brane modules, with pore sizes of 0.22 and 0.1 μm, were used to investigate their effects on virus rejection at the transmembrane pressure of 8.5 kPa. It was found that 0.1 μm membrane had complete rejection of virus, and 0.22 μm membrane had significant rejection of virus. In the longterm operation of this MBR, no significant difference was observed between both pore sizes because the virus concentrations of the effluent in both cases were in the same order. Effluent virus concentration at steady state of MBR running was less than 2 PFU/mL. The removal ratios of coliphage T4 in MF processes were more than 105.5. The membrane surface deposits played an important role in the rejection of virus. The formation of cake clay on the membrane surface was the main cause of high rejection of coliphage T4 with MF of 0.22 μm.

A pilot- scale(10 m3/d) anoxic/oxic membrane bioreactor(A/O MBR) was tested for dyeing wastewater treatment of woolen mill without wasting sludge in 125 days operation. Results showed that the effluent quality was excellent, i. e. effluent COD less than 25 mg/L, BOD5 under 5 mg/L, turbidity lower than 0.65 NIU, and colour less than 30 DT, and met with the reuse water standard of China. The removal rates of COD, BOD5, colour, and turbidity were 92.4%, 98.4%, 74% and 98.9%, respectively. Constant- flux operation mode was carried out in this study, and backwash was effective for reducing membrane fouling and maintaining constant flux. Membrane fouling had heavy impact on energy consumption. More attention should be paid on pipe selection and design for the sidestream MBR system, too.

A laboratory- scale membrane bioreactor(MBR) with a gravity drain was tested for dyeing and printing wastewater treatment from a wool mill. The MBR was operated with continuous permeate by gravity and without chemical cleaning for 135 days. Results showed that excellent effluent quality could meet the reuse water standard in China. The average concentrations of COD, BOD5, turbidity and color in the effluent were 36.9 mg l-1, 3.7 mg l-1, 0.2 NTU and 21 dilution times(DT), respectively. The average removal rates of COD, BOD5, turbidity and color were 80.3%, 95.0%, 99.3% and 58.7%, respectively. The membrane flux increased with increasing of aeration intensity, and its increasing rate was related to pressure- heads. The higher the pressure- head, the greater the impact of aeration intensity on membrane flux. Statistical analysis also showed that both the pressure- head and aeration intensity significantly affected membrane flux. Due to its compact design, simple operation and easy maintenance, MBR with a gravitational filtration system hs low energy consumption and is cost- effective to build and operate. If the life expectancy of the membrane is set for 3- 4 years and the membrane flux is set at 15 l/m2h, such a MBR would be very competitive.

Virus removal in a membrane bioreactor(MBR) by gravity drain was studied. Coliphage f2(mean size of 25 nm), which is similar in size to human enteric pathogenic viruses, was selected as a model virus. Two microfiltration membrane modules with pore sizes of 0.22 μm and 0.1 μ m were applied to investigate the effects of membrane pore size on the virus rejection. The MBR with these modules could reject virus in a range of 2.6- 5.1 logs. The experimental results showed that the mechanisms of virus removal in the MBR involved rejection and inactivation. The virus rejection depended mainly on the dynamic layer on the membrane surface(not membrane itself) because virus rejection by the membrane modules with pore sizes of 0.1 μm and 0.22 μm were similar. The microbial activity and the aeration oxidation were the two important factors for the virus inactivation. It was found that the inactivation of coliphage f2 was much more rapidly in activated sludge mixed liquor than in clean water, and the effect of aeration was significant.