Collisional granular flows with and without gas interactions in microgravity

We studied flows of agitated spherical grains in a gas. When the grains have large enough inertia and when collisions constitute the dominant mechanism for momentum transfer among them, the particle velocity distribution is determined by collisional rather than by hydrodynamic interactions, and the granular flow is governed by equations derived from the kinetic theory.; To solve these equations, we determined boundary conditions for agitated grains at solid walls of practical interest by considering; the change of momentum and fluctuation energy of the grains in collisions with the wall. Using these conditions, solutions of the governing equations captured granular flows in microgravity experiments and in event-driven simulations.; We extended the theory to gas-particle flows with moderately large grain inertia, in which the viscous gas introduces an additional dissipation mechanism of particle fluctuation energy. When there is also a mean relative velocity between the gas and the particles, the gas gives rise to particle fluctuation energy and to mean drag. Solutions of the resulting equations compared well with Lattice-Boltzmann simulations at large to moderate Stokes numbers. However, theoretical predictions deviated from simulations at small Stokes number or at large particle Reynolds number.; Using a method analogous to the integral treatment of laminar boundary layers, we derived averaged equations to study the development of granular and gas-particle flows in rectangular channels. The corresponding predictions of the stream wise evolution of averaged flow variables such as the mean particle velocity, the granular temperature, and the mean gas velocity agreed well with event-driven simulations.; Finally, we used our analyses to prescribe microgravity experiments in which to test theories for gas-particle interactions with large to moderate particle inertia and small gas inertia. We also predicted uncertainties in measuring the granular mean and fluctuation velocities from a computer vision analysis of video images of the flow.