A membrane bioreactor, mbr, is a wastewater treatment process that combines the biological reactor of conventional activated sludge with a microfiltration or ultrafiltration membrane barrier that separates the mixed liquor from the effluent, replacing the secondary clarifier altogether. Solids separation happens at the membrane, so the biology can run at mixed liquor concentrations three to five times what a clarifier can carry, the plant footprint shrinks by half or more, and the effluent is turbidity-free and largely disinfected, ready for reverse osmosis or direct reuse with little more than disinfection. In the submerged form, hollow fiber or flat sheet membrane cassettes sit directly in the bioreactor and pull permeate by slight suction; in the external form, a sidestream of mixed liquor pumps through tubular modules at crossflow velocity and returns to the tank.
This page treats the membrane bioreactor as a purchasing object: how flux and mixed liquor concentration size the membrane area, how scour aeration, relaxation, and chemical cleaning hold that flux against fouling, how sludge age and pretreatment protect both biology and membranes, and how the machine compares with the clarifier and dissolved air flotation trains it replaces. It stands between the bar screen that protects it and the reverse osmosis and ultraviolet steps that may follow it.
The decision concentrates in five places: the influent load and its variability, the effluent or reuse specification the membrane must deliver, the flux the fouling behavior of that wastewater will honestly sustain, the aeration and cleaning regime that holds it, and the pretreatment, screens and balancing, that keeps trash and grit from the membranes. Get those right and membrane area, blower duty, and tankage follow.
Chapter 1 / 06
What a Membrane Bioreactor Is and What Is Being Decided
The machine is a biological process with a membrane where the settler used to be. In conventional activated sludge, the mixed liquor flows to a secondary clarifier, a circular clarifier with a rake, where the flocs settle and the clear overflow leaves; the biology is limited by how well that settling works, because a clarifier carries perhaps 2 to 4 grams per liter of mixed liquor suspended solids before it overloads. In the membrane bioreactor, the mixed liquor, at 8 to 15 grams per liter or higher, is drawn through membranes with pores of 0.02 to 0.4 micrometers: everything larger than the pore, the whole biomass, the suspended solids, the protozoa, most bacteria and their aggregates, stays behind, and the permeate leaves with turbidity below 0.2 NTU without a settler at all. The membrane does not do the biological work, the biology does that; the membrane decides the solids inventory and the effluent quality.
Two consequences organize the whole value proposition. First, footprint: at the higher mixed liquor concentration, the biological volume for the same load shrinks, and the clarifier disappears, so mbr plants fit where conventional trains cannot, rooftop and basement installations, dense industrial sites, retrofits into old tanks. Second, effluent quality: the permeate is membrane filtered, low in suspended solids and associated pollutants, phosphorus bound to floc, pathogens removed to log values that leave only disinfection, uv sterilizer or chlorination, between it and reuse. Against these stand the costs: membrane replacement every 5 to 12 years, scour aeration energy well above conventional aeration, and a pretreatment discipline, fine screening, that conventional plants never needed.
What is being decided, in purchasing terms, is five things. First the load and its variability, biological and hydraulic, because membranes size on peak flux, not average. Second the effluent specification, discharge or reuse, with the downstream steps, reverse osmosis, uv, that depend on the permeate's quality. Third the sustainable flux for that wastewater, from pilot fouling data or honest references, which sets membrane area and therefore most of the capital. Fourth the membrane configuration, submerged hollow fiber, flat sheet, or sidestream tubular, which trades energy against robustness. Fifth the pretreatment and sludge strategy, screens, balancing, and the waste sludge handling that the high sludge age makes digestate-rich. The paragraphs that follow work these in order.
A scale tour grounds the machine. A 200 person hotel runs a packaged submerged flat sheet mbr in a basement, 50 cubic meters a day of reuse water for irrigation and flushing. A 50,000 person municipal plant runs hollow fiber cassettes in concrete basins at 15,000 cubic meters a day. An industrial park runs sidestream tubular trains on concentrated chemical wastewater at fluxes a submerged design would refuse. A landfill leachate plant runs mbr ahead of reverse osmosis because the membrane barrier is the only thing standing between the leachate and the ro. The biology is the same activated sludge in all four; the membrane is the machine's signature and its risk.
