Modeling of turbulent mixing inside a confined impinging jet mixer using Reynolds Averaged Navier-Stokes equations and the quadrature-based moment method

Confined impinging jet mixers (CIJM) are widely used for nano-particle suspension and precipitation due to their small-scale mixing efficiency but are difficult to accurately simulate using traditional Reynolds equation closure models and computationally expensive when using direct numerical simulation (DNS) and large eddy simulation (LES). In this work, binary nonreactive mixing of a CIJM is studied using Reynolds averaged Navier-Stokes (RANS) equations coupled with the extended quadrature method of moments and the Fokker-Planck molecular mixing model. The results of the flow simulation are compared to experimental and DNS simulation data at various inlet flow rates. We compared the results of multiple closure models for this study, but the model that produced the most favorable results was the Launder-Reece-Rodi turbulence model (LRR) due to its ability to handle flow separation and complex vortical structures. We then demonstrated that the Fokker-Planck molecular mixing model resolves the probability density function (PDF) at each grid point to determine the efficacy of mixing at the different inlet flow rates. The model used in this study is less computationally expensive than DNS and LES and is shown to be sufficient to accurately capture turbulent mixing inside CIJM. Thus, the present work provides a framework for industrial improvement to mixing processes using numerical analysis.