Computational modeling of microfluidic processes using dissipative particle dynamics

The objective of this dissertation is to develop a mesoscopic multiscale particle based computational method, extend its applicability to state-of-the-art problems and simulate microfluidic processes at mesoscale, An in-house comprehensive computational package based on dissipative particle dynamics (DPD) was developed in the last four years to model complex geometries and flows.;First. DPD was studied based on the coarse-graining of actual molecules. The governing equations were then expressed in non-dimensional forms and important parameters such as Reynolds number and Peclet number were identified. The scales of DPD were critically studied and their relation with real scales was established. Schmidt number and diffusivity were studied for a range of parameter and compared with the analytical solutions. The formulation was applied to water flow in microchannels and tested for a wide range of coarse-graining parameter. It was noted that diffusivity was low for high coarse-graining parameter, which resulted in higher values of Schmidt number.;The DPD methodology was then extended to simulate complex geometry. For the first time, non-orthogonal transformation method for particle based system was developed. Transformation function for position and velocity was used to relate physical and computational domains. This was done for each DPD particle at every time step as the points moved freely in the domain possessing the local property of the flow field. This method was benchmarked by simulating converging diverging nozzle for a range of geometric and flow parameters.;On the application side, DPD methodology was applied to model dynamics and deformation of red blood cell at microscale. A discrete particle model was used to construct the cell and minimum energy principle was implemented to achieve the optimized shape under different flow conditions. Deformation and dynamics of RBC in a capillary was simulated for a wide range of shear and flow rates and compared with the experimental and computational results reported in literature.