Biological nitrogen removal of high-strength ammonium wastewater was studied in this thesis, particularly, the nitrification process (the oxidation of ammonium to nitrate). This two-step reaction, catalyzed by two kinds of bacteria (AOB and NOB), can suffer serious inhibition problems due to ammonia and nitrous acid when dealing with highly concentrated ammonium wastewater and therefore it requires adequate process control. However, these inhibitions can be used to achieve partial nitrification (the oxidation of ammonium to nitrite), which coupled to a denitrifying process leads to significant benefits in terms of use of resources. A mathematical model describing the kinetics and the stoichiometry of the nitrification process was developed and calibrated. It considered the aforementioned inhibitions and took into account both kinds of nitrifying bacteria and also heterotrophic bacteria. Specific experiments were designed for parameter estimation and parameter identifiability tools were used to analyze and improve them. Optimal experimental designs were used to calibrate most of the model parameters and the obtained values were compared with values found in the literature. Affinity constants for substrate and substrate inhibition coefficients were estimated twice using different sludges and, as a result, different values were found indicating that they change depending on the biomass acclimation. This model was coupled to the hydraulic model of the experimental system (pilot plant) and was implemented in Matlab ®. Fluorescence in situ hybridization (FISH) was used for bacterial fractions detection and quantification. Several equipments were used for fluorescence detection: an epifluorescence microscope, a confocal microscope and a flow cytometer. Biomass fractions were determined with each of the equipment and also with simulations. Obtained results were compared and the advantages and disadvantages of the tested methodologies were discussed considering the accuracy of the results, the speed of the analysis and the availability of the equipment. FISH combined with confocal microscopy turned out to be the best technique for nitrifying biomass quantification although flow cytometry could not be extensively investigated. The start-up of a complete nitrification system was optimized by means of mathematical simulation using the previously developed and calibrated method. Two automatic control strategies were optimized and implemented in the experimental system by using sludge from a municipal wastewater treatment plant as inoculum. The controller was based on the measurement of the oxygen uptake rate (OUR) in the last reactor of the system and actuated over the nitrogen loading rate. Results were compared with a start-up performed with manual control and it was demonstrated that automatic control decreased the length of the start-up and increased its stability. Then, experimental results were simulated with the nitrification model. Model predictions agreed well with experimental data. The final model was useful for both long- and short-term prediction. The sludge enrichment in nitrifying bacteria was checked with FISH and confocal microcopy. The nitrifying sludge obtained after the last start-up contained both AOB an NOB and was used to achieve partial nitrification. Some environmental conditions and the automatic control strategy were changed in order to inhibit NOB and wash them out of the system. Partial nitrification with an effluent devoid of nitrate was achieved at 25 ºC, 1.1 mg O2 L-1 and pH of 8.3 using the appropriate OUR set point for the automatic controller. Partial nitrification was run for 120 days with an averaged nitrogen loading rate of 0.5 g N g-1 VSS d-1. FISH analysis demonstrated that NOB were completely washed out. The control strategy was improved by the addition of two expert rules and stable operation was maintained even when external disturbances were provoked. Finally, a model-based study was performed to test the partial nitrification start-up strategy under different conditions and system configurations.
Operation, Modeling and Automatic Control of Complete and Partial Nitrification of Highly Concentrated Ammonium Wastewater
Jubany Güell, I. (Author). 3 May 2007
Student thesis: Doctoral thesis
Student thesis: Doctoral thesis