Introduction

In conventional aerobic activated sludge systems, biomass separation from the treated liquor relies on sedimentation of aggregated mixed microbial floes. If biomass separation from the treated effluent is facilitated by physical retention within the bioreactor the need for flocculation is removed. The necessary operating conditions can be achieved with cross-flow membrane bioreactors where the membrane acts as a filter to provide better

Keywords than clarified effluent. Progress in the engineering of membrane bioreactors and the development of immersed membrane configurations, has produced compact, robust, cost-effective systems. Immersed membrane systems use either hollow fibre membranes or flat sheet membranes (Bussion et al. 1998)1 and consist of membrane modules submerged directly into the activated sludge compartment. Typically they operate in an outside-inside filtration mode at low suction pressures, (1-5 bar). Operation at low suction pressure means a reduction in the transmembrane pressure. This means reduced fouling of the membrane and slower formation of a cake layer, resulting in less frequent backwashing and lower operating costs, as smaller pumps are required. However, because the force driving the liquid through the membranes is relatively low, a large membrane surface is required to compensate for the low flux rates achieved. Sludge wastage is minimised by maintaining a low F/M ratio while the footprint of the plant is reduced by operating at high biomass concentrations, typically 15 - 20 g/1 (Cote and Pound 1997)2. In addition, biomass retention allows build-up of a waste-specific microbial population, of particular use when dealing with industrial effluents, thereby providing the most effective biological treatment. The main improvements arising from the coupling of membrane technology with biological wastewater treatment are summarised in Table 1.

Table 1. Conventional and membrane biological treatment comparisons.

Problems and limitations associated with conventional biological treatment

Improvement from the inclusion of membranes in the process

Variations in the pollutant loading

Acceptance of variation in concentration of activated sludge including high concentrations

Very slow rate of kinetic reaction

Natural selection and total retention of the bacterial population

Settlement of bacteria in clarifier can rate limit process

Removal of bacteria from effluent not dependent upon settling characteristics of bacteria

Post-treatment removal of viruses required prior to reuse

Significant removal of viruses from effluent make direct reuse possible

Insufficient contact time between macromolecules and micro-organisms

Increase in contact time allowing effective treatment of low biodegradable products

Rate of sludge production creates a problem for disposal

Sludge volume reduced with minimised cost of post-treatment

Irregular quality of treated water and absence of effective barrier to bacteria

Production of high quality effluent, free from bacteria, offering the possibility of recycling

Bulky

Foot-print reduced

1.1 Materials and Methods

1.1 Materials and Methods

Fine bubble aeration Coarse bubble aeration

Figure 1. Schematic diagram of the pilot membrane bioreactor.

All chemical analysis was carried out on site with a Dr Lange colorimeter using standard methods. These results and all microbiological analysis were verified at a NAMAS accredited laboratory using standard methodologies3-4'5. Bacteriological samples were monitored for E.Coli and Coliforms and viral samples were monitored for Human Enteric Viruses.

A ZenoGemĀ® pilot plant was fed with effluent from a chicken processing factory that was high in COD, BOD, ammonia, suspended solids and fat. It also contained blood and feathers from the culling process. The effluent was screened through a 2 mm drum and then through two crude fat removal tanks, (Figure 3). These consisted of lm3 tanks operated with a constant overflow, the feed being taken from the bottom of the tank using a small submersible pump. The pilot plant used in this study had a total working volume of 4 m3, (Figures 1 and 2). It consisted of two reactors in series (of 3 m3 and 0.8 m3, respectively), followed by a membrane compartment, and CIP (cleaning in place) tank, operated under aerobic conditions. Fine bubble aeration was supplied at the bottom of the first bioreactor and the second was mechanically stirred. Two hollow fibre membrane modules, (surface area 13.8m3 each, pore size 0.1 microns supplied by Zenon GmbH), were submersed in the third compartment. The membranes are chlorine resistant made from plastic, woven to obtain the required pore size. Hollow fibre membranes provide a higher surface area to volume ratio than flat plate membranes, thus occupying less reactor volume. Coarse bubble aerators were positioned at the bottom of the membranes supplying air to increase cross flow velocity and increase mass transfer in the vicinity of the membranes. The membranes were backwashed with permeate for 40 seconds after every 300 seconds of operation. This is to prevent surface fouling and dislodge the cake layer formed at the membrane surface. Provision was made in the pilot plant for chemical cleaning with sodium hypochlorite, however a chemical clean was not required during this study. A biomass recirculation pump was in constant operation transferring biomass from the membrane compartment back into the first reactor. The pH was maintained between 6.75 and 7.25 by acid and alkali dosing.

Figure 2. Picture of the pilot plant Figure 3. Picture of first grease trap

1.2 Feed Quality

The characteristics of the screened feed are given in Table 2. Table 2. Feed characteristics.

0 0

Post a comment