Rarefaction and temperature gradient effect on the performance of the Knudsen pump

The prediction of the multiscale flow in the Knudsen pump is important for understanding its pumping mechanism. However, there is little research on such interesting multiscale phenomenon in the Knudsen pumps. In this paper, a novel numerical analysis method combining the direct simulation Monte Carlo (DSMC) method with the smoothed particle hydrodynamics (SPH) method is presented for simulating the multiscale flow, which is often encountered in the application of the Knudsen pumps. Validity and accuracy of the new method are given by comparing its results with that of the previous research. Using the coupled multiscale approach, the rarefaction and the temperature drive are studied, which are two main factors on the performance of the Knudsen pumps. To investigate the effect of rarefaction on the performance of the Knudsen pump, various pump operation pressures are compared. The flow characteristics and pumping ability at different rarefaction are analyzed, and the phenomenon of the multiscale flow is also discussed. Several cases with different linear or nonlinear temperature gradients are set to investigate the effect of temperature gradient on the performance of the Knudsen pump. The flow characteristics of the Knudsen pump such as the velocity, pressure increase, and the mass flowrate are presented. A unique phenomenon, the reverse transpiration effect caused by the nonlinear temperature gradient is studied, and the reason of the significant pressure increase in the pump channel is also analyzed. Since the multiscale gas flow is widely encountered in the microflow systems, the above method and its results can also be greatly beneficial and provide significant insights for the design of the MEMS devices.