Chapter 2 / 06
How Flux, Scour Aeration, and Cleaning Do the Work
Flux is the machine's load rating: liters of permeate per square meter of membrane per hour, lmh in every datasheet. Design fluxes for municipal submerged duty run 12 to 25 lmh average with peaks to 40 for hours at a time; industrial and difficult wastes run 8 to 15; sidestream tubular systems claim 40 to 120 at the cost of pump energy. Flux times daily flow fixes membrane area, and membrane area is most of the purchase: at the scale of a small plant, every lmh of optimism buys a smaller rack that fouls sooner, so the flux line in a specification deserves the same scrutiny a vendor's words deserve everywhere else, and the honest basis is pilot data on the actual water.
Fouling is the physics the whole design fights. Solids, colloids, and organic matter deposit on and in the membrane: a cake layer builds on the surface, pores narrow, and the transmembrane pressure, tmp, the suction the permeate pump applies, climbs to hold flux. The defense is layered. Scour aeration bubbles up through the hollow fiber curtain or along the flat sheet, shaking the fibers and sweeping the cake back into the mixed liquor; this aeration is 40 to 60 percent of the plant's energy budget and its sizing, specific aeration demand per membrane area, is a headline number in every comparison. Relaxation, stopping permeate flow for a minute or two every ten, lets the scour strip the cake while flux is off. Backwash, reversing permeate through the membranes for a short pulse, adds mechanical cleaning on hollow fiber designs.
Chemical cleaning restores what the daily regime cannot. Recovery cleaning in place, cip, soaks the membranes in moderate chlorine, typically 200 to 1,000 milligrams per liter of free chlorine for polymeric fibers, or citric acid where scale dominates, at intervals of weeks to months, and the frequency and duration are the membrane's actuarial table. Maintenance dosing, chlorine in the permeate backwash on schedule, stretches the intervals. The cleaning chemistry must match the membrane polymer and the foulant: chlorine eats some polymers, oxidant tolerance is a purchase specification line, and greasy industrial foulants may want alkaline chemistries the vendor must approve. Cleaning records, chemicals, concentrations, durations, tmp recovery, are the membrane's service history and the warranty's language.
The biology under the membrane runs the same activated sludge kinetics as any plant, with one mbr signature: sludge age, the solids residence time, runs long, 15 to 40 days, because the membrane keeps every solid in the system until it is deliberately wasted. Long sludge age means low sludge production per kilogram of pollutant removed, nitrification robust against cold and shock, and a waste sludge that is denser and more digestible, hauled by the same sludge pump and dewatered by the same centrifuge or press as any plant's. It also means the mixed liquor carries the history of every upset for weeks, so toxicity that a conventional plant washes out in days stays home in an mbr.
Pretreatment is the membrane's bodyguard. Fine screening at 2 to 3 millimeters, finer than any conventional plant, keeps rags, hair, and fibers from wrapping the hollow fiber bundles and the rotors of recycle pumps; a bar screen or coarse screen at the headworks feeds it. Grit removal protects the coarse bubble diffusers and the membranes' potting from abrasion. Flow balancing blunts the hydraulic peaks that would demand peak flux. Fats and oils in industrial feed coat membranes and demand source control or pretreatment with dissolved air flotation ahead of the biology. The mbr vendor's warranty is written against the pretreatment's performance, and the plants that read that clause after the first wrapped cassette learn it the expensive way.
The chemistry of fouling deserves a paragraph because it decides the cleaning bill. Organic fouling, the work of soluble microbial products and colloids, answers to chlorine maintenance cleaning; inorganic fouling, struvite in digestion liquors, calcium carbonate in hard waters, answers to citric acid soaks; and mixed deposits answer to both in sequence. The upstream chemistry tools are cheap and underused: a ph meter trend on the mixed liquor catches the excursions that stress both biology and membrane charge, and coagulant or powder activated carbon dosing, judiciously applied, adsorbs the organics that would otherwise park in the pores. Some plants polish permeate with an ozone generator where reuse standards ask for trace organics control, a reminder that the mbr is one barrier in a train, sized against the whole specification, not its own.
