Numerical simulation of mixing at 1–1 and 1–2 microfluidic junctions

Enhancing mixing at a very low Reynolds number by a heaving square cylinder

“…In Sadegh Cheri and Latifi (2013), at Re ¼1 the input power coefficient was ≃ C 100 P , being the maximum mixing efficiency ∼60%. Finally, in Sarkar et al (2014), the optimum configuration gave an efficiency ∼63% with an input power coefficient ≃ C 18 P , although these data are for Re ¼ 50. As can be seen, depending on the mixing device, a high variety of efficiencies and input power coefficients can be found and it is not easy to compare one with the others.…”

“…Re ¼1. In that sense, in Chung and Shih (2007), where the mixing takes place in a rhombic micromixer and by means of numerical simulations with the software CFD-CAE and experimentally with the micromixer made of polydimethylsiloxane (PDMS), was found efficiencies between 20% and 55% for different configurations of their mixing unit; in Sadegh Cheri and Latifi (2013), where a microchannel with different mixing chambers with obstacles inside is proposed to enhance mixing, was obtained firstly by numerical simulations based on the finite element method and later by fabricating the optimal micromixer with a soft lithography technique on PDMS material, efficiencies between 56% and 59%; in Hsiao et al (2014), where the mixing is promoted by vortex generators, was studied both numerically, with the software CFD-CAE þ, and experimentally, fabricating the micromixer with PDMS, and finally the authors obtained efficiencies between 10% and 70%; in Sarkar et al (2014), where two fluid junctions are studied numerically, with the software COMSOL Multiphysics, looking for the best mixing downstream, the efficiency ranged between 20% and 85%; and in Parsa and Hormozi (2014), where the mixing was promoted by means of microchannels with sinusoidal side walls, was carried out a preliminary numerical study (based on the finite element method) to obtain the optimal micromixer design which later was fabricated via CO 2 laser micromachining, and they found that the efficiency was between 40% and 85%. Here, in this work, the efficiency ranges between 5% and 32%, a little lower than in most previously mentioned works, but it could be more than 10 times higher than when mixing takes place in a channel either without, or with a steady, square cylinder.…”

“…Many authors usually normalise the standard deviation to define a mixing performance or efficiency η, such that no mixing yields a value of 0 (0% mixing) and totally complete mixing a value of 1 (100% mixing). To that end, (24) is divided by the maximum standard deviation at the inlet s max , and then subtract it from unity (Sadegh Cheri and Latifi, 2013; Sarkar et al, 2014). η, in %, is then defined as…”

“…It is also important in the mixing process the way the fluids enter the mixing device (if through a T-junction, a W-junction, a Y-junction, etc.) in order to enhance the mixing (see Sarkar et al, 2014;Hsieh et al, 2013, among others).…”

Managing false diffusion during second‐order upwind simulations of liquid micromixing

“…The method is not limited to any particular computational scheme, but it does include the assumption that flow at the inlet and outlet to the micromixer is mainly unidirectional and that the gradient of concentration in the predominant flow direction is much smaller than in the primary mixing zone. Nevertheless, recent studies continue to use tetrahedral meshes because of their flexibility and ease of application [14][15][16]. Other higher-order methods (QUICK [12] and MUSCL [13]) achieved corresponding reductions of 6.9-7.6, reinforcing the idea that the SOU method can often perform nearly as well as its more complex, higher-order counterparts.…”