Rekha R. Rao, Ph.D.
Distinguished Member of Technical Staff
Sandia National Laboratory
Fri., Sept. 6, 3:00 pm
Sidney & Marian Green Classroom (3550 MEK)
ABSTRACT: Polyurethane structural foams are used as supports in a wide range of industries from furniture to refrigeration. Accurate models of the foam filling and curing and the resulting physical properties, from thermal conductivity to density and modulus, can be useful for designing both the mold, the formulation, and the foaming process. We have been developing detailed process-structure-property models of a moderate-density structural polyurethane, PMDI-10. The models begin with the reactive fluid mechanics of the foaming and filling process, which predict the solid properties and the voids and defects in the part. The properties and fields such as temperature and extent of reaction are then fed into a nonlinear viscoelastic solid model, to predict the manufacturing stresses and warpage over foam part lifetime. In this talk, we will focus on the fluid model, which is a coupled finite element discretization of the equations of motion with homogenized continuum foaming and polymerization kinetics. A level set method is used to capture the location of the free surface over time. The foam cures and foams simultaneously, making it very sensitive to processing parameters such as temperature, orientation, and foam injection location.
A validation study is presented comparing the results of the computational modeling of foam filling and curing to a flow visualization experiment of foam filling a clear mold with complex geometric features. This presentation uses a computational fluid dynamic model, which was published in a recent paper [1]. Various mold tilt angles and temperatures are evaluated to determine the best process variable to minimize defects. Metrics are developed to evaluate the efficacy of the filling process such as deviation from constant density and maximum void content. The experiment and models match well, giving confidence in the numerical predictions. Extensions of the model to include more bubble-scale information and improved interface capturing methods will also be discussed as well as coupling to the solid mechanics simulations.
BIO: Dr. Rekha Rao is a Distinguished Member of Technical Staff at Sandia National Laboratories. She came to Sandia in 1990 after earning her BS from UC Berkeley and Ph.D. from the University of Washington, both in Chemical Engineering. Rekha is a leading expert in the computational mechanics of complex fluids, including theoretical development, numerical algorithms, and finite element implementation. She is one of the founding authors of Goma 6.0, an R&D 100 winning software package for process flow modeling. Rekha is currently serving as a guest editor for a special issue of Computers & Fluids on free and moving boundary problems. Rekha’s research has spanned model development in support of energy-production, environmental issues, polymer processing, and manufacturing. Her work on foam process models have led to publications, collaborations with industry, and to a production computational capability impacting manufacturing yields. Rekha is active in the International Association of Computational Mechanics where she organizes symposia and serves as Chair of the Female Research Committee, mentoring and encouraging women in computational mechanics.