Role of Membrane and Compound Properties in Affecting the Rejection of Pharmaceuticals by Different RO/NF Membranes

Yang-ying Zhao,Fan-xin Kong,Zhi Wang,Hong-wei Yang,Xiao-mao Wang,Yuefeng F. Xie,T. David Waite
DOI: https://doi.org/10.1007/s11783-017-0975-x
IF: 6.048
2017-01-01
Frontiers of Environmental Science & Engineering
Abstract:Rejection of pharmaceuticals (PhACs) followed the order NF90 ≈ ESPA1>NF270>HL.Electrostatic effect had an important role in PhAC rejection by loose NF membranes.Effect of adsorption on rejection followed the order HL>ESPA1>NF270>NF90.High hydrogen bond formation potential of PhACs impaired the rejection by HL.This study was conducted to assess the merits and limitations of various high-pressure membranes, tight nanofiltration (NF) membranes in particular, for the removal of trace organic compounds (TrOCs). The performance of a low-pressure reverse osmosis (LPRO) membrane (ESPA1), a tight NF membrane (NF90) and two loose NF membranes (HL and NF270) was compared for the rejection of 23 different pharmaceuticals (PhACs). Efforts were also devoted to understand the effect of adsorption on the rejection performance of each membrane. Difference in hydrogen bond formation potential (HFP) was taken into consideration. Results showed that NF90 performed similarly to ESPA1 with mean rejection higher than 95%. NF270 outperformed HL in terms of both water permeability and PhAC rejection higher than 90%. Electrostatic effects were more significant in PhAC rejection by loose NF membranes than tight NF and LPRO membranes. The adverse effect of adsorption on rejection by HL and ESPA1 was more substantial than NF270 and NF90, which could not be simply explained by the difference in membrane surface hydrophobicity, selective layer thickness or pore size. The HL membrane had a lower rejection of PhACs of higher hydrophobicity (log D>0) and higher HFP (>0.02). Nevertheless, the effects of PhAC hydrophobicity and HFP on rejection by ESPA1 could not be discerned. Poor rejection of certain PhACs could generally be explained by aspects of steric hindrance, electrostatic interactions and adsorption. High-pressure membranes like NF90 and NF270 have a high promise in TrOC removal from contaminated water.
What problem does this paper attempt to address?