On the efficiency of convective mixing in a Y-shaped channel

Continuous-flow microfluidic devices are widely used in microbiology, fine organic synthesis, pharmaceuticals, biomedicine, etc. Most applications require rapid mixing of the fluids that pass through the microfluidic chip. The mechanism of natural diffusion is not always efficient due to limitations on the length of the channel. In this work, we numerically study the efficiency of using various mechanisms of natural convection for the mixing of fluids entering the microfluidic chip. Solutions typically differ in buoyancy and diffusion rates of dissolved components, making them sensitive to gravity-dependent instabilities such as Rayleigh-Taylor convection, double diffusion and diffusion layer convection. We consider a Y-shaped microchannel, which is, on the one hand, the simplest, and, on the other hand, a typical element of a microfluidic network. We assume that two miscible solutions independently flow into a common channel where they come into contact. For each type of instability, we numerically estimated the characteristic channel length, after which complete mixing of the solutions occurs. The simulations were performed in the framework of both 2D and 3D models. Finally, we compare the numerical results with the experimental data obtained recently.