Separation properties of y-alumina nanofiltration membranes compared to charge regulation model predictions

W.B. Samuel de Lint, Nieck E. Benes*

*Corresponding author for this work

Research output: Contribution to journalArticleAcademicpeer-review

15 Citations (Scopus)
4 Downloads (Pure)


The separation behaviour of asymmetric nanofiltration alumina membranes is determined experimentally for binary NaCl and CaCl2 electrolyte solutions and a ternary NaCl–CaCl2 mixture as a function of pH and pressure. Experimental data suggests that the supported alumina membranes are chemically stable over the pH range 4–10. The measured separation behaviour is compared to the results of a predictive charge-regulation transport model with no adjustable parameters. The model predictions are in good agreement with the experimental data both for the binary as well as for the ternary solution. For pH<6 the retention of CaCl2 is relatively insensitive to the value of the adsorption constant of Ca2+. Obtaining the membrane adsorption parameters by fitting a model to retention data, as is often done in nanofiltration literature, can therefore be delicate. Instead, the acquisition of parameters by independent measurement techniques is clearly preferential. Supports with small pore sizes enhance the membrane's mechanical strength but they can exhibit retention and reduce the flux, leading to a decrease of the overall membrane retention. Two supports with pore sizes of 0.20 and 0.12 μm are compared. The support with the larger pores increased the trans-membrane flux by ≈40% and the overall membrane retention by ≈30%, compared to the support with the smaller pores.
Original languageEnglish
Pages (from-to)149-159
Number of pages11
JournalJournal of membrane science
Issue number1-2
Publication statusPublished - 2005


  • n/a OA procedure
  • Nanofiltration
  • Predictive modelling
  • Charge regulation
  • Maxwell–Stefan
  • Mass transfer
  • Ion exchange


Dive into the research topics of 'Separation properties of y-alumina nanofiltration membranes compared to charge regulation model predictions'. Together they form a unique fingerprint.

Cite this