A systematic framework for computer -aided design of engineering rubber formulations

Prasenjeet Ghosh, Purdue University

Abstract

This thesis considers the design of engineering rubber formulations, whose unique properties of elasticity and resilience enable diverse applications. Engineering rubber formulations are a complex mixture of different materials called curatives that includes elastomers, fillers, crosslinking agents, accelerators, activators, retarders, anti-oxidants and processing aids, where the amount of curatives must be adjusted for each application. The characterization of the final properties of the rubber in application is complex and depends on the chemical interplay between the different curatives in formulation via vulcanization chemistry. The details of the processing conditions and the thermal, deformational, and chemical environment encountered in application also have a pronounced effect on the performance of the rubber. Consequently, for much of the history of rubber as an engineering material, its recipe formulations have been developed largely by trial-and-error, rather than by a fundamental understanding. A computer-aided, systematic and automated framework for the design of such materials is proposed in this thesis. The framework requires the solution to two sub-problems: (a) the forward problem, which involves prediction of the desired properties when the formulation is known and (b) the inverse problem that requires identification of the appropriate formulation, given the desired target properties. As part of the forward model, the chemistry of accelerated sulfur vulcanization is reviewed that permits integration of the knowledge of the past five decades in the literature to answer some old questions, reconcile some of the contradicting mechanisms and present a holistic description of the governing chemistry. Based on this mechanistic chemistry, a fundamental kinetic model is derived using population balance equations. The model quantitatively describes, for the first time, the different aspects of vulcanization chemistry. Subsequently, a novel three-dimensional, tensorially valid constitutive equation is developed that predicts the mechanical response of the material under simultaneous deformation, and thermal and chemical aging. The equation is based on the inclusion of the kinetic description of the chemical microstructure in the traditional hyperelastic constitutive models. The inverse problem is addressed as a combinatorial optimization problem using genetic algorithms, and it is demonstrated that genetic algorithms can successfully identify optimal formulations for various complex target properties.

Degree

Ph.D.

Advisors

Caruthers, Purdue University.

Subject Area

Chemical engineering|Materials science|Polymer chemistry

COinS