Biofouling potentials of microporous polysulfone membranes containing a sulfonated polyether-ethersulfone/polyethersulfone block copolymer: correlation of membrane surface properties with bacterial attachment

Multivariate methods were used to identify relationships between bacterial attachment (biofouling potential), water transport, and the surface propert ies of nine modified polysulfone (MPS) membranes comprising blends of polys ulfone (PS) with a sulfonated polyether-ethersulfone/polyethersulfone block copolymer. The topology of the microporous MPS membranes, including surfac e roughness, surface height, pore size and pore geometry were determined by atomic force microscopy (AFM) and digital image analysis. Other measuremen ts included relative surface hydrophobicity by captive bubble contact angle , surface charge (i.e., degree of sulfonation) by uranyl cation binding, wt % solids, porosity, membrane thickness, water flux, and the affinity of mem branes for a hydrophilic Flavobacterium and hydrophobic Mycobacterium speci es. The mycobacteria attached best to the MPS membranes, but the attachment of both organisms was inversely correlated with the mean aspect ratio of p ores, suggesting that irregular or elliptic pens discouraged attachment. Mu ltivariate regression analyses identified the pore mean aspect ratio, mean surface height, PS content, and the n-methylpyrrolidone +propionic acid (NM P-PA) solvent concentration as influential factors in Mycobacterium attachm ent, whereas membrane thickness, surface roughness, pore mean aspect ratio, porosity, and the mean pore area/image area ratio influenced Flavobacteriu m attachment. Cluster analyses revealed that Mycobacterium attachment was a ssociated with hydrophobic determinants of the MPS membranes, including PS content, wt% solids, and air bubble contact angle. In contrast, Flavobacter ictm attachment was primarily associated with membrane thickness and charge (i.e., uranyl cation binding or degree of sulfonation). Membrane flux was inversely correlated with surface hydrophobicity and PS content, but (in co ntrast to cell attachment) positively correlated with most pore geometry pa rameters including the mean aspect ratio, suggesting that pore geometry can be optimized to minimize cell attachment and maximize water transport. Oth er variables influencing water flux included the NMP-PA solvent concentrati on and membrane roughness. The results should facilitate the design of nove l microporous PS membranes having reduced biofouling potentials and greater water fluxes. (C) 1999 Elsevier Science B.V. All rights reserved.