Abstract
Spring-back of poly(methyl methacrylate) (PMMA) at large strains, various embossing temperatures, and release temperatures below glass transition is quantified through modified unconfined recovery tests. Cooling, as well as large strains, is shown to reduce the amount of spring-back. Despite reducing the amount of spring-back, these experiments show that there is still a substantial amount present that needs to be accounted for in hot embossing processes. Spring-back is predicted using finite element simulations utilizing a constitutive model for the large strain stress relaxation behavior of PMMA. The model's temperature dependence is modified to account for cooling and focuses on the glass transition temperature region. Spring-back is predicted with this model, capturing the temperature and held strain dependence. Temperature assignment of the sample is found to have the largest effect on simulation accuracy. Interestingly, despite large thermal gradients in the PMMA, a uniform temperature approximation still yields reasonably accurate spring-back predictions. These experiments and simulations fill a substantial gap in knowledge of large strain recovery of PMMA under conditions normally found in hot embossing.