Nanofiltration of concentrated amino acid solutions
Desalination 240 (2009) 78-88
Authors
Abstract
Permeation experiments of aqueous solutions of diprotic amino acids (L-glutamine and glycine) were carried out with numerous commercial polymeric nanofiltration (NF) and tight ultrafiltration (UF) membranes. The aim of this study was to determine the permeate flux and the amino acid rejection as a function of increasing feed concentration and ionization state of the amino acids. So far, apart from a few limited studies, highly diluted solutions have been considered in the literature, although separation and purification of concentrated systems possess particular industrial interests. The concentration of amino acids in the whole range of their solubility was studied with a stepwise pH scan ranging from 0 to (/1 total net charge. Differences in separation behavior in between high feed concentration area and diluted systems are discussed. Considerable higher rejection and flux drop over the concentration was observed in higher pH range, where amino acids are present in dissociated form. Membranes with different type of active layer material show similar concentration dependent tendency in the permeation behavior. This phenomenon can be explained by the dissociation dependency of the osmotic pressure. An altered form of the van’t Hoff law is reported to calculate the pH dependency of the osmotic pressure and verified by vapor pressure osmometry and reverse osmosis (RO) experiments.
Conclusion
Permeation experiments using several polymeric NF and tight UF membranes were carried out with diprotic amino acids for a wide range of feed concentration and for total net charge ranging from 0 to (1. / 1. At low concentration range, commonly below app. 0.2 mol/L, a considerably higher rejection for charged than for zwitterionic amino acids were measured. 2. At higher concentration range, charged amino acids have a much pronounced rejection drop over the concentration than zwitterions. As a result, even lower rejections for anionic amino acids were observed at the applied transmembrane pressure. 3. At low concentration range, depending on the rejection generally below app. 0.1 Á 0.2 mol/L, no flux difference was observed. Above this concentration range, more stressed decrease in flux with increasing concentration occurs for negatively charged amino acids than for zwitterions. 4. The pH dependency of osmotic pressure has not been taken into account in the literature yet. Membranes with different active layer material show identical tendency according to their response for increasing feed concentration. The pH dependency of osmotic pressure is identified as major parameter affecting the separation behavior in higher concentration regime. 5. Altered form of van’t Hoff law is reported for the pH dependency of the osmotic pressure of diprotic amino acids and was confirmed experimentally with vapor pressure osmometry and reverse osmosis. Based on the series of our permeation experiments, we can conclude, that a great care in NF process design for concentrating amino acids is needed if it is based on experimental results of diluted systems, because the separation performance in more concentrated systems is notably influenced by the osmotic pressure, which is a strongly ionisation (i.e. pH) dependent quantity. The altered form of the van’t Hoff law can be used for practical engineering calculations, it can be favorably applied, for instance, in irreversible thermodynamics models. Simulations with Kedem Á Katchalsky equations are under investigation in our laboratory.
Tags
Amino acid, Concentration, Nanofiltration, Osmotic Pressure, PH
Source: http://www.desline.com/articoli/10048.pdf