Chapter 3 / 06
Submerged Hollow Fiber, Flat Sheet, and Sidestream Designs and Duty Boundaries
The submerged hollow fiber design is the municipal and package-plant standard: curtains of vertical fibers, potting top and bottom into headers, mounted in cassettes of hundreds of square meters each, set directly into the bioreactor or a dedicated membrane tank. Permeate is drawn by suction through the fibers from an overhead manifold; scour air rises between the curtains from coarse bubble diffusers at the base; backwash pulses run the reverse. Hollow fiber racks pack area densely, at the best cost per square meter, and modular cassettes let plants grow by dropping in rows. Their sensitivities are fiber breakage, which sends solids straight to the effluent until repaired, and wrapping, hence the fine screen discipline.
The submerged flat sheet design mounts vertical panels of membrane on both sides of a rigid plate, stacked into cassettes with scour air rising between plates. Flux runs a step lower than hollow fiber for the same duty, and energy per cubic meter is comparable or slightly higher, but the geometry tolerates dirtier water, higher solids, and upsets that would blind fibers, and the panels are robust against breakage and easy to inspect and clean one by one. Flat sheet mbrs dominate the small package and industrial niche, the hotel basements and the food plants, and hold ground in municipal mid sizes where their tolerance pays for the extra area they ask.
The external or sidestream design pumps mixed liquor out of the bioreactor through tubular membranes at 2 to 4 meters per second crossflow and back, with permeate leaving the tube walls under modest pressure. Pumping crossflow costs energy, several times the submerged designs' scour, but buys fluxes that nothing submerged sustains, tolerance of high viscosity and high solids liquors, and a membrane train that sits above ground, accessible, cleanable, and expandable by bolting on modules. Sidestream mbr serves concentrated industrial waste, leachate, and sludge digestion liquor duty, where the water is too mean for a submerged cassette and the energy bill is smaller than the alternative.
Duty boundaries against the neighbors deserve plain language. The mbr is not a clarifier or dissolved air flotation replacement by preference but by economics: where land is cheap, discharge limits modest, and the flow large, the conventional train, biology plus clarifier plus multimedia filter if needed, still wins on lifecycle cost, and the mbr wins as footprint, reuse quality, or automation demands rise. It is not reverse osmosis: the mbr membrane takes solids and microbes, not dissolved salts, and the two run in series when reuse targets total dissolved solids. Downstream it hands a uv sterilizer or chlorination step an effluent they can actually disinfect, free of the suspended solids that shield pathogens in conventional effluent.
Configuration
Typical flux
Energy signature
Where it wins
Watch-outs
Submerged hollow fiber
12 to 25 lmh
Scour air 0.2 to 0.6 kWh/m3 total plant
Municipal scale, lowest area cost, modular
Fine screens mandatory; fiber integrity monitoring
Submerged flat sheet
10 to 20 lmh
Comparable to slightly higher
Dirty waters, small packages, robustness
More area for same flow; panel handling
External sidestream tubular
40 to 120 lmh
Crossflow pump energy multiples higher
Concentrated industrial, leachate, high solids
Energy per cubic meter; pump wear
Chapter 4 / 06
Key Specification Parameters
The specification is built from the effluent requirement backward. The process set states influent flow and load, bod, cod, nitrogen and phosphorus, temperature range, and the industrial character, oils, salinity, toxicity risk. The performance set states effluent bod and cod, total nitrogen and phosphorus, turbidity, and log removal targets for indicator organisms, plus any reuse standard by name. The design set states membrane configuration, design and peak flux, membrane area, mixed liquor concentration, sludge age, and the aeration split between biological fine bubble and membrane scour. The mechanical set states screen class, blower selection and redundancy, permeate and recycle pump materials, and the cleaning system capacity. The commercial set states membrane life guarantee, replacement membrane pricing, and the clean-in-place chemical program.
