Layer-by-layer assembly of bio-inspired borate/graphene oxide membranes for dye removal

Dye wastewater is harmful to the ecological environment because of its potential biological toxicity, teratogenicity, carcinogenicity, and mutagenicity. We fabricated a layered graphene oxide (GO) membrane through layer-by-layer (LBL) self-assembly. We used borate to crosslink with GO on a polyethyleneimine (PEI)-coated hydrolyzed polyacrylonitrile (hPAN) support. Fourier transform infrared (FTIR) spectrometry, Raman spectra, and x-ray photoelectron spectroscopy (XPS) confirmed the presence of a crosslinking reaction. The dynamic thermomechanical analysis (DMA) results indicated that the introduction of borate can significantly improve the mechanical properties of the membrane. The Young’s modulus, ultimate tensile strength, and proportional limit of borate that was assembled twice as the outermost layer were increased by 110.81%, 62.37%, and 53.72%, respectively, as compared to those of a single-layered GO membrane. Moreover, the pure water fluxes of the layered GO membrane did not obviously decrease with an increase in the number of layers. The flux of the membrane with an outermost layer of borate was greater than that of the previous GO layer. The salt and dye rejection of the membranes was augmented with an increase in the number of layers. For the GO membrane assembled three times, rejection to methyl orange (MO), methylene blue (MB), NaCl, MgCl2, and MgSO4 reached 74.02%, 88.56%, 14.55%, 27.50%, and 41.95%, respectively. The use of borate as an inorganic crosslinker can avoid the environmental pollution caused by organic agents, and improve the mechanical properties as well as the filter capability of the layered GO membrane. Therefore, this study presents a novel method of membrane preparation for dye removal. •A layer-by-layer strategy for borate/GO membranes with bionic structures is proposed.•Borate/GO membranes exhibit good mechanical properties and permeation flux.•The removal of dyes and salts may be explained by size sieving and Donnan exclusion.