INTRODUCTION
The membrane bioreactor (MBR) is a very interesting solution for wastewater treatment. It offers many benefits compared to the traditional activated sludge system where clarifiers are used. However, one disadvantage for MBRs is the vulnerability to membrane fouling. Because of this disadvantage the use of MBRs is rather limited in industry.
In the framework of IWT TETRA-project conducted major at the UCWF:”advanced water filtration by using nanofibers” the water filtration capacities of an electrospun nanofibre membrane (produced by department of textiles, Ghent University) are being tested. The nanofiber membrane will be used in an activated sludge MBR, with and without a coagulation product (MPE50) but also in a biofilm MBR. Based on the performed experiments, it was possible to improve the membrane (production process).

RESULTS AND DISCUSSION
From the obtained results of the removal efficiency (Figure 2), it’s possible to conclude that the AS-MBR has the best removal efficiency.
Figure 2: Removal efficiency of the three setups and a MBR from the literature
Although, when looking at the limitation of membrane fouling, the BF-MBR performs better. This can be seen from Figure 3: the flux for AS-MBR and AS-MBR + MPE50 decreases fast comparing to the flux of the BF-MBR. This decrease in the flux can be explained by membrane fouling. When the fouling of the membrane increases less water can permeate. If less water can permeate trough the membrane the flux will drop. The amount of fouling determents how fast or slow the decrease in flux will be. When observing Figure 3 it can be noticed that at the end of the filtration period for BF-MBR the flux stabilizes.
Figure 3: comparing the flux for the three setups
A major problem that has been encountered when working with the electrospun membrane was delamination. This problem is nothing more than a layer of the membrane which detaches so that there is loss in filtration efficiency because of the reduction in tortuous network. Furthermore it also leads to a reduced tensile strength on that spot. During the filtration process the membrane obtains a concave shape induced by under pressure. When back-flushing the membrane obtains a convex shape. During this constant changing of membrane shape it is possible that the fibers get stretched. Probably the stretching of these fibers promotes delamination. Eventually a good solution was found for the problem by using support membranes. When using those support membranes the set-up was able to run continues for a minimum of 48 days (working with BF-MBR set-up). When the set-up was stopped it was seen that the flux was above the end flux of previous setups (Figure 3).
From experiments it appeared that NaOCl and oxalic acid were the best cleaning agents. These findings were supported by literature (Kimura et al., 2004). After analyzing the membrane with a FT-IR, same peaks were found as in the literature. This means that the same components were on the membrane under study: polysaccharides and protein.