A wastewater plant is a building full of pumps, and the largest electrical load in it is a machine pushing air into water. That is a peculiar thing to spend most of a power budget on, and it stays peculiar right up until you look at how little of the air ever dissolves.
This post works through what oxygen transfer efficiency measures, what a change in it is worth arithmetically, and — the part usually skipped — why a better transfer number does not by itself entitle anyone to claim a saving on your bill.
Why is aeration the biggest line on a treatment plant's power bill?
Aeration is the largest electrical load at a conventional activated-sludge plant, reaching up to about 60% of the energy bill in published research. The reason is duty rather than size. Blowers run continuously to hold dissolved oxygen against a biological demand that never stops, while pumps, mixers, and dewatering equipment cycle on and off.
The load is structural. Activated sludge is an aerobic process: the organisms doing the treatment need oxygen delivered at the rate they consume it, and they consume it around the clock. Turning the air down is not an efficiency measure, it is a treatment decision, and the permit is written on the treatment.
It is a share of a plant's electricity, in published research and not from our own measurement, and it is an upper end rather than a typical value.
Scope the figure carefully, because it is easy to inflate. The share moves with plant type, loading, effluent limits, and how much of the site's other equipment is electrified. A lagoon and a tightly loaded nutrient-removal plant do not have the same bill or the same split.
What makes it interesting is the concentration of spend. A single process, running one class of equipment, on one physical mechanism — dissolving gas into liquid. That is a large enough share of one bill to be worth examining the mechanism, which is what the rest of this post does.
What is oxygen transfer efficiency, and what is typical?
Oxygen transfer efficiency is the fraction of injected gas that dissolves rather than escaping at the surface. In published research, above 85% has been reported for nanobubble delivery against roughly 15% for conventional diffused aeration. Both figures move with basin depth and geometry, so read them as two delivery regimes rather than as constants.
The physics behind the spread is contact time and interfacial area. A coarse bubble rises quickly, presents little surface per unit of gas, and reaches the atmosphere with most of its oxygen still inside. A smaller bubble rises slowly, carries far more interface per unit of gas, and has time to give that oxygen up. Depth is the other half: a deep tank gives any bubble a longer trip. Quoting either figure without the geometry it came from is quoting half a measurement.
It is also worth saying what transfer efficiency is not. It is a property of the injection process — what fraction of the gas you paid to compress ends up dissolved. It says nothing about oxygen crossing a root or a gill or a floc boundary layer, which is a diffusion problem at a surface no injector touches. The difference between concentration and delivery is a separate post because the two get conflated constantly.
Ratings in the field are lower than ratings on paper, and the derating chain is where plants lose money quietly. Manufacturers publish standard oxygen transfer in clean water: in published research, roughly 1.8–2.4 kg O₂/kWh for fine bubble, 0.9–1.3 for coarse bubble, and 1.2–1.8 for mechanical surface aerators. Process water then applies an alpha factor of about 0.5–0.65 for clean municipal fine-bubble service and 0.30–0.50 for industrial water with a surfactant load, before temperature, salinity, fouling, and the dissolved oxygen deficit are accounted for.
Where does the wasted energy actually go?
Diffused aeration spends most of its energy moving air that never dissolves. A blower pays to compress the entire volume of air against the water column, and a bubble reaching the surface with its oxygen still inside carries that compression work away. The rest disappears in blower and motor losses before the air reaches the diffuser.
The key asymmetry is that compression is paid on volume and treatment is paid in dissolved mass. The blower does not know which molecules will dissolve; it compresses all of them to the pressure at the diffuser depth. At a transfer efficiency near 15%, roughly six sevenths of that compression work leaves through the water surface.
Diffuser condition then makes it worse over time, in a way that is invisible without instrumentation. Membranes foul, pores narrow, back-pressure rises, and the bubbles coarsen. The blower works harder, the transfer efficiency falls, and both effects move the cost per kilogram of oxygen in the same direction. A plant can be paying materially more per unit of treatment than its design documents say and have no line item that shows it.