Parameter
Typical range
Why it matters
Design flux
8 to 25 lmh by duty
Sets membrane area and most of the capital
Peak to average flux ratio
1.5 to 2.0 for hours
Peak day sizes the rack, not the average day
MLSS
8 to 15 g/l
Biology density; scour and viscosity follow
Sludge age
15 to 40 days
Nitrification, sludge yield, upset memory
TMP operating band
5 to 30 kpa rising to clean
The fouling gauge the whole machine runs on
Scour aeration demand
0.15 to 0.6 nm3/h per m2 membrane
The largest single energy line
Screen opening
2 to 3 mm fine
Warranty condition for hollow fiber racks
Membrane life
5 to 12 years
The lifecycle replacement line
Two lines in that table decide the purchase's honesty. Flux, because every lmh claimed above the wastewater's true fouling behavior converts directly into a smaller rack that spends its life in cleaning and its warranty in argument; the protection is a pilot or a reference on comparable water with numbers attached. Membrane life and replacement pricing, because the membranes are the consumable the capital hides: a guarantee of 8 years backed by a published replacement price is a lifecycle number a bid can be compared on, and the absence of either is a message.
Instrumentation closes the machine: permeate flow meter, tmp by pressure transmitter pair or direct, turbidity meter on the permeate as the integrity alarm, dissolved oxygen meter in the biology, level transmitter on the tanks, and the vfd driven blowers and permeate pumps under plc sequence control running the relax, backwash, and cip cycles. Hollow fiber plants add periodic integrity testing, pressure hold or bubble point on the rack, because a broken fiber is invisible to everything except the turbidity meter until it is found on purpose. The scada trend of tmp slope per hour is the fouling rate, the single number that tells the operator whether tomorrow's cleaning can wait.
Waste sludge handling deserves its own numbers because the long sludge age changes its character and its route. At 15 to 40 days of solids residence, the yield falls to roughly 0.3 to 0.5 kilograms of dry solids per kilogram of bod removed, well below the conventional plant's 0.6 to 0.8, and the sludge arrives stabilized and denser, 8 to 12 grams per liter straight from the mbr instead of 3 to 5 from a clarifier underflow. Dewatering follows the same logic as anywhere: small plants thicken and haul liquid, mid plants run a decanter centrifuge or a filter press to 20 percent cake, and the digestion option rewards the stabilized character with gas rather than instability. The sludge line, wasting pump, storage, and dewatering, belongs on the mbr specification because its omission has stranded more than one package plant with a full tank and no plan.
Chapter 5 / 06
Selection Decision Factors and RFQ Language
Selection starts with the wastewater, characterized for the membrane's sake, not just the biology's. Beyond the usual bod and nitrogen: oils and grease, which coat membranes; fibers and trash, which wrap them, hence the screen audit; salinity and hardness, which scale them; and the industrial shocks, ph swings, solvents, surfactants, which the long sludge age will remember for weeks. A wastewater characterization that names those four is the difference between a flux that lasts and a rack that cleans itself to death, and the pilot that answers them is the cheapest line in the project.
The configuration then follows the water and the site. Municipal scale with decent pretreatment: submerged hollow fiber on area economics. Small package, dirty variable water, minimal attention: flat sheet on robustness. Concentrated industrial or leachate: sidestream tubular on tolerance. Sites with cheap power and expensive labor lean sidestream's way; sites with expensive power lean submerged. The blower selection threads through it, roots blower for small plants, turbo blower for large ones with its magnetically suspended efficiency, and the aeration energy line deserves the same comparison discipline the membrane price gets.
The reuse question stands in front of the whole selection: what is the permeate for. Discharge to a moderate limit may not justify any mbr, and the honest comparison is the conventional train with its clarifier and filter on lifecycle cost. Reuse, unrestricted urban reuse, boiler or cooling makeup, ro feed, makes the membrane's product the point, and then the downstream steps, reverse osmosis for dissolved solids, uv sterilizer or chlorination for the final barrier, belong in the same specification so interfaces, permeate tankage and pumping, are designed once.
