Cavitation, gas dissolution and fine bubbles

How a Venturi draws in gas

A Venturi aerates without a compressor. Water speeds up through the throat and its pressure falls, by Bernoulli's principle Zheng et al., 2022. Even a small pressure difference between inlet and outlet produced suction at holes in the throat; the air injection rate depended on the flow's Reynolds number (a measure of turbulence), the air-hole and inlet diameters, and the pipe downstream Baylar et al., 2007.

Orifice-type generators work the same way. In one design, the pressure just downstream of the orifice turned negative and drew air in through a porous pipe Sadatomi et al., 2012. The pump that drives the water supplies all the energy.

How shear breaks gas into micro- and nanobubbles

The flow tears the incoming gas apart. In the orifice generator, high shear and strong turbulence broke the drawn-in air into a very large number of microbubbles. More than 70% of the measured bubbles were smaller than 0.1 mm (bubbles under 0.01 mm could not be measured). At 1.0 L/min of air, the mean diameter was about 0.12 mm. Producing microbubbles alone required a mean water velocity through the orifice above about 10 m/s Sadatomi et al., 2012.

Cavitation devices can also leave nanobubbles, smaller than 1 µm. A Venturi-type device recirculating water at 2 to 5 bar for 5 to 20 minutes produced bulk nanobubbles with mean diameters of 180 to 210 nm; a sample made in 15 minutes at 3 bar persisted for 72 hours Li et al., 2021. In a self-aspirating Venturi, freshly treated water was milky with microbubbles; on standing, they coalesced, rose and burst within minutes, and the water cleared Zhou et al., 2022. Why nanobubbles persist, and how to tell them from microbubbles, is covered in Generation mechanisms and stability and Nanobubbles vs. microbubbles.

Pressure recovery and dissolution

Downstream of the throat, the pressure recovers suddenly Zheng et al., 2022. Bubbles carried on are now at a higher pressure, which favors dissolution. By Henry's law, the equilibrium ratio between a gas's abundance in the gas phase and in the water is constant for a dilute solution Sander, 2023, so more gas dissolves at a higher partial pressure, whether from pressure recovery or tank depth. Fine bubbles also offer a large gas–liquid interface and rise slowly, which lengthens contact time. How interface, bubble size and contact time set the transfer coefficient (kLa) is covered in Diffusion and gas–liquid mass transfer.

Only part of the gas dissolved in the one generator that measured it: 25 to 30% of the oxygen supplied, roughly independent of air flow and generator type, in a 1.2 m deep tank of tap water at 20 °C Sadatomi et al., 2012.

Measured oxygen transfer

Two measures answer different questions. Standard oxygen transfer efficiency (SOTE) is the share of supplied oxygen that dissolves. Standard aeration efficiency (SAE) is the oxygen dissolved per kilowatt-hour of power, pump included.

Modules of Venturi air injectors tested in water reached kLa values of 9.67 and 5.93 h⁻¹ for two- and three-injector parallel modules, against 3.63 to 4.54 h⁻¹ in series, and aeration efficiencies of 0.10 to 0.14 kg O₂/kWh in parallel against 0.06 to 0.07 in series Zhu et al., 2007. A 1.1 kW submersible pump with a Venturi injector reached 1.166 kg O₂/kWh at its best setting (14 mm nozzle, 60 cm depth, 45° angle); nozzle diameter made a significant difference, depth and angle did not Ghomi et al., 2009. The studies used different pumps and tanks, so these figures do not rank designs.

At pilot scale, an HC setup with a 1.86 m³ open tank, treating sedimented municipal wastewater at 23 °C, raised dissolved oxygen (DO) from 4.5 to 8 mg/L within 4 minutes; DO then held between 7.02 and 7.86 mg/L to the end of the 24-minute run Janetasari & Bokányi, 2026.

Pump power decides the comparison. At full scale, jet aerators in thick digester sludge failed their standard tests by substantial margins. They transferred a similar share of oxygen as coarse-bubble diffusers, but the diffusers had substantially higher aeration efficiency once the jet pump's power was counted, and fine bubbles coalesced in the thick sludge Steele et al., 2023.

Degassing versus dissolving

The low pressure that draws gas in can also pull gas out of solution. In a Venturi tunnel with controlled air content, large cavities were mostly water vapor, but the share of dissolved air diffusing into them rose as they shrank, and dissolved air left more gas bubbles behind after collapse Tian et al., 2025. Whether that gas redissolves or escapes depends on the pressure and contact time downstream.

Held at low pressure, a Venturi can degas water almost completely. A Venturi-nozzle bubbler in a vessel kept at 1 kPa, without heating, brought dissolved oxygen in 360 to 400 L of tap water close to zero Jun, 2023.

One study saw both directions. In 1 L of ultrapure water, a self-aspirating Venturi raised DO from about 8.5 mg/L to a peak of 11.3 to 11.8 mg/L at about 100 seconds, after which DO fell gradually while the device kept running; the water warmed by under 1.5 °C, too little to explain the fall. DO rose and fell in step with the number of microbubbles, the peak was lower when the added air flow exceeded 10 mL/min, and DO dropped again on standing as the microbubbles vanished Zhou et al., 2022. Water above saturation (supersaturation) tends to lose the excess, as explained in How gases dissolve in water and Water as a gas carrier.

Where this leads

Cavitation is one of the main ways fine bubbles are made, and the Micro- and nanobubbles course starts from there with Generation mechanisms and stability. How industries use cavitation and the gas it carries is covered lesson by lesson in agriculture, aquaculture, water treatment, mining and energy.