Dynamic Modeling of Nitrifying Trickling Filters
Licentiatavhandling, 1996

Trickling filters, one kind of biofilm reactors, for removal of nitrogen in municipal wastewater are presently increasing in number in Sweden. For improved control and operation of wastewater treatment plants, where such reactors are used, dynamic models describing their behavior are necessary. In this thesis a physical dynamic model of cross-flow nitrifying trickling filters (NTFs), based on a general multi-species biofilm model, is presented. The model predicts effluent concentrations of ammonium, nitrite, nitrate, and alkalinity as functions of the corresponding influent concentrations, the water temperature, the flow, and the present state of the distribution of the nitrifying bacteria {\it Nitrosomonas} and {\it Nitrobacter} in the biofilm of the NTF. Efficient methods to solve the model equations, also in steady state, are presented. Experimental data achieved on a large pilot scale trickling filter are compared with model simulations, and model assumptions are experimentally validated. From the modeling and experiments it is concluded that the dynamics of the NTF can be divided into two modes: One fast mode that can be assumed to depend only on the mixing in the bulk, and one slow mode that depends on the bacterial growth and decay in the biofilm. The settling times of the two modes are separated by a factor of order 1000, which considerably simplifies model simulations. Comparisons between experimental data and a simplified version of the model show that it takes less than a few minutes for the nitrifying bacteria in the biofilm to change their substrate uptake rate after changes in substrate bulk concentrations. From pulse experiments and laminar flow theory analysis it is shown that the flow through the NTF is turbulent and, hence, significant mixing occurs inside the NTF. The residence time distribution can be approximated by a model of continuously stirred tanks in series. Comparisons between simulations and semi-stationary data show that the substrate flux into the biofilm is enhanced by an increase in flow, probably due to increased turbulence in the bulk. Analysis and simulation of steady-state multi-species biofilms in general indicate that bacterial coexistence is not only dependent on the bulk water substrate concentrations, but also on the biofilm thickness, which means that control of the biofilm thickness may be a way of controlling the bacterial composition in biofilms.

Distributed parameter system

Trickling Filter


Process Model





Torsten Wik

Institutionen för reglerteknik


Övrig annan teknik


Oceanografi, hydrologi, vattenresurser



Technical report L - School of Electrical and Computer Engineering, Chalmers University of Technology.