Carboxylic acid concentration by forward osmosis processes: Dynamic modeling, experimental validation and simulation

Driven by osmotic pressure, forward osmosis has attracted growing attention in desalination, water purification and wastewater reuse application. The performance behavior of FO process relies upon not only the membrane properties but also the process operation parameters. In addition, the desired process optimization can be defined from these operation parameters. To profoundly comprehend the forward osmosis performance, related to the process operation parameters during concentrating carboxylic acid, the mathematical model was developed by thoroughly enumerating and integrating each single logical phenomenon equation with pertinent variables during the operation of FO process, which aims to emulate a large-scale FO process with a plate-and-frame module configuration. By means of Levenberg-Marquardt algorithm, the 51 precise dependent process variables were simultaneously determined, as functions of time (Dynamic Simulation Model). FO experimental process was carried out to verify the developed model under the same operating conditions by using Thin-Film Composite (TFC) FO membrane whereas feed solution was varied types of carboxylic acids. The good agreement between model predictions and experimental data was observed. The concentration performances of FO process at 30 h system operation for acetic, butyric, valeric and lactic acid were 1.65, 2.2, 2.3 and 2.5 fold increase, respectively, quantified by developed model. As 3D simulation, not only is the developed model aimed to forecast the optimum initial conditions of draw solution for the best concentration performance of FO process, but the results can also be used as guidelines to select the effective regions of initial conditions to achieve the nearby maximum concentration performance.