Abstract

Membrane Distillation and Crystallization (MDCr) processes are notable in membrane engineering for their capacity to recover valuable resources from hypersaline brines while water recovery, positioning itself as a sustainable alternative to conventional evaporative techniques. This study introduces a phenomenological model for the MDCr process that addresses challenges such as modeling multicomponent hypersaline brines, integrating mass and heat transfer phenomena, salt crystallization, a semi-empirical representation of membrane fouling, and crystal nucleation and growth kinetic. The methodology encompasses developing and validating the model against experimental data through a meticulous calibration process. The results demonstrate the predictive capability to dynamically simulate this process and accurately estimate the main variables, such as transmembrane flux, temperatures, and crystallization kinetics. The model validation demonstrates good precision in the prediction of transmembrane flux, with an average root mean square deviation (RMSE) and mean absolute percentage error (MAPE) of 0.29 kg/m2h and 8.9 %, respectively. The outlet temperatures of the membrane module for brine and distillate exhibit an average RMSE of 1.24 and 0.97 ◦C and an average MAPE of 2.5 and 3.2 %, respectively. Although the crystal size distribution (CSD) displays a higher level of variability in prediction, the estimation remains satisfactory in terms of order of magnitude and trends. These findings underscore the model’s capability as a robust tool for designing, optimizing, and scaling MDCr processes in various industrial settings.