Fractionation of natural organic matter (NOM) in water using prepared porous silica based materials as size exclusion (SEC)/GEL permeation chromatography (GPC) stationary phases

Abstract

Natural organic matter (NOM) is a diverse blend of decomposed animal and plant material found in different natural water sources. Due to its large and complex structure, NOM is difficult to both remove and characterize in water. Therefore, there is a need to separate NOM into its components before it can be characterized. The aim of this project was to fractionate NOM through a novel size exclusion chromatography (SEC) composite (poly (styrene-divinyl benzene) (PS-DVB) and Polysilsesquioxane (PSQ)) packed column. Raw and final water samples from Mid-Vaal (MV), Olifantspoort (LO), Mtwalume (MT) and Preekstoel (P) were investigated. Poly (styrene-divinyl benzene) (PS-DVB) and polysilsesquioxane were both synthesized and optimized at various temperatures, compositions and time periods. An end-capping material such as hexamethyldisilizane (HMDS) was added on the PSQ to prevent active silanol groups on the polysilsesquioxane (PSQ) from reacting with active sites of NOM (our analyte). The E-PSQ (end-capped PSQ) and PS-DVB materials were packed in eight different SPE cartridges first, before the materials could be packed in the SEC column. This packing was done to check for the best mass composition of the E-PSQ and PS-DVB. From the obtained SPE results, both the EPSQ and PS-DVB were packed in one SEC/GPC column at a ratio of 1:1 in order to form the composite hybrid material. The packed SEC column was connected to an HPLC instrument and various column efficiency tests were evaluated. The results for the test of interactions with acidic compounds implied that the column can be used for the acidic analytes such as those forming NOM composition (humic acids, fulvic acids) and the column had minimum silanol groups. For hydrophobic interactions the stationary phase strength was different to that of the commercial columns and it could selectively elute molecules based on their different masses. The steric selectivity test showed that the stationary phase could separate and distinguish between molecules with similar hydrophobicity and structure but different shapes (o-terphenyl and triphenylene). The Hydrogen bonding capacity (HBC) test showed that the column had minimum silanol groups and the end-capping was successful on the E-PSQ. After fractionation of all the water samples, the MT raw showed NOM peaks around 1.8 mins, 3.4 mins and 5.3, and the final showed NOM peaks around 1.8 mins and 5.5 mins. The Mid-Vaal (MV) raw and final samples shows NOM peaks at around 1.8 mins and 6 mins. The Preekstoel (P) final water had one NOM peak at around 1.8 mins and raw samples had two NOM peaks around 1.8 mins and 6 mins.