| Membrane BioReactor (MBR) Sheets |
|
|
|
13-jun-2006, STOWA |
|
Country |
|
Japan - The Netherlands |
|
Stage of development |
Full scale |
|
|
Process - |
Line |
Water |
|
|
Function |
Secondary treatment and clarification |
|
|
Input |
Effluent from primary treatment |
|
|
Ultra- and microfiltration using porous membranes |
|
|
Keywords: decrease area demand; improve effluent quality; water line; SS removal effluent |
||
The term membrane bioreactor (MBR) defines a combination of an activated sludge process and membrane separation. Due to recent technical innovations and significant cost reductions the applicability for the MBR technology in municipal wastewater treatment has sharply increased.
The MBR process can be employed in activated sludge processes, using the membranes as liquid-solid separation instead of the usual settling. Suspended solids can be removed completely and bacteria-free treated water produced. The sludge concentration and hydraulic loading rates are considerably higher than in conventional treatment.
Pre-treated, screened influent enters the membrane bioreactor, where biodegradation takes place. The mixed liquor from the bioreactor is withdrawn and pumped along submerged or semi-crossflow filtration membrane modules. The permeate from the membranes constitutes the treated effluent.
The reject stream, consisting of concentrated biosolids, is returned to the bioreactor. Excess biosolids are wasted from the bioreactor or from the return line. Due to the membranes acting as an absolute barrier to solids, it is possible to improve the effluent quality, especially with regard to suspended solids as well as bacteria and viruses.
The separation of hydraulic and solids retention times provides a good control of biological reactions. Due to the high biomass concentration in the bioreactor, the reactor can be much more compact compared to conventional activated sludge systems.
|
Membrane geometry |
flat plate |
|
Process configuration |
|
|
Material |
hydrophilic polypropylene |
|
Pore size |
0,4 mm |
|
Surface area of one plate |
0,8 m2 |
|
Distance between plates |
7mm |
|
Process mode |
Ralaxation (optional) |
|
Cleaning procedure |
intensive cleaning every 6 to 12 months |
|
Design flux |
30 - 40 l/m2.h |
|
Membrane geometry |
hollow fibre |
|
Process configuration |
|
|
Material |
hydrophilic polyethylene |
|
Pore size |
0,4 mm |
|
Outer diameter of fibre |
540 mm |
|
Inner diameter of fibre |
360 mm |
|
Process mode |
relaxation and back-flush |
|
Cleaning procedure |
maintenance cleaning every week |
|
|
intensive cleaning every 3 to 6 months |
|
Design flux |
10 -20 l/m2.h |
Design and operational parameters as well as removal efficiencies are strongly dependent on the wastewater characteristics. The membranes described above have been extensively tested on pilot scale for the Dutch wastewater situation. In the table below some technical parameters are given (WWTP Beverwijk).
|
Parameter |
Range |
Unit |
|
Sludge load |
0.05 - 0.1 |
kg COD/(kg MLSS.d) |
|
Sludge load |
0.008 - 0.012 |
kg N/(kg MLSS.d) |
|
MLSS conc. |
10 – 12 |
g/l |
|
Temperature |
5 - 25 |
°C |
|
Sludge age |
30 - 60 |
days |
Some general indications of the removal efficiencies of the MBR systems were given by [3]:
· In general only COD, BOD, SS and nutrient removal in MBR systems are described in literature. The results on nutrient removal were often very good due to the very low loading of the systems. The SS permeate concentration was usually below detection level. For a number of (full scale) installations the microbial parameters were measured and published. In general, the removal of total and faecal coliforms was around log 6, the removal of viruses was varying between log 2 - 4. An advantage of the MBR is that the micro-organisms are totally removed and not just inactivated with a risk of reactivation.
· Only few publications were found in which attention was paid to the removal of micro-pollutants. In one publication a research was presented in which AOX's, medical substances, pesticides and endocrine substances were measured in both the MBR's permeate and a conventional WWTP's effluent. The AOX level in the MBR permeates was equal to the conventional effluent. For all the other substances significant lower concentrations were measured in the MBR permeate.
Removal efficiencies for the most common parameters are given in the table below. In a MBR all of the suspended solids are removed. As a consequence the removal of heavy metals and micropollutants attached to the suspended solids is also improved.
|
Parameter |
Unit |
Conventional treatment |
|
|
mg/l |
0 |
10 |
|
|
mg/l |
30 |
40 - 50 |
|
|
NTotal |
mg/l |
10 |
10 |
|
PTotal |
mg/l |
0,15 - 0,5 |
0,5 - 1,0 |
From experiences on several municipal MBR plants it is known that membrane systems can be very sensitive for macro-fouling by debris. The occurrence of debris has led to problems regarding throughput and eventual membrane damage. The main purpose of the pre-treatment is to remove solids (screenings) which are harmful for membranes, such as coarse solids (plastics, leaves, seeds, sand particles), oils, fats and hairs.
