A comparison of surface water natural organic matter in raw filtered water samples, XAD, and reverse osmosis isolates

This research compared raw filtered waters (RFWs), XAD resin isolates (XAD-8 and XAD-4), and reverse osmosis (RO) isolates of several surface water samples from McDonalds Branch, a small freshwater fen in the New Jersey Pine Barrens (USA). RO and XAD-8 are two of the most common techniques used to isolate natural organic matter (NOM) for studies of composition and reactivity; therefore, it is important to understand how the isolates differ from bulk (unisolated) samples and from one another. Although, any comparison between the isolation methods needs to consider that XAD-8 is specifically designed to isolate the humic fraction, whereas RO concentrates a broad range of organic matter and is not specific to humics. The comparison included for all samples: weight average molecular weight (Mw), number average molecular weight (Mn), polydispersity (rho), absorbance at 280 nm normalized to moles C (epsilon280) (RFW and isolates); and for isolates only: elemental analysis, % carbon distribution by 13C NMR, and aqueous FTIR spectra. As expected, RO isolation gave higher yield of NOM than XAD-8, but also higher ash content, especially Si and S. Mw decreased in the order: RO > XAD-8 > RFW > XAD-4. The Mw differences of isolates compared with RFW may be due to selective isolation (fractionation), or possibly in the case of RO to condensation or coagulation during isolation. 13C NMR results were roughly similar for the two methods, but the XAD-8 isolate was slightly higher in 'aromatic' C and the RO isolate was slightly higher in heteroaliphatic and carbonyl C. Infrared spectra indicated a higher carboxyl content for the XAD-8 isolates and a higher ester:carboxyl ratio for the RO isolates. The spectroscopic data thus are consistent with selective isolation of more hydrophobic compounds by XAD-8, and also with potential ester hydrolysis during that process, although further study is needed to determine whether ester hydrolysis does indeed occur. Researchers choosing between XAD and RO isolation methods for NOM need to consider first the purpose of the isolation; i.e., whether humic fractionation is desirable. Beyond that, they should consider the C yield and ash content, as well as the potential for alteration of NOM by ester hydrolysis (XAD) or condensation/coagulation (RO). Furthermore, the RO and XAD methods produce different fractions or isolates so that researchers should be careful when comparing the compositions and reactivities of NOM samples isolated by these two different techniques.