Chemical research · Research insights
Colour and video reveal where a stirred tank mixes slowly
Automated colour-video analysis measures local as well as overall mixing time. Learn why reproducibility is a meaningful advance for reactor studies.
A stirred tank can look busy while parts of its contents remain poorly mixed. A single overall mixing-time number may conceal those slow regions. Measuring where the last differences disappear is therefore valuable when comparing impellers, operating conditions or numerical flow predictions.
Latest breakthrough
A June 2026 Applied Microbiology and Biotechnology paper combines a colour-based mixing experiment with video capture and automated Python analysis. The method estimates global and local mixing times while varying stirring and aeration. The authors describe a multi-laboratory round-robin effort and revisions made after initial reproducibility problems were identified. That wider exercise was still in progress when reported. The contribution is a transparent measurement and analysis framework that can reveal spatial heterogeneity, rather than a claim that every laboratory already obtains identical results.
Barth and colleagues. Applied Microbiology and Biotechnology (2026). Read the original study.
How the method works
A colour-based experiment provides a visible marker of how mixing develops, while video records changes at different positions over time. Automated analysis turns those image changes into estimates of local and global mixing time. The distinction is useful: one overall number can hide a poorly mixed region. Varying stirring and aeration explores operating effects, and comparison between laboratories tests whether the procedure gives reproducible information. The reported revisions after reproducibility problems are part of the method's development, not evidence that the wider validation was already complete.
A simple analogy
Stirring coloured syrup into a glass may make the top look uniform while a darker pocket remains near the bottom. Recording several regions instead of glancing at the whole glass helps reveal that difference. Industrial vessels have more complicated flows and sometimes gas or biological material, so a visible colour pattern is a measurement tool rather than a complete account of every process.
A tank contains several flow environments
Fluid close to an impeller can move and exchange rapidly, while a remote corner may respond more slowly. Gas bubbles and vessel geometry add further variation. An average can hide this distribution. A spatial view lets an engineer ask whether a particular region limits the process or whether a measurement location simply gives an overly optimistic impression.
Why better measurement can be a major practical step
Our interpretation is that comparable measurements are essential before claiming that one mixer is better than another. A repeatable method can make experiments more useful for validating computational fluid dynamics and planning scale-up. It can also reduce time spent resolving disagreements caused by analysis choices rather than the equipment itself. The gain is stronger evidence for design decisions.
A coloured tracer is still a proxy
Optical access, lighting and the chosen endpoint influence the measurement. A transparent laboratory liquid is also different from every industrial broth, suspension or non-Newtonian material. The method should be calibrated and interpreted for the system being studied. The ongoing round-robin work is an important part of establishing how broadly the approach can be reproduced. Spatial variation also matters when supplying heat to a reactor, as illustrated by the electrically heated catalyst-bed demonstration.
Explore circulation around the impellers
The jacketed-tank model shows two radial impellers on a shared shaft, a drive assembly and a cutaway cooling jacket. Compare the region near each impeller with the spaces above, below and between them. This helps frame questions about mixing uniformity. The viewer displays geometry rather than a solved flow field and does not calculate the study's local mixing indices.
Related articles
An electrically heated foam brings heat inside a catalyst bed
Can a shorter cyclone still separate particles effectively?
A pilot heat pump reuses distillation heat and reveals a trade-off
Shaping a liquid interface produces high-rejection filter membranes
Sludge dewatering study shows why pressing harder is not enough