(142r) Simulation of Reaction and Mixing Processes in Turbulent Aqueous CaCO3 Precipitation

Simulation of reaction and mixing processes in turbulent aqueous CaCO₃ precipitation

Derek D. Harris, David O. Lignell

Abstract – AIChE 2012 Annual Meeting

Particulate systems in turbulent flow are computationally expensive to model accurately due to nonlinear interactions between multiphase mixing and reaction processes. In aqueous carbonate precipitation, surface energies, activity coefficients, and the particle size distribution are needed to accurately calculate the rates and timescales of nucleation and growth processes. Crystals formed often have different polymorphic structures which exhibit different equilibria with the liquid phase. Liquid Schmidt numbers are 500 times larger than those found in typical gaseous systems, necessitating a high grid density to resolve the turbulent fluctuations. One-dimensional turbulence (ODT) has proven to be a useful tool in resolving turbulent mixing fluctuations and is used to show the effects of mixing on the precipitation of polymorphic particles. These fluctuations are important to resolve because of how non-linear the nucleation and growth rates are with super saturation.

A kinetic model for carbonate formation has been devised. Model results are presented of mixing between two adjacent aqueous streams of NaCO₃ and CaCl₂. Flow and particle statistics are recovered using multiple realizations of the stochastic mixing process in ODT. The quadrature method of moments (QMOM) is used to model the solid phase particle balance equation for each of three anhydrous CaCO₃ polymorphs and a common amorphous phase. The non-ideal aspects of this system are taken into account using a Debye-Huckel-type model.

A parametric study is presented by varying the temperature of a single stream, the Reynolds number, and the NaCl concentration. These parameters have a strong effect on the timescales of the nucleation and growth kinetics, and the diffusive, and turbulent mixing processes. The interaction between the transport and reaction processes, and their associated timescales is quantified. The results of these detailed simulations will aid in model development for complex flow configurations (using e.g., LES and RANS simulations) and uncertainty quantification in crystal systems.