Optimizing size-exclusion chromatographic conditions using a composite objective function and chemometric tools: application to natural organic matter profiling

Abstract

A novel strategy is suggested for developing a size-exclusion chromatography (SEC) method in order to assess the molecular size distribution of natural organic matter (NOM). This strategy is based on the use of a chromatographic response function (CRF) and a response surface based on a central composite experimental design. The CRF qualifies the resolution degree attained under different SEC conditions with respect to global resolution of the chromatogram, number of distinguishable peaks, and analysis time. Based on a SEC study with a mixture of known organic compounds, predictive polynomial and linear models were developed according to the practice of design of experiments. Additionally, a chromatographic response surface was used, describing the effect of pH, amount of organic solvent, and salt concentration of the mobile phase on the values of the CRF. This new strategy predicted a single linear effect of salt concentration over the quality of the chromatographic separation, and the influence of this experimental variable on the quality of size-exclusion separations of various NOM samples was further discussed. For the analysis of NOM, the optimum settings for the mobile phase composition were as follows: pH 8.5, 11% of acetonitrile, and 15 mM of sodium chloride. Under such analytical conditions, the measured molecular size distributions of water-soluble organic matter from atmospheric particles and aquatic fulvic acids compare reasonably well with the published molecular weight data. The newly developed strategy is a reliable alternative for accomplishing an accurate description of the size-exclusion profiles of NOM from different environments.