Comparison and development of advanced techniques for organic matter characterisation in water and wastewater processing

Abstract

Organic matter (OM) is one of the most complex natural mixtures on Earth. It is ubiquitous in all natural environments and plays an important role in a number of natural processes. However, despite the rich literature on its role and function in many environments, key aspects of OM composition remain unclear. What is now appreciated, is that OM exists as a complex mixture rather than as a single, defined material, whose structure depends on the source of the OM, the environment in which it was produced or transported to, and its stage of degradation. There are different analytical techniques currently available that are used to study the composition and structure of OM. The more sophisticated techniques, including spectroscopic and chromatographic techniques, can provide valuable chemical or structural information on a sample, and have been widely applied to the study of OM. Characterisation of OM now commonly involves the combined use of techniques for an informative analysis. However, when they are used together, it is unclear how much information provided by the different techniques is the same, and how much is unique. The focus of this study was to develop a protocol to quantitatively compare the information provided by different techniques when used in combination to characterise OM. This approach to assess the extent of technique complementarity relies on multivariate statistics. It involves the use of ordination plots to assess the information qualitatively, and the Spearman Rank Correlation method to assess the information quantitatively. The sophisticated analytical techniques chosen for the multi-technique approach are at the forefront of OM characterisation and include (i) solid-state ¹³C nuclear magnetic resonance (NMR) spectroscopy, (ii) flash pyrolysis-gas chromatography mass spectrometry (py-GCMS), and (iii) high performance size exclusion chromatography (HPSEC). The protocol was initially developed for the combination and comparison of solid-state ¹³C NMR spectroscopy and flash py-GCMS data. The approach was demonstrated on a set of well understood plant residues, where NMR and py-GCMS results could be compared to the literature. This methodology was further demonstrated on a different set of sediment organics. The approach was expanded to include HPSEC data in the analysis of pulp and paper mill water and wastewater (WW) organics. The three-technique approach was then applied to the final set of samples in this thesis, consisting of partially and fully treated sewage effluent OM. In each case, the ordination plots were able to help determine and compare how the different techniques differentiated between the organics in the sample sets. What is novel about the protocol developed is the quantitative comparison of this information. The Spearman Rank Correlation method was able to determine that two techniques in each study provided some complementary information to the analysis, and when the third technique was used, one technique provided unique information only. From this, it was determined that the most beneficial combination of techniques was when some complementary information and some unique information were provided. There was little benefit to the analysis when a large degree of complementary information was provided by two techniques. With the ability to determine how much complementary information is provided, analytical techniques can be more appropriately applied to OM characterisation therefore improving the allocation of resources including time and money. Importantly, the degree of technique complementarity varied with each study. This was a promising result, as the complexity and variability of OM was therefore reflected in the analysis. The limits to the protocol were thought to have been reached in the final study of sewage effluents, as the ordination plots were thought to reflect primarily random variation due to the high degree of similarity between the spectral results. However, these results were put into context by the addition of reference organics to the ordination plots.