Laboratory Methods for the Advancement of Wastewater Treatment Modeling

This thesis explores potential relationships between dissolved organic matter and organic phosphorus and methods of removing organic phosphorus from waste water. This thesis also investigates current limitations of pH simulation in the wastewater treatment process. In order to optimize any type of nutrient removal pH simulation must be as advanced as the wastewater treatment technologies.;I. The first study in this thesis represents an effort to probe the relationship between organic phosphorus removal and the molecular nature of dissolved organic matter in wastewater. The results obtained by fluorescence characterization of dissolved organic matter (DOM) in 44 wastewater samples collected from 12 wastewater treatment plants and several different technologies are presented. Samples within treatment plants at various steps in the treatment process allow for the observation of the effects of treatment processes on DOM and phosphorus. PARAllel FACtor analysis (PARAFAC) was used to resolve the fluorescence spectra into three components. It was found that proteinaceous fluorophores, tyrosine (Tyr) and tryptophan (Trp), correlate well with nonreactive phosphorus (nRP) removal. The correlation between Tip, or Tyr, and nRP improved with use of biological treatment. The steepest correlation was determined to be between Tip and nRP for a plant using tertiary biological treatment (R2=.810, r =.900, p<0.01). Secondary biological treatment plants have a more moderate slope and a correlation coefficient R2=0.642, p<0.01 for nRP and Tip. Humic substances (HS) fluorescence was found to have no correlation with nRP removal. It was found that as nRP decreased, HS fluorescence stays relatively the same.;II. In hopes of achieving low level phosphorus in effluents, wastewater treatment plants are potentially interested in implementing advanced oxidative processes (AOPs) to break down non-reactive (or organic) phosphorus to reactive phosphorus which is easier to remove. Chapter 4 explores six wastewater treatment processes in hopes to discover methods of breaking down non-reactive phosphorus. These methods have been used in the past for wastewater disinfection but have never been tested for phosphorus removal. These treatments included AOPs hydrogen peroxide (H2O2), ultraviolet (UV) photolysis, ferrate (FeO42-), ozone (O3) and combined H2O2 and UV photolysis. Absorption chemistry was also investigated using activated carbon, which was discovered to be a source of phosphorus and thus not useful for nRP removal. Changes in non-reactive phosphorus were observed in all AOP treatments except ozone where no change occurred. When used separately, UV photolysis was found to decrease non-reactive phosphorus by approximately 26%. In the combined UV/H2O2 treatment, non-reactive phosphorus was decreased by 18%. Ferrate, the final AOP investigated in this study, was found to decrease total phosphorus by approximately 35%. With the exception of activated carbon, the different treatments investigated show promise of conversion of non-reactive to reactive phosphorus. Ferrate treatment could be very useful due to it being a combination treatment, combining oxidation and chemical phosphorus precipitation.;III. The final chapter of this thesis focuses on the improvement of a pH prediction model for wastewater. Current modeling determines pH based on the concentrations of strong base cations, strong acid anions, weak acids and ammonia. This causes an underestimation of pH because the modeling does not take into account positively charged surface reactive sites in wastewater solids. The effect on proton concentration can be seen in the following rearranged electroneutrality equation: [H+] = sum[Strong acid anions] - 765428sum[Strong base cations] + sum[Weak acid anions] - sum[ positive surface sites]. Characterization of the terms highlighted in bold could lead to the improvement of phosphorus removal modeling. Acid base titrations are an excellent method to probe surface reactivity in terms of proton binding affinities (pKa) and capacities. For each type of reactive surface group, proton binding affinity and ionizable site concentrations are unique. Data obtained from acid-base titrations can be used to determine reactive site concentrations at certain pKa values. This study uses linear programming to calculate reactive site concentrations at various pKa values. A synthetic wastewater recipe was used since characterization of surface reactive sites would lead to an improvement in wastewater treatment modeling. High solid titration data agreed with the model after the addition of two positively charged surface reactive sites to the pH modeling. The first positively charged site had a pKa of around 8 while the second site had a pKa around 10.2. The pKa value of 8 agreed with pKas found for hydrous ferric oxides in literature as well as the pKa spectra calculated using titration data and discrete site analysis for the high solids system. (Abstract shortened by UMI.).