Modelling the Removal of Airborne Contaminants in Swine Facilities by a Biotrickling Filter
Date
2014-01-28
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
Type
Degree Level
Doctoral
Abstract
The overall objective of this dissertation work was to optimize the performance of biotrickling filters in reducing emissions of odour and harmful substances from swine facilities. The parameters and operating conditions that have significant impact on the treatment process were identified through a modelling study.
Key odour components were selected to serve as model pollutants, which were identified from linear relationships between the logarithm of odour emission and the logarithm of pollutant emission/odour intensity and from odour indices. The potential model pollutants identified were ammonia, dimethyl sulphide, and p-cresol.
Different sets of shake-flask experiments were conducted to assess different inocula, to determine the optimum pH, and to estimate the biokinetic parameters for the biodegradation of ammonia and p-cresol. Among the three inocula evaluated, the complex inoculum taken from an existing biotrickling filter showed the best performance in terms of p-cresol and ammonium reduction. The results also showed that the highest p-cresol uptake and reduction rates and NO3- production rate were at pH 7. Moreover, it was found that the biodegradation of p-cresol was better described by the Monod equation (R2 = 0.96) with estimated values of 0.10 h-1 for µm and 103.4 mg L-1 for Ks. The biodegradation of ammonia, on the other hand, was better described by the Haldane equation (R2 = 0.72) with estimated values of 0.17 h-1 for µm, 11.9 mg L-1 for Ks, and 617.9 mg L-1 for Ki.
Mass balance equations were formulated to describe the processes occurring in the gas, liquid, and biofilm phases of the treatment system. The differential equations were solved using the finite difference numerical analysis method. A one-at-a-time sensitivity analysis was conducted to identify parameters that have significant impact on ammonia removal. Calibration and validation results showed good agreement between predicted and measured values; based on the fractional bias (FB) results, the normalized model’s prediction errors were within ±1 to 7%. After model calibration and validation, a simulation study was conducted using the model to evaluate the impacts of selected process and design parameters for a biotrickling filter system.
Description
Keywords
biotrickling filter, modelling, swine air, ammonia
Citation
Degree
Doctor of Philosophy (Ph.D.)
Department
Agricultural and Bioresource Engineering
Program
Agricultural and Bioresource Engineering