| Summary: | 碩士 === 國立屏東科技大學 === 環境工程技術研究所 === 83 ===
This research is intended to unravel the mystery of substrate competition between suspended and attached microorganisms in a completely mixed reactor, called hybrid reactor. Results obtained from this study was also used for evaluating a model (Lee, 1992) developed for describing the competition behaviors. There were two experimental phases conducted in this research: estimation of the corresponding kinetic coefficients and evaluation of performance f a laboratory-scale hybrid reactor.
Based on Chemostat technique, the estimation of standard Monod coefficients of k, Ks, Y, and bd were found as 4.13 mg TOC/mg VSS-day, 10.68 mg TOC/L, 0.79 mg VSS/mg TOC, and 0.19 day-1, respectively. The biofilm detachment rate perofrmed with a biofilm reactor (submerged rotating biological contactors) appeared independent to the disc speed as controlled at 5, 10, and 15 rpm. However, it was related to the product of Xf (biofilm density) and Lf (biofilm thickness); best fit of the detachment coefficient was 0.216 day-1. In additiion, experimental results also showed that the value of Xf varied in the range of 15-25 g/L. Sensitivity analyses of the above parameters revealed that k, Y, and bs are the most important. Higher values of k and Y encourage suspended growth to compete for substrate which yields better effluent quality; while the increase of bs would promote suspended growth through which effluent quality is deteriorated.
Results of continuous operation of the hybrid reactor indicated that the substrate competition by the two growths depended on substrate loads (So), hydraulic detention time (HRT), and solids retention time (SRT). Both effects of heavy loads of So and longer HRT brought about the suspended growth more competitive; while SRT did not play important role in affecting the competing behaviors. When the data obtained from the experiments were compared with the model predictions, a little discrepancy was found, such as that the predicted effluent levels agree with the measured values only occurring around 4 mg TOC/L. Furthermore, the amount of suspended biomass as well as the fraction of suspended growth are overestimated which result in a lower predicted competition of attached grwoth. Nevertheless, in general the model successfully predicts the competing results at the operating conditions. Especially, when the suspended growth is not predominant the predictions are very close to the measured values.
|
|---|
