Application of factorial design for modeling reverse osmosis process using thin film composite polyamide membrane: a theoretical analysis and experimental validation

In this paper, Regression model using a full factorial experimental design with four factors including operating pressure (P-f), salt concentration (C-f), feed flow rate (Q(f)), and feed temperature (T-f) at two levels were established to model and optimize the reverse osmosis (RO) process performance. Analysis of variance (ANOVA) technique was used to identify the main and interaction effects of the considered factors on the permeate flux (J(p)). Comparisons of model predictions against analytical and experimental data of J(p) are analyzed and plotted on 3D response surface and 2D contour plots. The results show that the effect of operating pressure and feed flow rate were significantly influencing factors on the permeate flux at comparatively high feed water temperature and low salt concentration. The chosen model exhibit a satisfactory agreement compared with theoretical and experimental results. An optimal combination of P-f, C-f, Q(f), and T-f were found to be 6.3 MPa, 25.6 g/L, 18.0 L/min and 34.6 degrees C, respectively, resulting in a maximum J(p) of 44.756 L/h.m(2). Within these optimum conditions, the observed permeate flux is closer to that predicted.