New and conventional pore size tests in virus-removing membranes

Microorganisms are retained by ultrafiltration (UF) membranes mainly due to size exclusion. The sizes of viruses and membrane pores are close to each other and retention of viruses can be guaranteed only if the precise pore diameter is known. Unfortunately and rather surprisingly, there is no direct method to determine the membrane pore size. As a result, the UF membranes are not trusted to remove the viruses, and the treatment plants are required to enhance viral disinfection. Here we propose a new, simple and effective method for UF pore size determination using aquasols of gold and silver nanoparticles. We synthesized highly monodispersed suspensions ranging in diameter from 3 to 50 nm, which were later transferred through polymer and ceramic UF membranes. The retention percentage was plotted against the particle diameter to determine the pore size for which a membrane has a retention capability of 50, 90 and 100%. The d(50), d(90) and d(100) values were compared with data obtained from conventional transmembrane flux, polyethylene glycol, and dextran tests, and with the retention of phi X 174 and MS2 bacteriophages. The absolute pore size, d(100), for the majority of tested UF membranes is within 40-50 nm, and can only be detected with the new tests. The average 1.2 log retention of hydrophilic phi X 174 was predicted accurately by models based on the virus hydrodynamic radii and d(100) pore size. The 2.5 log MS2 retention suggests hydrophobic interactions in addition to simple ball-through-cylinder geometry.