MBR 공정에서 가장 문제시되는 부분은 실제 공정상에서 막 표면에 오염이 발생하는 문제이다. 일반적으로 막 오염은 활성 슬러지들이 막 표면에 퇴적되어서 일어나며 이로 인해 심각한 투과 유량의 감소를 야기하게 된다. 본 연구에서는 막 오염 저항성이 뛰어난 나노 입자를 분리막 표면에 함침시켜 MBR 막을 제작하였으며 이 입자들은 막 표면에서 활성 슬러지들이 쉽게 달라 붙지 못하는 역할을 수행하게 된다. 즉, 뛰어난 투수량을 지닌 정밀여과막에 나노 입자를 첨가함으로서 실제 MBR 공정에서 막 오염을 저감시킴으로서 투수량을 보전할 수 있게 하였다. 이들을 이용하여 MBR 공정에서 막 오염이 휠씬 적게 일어나는 것을 확인하였으며 이를 바탕으로 현장에 적용하여 막 오염 저항성을 확인하였다.

십자류 흐름 평판형 cell에서 비대칭 cellulose acetate 막으로 PEG #6000 및 dextran 70T 용액의 한외여과를 압력차와 도입용액 농도 변화에 따라 실험하여 막투과량 및 용질 배제도를 측정하고, 막투과량에 미치는 농도분극층 저항을 고찰하였다. 이 결과 농도분극층 저항 Rb1을 농도분극층내의 고분자 용액의 평균 농도 Cb1과 상관시켜 다음의 무차원 관계식으로 나타낼 수 있다. Rb1Rm=α[ρb1Cb1]β

The aim of this study is to measure the exact average specific cake resistance in cake filtration. Using the new experimental method “filtration-permeation,” the average specific resistances were measured for various conditions as follows. First, the experiment with changing pre-sedimentation time and the filtration of sediment were conducted to determine the influence of sedimentation on the average specific resistance. There was a wide difference between the values of αav with various sedimentation times. Regarding the variation in the pre-sedimentation time, there was maximum 2.9x difference between the measured values of the average specific resistances for the filtration of calcium carbonate suspension. The “filtration-permeation” of bentonite floc and that of biological flow were also performed to establish a method for obtaining the average specific resistance of flocs.

The resistance in series model has been frequently used for determination of various filtration resistance to correctly understand the membrane fouling behaviour in MBR (membrane bio-reactor) for wastewater treatment. The cake layer resistance (Rc) which is commonly determined by calculation of flux dataset that are obtained empirically before and after removing the cake layer on membrane surface. However, the calculated Rc values are very dependent on the cleaning methods adapted for removal of cake layer. This study investigated how the various cleaning options affect Rc. Seven different cleaning methods were employed: i) ultrasonication (100 W, 10 min), ii) ultrasonication (200 W, 60 min), iii) ultrasonication (400 W, 120 min), iv) water rinsing in a shaker (100 rpm, 10 min), v) water rinsing in a shaker (300 rpm, 60 min), vi) water rinsing, vii) sponge scrubbing. For the hydrophilic PES membrane, the cake layer removal efficiencies ranged from 64% to 10%, indicating that the removal of cake layer was highly dependent on the cleaning options. For the hydrophobic PVDF membrane, the cake layer removal efficiencies ranged from 79% to 97%. Consequently, a standardized method for cake layer removal to determine cake resistance (Rc) is needed for correct interpretation of the fouling phenomena.

Research results for the pressure drop variance depending on operation conditions such as change of inlet concentration, pulse interval, and face velocity, etc., in a pulse air jet-type bag filter show that while at 3kg/cm2 whose pulse pressure is low, it is good to make an pulse interval longer in order to form the first layer, it may not be applicable to industry because of a rapid increase in pressure. In addition, the change of inlet concentration contributes more to the increase of pressure drop than the pulse interval does. In order to reduce operation costs by minimizing filter drag of a filter bag at pulse pressure 5kg/cm2, the dust concentration should be minimized, and when the inlet dust loading is a lower concentration, the pulse interval in the operation should be less than 70 sec, but when inlet dust loading is a higher concentration, the pulse interval should be below 30 sec. In particular, in the case that inlet dust loading is a higher concentration, a high-pressure distribution is observed regardless of pulse pressure. This is because dust is accumulated continuously in the filter bag and makes it thicker as filtration time increases, and thus the pulse interval should be set to below 30 sec. If the equipment is operated at 1m/min of face velocity, while pressure drop is low, the bag filter becomes larger and thus, its economics are very low due to a large initial investment. Therefore, a face velocity of around 1.5 m/min is considered to be the optimal operation condition. At 1.5 m/min considered to be the most economical face velocity, if the pulse interval increases, since the amount of variation in filter drag is large, depending on the amount of inlet dust loading, the operation may be possible at a lower concentration when the pulse interval is 70 sec. However, for a higher concentration, either face velocity or pulse interval should be reduced.