Year

2010

Degree Name

Doctor of Philosophy

Department

School of Civil, Mining and Environmental Engineering

Abstract

This research investigates the removal of trace organics by membrane bioreactors (MBRs) under various operating conditions to elucidate the removal mechanisms and factors affecting the removal efficiency of these contaminants. The reported results show excellent performance of a laboratory scale MBR system regarding the removal of basic biological performance with stable removal efficiency throughout over two years. Changes in operating conditions including mixed liquor pH, temperature, salinity could affect the performance of the MBR system at extreme conditions. However, the MBR system could quickly recover after returning to a normal operating condition. In contrast, mixed liquor suspended solid (MLSS) and dissolved oxygen (DO) concentration could not exert any discernible effects on the performance of the MBR.

A set of 40 trace organics was selected to investigate the removal efficiencies and the associated removal mechanisms under a stable operating condition. The study demonstrated an obvious correlation between molecular features and the removal of trace organic contaminants by MBR system. The results revealed high removal efficiencies (>85%) of very hydrophobic trace organic compounds (log D at pH 8 > 3.2). Moreover, the occurrence of electron withdrawing or electron donating functional groups was found to be another important factor governing their removal. All hydrophilic and moderately hydrophobic (log D < 3.2) compounds possessing only electron withdrawing functional groups consistently showed removal efficiency of well below 20%. In contrast, high removal efficiency was observed with most compounds bearing electron donating functional groups such as hydroxyl groups and carboxylic groups.

The effect of mixed liquor pH (between pH 5 and 9) on removal of 10 selected trace organics was investigated. The results showed that ionisable trace organics (sulfamethoxazole, ibuprofen, ketoprofen and diclofenac) was strongly pH dependent. High removal efficiency of these compounds was observed at pH at 5 due to predominant hydrophobic form. This results in readily adsorb to sludge compared to basic condition. On the other hand, the removal efficiencies of bisphenol A and carbamazepine which are non-ionisable compounds were independent of the mixed liquor pH.

The effect of operating conditions on the MBR performance regarding to removal of trace organic compound was determined. Four important operating conditions including hydraulic retention time (HRT), DO concentration, temperature and MLSS concentration were selected for this study. The results showed that higher HRTs had a large impact on the removal of compounds which were biodegradable whereas HRT could not influence the removal efficiency of high persistent biodegradable compounds and hydrophobic compounds. DO concentrations ranging 2 to 5 mg/L had no discernible influence on the removal of compounds exception for bisphenol A and estrone. Temperature had also no impact on the removal efficiency of model compounds exception for carbamazipine with 50% removal at higher temperature of 25°C. There was some fluctuation of removal efficiencies of compounds during MLSS experiment. The reported results were not conclusive to ascertain whether MLSS concentration would have any effects on the removal of trace organic compounds. This might be because of the unstable sludge retention time (SRT) in the MBR system.

Change in salinity influent ranging of 1 to 12 g/L in MBR system appeared to exert a small effect on the removal of basic biological performance when exposed to higher salt concentration. The soluble microbial product (SMP) in form of protein increased resulting from disturbance in microbial behaviour. In contrast, with the exception of bisphenol A, no impact of salinity on removal of selected trace organic contaminants could be observed.

Download

Share

COinS
 

Unless otherwise indicated, the views expressed in this thesis are those of the author and do not necessarily represent the views of the University of Wollongong.