Minimization Of Turn Induced Dispersion In Capillary Electrophoresis Channel

First, the Lattice Boltzmann model for the motion of samples in the channel flows was established. The velocity and concentration boundary conditions in LBM were investigated. In order to achieve improved accuracy for no-slip velocity boundary conditions, a new method based on the idea of standard bounce-back scheme was proposed. The pure diffusion and convection-diffusion of particles in the Poiseuille flow have been numerically studied with LBM model. The result is consistent with the Taylor-Aris dispersion theory.Second, the electroosmotic flows, which are becoming one of the most successful technologies of driving and controlling microfluids in capillary electrophoresis, were studied in this paper. The mathematical model for it was presented and solved by vortex-stream function method. Numerical simulation of electroosmotic flows in a straight channel and a constant radius turn were carried out.Third, turn induced dispersion in capillary electrophoresis channel and reasons for it were analyzed in this paper. Comparison of several methods for minimizing the dispersion proposed by others was discussed. In order to conquer their limitations, a new approach based on the idea of altering ζ potential at turns was developed to lower the dispersion. Also, this paper presents a new mathematical model describing the dispersion caused by turn. In this model the motion of advection-diffusion of samples is replaced by the motion of pure advection of passive particles, and the dispersion can be quantitatively described by the scatter and symmetry of the particles. Based on this model we develop an optimization algorithm to determine the optimal ζ potential distribution in the turns. Numerical results showed that turn induced dispersion after optimization is only 10% of that before optimization. The model for the dispersion in this work was validated again by applying it to other methods mentioned above.