INCITE Project

Rheology deals with flow and deformation of materials. Rheology of dense suspensions is a complex phenomenon encompassing multiple length and time scales, and diverse physics ranging from hydrodynamics to electrostatics. Dense suspensions have applications in many industrial processes ranging from ceramics to polymers, and food industry to pharmaceuticals. Innate in these suspensions is the complex microstructure that includes solid particles with broad size and shape distributions, a highly non-ideal solvent phase, and additives like polymers, and surfactants that act as binders and lubricants. Macroscopic rheology has a complex dependence on all of these factors and it is imperative that they are accurately taken into consideration. Understandably, computational modeling of such a system is complex and intensive, and calls for an elaborate computational exercise. Most of the understanding in this area at the present time is empirical with a heavy experimental bias. Theoretical and computational work in this area has focused on idealized systems.

The work proposed here is part of a long-term effort to develop theoretical understanding and computational tools to study the rheology of dense suspensions that are comprised of nonspherical oxide particles, with a large distribution in size, in a complex aqueous organic solvent. A bottom-up, multi-scale approach is being espoused wherein relevant phenomena at atomistic and mesoscopic regimes are modeled and appropriately coupled to render macroscopic rheological properties. We are developing a generalized "Dissipative Particle Dynamics" (DPD)-based computational code to accomplish this. The DPD formulation deals with phenomena in the mesoscopic regime, which is the most relevant regime here. We are also employing novel atomistic calculations using the molecular dynamics (MD) technique to calculate force fields for the mesoscopic DPD formulation. For performing MD calculations, we are using a well-established code called LAMMPS, while the DPD calculations use a Corning-developed research code.

We are requesting INCITE resources to extend the development and validation of the DPD code for realistic suspensions, and then use the code to study the rheology of these systems by conducting large amounts of intensive computations. An integral part of this effort is the development of accurate force fields from atomistic calculations. To accomplish this, we plan to use INCITE resources to run the LAMMPS MD code. With the availability of INCITE resources, many realistic suspensions can be analyzed under conditions that prevail in real operations. Since much of the understanding in this field is still far from complete, with the availability of INCITE type facilities we can analyze many theories and algorithms for a large number of samples to ascertain the best among them. This program can make significant contributions to the entire field of computational rheology of complex fluids, which is still in its infancy.