Observations on solvent flux and solute rejection across solvent resistant nanofiltration membranes

Organic solvent nanofiltration is an emerging technology made possible by the recent development of solvent resistant nanofiltration (SRNF) membranes. These membranes have many potential applications from continuous operation over many months in refinery systems [1,2] to short term operation for a few hours in batch chemical processes [3]. In this paper, solvent flux decline and membrane separation properties are investigated (including their dependence on pressure), using methanol with quaternary alkyl ammonium bromide salts with molecular weights (MW) in the range 322 to 547 Daltons as solutes. The membranes are characterised in terms of an equivalent uniform pore size using three simple pore flow models: Ferry model, Steric Hindrance Pore (SHP) model and Verniory model.

For permeation of pure solvent, the data suggests a compaction under pressure reaching a maximum level, after which the flux is steady. Only after this initial flux decline can membrane behaviour be considered reproducible. Therefore, prior to use, each membrane should be pre-conditioned with pure solvent until a steady flux is obtained. The effect of different solvents on the flux decline and MWCO curves still remains to be investigated; toluene is currently under investigation. This is particularly important in the case of STARMEM 122 (which we have only investigated in methanol so far) as this type of polyimide SRNF membrane was developed for work in more apolar solvents. The membranes have been characterised using three simple pore flow models in terms of an equivalent (uniform) pore size. The predicted pore size varies with applied pressure and solute size, though these variations are small enough that they may be neglected. Thus, the membrane pore size can be quoted on the basis of an average over all pressures and solutes. Reasonable estimates are obtained for a nanofiltration membrane (0.5–0.8 nm pore radius, corresponding to a porosity of 1–2% for the polyimide membrane) which is expected to effect separations for solutes in the nanometre size range, although, in reality a pore size distribution will exist. The results are also consistent with the results of Bowen et al. [13] who calculate the pore radius of a polyethersulphone nanofiltration membrane as 0.72 nm (calculated from fluxes based on membrane area, that is, setting the porosity to 100% for the calculation). It should be noted that the data presented leads to a rough estimate of the pore size of a membrane, and in reality the pore size distribution should be taken into account. Finally, we note that the data obtained here is for short time periods and should not be used to predict long time membrane performance. However, operating times of 3–12 h may not be uncommon during batch chemicals manufacture and the data does show that it is important, in such short term operations, to condition the membrane by pre-fluxing a sufficient volume of pure solvent prior to filtration.