Modeling and optimization of spinning conditions for polyethersulfone hollow fiber membrance fabrication using non-dominated sorting genetic algorithm-II

Abstract

Optimization of spinning conditions plays a key role in the development of high performance asymmetric hollow fiber membranes. However, from previous studies, in solving these spinning condition optimization problems, they were handled mostly by using an experimentation that varied one of the independent spinning conditions and fixed the others. The common problem is the preparation of hollow fiber membranes that cannot be performed effectively due to inappropriate settings of the spinning conditions. Moreover, complexities in the spinning process have increased where the interaction effects between the spinning conditions with the presence of multiple objectives also affect the optimal spinning conditions. This is one of the main reasons why very little work has been carried out to vary spinning conditions simultaneously. Hence, in order to address these issues, this study focused on a non-dominated sorting genetic algorithm-II (NSGA-II) methodology to optimize the spinning conditions during the fabrication of polyethersulfone (PES) ultrafiltration hollow fiber membranes for oily wastewater treatment to maximize flux and rejection. Spinning conditions that were investigated were dope extrusion rate (DER), air gap length (AGL), coagulation bath temperature (CBT), bore fluid ratio (BFR), and post-treatment time (PT). First, the work was focused on predicting the performance of hollow fiber membranes by considering the design of experiments (DOE) and statistical regression technique as an important approach for modeling flux and rejection. In terms of experiments, a response surface methodology (RSM) and a central composite design (CCD) were used, whereby the factorial part was a fractional factorial design with resolution V and overall, it consisted of a combination of high levels and low levels, center points, as well as axial points. Furthermore, the regression models were generated by employing the Design Expert 6.0.5 software and they were found to be significant and valid. Then, the regression models obtained were proposed as the objective functions of NSGA-II to determine the optimal spinning conditions. The MATLAB software was used to code and execute the NSGA-II. With that, a non-dominated solution set was obtained and reported. It was discovered that the optimal spinning conditions occurred at a DER of 2.20 cm3/min, AGL of 0 cm, CBT of 30 °C, BFR (NMP/H2O) of 0/100 wt.%, and PT of 6 hour. In addition, the membrane morphology under the influence of different spinning conditions was investigated via a scanning electron microscope (SEM). The proposed optimization method based on NSGA-II offered an effective way to attain simple but robust solutions, thus providing an efficient production of PES ultrafiltration hollow fiber membranes to be used in oily wastewater treatment. Therefore, the optimization results contributed by NSGA-II can assist engineers and researchers to make better spinning optimization decisions for the membrane fabrication process.