Define the model and its assumptions
For a well-mixed batch with no significant oxygen consumption or inflows and outflows, dC/dt = kLa × (C* − C). C* is the equilibrium concentration for the test gas and conditions. When kLa is in h⁻¹, time must be expressed in hours. Fit a time series rather than selecting two convenient readings.
Equilibrium concentration depends on gas oxygen fraction, pressure, salinity and temperature. A pure-oxygen test does not use air saturation as C*. Check whether the model describes the measurements and whether mixing is sufficient. If conditions change during the curve, one fitted coefficient may conceal that change rather than explain it.
Prepare reproducible measurements and controls
Record volume, geometry, recirculation, gas and gas-flow reference conditions. Calibrate the probe, check its range and consider response time. Avoid positions where a local jet or attached bubbles prevent the reading from representing the tank. Repeat tests with independent samples and retain the original time series.
Initial deoxygenation and any reagent used can alter the water. The laboratory or competent test team should define the method and controls; an article’s procedure should not be transferred directly into a biomass or product installation. The method must preserve a meaningful comparison and the integrity of the system.
Apply temperature correction in the right direction
A common convention is kLa20 = kLaT/θ^(T−20). The cited references use θ = 1.024 as an empirical factor; justify it for the protocol rather than treating it as a universal constant for every water. Also determine equilibrium concentration for the actual temperature and other test conditions.
Illustrative example: 6.0 h⁻¹ measured at 25 °C with θ of 1.024 corresponds to approximately 5.33 h⁻¹ at 20 °C. Multiplying rather than dividing would incorrectly increase the normalised coefficient. Temperature correction cannot remove probe errors or compensate for inadequate mixing in the tank.
Separate transfer, respiration and energy
For biologically active water, the balance includes oxygen uptake and any inflow and outflow terms. Estimate respiration through an appropriate method or fit a validated model. A slowly rising DO curve can combine high transfer with high consumption. Clean-water results cannot be transferred directly to wastewater.
Measure total power and gas supply when comparing efficiency. Higher kLa may require greater recirculation energy. The cited studies apply to their specific equipment and matrices, rather than being GENESIS™ performance data. Report uncertainty, conditions and the process outcome alongside the coefficient and energy boundary.
Common questions
Can kLa be obtained from one DO reading?
No. It needs a time series and a justified equilibrium concentration, together with checks on mixing, probe response and possible oxygen consumption.
Does correction to 20 °C turn wastewater results into clean-water results?
No. It normalises a thermal effect within a model. Matrix constituents, surfactants and biomass can affect transfer and uptake and require separate assessment.
Sources and documentation
- US EPA — Development of Standard Procedures for Evaluating Oxygen Transfer Devices, dynamic tests and temperature correction
- Arias-Torres and colleagues — Comparative Analysis of Volumetric Mass Transfer Coefficients for Oxygen Uptake and Desorption with Nanobubbles, Water, 2025
- Messina and Christensen — Aeration of wastewater with a ceramic membrane nanobubble generator, Chemical Engineering Journal, 2025