Put the derating chain together and the real number appears. A fine-bubble grid rated at 1.8–2.4 kg O₂/kWh in clean water, run in municipal process water at an alpha of 0.5–0.65, is delivering somewhere near 0.9–1.6 kg O₂/kWh in service. That figure — kilograms of oxygen actually transferred per kilowatt-hour actually consumed — is the only one that belongs in an energy comparison.
What is a percentage point of transfer efficiency worth?
A percentage point of oxygen transfer efficiency is worth what it displaces, which depends on where you start. Air per kilogram of oxygen scales as one over transfer efficiency, so moving from 15% to 16% cuts air demand by about a sixteenth, while the same point at 85% cuts it by one part in eighty-six. Same point, different money.
Work it in that direction and the shape is obvious. At 15%, a plant pushes a little under seven units of oxygen through the diffusers for every one it dissolves; at 16% it pushes about six and a quarter. That is a reduction of roughly 6% in air moved for the same oxygen delivered. Repeat the exercise at the top of the range and the same single point buys about 1%. Percentage points of transfer efficiency are not fungible: the same point bought at the bottom of the range and at the top are different quantities of air, and averaging the two describes neither.
Now the step that gets skipped, and it is the one that decides whether any of this is a saving.
A better transfer efficiency is not a discount on your current bill.
It is a different way of buying oxygen, and it comes with its own power draw: an oxygen concentrator separating nitrogen out of air, and a circulation pump moving the side stream. The honest comparison is two-sided — your derated cost per kilogram of oxygen transferred, against ours, both sides fully loaded.
Run that comparison honestly and it does not always come out our way. On a well-run fine-bubble municipal plant, oxygen made and delivered our way can cost more per kilogram than oxygen the existing blowers transfer, and our own wastewater calculator returns exactly that answer rather than hiding it. Where the argument does hold is coarse-bubble grids, mechanical surface aerators, lagoons, and high-strength industrial basins — the systems whose transfer efficiency is genuinely poor.
So the transfer-efficiency pair is a description of two delivery regimes, not an energy conclusion. Anyone who multiplies your aeration bill by a ratio of those two numbers has skipped the only step that matters, and the result will be a saving that does not appear on the meter.
What would change in an existing basin?
A retrofit changes the oxygen source and the delivery path rather than the tank. A side-stream loop draws basin water, dissolves oxygen into it under pressure, and returns it, so the existing diffuser grid can be turned down instead of removed. What moves on the meter is blower runtime, offset by the concentrator and pump that replace it.
Sizing it starts with what the basin actually needs, not with what is installed. Oxygen demand comes from the loading and the permit; installed blower capacity is a historical decision. The gap between those two is the first thing an assessment should produce, because a basin with substantial excess installed capacity has a different economic case from one running at its limit.
Then measure before you model.
- Aeration energy. What is the plant's aeration energy in kilowatt-hours, separated from the rest of the site load?
- Alpha factor. What is the basin's alpha factor, measured rather than assumed?
- Diffuser condition. How old are the diffusers and what is the current back-pressure?
Those three answers move the case more than any headline transfer-efficiency number, and two of them are usually unknown at the start.
One engineering constraint is easy to forget. Diffused air does two jobs — it delivers oxygen and it mixes the basin — and a basin that has its air turned down may still need mixing energy to keep solids in suspension. Turning the blowers down to a level the oxygen demand allows, and then discovering the tank needs mixers, converts a saving into a capital item.
Redundancy is the last one, and it is not negotiable in a permitted process. The existing aeration stays as the fallback, sized and maintained, whatever the new equipment does on a normal day. Our wastewater page sets out which basin types the argument holds for, the calculators will run the two-sided energy comparison on your own tariff and loading, and the classroom section on physico-chemical properties carries the mass-transfer background.