Tracking Inorganic Solids in Biological Enhanced Phosphorus Removal Wastewater Treatment Systems

Master Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
The modelling and tracking of solids in wastewater treatment plants (WWTPs) has historically been focussed on the different types of organic solids. Conversely, there is currently limited information on the composition of inorganic solids, and how they behave across different WWTP unit operations. For the biological removal of organics or nitrogen, this limited information on inorganic solids has no implications as the inorganic solids (sediments) do not interact with the biochemical processes of the WWTP units. For enhanced biological phosphorus removal (EBPR) systems, biochemical processes result in the generation of additional forms of inorganic solids. This includes the formation of polyphosphate by phosphorus accumulating microorganisms (PAOs) and precipitation of inorganic minerals (e.g., struvite). Up until now, the only test that provides information on inorganic settleable or suspended solids (ISS) is the total solids test which yields the total lumped ISS mass of sewage or sludge samples. Furthermore, although there are several models that can provide useful information on the forms and concentrations of ISS in WWTP units (Musvoto et al., 2000; Ekama et al., 2006; Kazadi et al., 2015), no experimental tests exist that can verify the assumptions or predictions of these models. The ISS in EBPR systems can potentially be made up of the following: sediments (clay, silt and sand), polyphosphate (polyP) and mineral precipitates. Based on the ionic concentrations, as well as the conditions in WWTP units, the main forms of mineral precipitates present in EBPR systems are struvite and amorphous calcium phosphate (ACP). For sludge treatment processes downstream of EBPR activated sludge (AS) systems, several assumptions are made on the forms of inorganic solids present in the sludge. For instance, prior to the anaerobic and aerobic sludge digesters, mineral precipitation is currently not simulated, and the polyP molecules are assumed to remain unchanged from the AS system to the anaerobic digesters (AD). Post anaerobic digestion, it is assumed that all polyP has been hydrolysed and the released ions participate in mineral precipitation. The objective of this research is to develop and evaluate an ISS characterisation procedure. The output of the ISS characterisation procedure is a fully characterised ISS i.e., the individual ISS components and their respective concentrations. This ISS characterisation procedure was made up of two major components: an experimental test (named as an ISS characterisation test (ICT)) and a data modelling procedure (DMP). The former is a laboratory test that can be applied to a sample to provide input data for the DMP. The approach taken towards the development of the ICT was to identify base methods from previous literature, and thereafter modifying them to provide analytical measurements to the DMP. Several potential base methods for the ICT were identified in the literature and assessed using a multicriteria decision analysis. The base method was then modified to an ICT. The DMP was developed based on the expected effect of the ICT on the sludge samples, as well as stoichiometric ratios of the different inorganic compounds. To evaluate the performance of the ICT and DMP, the sample set was made up using solution with (i) known amount of mineral precipitate, (ii) EBPR sludge, (iii) a mixture of EBPR sludge and mineral precipitate. The EBPR sludge was obtained from a parallel study that aimed at growing an enhanced culture of PAOs in a University of Cape Town (UCT) configured AS system. Through the multicriteria analysis it was deduced that the cold perchloric acid (PCA) fractionation procedure by De Haas et al. (2000) was the most suitable base method for the ICT. Short extraction times of EBPR sludge with cold PCA acid was proven to be efficient in dissolving mineral precipitates without hydrolysis of polyP molecules (De Haas et al., 2000). The cold PCA fractionation procedure was modified (to the ICT) to include more analytical measurements such as magnesium (Mg), potassium (K), calcium (Ca), free and saline ammonia (FSA) and orthophosphate (OP), such that these measurements can be used in the DMP to characterise ISS. The DMP I (denoted as DMP I for being the first attempt) was developed based on the expected effect of the cold PCA fractionation procedure on the EBPR sludge sample with mineral precipitates. The analysis of the results from the application of the ICT on the EBPR sludge samples showed significant findings. It was demonstrated that the ICT was successful in dissolving the maximum expected concentration of mineral precipitates in its PCA extracts. The application of the ICT on the EBPR sludge sample showed that the hydrolysis of polyP did occur in the PCA extract. PolyP phosphate and counter-ions (Mg, K and Ca) were released into the PCA extract. Results from the application of the ICT on a mixture of EBPR sludge and mineral precipitates showed that the PCA extract contained both mineral precipitates ions and polyP ions. The FSA measurements on the PCA extract showed that the FSA was from the dissolution of struvite only, and there was negligible N from the hydrolysis of biomass. The release of polyP phosphate was consistent for all tests carried on sludge samples. On the other hand, the release of polyP counter-ions Mg and Ca were varied amongst the test cases with sludge samples. The results of the ICT were used as input to the DMP. The output of the DMP I was significantly inaccurate. The concentration of struvite and ACP were overpredicted by 66% and 129%, respectively. The concentration of polyP (as mgP/L) was underpredicted by 28%. It was deduced that the release of polyP counter-ions into the PCA extract led to the overprediction of the mineral precipitate and the release of polyP phosphates into the PCA extract led to the underprediction of polyP phosphate concentration. Following this analysis, a second DMP was developed (DMP II) and evaluated. Using the DMP II, the error in struvite and ACP prediction improved to 15% and 25%, respectively. The error in prediction in polyP concentration increased slightly to 31%. The ICT was not successful in separating the polyP and mineral precipitates due to the hydrolysis of polyP molecules. Hence the percentage error in the prediction of the ISS constituents concentrations were high using both DMPs. The DMP II showed an improvement in the prediction of mineral precipitate concentration. However, there are several caveats to the application of DMP II. First, it depends largely on a few parameters (such as the soluble ammonium) and thus inaccurate measurements of these parameters can result in significant mispredictions form the model. Second, it depends on parameters (namely the Mg:P ratio in polyP and extent of polyP hydrolysis) that need to be calibrated for each experimental investigation. Further research is required to determine what affects these ratios and how it can be parametrised into the DMP II. Although, this research has successfully identified, developed and evaluated a potential ISS characterization procedure for EBPR systems, there is still further research that still needs to be performed on this topic to achieve a fully calibrated and validated procedure.