Lap-on-a-chip system is one of challenging parts in nano and bio engineering fields, for instance, microfluidic channels on the chip are useful for selecting a target particle and mass transferring of boiomolecules in fluid. However, since experimental approach is highly expensive both in time and cost, alternative reliable methods are required to conceive optimized channels. The purpose of this research is to simulate a nanoparticle focusing lens in a microfluidic channel from nanoparticle control point of view. A promising immersed finite element method is expanded to estimate the path of randomly moving nanoparticles through a focusing lens. The channel flow is assumed as incompressible viscous fluid and Brownian motion effects as well as initial position of particle are quantitatively examined. As a representative result, while the nanoparticles with/without Brownian motion were focused along the center of the channel, the concentration factor representing focusing efficiency was calculated. Therefore, it is expected that the newly proposed numerical method considering Brownian motion will be efficiently applicable to design the microfluidic channel containing various particles, molecules and so forth in the near future.