Two main types of cleaning can be distinguished: chemical cleaning and mechanical cleaning. Chemical cleaning is essentially a physical-chemical reaction between the cleaning chemical and the foulant. Mechanical cleaning is a term used for the physical removal of suspended solids from the membrane material. Mechanical cleaning is usually based on turbulence and fluid mechanics, caused by (coarse) bubble aeration, permeate back-flushing, relaxation and circulation.
Cleaning of the membranes, irrespective of the procedure used, has a knock on effect to the membrane life-expectancy and operational replacement costs. Moreover, the chemical usage required for cleaning purposes has an impact on the biological system and the membrane integrity.
In municipal MBR systems mainly bio-fouling, and cake formation occur. Bio-fouling can be treated with a NaOCl cleaning eventually followed by an acid cleaning. Cake formation can be treated hydraulically.
The most important aspect is the operation of the MBR to prevent the forming of a cake on the membranes, followed by the removal of the cake layer and the cleaning of the membranes with chemicals. Some basic rules to prevent cake formation on the membranes are given below:
· use a membrane with a high maintainable permeability; lower suction pressures on the membrane surface prevent cake formation
· reduce dead zones in the membrane module/tank, and avoid a too higher packing density of the membranes
· select responsible operating fluxes, however, during peaks temporary higher fluxes can be maintained
· generate larger sludge flocs to allow a more permeable cake layer on the membrane
· install and maintain good biological conditions thus preventing the formation of EPS and filamentous organisms
The transfer of oxygen from air into water depends on the characteristics of the transfer layer, air bubble size and mean residence time of the bubble in the water. These last two parameters depend strongly on the viscosity of the solution to be aerated. Viscosity itself can be influenced by reactor configuration, the mixing/aeration device, and the concentration and characteristics of the activated sludge. In general the concentration and as a result the viscosity of MBR sludge is higher than in a conventional sludge system.
In comparison with conventional waste activated sludge the MBR sludge contains smaller flocs and has a somewhat higher viscosity. Based on tests executed [3], it can be concluded that MBR sludge thickening and de-watering shows results comparable with conventional activated sludge.
Information on costs is available from different feasibility studies in which the costs for a MBR- and traditional installation under different circumstances were compared. In the figure below investment costs for a new WWTP are compared [4].

The
annual costs of a MBR installation are strongly influenced by the depreciation
for the mechanical equipment and the replacement of the membranes. At the
moment the latter is still difficult to predict, practical experiences from 3
to 8 years being not enough for a realistic estimation. With manufacturers, the
discussion around the membrane lifetime has led to capacity guarantees.
Regarding variable costs, especially the energy consumption is an interesting item, which is mainly determined by the alpha-factor and consequently the energy for aeration. In the feasibility studies mentioned before, alpha‑factors of 0.40 - 0.45 have been used for calculation purposes. Based on the results of pilot scale measurements, side study 3 and full scale measurements at WWTP’s in Germany it is expected that these values are too low. Cost differences between a MBR and traditional WWTP concerning manpower, chemical consumption and sludge treatment are supposed to be minimal.
About 600 Kubota reference installations are reported world-wide for industrial and municipal applications. Of the 21 installations in Europe, eight are treating municipal sewage. Hydraulic capacities are ranging from 200 to 13.000 m3/d. The largest installation realised so far is the WWTP Swanage (23.000 p.e./1999) in the UK.
About 700 Mitsubishi reference installations are reported, most of them are situated in Japan. A third of these installations is used for the treatment of domestic wastewater.
Kubota Corporation
London Office, Membrane System Department
8 Hanover Street - London W1S 1TYE - UK
Tel: +44 - 20 7290 2730 - Fax: +44 20 7290 2733 - E-mail : membrane@kubota.co.jp - Web: www.kubota.co.jp/english/index.html
Kubota has a co-operation agreement with SOLIS Engineering in the Benelux.
Mitsubishi Rayon Co.,Ltd
Membrane Products Department
6-41, Konan 1-chome - Minato-ku - Tokyo - Japan
Tel : +81 - 3 5495 3152 - Fax : +81 3 5495 3217 - E-mail : membrane@mrc.co.jp - Web: www.mrc.co.jp/english/
|
[1] |
Websites owned by Kubota and Mitsubishi Corporation |
|
[2] |
DHV Visit report Japan |
|
[3] |
Magazine H2O: MBR Special.11/2001. |
|
[4] |
Stowa report 2002-11A: MBR for municipal wastewater treatment. Pilot plant research Beverwijk WWTP. 2002. |
|
|
|
|
|
Kubota - process design |
Mitsuibishi - process design |
|
|
|
|
|
|
Kubota - top view |
Mitsubishi - top view |
|
| |