The objectives of the proposal are to 1) implement a new improved boundary treatment method (refer to as Multiple Boundary Tangent (MBT) developed by PI) to simulation of complex geometries such as flow over an airfoil, and 2) develop an incompressible multiphase SPH (IMSPH) algorithm, and 3) study feasibility of modeling single crystal growth (CG) processes with SPH. In literature, benchmark simulations using previously reported boundary treatments can suffer from particle penetration and may corrupt simulations near solid boundaries. Current SPH boundary treatments do not properly treat curved boundaries in complicated flows. These drawbacks are remedied in MBT. Previously solved benchmark problems using MBT were relatively simple. To understand full power and limitations of MBT, PI plans to solve complex flow problems such as flow over an airfoil and free surface flow. PI’s current incompressible SPH code will be further improved to develop an IMSPH code. In this direction, a novel multiphase projection formulation will be developed to treat sharp changes in transport properties and density. Several benchmark problems will be solved for validation. Resin filling process in polymer composite manufacturing by RTM will be simulated as a multiphase flow. SPH is convenient for simulations involving moving interfacial surfaces such as semiconductor CG. In CG, mesh-dependent methods are difficult to implement, as computational mesh is required to adapt to CG surface. In modeling of ternary alloy crystals, governing equations must be solved simultaneously in liquid and solid phases to resolve growth interface correctly. This requires iterations at liquid/solid interfaces, slowing down simulations notably. In SPH, there is no need for interface iterations; all domains are solved as a single domain. However, SPH is still slow compared with the finite volume method. It requires further research and developments for computational efficiency.