Make the cleaning cycle longer, not the membrane cheaper
Fouling is the hidden cost of every membrane plant: lost flux, chemical cleaning, downtime and eventually premature replacement. There are two places to attack it — the surface, and the feed.
Where the money goes
In an MBR, aeration for membrane scouring alone consumes more than half of the plant's operating energy, and conventional millimetre bubbles do a poor job of it: they rise away from the surface instead of scouring it. Meanwhile every chemical clean costs money, produces a waste stream and shortens membrane life.
Two routes, and they are not alternatives
Scour with nanobubbles
Finer bubbles at the same gas flow create far more surface interactions and a looser cake. Installed on the existing aeration line, a nanobubble generator raises the number density by orders of magnitude at the same energy, and the cake becomes easier to remove rather than being compacted onto the surface.
Cut the load before the membrane
Every kilogram of oil and solids removed upstream is a kilogram the membrane does not have to store. Flotation ahead of the membrane is usually the cheaper intervention, and it is the one most plants skip because the membrane is already installed.
The two compound. At Cui Jiagou, nanobubble flotation removes the coal fines first, and the 20 nm SiC membrane behind it then holds 0.18 NTU with a cleaning interval measured in months rather than days.
Measured at full scale
| Project | Configuration | Result |
|---|---|---|
| Ningbo seawater RO pre-treatment | 1160 m³/h flat-sheet SiC with nanobubble pre-treatment | Cleaning cycle held at 20–25 days; effluent 0.2 NTU |
| Cui Jiagou mine water | 2×75 m³/h, nanobubble flotation + 20 nm SiC | 0.18 NTU, 94% TSS removal, accepted 2025 |
| Oilfield produced water | Hypergravity swirl + SiC, TDS 30,000–60,000 mg/L | <10 mg/L oil for reinjection |
Source: Ningbo project test report (May 2024) and acceptance report (November 2025); Cui Jiagou scheme ARFU-2025-083. The comparison baseline is the site's own previous cleaning frequency, not a generic industry figure.
What fouling control cannot fix
- A membrane already at end of life. Scouring does not restore a delaminated skin layer.
- Scaling from supersaturation. If the concentrate is above saturation for calcite or gypsum, no amount of bubble scouring prevents crystallisation inside the module.
- Biofouling inside a spiral-wound element. Once a biofilm forms in the feed spacer, cleaning is the only route; prevention has to happen upstream.
How to size the intervention
- Measure the current cleaning interval and the cost per clean (chemicals + labour + lost output).
- Identify whether the foulant is organic, biological, colloidal or crystalline. The answer decides whether you need scouring, oxidation or pre-treatment.
- Set the success metric as cleaning interval, not flux. Flux recovers after a clean; interval is what you are buying.
What is your current cleaning interval?
Give us that plus the foulant analysis, and we will tell you which of the two routes applies — or whether neither will move the number.
Send the data → SiC membrane specifications