Mixer-Settlers

A mixer-settler is a stage, not a machine: an agitated box where two immiscible phases are contacted, followed by a quiet compartment where they separate by gravity. Stages are repeated in series counter-currently, and everything about the design follows from two ratios — the mixer-to-settler volume ratio and the stage count. This range covers 100 mL to 100 L per mixer, mixer-to-settler ratios of 1:3 or 1:4, and five vessel materials, so the same process can be developed on a bench and run as a continuous pilot skid.

Configure it in four decisions

Material is a temperature and visibility trade-off

Transparent PMMA or PVC is what you want for method development. You can watch where the dispersion band sits, how long it takes to collapse, and whether an interphase crud layer is building — none of which you can see through PP. The limit is temperature: ambient to about 60 °C with PMMA/PVC, up to about 80 °C with PP/PPH.

PP and PPH are the workhorse materials for pilot and small commercial stages — welded, so repairable, and good to 80 °C. PTFE is for the chemistry that attacks everything else, and it is usually bought as an integral-moulded small stage rather than a large welded box, because PTFE does not weld into a pressure-tight seam easily.

Fabrication is not a detail. Adhesive bonding is how PMMA and PVC boxes are made and it is fine for aqueous duty at ambient temperature; hot gas welding is how PP/PPH boxes are made and it holds at 80 °C; integral moulding removes the seam entirely, which is why it is used for the small PTFE and PP stages where a leak path at the corner would be a permanent nuisance.

Why the settler is three to four times the mixer

The mixer's job is mass transfer: it needs enough volume for the droplets to be created and for the reaction to go to completion, and that is measured in seconds to a couple of minutes of residence. The settler's job is phase separation, which is governed by the settling rate of the dispersion and therefore by the horizontal area available, not by volume alone. A 1:3 or 1:4 ratio is the practical outcome of that difference; if a test shows a slow, rag-like dispersion, the ratio goes up or the throughput comes down.

That is also why throughput is quoted in L/h of aqueous feed and marked "to be confirmed by test work". The stage that limits a train is almost always the settler whose dispersion band does not collapse fast enough, and no catalogue number replaces a batch settling test on your own liquor.

Bench, pilot or continuous skid

• 100 mL – 5 L — method development: O/A ratio, stage count by McCabe-Thiele, strip and scrub ratios, crud behaviour. Run single stages, then a cascade.
• 15 – 100 L per mixer — pilot stages for flow confirmation, with per-stage agitators and interstage pumps.
• Complete skid — feed, scrub and strip stages with PP storage tanks, level gauges and a steel frame. This is the configuration that produces the numbers a feasibility study needs.

Questions buyers ask before ordering

How many stages do I need? From the extraction isotherm: draw the operating line at your chosen O/A, step off the stages, then add one for the scrub and one or two for strip. A pilot cascade is normally built 3–6 stages because stage count is easy to add and expensive to remove.

Can I see the interface? Only with PMMA or PVC. If watching the dispersion band matters to your test programme, choose transparent material and stay inside its temperature limit.

What is the smallest volume that still means something? About 100 mL of mixer, if the O/A ratio and mixing intensity are representative. Below that, wall effects and the difficulty of measuring a dispersion band start to distort the result.

Do you supply the pumps and tanks? On the skid configurations, yes — interstage pumps with independent flow control, PP feed/scrub/strip tanks, and one-to-one or touchscreen control. On single stages, the stage only.

Send the isotherm, not the drawing. O/A ratio, stage count, temperature and the worst-case chemistry (chloride, fluoride, organics) determine material and ratio. With those four we will propose a configuration; without them any quote is a guess.