Full Scale Denitrification Filtration and Process Modeling Using a Steady State Mathematical Framework

Abstract

Biological denitrification is a process where nitrate is reduced to nitrogen gas in a step-wise manner by electron donors (usually methanol, acetic acid, or ethanol) in the absence of oxygen. A fixed-film denitrification process is basically a biofilm process, which fundamentally consists of two simultaneous steps, substrate diffusion and biological reaction. The mathematical framework for both steady state and dynamic one-dimensional models is well established. However, there is limited modeling and simulation of full scale plants with tertiary denitrification filters with external carbon addition. In this work, performance data were reviewed for the denitrfication filters at Central Johnston County Wastewater Treatment Plant, Smithfield, North Carolina. A process model was established using AquaNET Aquifas (Aquaregen, Mountain View, CA). The model successfully predicted the process performance and the nitrate profiles in a pilot denitrification study and subsequently the model was applied to the denitrification system at Central Johnston County Wastewater Treatment Plant typically during the summer and winter seasons. The model was successful in predicting effluent nitrate concentrations but failed to predict the effluent soluble chemical oxygen demand concentrations in the winter season when the filters were overdosed with methanol. It was demonstrated that the model can be applied to situations with a wide range of nitrate loading from 0.07 lbs/ft 2 /day (0.34 kg/m 2 /day) to 1.02 lbs/ft 2 /day (4.98 kg/m 2 /day). It was hypothesized that methanogens were produced under the anoxic environment and consumed the excessive carbon source. The temperature effect was also discussed and an Arrhenius coefficient of 1.03 was found appropriate for process simulation.