The virtual impactor, as an atmospheric particle classification chip, provides scientific guidance for identifying the characteristics of particle composition. Most of the studies related to virtual impactors focus on their size structure design, and the effect of temperature in relation to the dynamic viscosity on the cut-off diameter is rarely considered. In this paper, a new method that can reduce the cut-off particle size without increasing the pressure drop is proposed. Based on COMSOL numerical simulations, a new ultra-low temperature virtual impactor with a cut-off diameter of 2.5 μm was designed. A theoretical analysis and numerical simulation of the relationship between temperature and the performance of the virtual impactor were carried out based on the relationship between temperature and dynamic viscosity. The effects of inlet flow rate (Q), major flow channel width (S), minor flow channel width (L) and split ratio (r) on the performance of the virtual impactor were analyzed. The collection efficiency curves were plotted based on the separation effect of the new virtual impactor on different particle sizes. It was found that the new ultra-low temperature approach reduced the PM2.5 cut-off diameter by 19% compared to the conventional virtual impactor, slightly better than the effect of passing in sheath gas. Meanwhile, the low temperature weakens Brownian motion of the particles, thus reducing the wall loss. In the future, this approach can be applied to nanoparticle virtual impactors to solve the problem of their large pressure drop.
Keywords: PM; air−microfluidic chip; computational fluid dynamics; dynamic viscosity; virtual impactor.