Search

Enhanced Biological Phosphorus Removal (EBPR) is an established process in wastewater treatment that uses bacteria to reduce phosphorus levels below regulatory discharge limits. Recently, in light of growing political concern over phosphorus sustainability, EBPR has also been recognised as a platform from which phosphorus may be recovered and recycled onto land as fertiliser. Operating EBPR to optimise performance and efficiency is therefore extremely important, but remains a challenge due to poor understanding of these bacterial ecosystems. Raman spectroscopy is a non-invasive, label-free, culture-independent technique capable of analysing live, single cells. Despite its advantages, Raman spectroscopy has been applied to study EBPR bacteria in just a handful of studies and thus has a low profile in this field of research. More work is required to investigate potential areas of application for Raman spectroscopy in EBPR research. The principal thesis presented here is that Raman spectroscopy could be used as a tool to improve EBPR. The Raman spectra used for this investigation were acquired from individual EBPR bacteria dried onto a calcium fluoride substrate. The bacterial samples were collected from three different sources, namely lab-scale sequencing batch reactors located in Edinburgh (University of Edinburgh, UK) and Boston (Northeastern University, USA), and a full-scale EBPR plant in Slough (Thames Water, UK). Using these spectra, some potential applications and limitations of Raman spectroscopy for improving EBPR were explored. In this foundation work, a particular emphasis on spectral analysis methods was kept in light of the benefits of automating analysis as well as the need for standardisation to be able to compare results between different studies and groups. Nine methods were compared for baselining Raman spectra of individual EBPR bacteria. From these, the “small-window moving average” (SWiMA) method was determined to be the best baselining technique for our purposes at the current time. In agreement with earlier studies, the Raman spectroscopic signatures of three key EBPR metabolites – polyphosphate, polyhydroxyalkanoate (PHA) and glycogen – were shown to be clearly identifiable in individual EBPR bacteria when present. The Raman shifts of characteristic spectral bands arising from polyphosphate were shown to vary significantly between samples and the implications of this were discussed. Examples of how the Raman spectra of individual bacteria can be modelled with multivariate tools to open up new areas for research were given. MCR modelling was demonstrated to offer a novel way to normalise the Raman spectra of individual EBPR bacteria prior to quantitative analysis. With the instrumental set-up in this work, the limit of detection (LOD) of aqueous polyphosphate samples was estimated to be approximately 0.08 M and 0.02 M for 10 second and 200 second acquisitions respectively. Future work is required to research ways in which a more comparable form of polyphosphate ‘standard’ might be prepared so that direct correlation can be drawn between measurements made on such a standard and measurements made in bacterial cells. Overall, several applications and challenges of Raman spectroscopy for the investigation of EBPR bacteria are presented in this work together with recommendation for how to process the spectral data. The conclusions drawn from this work indicate that Raman spectroscopy could be used as a tool to improve EBPR but further work is required to refine and apply these methods.