对以聚乙烯微孔管为膜组件的过滤/曝气两用型膜-生物反应器进行了研究。通过清水过滤试验发现，PE21和PE24（孔径分别为70～120μm和5～10μm）2种微孔管均具有较高的清水通量,单位水头下高达250L/m2·h。PE24微孔管对活性污泥混合液的过滤性能略优于PE21，初始膜通量的选择对二者的过滤性能影响很大，但污泥浓度的影响则不甚显著。对膜组件进行过滤/曝气交替运行，可有效地清除膜组件表面的泥饼层，较好地保持膜过滤性能的稳定。分别改变过滤/曝气运行周期和初始膜通量，考察了膜过滤性能的变化，发现交替运行周期在015～3h，初始膜通量在60～140L/m2·h时，系统在连续运行过程中每一运行周期内的平均膜通量先经历了初始的下降阶段，然后基本稳定在50～80L/m2·h的水平，但低的初始膜通量可以使平均膜通量表现得更为稳定。过滤/曝气两用型膜-生物反应器用于处理生活污水，可获得与传统膜-生物反应器相似的出水水质，并且对冲击负荷具有较好的承受能力。大小2种孔径的微孔管对污染物的去除效果没有明显差异。

通过静态吸附试验表明，粉末活性炭（PAC）为天然腐殖质类物质饱和后仍可吸附二氯酚（DCP），并在连续运行试验中进一步验证了PAC-MBR抗DCP冲击负荷的能力强于投加普通填料的MBR。通过测定进、出水有机物的表观分子质量分布，发现PAC-MBR 能有效去除1～4K（1K=1000u）的有机污染物，对4K～0.45μm有机物的去除与PAC的饱和程度有关，去除率随运行时间延长而逐渐降低。PAC-MBR和普通填料-MBR均能有效去除可生物降解有机物，前者对可吸附有机物的去除率受PAC饱和程度的影响较大。采用磷脂法可测定PAC上附着的活性微生物量，连续运行中PAC-MBR内的微生物量为0.6×10 8个活细胞PmL，与普通填料- MBR相当。耗氧速率测定结果表明，PAC-MBR内的微生物活性略高于投加普通填料的MBR。

自生动态膜-生物反应器是一种将粗网材料制成的组件置于反应器中进行泥水分离的生物处理工艺。粗网材料表面形成的动态膜不仅能有效去除浊度，还具有生物活性，可以降解污染物。通过微电极技术，可以研究动态膜的生物活性。为便于微电极在线测量，构建微型反应器模拟动态膜的实际过滤过程，结合微动平台，采用溶解氧微电极测量动态膜内的溶解氧分布。结果表明，动态膜在反应器运行过程中结构发生变化，出现气泡。动态膜内溶解氧浓度随深度的增加而降低，在210～215mm深度处降至0。动态膜表层微生物的比耗氧速率在反应器运行第1、5 和8天分别为3413、1016 和1214mg/（g.h）。

Hydraulic characteristics in a membrane bioreactor are of significance for retarding membrane fouling and maintaining stable operation. Cross flow velocity along the membrane surface and its effect on transmembrane pressure change were experimentally.

In order to clarify the dependence of apparent adsorption constant Ks in the Langmuir-Hinshelwood (L-H) model on the incident light intensity, photocatalytic degradation kinetic characteristics were experimentally investigated at different light intensities using para-chlorobenzoate (4-CBA) as a model compound. In all cases 4-CBA initial degradation rates showed dependence on 4-CBA initial concentration, which can be described by the L-H model. However, the adsorption constant Ks and rate constant kr obviously varied with light intensity. To account for the experimental finding, slight modification of the classic kinetic model developed by Turchi and Ollis was tentatively proposed by assuming insufficiency of H2O/OH-available for photo-activated holes' scavenging. Such a kinetic model predicts that both kr and Ks-1 are linearly proportional to the square root of the intensity in a rather large intensity range. The validity of the modified model was proved by fitting it to the experimental data. #2002 Elsevier Science Ltd. All rights reserved.