Deconvolution of Size Exclusion Chromatograms: New Insights into the Molecular Weight Distribution of Dissolved Organic Matter in Ozone and Biological Activated Carbon
Minkyu Park, Doorae Lee, Shane A. Snyder
Abstract
Size exclusion chromatography (SEC) is a useful tool for understanding the apparent molecular weight (AMW) distribution of dissolved organic matter (DOM). However, the limitation that a single analytical column generally cannot resolve the entire AMW range of DOM from hundreds of daltons to hundreds of kilodaltons often oversimplifies the interpretation of the AMW distribution. This study applied the nonnegative matrix factorization to deconvolute SEC chromatograms and cross-validated deconvoluted components. The algorithm was further applied to SEC chromatograms of samples collected from a pilot-scale ozone–biological activated carbon system. The trimodal AMW distribution of chromatograms was deconvoluted into six components: three from the biopolymer and the other three from humic and low-molecular weight (humic/LMW) substances. The deconvolution process captured the formation of a biopolymer fraction that was not noticeably observed without chemometric deconvolution. Also, the increased level of removal of the lowest-molecular weight fraction at the elevated ozone dose from 4.1 to 6.4 mg of O3/L suggests that ozone enhanced biodegradability selectively for low-molecular weight fractions. The proposed chemometric method can be a useful tool for providing novel insights into SEC chromatogram interpretations. MATLAB-based GUI software for SEC chromatogram deconvolution was developed and is available online.