A complete RFQ states, on one process data page: influent flow profile and loadings with seasonal extremes; wastewater character including oils, fibers, scaling species, and shock risks; required effluent specification by parameter and any reuse standard; configuration preference or openness; design and peak flux with its basis; mixed liquor and sludge age targets; pretreatment provided, screen class and grit; aeration utilities available and power price; cleaning chemical constraints; membrane life guarantee and replacement pricing; redundancy philosophy, cassette or train out of service at what flow; and the automation level expected. Vendors quoting against that page quote the same plant, and their areas, energies, and membrane replacement economics become comparable line items.
Balance of plant belongs in the same decision. Return activated sludge and wasting, a sludge pump and a small dewatering or hauling arrangement, close the biology's mass balance; the permeate tank and its pumps serve reuse; the chemical room stores chlorine and citric in the quantities the cip calendar demands. Sewage pump stations and submersible pump wet wells feed the headworks; the bar screen and fine screen, the grit system, and flow balancing upstream are as much a part of the mbr as the cassettes are, and the warranty says so in writing.
Retrofit deserves its own selection note because so many mbrs are born inside old tanks. An existing conventional plant's aeration basins become the bioreactor with modest baffle work; the clarifier becomes equalization, sludge handling, or membrane tank; the blowers get company in the form of scour duty. The retrofit economics usually beat greenfield because the civil work exists, but three audits gate the idea: tank geometry versus cassette dimensions, blower capacity and age against the new duty, and the headworks against the fine screen the warranty demands. Vendors publish retrofit reference lists precisely because this path is half the market, and the honest bid asks for the retrofit references specifically.
Chapter 6 / 06
Installation, Standards, Failure Modes, and Maintenance
Installation is tank and skid work with a cleanliness finish: membranes are the last thing installed, into basins cleaned of construction debris that would have wrapped and scored them, with the fine screens commissioned and proven before biomass ever arrives. The first months establish the mixed liquor and the fouling baseline together, and the tmp trend from commissioning is the reference curve the plant will judge itself against for years. The acceptance test states permeate quality, tmp behavior at design flux, and specific energy, and it is run long enough for the first fouling cycle to show its slope.
The governing references are engineering practice and reuse standards rather than a single machine standard: the EPA's membrane bioreactor fact sheet documents the technology's basis and performance record, vendor documentation from Kubota and DuPont's Mempulse represents the two dominant submerged practices, and the state or national reuse standards define the effluent targets the permeate answers to. Membrane integrity testing practice, pressure hold and bubble observation, comes from the membrane filtration body of practice. The pilot and the reference plant remain the honest selectors, and the warranty document, with its pretreatment and cleaning clauses, is the legal boundary of the machine's promises.
Failure modes are fouling, breakage, and biology. Chronic tmp climb means the flux or the cleaning regime is mismatched to the water, answered by flux reduction, more frequent maintenance cleans, and a foulant review, not by heroic chemicals. Sudden turbidity on hollow fiber means a broken fiber, found by integrity test and isolated by pinning or replacing the element. Foaming, bulking, and nitrification loss are the biology's complaints, standard activated sludge failures with the mbr twist that they arrive at the membrane as fouling. Scour air decline, a worn blower or a diffuser buried in mixed liquor solids, fouls racks quietly, and the specific aeration demand trend catches it.
