Wait, What? If you have 30 mL of extraction solvent, using all 30 mL at once can recover less solute than using three 10 mL portions.
Liquid–liquid extraction is not a washing ritual. It is repeated equilibrium. A solute distributes between two immiscible phases according to its relative affinity for them, so every fresh portion of solvent gives the remaining solute another chance to repartition.
The mechanism: distribution between phases
For a solute X at fixed temperature and chemical form, a simple partition coefficient can be written K = [X]organic/[X]aqueous. The model assumes the relevant molecular form is comparable in both phases and equilibrium has been approached.
The Royal Society of Chemistry demonstrates iodine distributing between aqueous and organic solvents, making the two-phase equilibrium visible through colour. Royal Society of Chemistry practical experiments
Why several smaller extractions can win
Suppose K = 4, the aqueous phase is 20 mL, and 100 arbitrary units of solute begin in water. One 30 mL extraction leaves a fraction Vaq/(Vaq + K Vorg) = 20/(20+120) = 0.143, so about 14.3 units remain.
With three fresh 10 mL extractions, each leaves 20/(20+40)=1/3 of what was present. After three rounds, (1/3)³ × 100 ≈ 3.7 units remain. Same total solvent volume, much better recovery.
Equilibrium requires contact, then separation
The phases must contact sufficiently for solute transfer. Shaking increases interfacial area, but violent shaking can create emulsions that separate slowly. Vent a separatory funnel appropriately when pressure can build, and follow laboratory risk assessment for the solvents used.
Know which layer is which
“Organic layer is always on top” is false. Layer order depends on density. Identify phases from known solvent densities or a small drop test where appropriate, not from a memorised rule.
Chemical form can change the partition
Acids and bases may be neutral in one pH range and ionic in another. Ionised species are often more water-compatible than neutral forms. Acid–base extraction deliberately changes chemical form to move compounds selectively between phases. A single K value cannot describe a system whose chemistry changes during extraction.
Observation versus inference
Observation: “The upper layer became purple while the lower layer became paler.” Inference: “More coloured solute is present in the upper phase after mixing.” Stronger quantitative claims require concentrations, phase volumes and calibration; colour intensity alone is not a partition coefficient.
Quantitative window: extracting K
If equilibrium concentrations are measured as 0.080 mol dm⁻³ in the organic phase and 0.020 mol dm⁻³ in the aqueous phase, K = 4.0 under those conditions. But if the solute reacts, associates or ionises differently between phases, the simple ratio may need a more complete distribution model.
Failure modes
- Discarding the wrong layer.
- Using colour as though path length and concentration were identical.
- Insufficient mixing or separation time.
- Emulsion formation.
- Ignoring solvent remaining in the opposite phase.
- Assuming K is universal across temperature, pH and chemical form.
Unfamiliar transfer: caffeine and environmental pollutants
The same logic appears in extracting natural products, sample preparation and pollutant transport between water, soil organic matter and biological lipids. The transferable question is: what species is present, what phases are available, and where does equilibrium favour it?
Secondary → JC → deeper Chemistry
Secondary: distinguish miscible and immiscible liquids and separate layers safely. JC: calculate partition, extraction fraction and repeated-extraction efficiency; include acid–base speciation. Deeper Chemistry: distribution ratios, activity, multistage counter-current extraction and chromatographic partition extend the same principle.
Checkpoint
You have 20 mL aqueous solution and 20 mL organic solvent. K = 3 in favour of the organic phase. Would splitting the organic solvent into two 10 mL extractions increase recovery?
Answer key and WHY reasoning
Yes. Each fresh 10 mL portion re-establishes equilibrium with the solute remaining in water. Sequential equilibrations reduce the residual fraction multiplicatively.
How we know and evidence boundaries
A measured distribution supports a partition model for the specified solute form, solvent pair, temperature and concentration range. It does not establish that the same coefficient applies to a different pH, solvent composition or reacting system.
Teaching Guide
Give students a fixed solvent budget and ask them to choose one large extraction or several small ones before calculating. Then make them explain the result from equilibrium rather than treating repeated extraction as a recipe.