Understanding and Modeling the Microbial Interactions in a Novel Nitrogen Removal Process

Xueming Chen, Bing-Jie Ni (Main supervisor), Zhiguo Yuan (Supervisor)

Research output: Book/ReportPh.D. thesis

Abstract

Because of its economic benefits, the combined partial nitritation-anaerobic ammonium oxidation (PN-Anammox) process is increasingly adopted/recommended for efficient nitrogen removal, despite its application limitations such as the concomitant nitrate production. The discovery of denitrifying anaerobic methane oxidation (DAMO) processes offers a potential solution to overcome the limitations of the PN-Anammox process and enables complete/high-level nitrogen removal through coupling PN-Anammox-DAMO, which could be favourably supported in membrane biofilm reactor (MBfR) systems. However, this novel integrated process involves complex microbial interactions between functional microorganisms including ammonium-oxidizing bacteria (AOB), nitrite-oxidizing bacteria (NOB), Anammox bacteria, DAMO archaea, and DAMO bacteria, which determine the treatment performance. Although mathematical modeling is widely applied to generate multifaceted analysis of emerging technologies, limited work has been dedicated to investigating the integrated PN-Anammox-DAMO process. Hence, this thesis aims to develop mathematical models to understand/evaluate the novel MBfR-based complete/high-level nitrogen removal technology with focus on the system performance and microbial interactions under different conditions of operations and reactor configurations. To this end, a mathematical model was firstly developed to describe the coupled Anammox-DAMO process in a lab-scale MBfR. The MBfR with an Anammox-DAMO dominated biofilm was fed with methane through gas-permeable membranes, while nitrate and ammonium fed in the bulk liquid outside the biofilm. The key stoichiometric and kinetic parameters of DAMO microorganisms were calibrated using the long-term dynamic experimental data, and the model was successfully validated using two independent batch tests at different operational stages of the MBfR. The developed model was then extended with nitrite inhibition terms and further verified by another sets of batch experimental data obtained from the MBfR with different practically applied feeding compositions (i.e., ammonium and nitrite/nitrate). The verified model was applied to assess the feasibility of achieving complete nitrogen removal by a partial nitritation reactor followed by an MBfR performing Anammox and DAMO. The optimum NO2-/NH4+ ratio produced from the preceding partial nitritation for the Anammox-DAMO MBfR was found to be 1.0 in order to achieve the maximum total nitrogen (TN) removal of over 99.0%, irrespective of the TN surface loading applied, while the corresponding optimal methane supply increased with the increasing TN surface loading, accompanied by the decreasing methane utilization efficiency. Through coupling the verified model with the well-established stoichiometry and kinetics of AOB and NOB, the feasibility of integrating PN-Anammox-DAMO into a one-stage MBfR for high-level nitrogen removal was tested through controlling the bulk liquid dissolved oxygen (DO) concentration in the system. The maximum TN removal was found to be achieved at a low bulk DO concentration (depending on process parameters), with AOB, Anammox bacteria, and DAMO microorganisms coexisting in the biofilm. The coupling PN-Anammox-DAMO process could potentially treat anaerobic digestion liquor through utilizing the dissolved methane remaining, avoiding the dissolved methane stripping and thus reducing the carbon footprint of wastewater treatment. A single-stage MBfR was therefore proposed for simultaneous ammonium and dissolved methane removal from side-stream anaerobic digestion liquor through integrating PN-Anammox-DAMO. In such an MBfR, ammonium and dissolved methane are provided in the bulk liquid, while oxygen via gas-permeable membranes. The previously verified model was applied to assess the MBfR system under different operational conditions. The simulation results demonstrated that both influent and oxygen surface loadings significantly influenced the TN and dissolved methane removal. The maximum simultaneous removal efficiencies of TN and dissolved methane could reach up to 96% and 98%, respectively, by adjusting the influent and oxygen surface loadings whilst maintaining a sufficient and suitable biofilm thickness (e.g., 750 μm). The counter-diffusional supply via the biofilm and the concentration gradients of substrates caused microbial stratification in the biofilm, where AOB attached close to the membrane surface where oxygen and ammonium were available, while Anammox and DAMO microorganisms jointly grew in the biofilm layer close to the bulk liquid where methane, ammonium, and nitrite were available with the latter produced by AOB. In addition to ammonium and dissolved methane, anaerobic digestion liquor might also contain sulfide which needs to be managed properly. A mathematical model was therefore developed through incorporating sulfide related metabolisms into the previously verified model and applied to evaluate the system performance and the associated microbial community structure of the single-stage MBfRs, which integrate desired microbial consortia to treat main-stream and side-stream anaerobic digestion liquors containing ammonium, dissolved methane, and sulfide simultaneously. The simulation results showed that the dissolved methane and sulfide remaining could be utilized as electron donors by DAMO bacteria and sulfide oxidizing bacteria (SOB), respectively, to further enhance the overall nitrogen removal. The high-level (>97.0%) simultaneous removal of ammonium, dissolved methane, and sulfide could be achieved by adjusting the influent and oxygen surface loadings. AOB, DAMO bacteria, aerobic methane-oxidizing bacteria (MOB), and SOB dominated the biofilm of the main-stream MBfR, while AOB, Anammox bacteria, DAMO bacteria, and SOB coexisted in the side-stream MBfR and cooperated to remove ammonium, dissolved methane, and sulfide simultaneously.
Original languageEnglish
Number of pages146
DOIs
Publication statusPublished - 2017
Externally publishedYes

Fingerprint Dive into the research topics of 'Understanding and Modeling the Microbial Interactions in a Novel Nitrogen Removal Process'. Together they form a unique fingerprint.

Cite this