Matrix Characterization and Development for the Vacuum Assisted Resin Transfer Molding Process

The curing kinetics and viscosity of an epoxy resin system, SI-ZG-5A, have been characterized for application in the vacuum assisted resin transfer molding (VARTM) process. Impregnation of a typical carbon fiber perform provided the test bed for the characterization. Process simulations were carried out using the process model, COMPRO [8], to examine heat transfer and curing kinetics for a fully impregnated panel, neglecting resin flow. The predicted viscosity profile and final degree of cure were found to be in good agreement with experimental observations. INTRODUCTION The VARTM process has been developed over the last ten years for application in both commercial and military, ground-based and marine composite structures [1-3]. The process has advantages over conventional RTM by eliminating the costs associated with matched-metal mold making, volatiles emission, and allowing high injection pressures [4]. VARTM is typically a three-step process including lay-up of a fiber preform, impregnation of the preform with resin, and cure of the impregnated panel. The reinforcement, in the form of woven carbon or glass fabric, is laid onto a rigid tool surface. The matched metal top commonly found in RTM is replaced in the VARTM process by a formable vacuum bag material. The resin is injected through a single or multiple inlet ports depending upon part size and shape. A vacuum port allows the fiber preform to be evacuated prior to injection and provides the mechanism for transfer of the resin into the part. In addition to the pressure 1 This paper is declared a work of the U.S. Government and is not subject to copyright protection in the United States 2 NASA Langley Research Center, Hampton, VA 3 Old Dominion University, Norfolk, VA 4 Virginia Polytechnic Institute and State University, Blacksburg, VA gradient caused by the vacuum pressure, gravity and capillary flow effects must be considered [5]. The preform infiltration time is a function of the resin viscosity, the preform permeability and the applied pressure gradient. The infiltration time can be greatly reduced by utilizing a distribution medium with a higher permeability than the preform [6,7]. Consequently, the resin flows in the medium first and then the infiltration process continues through the preform thickness. Work at NASA Langley Research Center has focused on further developing of the VARTM process for fabrication of aircraft-quality composite parts. In order to succeed, it is important to achieve the high mechanical properties and dimensional tolerances required in these applications. The development or selection of the matrix material for application in advanced composite structures cannot be divorced from the manufacturing processes and the specific application geometries. Rather, it is necessary to develop material systems that meet a variety of requirements to ensure successful applications. In addition to the required strength and durability of the polymer matrix, properties that govern the processing characteristics must be considered. Traditionally, development or selection of the matrix for a particular application has been based on a limited number of factors like toughness, glass transition temperature (Tg) and viscosity. Consequently, process difficulties have often been encountered which prevented successful application of the matrix in the structure of interest. The use of process models allows sensitivity analyses that determine the influence of a larger number of polymer properties on final part quality. Material development and characterization efforts can then be focused on the most important parameters for a given application. For example, it was found that the interaction between resin modulus development and cure shrinkage has a profound effect on part dimensional stability [8]. In the present work, the cure kinetics and viscosity of a typical VARTM resin system were characterized. The cure kinetics and viscosity models were validated by processing a carbon fiber composite panel with a typical process cycle. Finally, a processing model was used to study the resin curing behavior for different process cycles and panel thicknesses. MATRIX CHARACTERIZATION Accurate prediction of many of the key material properties required in composites process models such as resin viscosity, modulus development and cure shrinkage depend on an accurate knowledge of the cure state of the resin during processing. Furthermore, an understanding of resin viscosity behavior is also required to predict the flow of resin during VARTM infiltration. Cure kinetics and viscosity models are obtained for the resin using a combination of isothermal and dynamic differential scanning calorimeter (DSC) and parallel-plate rheometer scans, respectively. In this work, SI-ZG-5A, a commercially available epoxy blend VARTM resin developed at A.T.A.R.D Laboratories [9] was selected. ∗ Use of trade names or manufacturers does not constitute and official endorsement, either expressed or implied, by the National Aeronautics and Space Administration Cure kinetics model All tests were preformed on a Shimadzu DSC-50 differential scanning calorimeter. The total heat of reaction (HR) was measured from dynamic scans at 1.1°C/minute from room temperature up to 250°C. The isothermal tests were preformed at temperatures ranging from 50°C to 140°C. In these tests, the specimens were heated rapidly to the desired temperature where they were maintained for a total of 2 hours, and then rapidly cooled. The isothermal tests were followed by a dynamic scan at 1.1°C/minute to measure the residual heat of reaction. Raw data from the DSC experiments consisted of measurements of heat flow and total resin heat of reaction as calculated by the apparatus software. The dynamic runs produced very similar heat flow curves (Figure 1) and the measured HR was nearly the same in all cases: 350 kJ/kg. From the baseline heat flow ( baseline q& ) and the total heat flow, the resin cure rate was then determined using: