An ex-situ material state monitoring of curing based on viscoelastic properties in polymer composites
It is well known that time and temperature are not direct measurements of either material state during cure or mechanical properties after cure in polymer composites. Rather, in current practice, the viscoelastic properties and mechanical properties during cure are merely assumed based on an extensive statistical data of time and temperature history of the material. This practice is time-consuming and costly. In this study, an ex-situ estimation of the actual material states using advanced analytical instruments has been proposed and validated. An encapsulated sample rheometer was used as the main ex-situ instrument capable of measuring with high repeatability and robustness necessary for the validation of viscoelastic curing models of composites (especially prepregs) at production level. This rheometer was coupled with Differential Scanning Calorimetry (DSC) to obtain a correlation of the key variables of curing and the actual material states. Experimental and analytical modeling studies of the viscoelastic properties and thermal properties of four commercial prepregs were conducted using these instruments. These key cure variables were directly correlated with the viscoelastic states of the material during cure. The viscoelastic properties such as storage modulus, loss modulus, and tanδ and the glass transition temperature of the Advanced Composite Group (ACG) MTM45 and MTM45-1 prepregs and Cytec 977-2 PW and 977-2 UD prepregs were measured using the rheometer during different isothermal cure cycles below the final glass transition temperature (g∞ T). Thermal analysis of 977-2 PW and 977-2 UD prepregs was obtained using the DSC and these thermal results were correlated to the rheometry measurements. Glass transition temperatures (g T) of the cure cycles were measured using both rheometry and DSC techniques. A semi-empirical curing model, based on the viscoelastic properties of prepregs, was developed and compared with the experimental data collected at a constant frequency. A process engineer could use this curing model to monitor, control, and optimize a cure process, and to aid in the curing of parts that have time and temperature history discrepancies. This model can be statistically correlated to critical composite properties and can be validated with time and temperature feedback. Therefore, temperature sensors, such as thermocouples, would remain as the primary in-situ sensors and there would be no need for material state sensors inside the autoclave or other processing units.
Thesis (M.S)-- Wichita State University, College of Engineering, Dept. of Mechanical Engineering