Preparation and properties of PVDF/SiO2@GO nanohybrid membranes via thermally induced phase separation method

Graphene oxide (GO), one of the most promising filter materials, is intensively employed to prepare the membranes via non-solvent induced phase separation(NIPS) method, but it has not been found to modify polymeric membranes via thermally induced phase separation (TIPS) method. In this work, the derivative of GO, SiO2@GO nanohybrid, is fabricated and employed to synthesize PVDF/SiO2@GO nanohybrid membranes via TIPS method for the first time. The results indicate that the PVDF/SiO2@GO nanohybrid membranes experience liquid-liquid phase separation mechanism, and exhibit bi-continual and asymmetric structure. The included SiO2@GO nanohybrid is uniformly dispersed in the membrane matrix. With the addition of SiO2@GO, the top surface becomes denser and the pore size decreases; but overhigh SiO2@GO addition for membrane M-5 triggers in the adverse trend. This is caused by the combined actions of nucleation and growth of PVDF and viscosity increase of cast solution due to the addition SiO2@GO nanohybrid. The BSA rejection of membrane gradually increases with the addition of PVDF/SiO2@GO nanohybrid accompanied with the decline of pure water permeation flux. As the SiO2'@GO content increases to 0.9 wt%, the PVDF/SiO2@GO nanohybrid membrane M-4 presents the highest BSA rejection of 91.7% and the lowest permeation flux of 182.6 L m(-2) h(-1) bar(-1). However, the overhigh SiO2@GO addition (1.2 wt%) leads to the outstanding pure water permeation flux of 679.1 L m(-2) h(-1) bar(-1) and a much lower BSA rejection. The addition of PVDF/SiO2@GO nanohybrid evidently improves the surface hydrophilicity and antifouling ability of resultant membranes. The XRD patterns and FTIR spectra of membranes verify the exclusive oc-phase of PVDF, and the melting temperature(T-m) and crystallinity(x(C)) evidently increase with the addition of SiO2@GO nanohybrid up to 0.9 wt% in the dopes. Overhigh SiO2@GO addition (1.2 wt%) leads to small decline for these parameters. (C) 2016 Elsevier B.V. All rights reserved.