investigated the permeability of thinfilm polymers after pre-conditioning the samples.Nettles [9,10] has made a significant contribution inestablishing optimal conditions for permeability experiments. The permeability of laminated composites afterexperiencing impact loads was determined using the volumetric method [10]. Moreover, the study investigated thevarious testing conditions that influence the permeabilityresults. When the glass capillary tube is mounted either vertically or horizontally, the variation in permeability results783was found insignificant. The permeability tests were performed using various types of liquid indicators in the capillary tube, and again the variation in permeability resultswas found to be insignificant. Also, the length of liquidindicator does not affect the permeability results. However,glass capillary tubes with inner diameter of 0.4 mm underestimated the permeability than capillary tubes with1.2 mm and 3 mm diameters. This may be due to the factthat very narrow tubes add to the flow resistance and thusindicate a lower apparent permeability. Nettles also investigated the permeability of composite laminates and bonding materials used for feedline components of a spacevehicle before and after cryogenic cycles. Glass et al. [11]determined the permeability of core materials for composite sandwich structure when shear loads are applied on thesurface. They used Hexcel HRP honeycomb and DupontKorex honeycomb materials in the sandwich structures.The purpose of the present experimental study is to measure the gas permeability of composite laminates used forcryogenic storage tank application. Some of the specimenswere subjected to cryo-cycling. It is found that the permeability initially increased rapidly with cryo-cycling, butbecomes a constant after several cryogenic cycles. Textile(plain woven) composite specimens exhibited lower permeability and also retained the low values after cryogeniccycling compared to laminated composites. Microscopicexamination of specimen cross-sections offers some explanation of behavior of various types of laminates.2. Standard test method for determining gas permeabilityThe permeability is defined by the amount of gas thatpasses through a given material of unit area and unit thickness under unit pressure gradient in unit time. The SI unitof the permeability is mol/s/m/Pa. The standard testmethod for determining gas permeability is documentedin ASTM D14382 (Re-approved in 1997) ‘‘Standard TestMethod for Determining Gas Permeability Characteristicof Plastic Film and Sheeting [5]”. The permeability canbe measured by two experimental methods, monometricdetermination method and volumetric determination methods. The experimental setup for the monometric determination method is shown in Fig. 1 [4,5]. The lowerpressure chamber beneath the specimen in Fig. 1 is initiallyvacuumed and the transmission of the gas through the testspecimen is indicated by an increase in pressure. The permeability can also be measured using volumetric determination as shown in Fig. 2 [5]. The lower pressurechamber is exposed to atmospheric pressure and the transmission of the gas through the test specimen is indicated bya change in volume. The gas volume-flow rate is measuredby recording the rise of liquid indicator in a capillary tubeper unit time. The gas transmission rate (GTR) is calculated using the ideal gas law. The permeance is determinedas the gas transmission rate per pressure differential acrossthe specimen. And, then the permeability is determined bymultiplying permeance by the specimen thickness.

784S. Choi, B.V. Sankar / Composites: Part B 39 (2008) 782–791Fig. 1. Permeability experimental setup for monometric determinationmethod [4].Fig. 2. Permeability experimental setup for volumetric determinationmethod [4].The monometric determination method was not considered for this study since the mercury compound used in theexperiments requires special safety and handling procedures. Therefore, the permeability facility was constructedbased on the volumetric determination method as shown inFig. 2 [5].3. Permeability apparatusThe permeability experimental apparatus basically consists of two chambers between which the specimen is placedas shown in Fig. 3. The permeant gas is pressurized in theupstream chamber. The gas permeates through one side ofthe specimen and escapes out of the other side. The escapedgas is collected in the downstream chamber and flows intoa glass capillary tube. The amount of gas escaping per unittime is measured. The permeance is determined by gastransmission rate and the pressure differential across thespecimen. The permeability P is defined by the product ofpermeance P and the specimen thickness h.The gauge pressure of the gas in the upper upstreamchamber is measured using a pressure transducer (P-303Afrom the Omega Engineering Inc.). The ambient pressureis measured by a barometric sensor (2113A from the PascoScientific). A precision pressure regulator provides constantgas pressure to the upstream chamber. The ambient temperature is measured using a glass capillary thermometer.The specimens are mounted horizontally between theupstream and downstream chambers and clamped firmlyby applying a compressive load (approximately 300 lbs)as shown in Fig. 4. The specimen is sealed with a gasketand an O-Ring (38 mm inner diameter). A force gaugemounted at the top measures the compressive load toensure that the same amount of compressive load is appliedon the specimens for every test. The compressive loadshould be enough to prevent gas leakage, but should notdamage the specimen. The upstream chamber has an inletvent and an outlet vent. The inlet vent allows the gas flowinto the upstream chamber and the outlet vents is used topurge the test gas to atmosphere (see Fig. 4). The downstream chamber has two outlet vents. One is used to purgethe test gas to atmosphere and the other allows the gas flowto the glass capillary tube for measurements. The sensitivity of permeability measurement can be improved byincreasing the gas transmitting area of a specimen.The glass capillary tube is mounted on a rigid aluminumbase horizontally to minimize the gravity effect on the capillary indicator and for easy reading of the scale marks onthe capillary tube. Nettles [9] found that there was no significant difference in the volumetric flow rate when the capillary tube is placed vertically or slanted. The innerdiameter of the glass capillary tube is 1.05 mm and thelength is 100 mm. A magnifying glass is used to read thescale marks at the top of the meniscus of the liquidindicator.The liquid indicator in the glass capillary tube is used tomeasure the rate of rise of the liquid indicator. The rate isused to calculate the volume-flow rate of the escaped gasacross the specimen. Nettles [9] investigated the effects onthe volume-flow rates by using various types of liquid.The volume-flow rates obtained using water; alcohol andalcohol with PhotoFloÒ were not significantly different.In this study methyl alcohol is chosen as the liquid indicator since alcohol has low viscosity and density. The methylalcohol is colored with a blue dye to obtain precise readingson the scale marks.The primary objective of this investigation is to studythe hydrogen permeability of laminated composites. However, hydrogen gas is highly flammable and explosive whenit mixes with air, and it needed to be handled with extremecare during the test. Hence, other permeate gases were considered as a substitute for the hydrogen gas. The molecular

S. Choi, B.V. Sankar / Composites: Part B 39 (2008) 782–791785Fig. 3. Permeabili