Symptom
Likely cause
First action
TMP climbing faster each week
Flux too high, fouling regime mismatched
Reduce flux or increase maintenance cleans; foulant autopsy; review pretreatment
Permeate turbidity spike
Broken hollow fiber or seal
Run integrity test; isolate and pin or replace element; resample
Foaming events recurring
Biology upset, surfactant load
Source control survey; antifoam compatible with membranes; review f/m ratio
Scour air uneven across rack
Diffuser fouling or header blockage
Drain and inspect diffusers; restore air balance between cassettes
Cleaning no longer restores tmp
Irreversible fouling or aging membranes
Escalate chemical review with vendor; plan membrane replacement from life model
Routine maintenance is the discipline of the trend charts: daily tmp slope and turbidity review, weekly integrity testing on a rotating cassette schedule, monthly cip per the calendar or the trend, whichever speaks first, quarterly screen inspection and rake or screen wear attention, and the blower program, filter changes on the roots blower, condition monitoring on the turbo blower. The permeate and recycle pumps, vfd driven with ac motor drives running constant duty, follow standard rotating equipment care. Records are the maintenance: every clean logged with chemicals and recovery, every integrity test with results, because the membrane's warranty and its remaining life are both argued from them.
The institutional item is source control, the quiet duty of every mbr operator. The long sludge age remembers every solvent slug for weeks, the membranes remember every grease load with a cleaning bill, and the fine screens remember every rag with a rake jam. A standing surveillance of what enters the sewer, industrial discharges, hauler loads, and the occasional kindergarten of wet wipes, protects the machine more than any chemical program, and costs a phone call.
The energy audit is the maintenance item that pays the plant back. Twice a year, measure and trend specific air demand per membrane area, total kilowatt hours per cubic meter, and the split between scour, biology, and permeate pumping: the first reveals diffuser fouling and air imbalance that cost flux, the second catches the slow drift that turning off one oversized blower would fix, and the third reminds everyone that the vfd constants set in commissioning were opinions. Plants that run this audit on schedule routinely shave 10 to 20 percent off the aeration bill, which over a membrane life pays for a cassette.
FAQ
What does the membrane bioreactor replace in a conventional plant?
The secondary clarifier and usually the tertiary filter. Solids separation happens at the membrane, so the biology runs at much higher mixed liquor concentration in smaller tanks, and the permeate is already low-turbidity and largely disinfected, ready for ultraviolet or chlorine polishing and reuse.
How much energy does an mbr use compared to conventional activated sludge?
Typically one and a half to three times the aeration energy, because membrane scour air is added on top of biological aeration, bringing total plant energy to roughly 0.3 to 1.0 kilowatt hour per cubic meter by scale and configuration. Sidestream designs pay multiples more in crossflow pumping. The premium buys footprint and effluent quality, and the honest comparison prices those explicitly.
What flux should I design at?
The one your wastewater sustains, from pilot data or references with numbers: 12 to 25 lmh for municipal submerged duty, 8 to 15 for difficult industrial waters, 40 and above for sidestream tubular. Every lmh above the honest value buys a smaller rack that spends its life cleaning and its warranty in dispute.
Why do mbr plants need finer screening than conventional plants?
Because hollow fiber curtains and flat sheet channels wrap and blind with rags, hair, and fibers that a clarifier would shrug at; 2 to 3 millimeter fine screening is the warranty condition for most submerged designs. Pretreatment failure shows up as wrapped cassettes and manual cleaning shifts, and it is the most common root cause behind mbr horror stories.
How often do membranes need cleaning and replacement?
Maintenance cleaning in place runs weekly to monthly by trend, recovery cleans a few times a year, and replacement lands at 5 to 12 years depending on water, cleaning discipline, and polymer. The lifetime economics ride on published replacement pricing and honest life guarantees, which belong in the bid comparison, not the fine print after it.
Can an mbr effluent go straight to reverse osmosis?
Yes, that pairing is one of the mbr's strongest niches: the permeate's low suspended solids and silt density index protect ro membranes that conventional effluent would foul, and leachate and reuse plants run exactly this train. The mbr takes solids and microbes; the ro takes the dissolved salts behind them.
When is a conventional clarifier plant still the better choice?
Where land is cheap, discharge limits are modest, flow is large, and energy is priced: the conventional train with a clarifier and multimedia filter wins on lifecycle cost. The mbr earns its premium when footprint, reuse quality, automation, or discharge limits push beyond what settling can deliver.