What are you working on?

Find a calculator, explore a formula, or learn the method.

THE ENGINEERING REFERENCE / GUIDE

Single-Stage Liquid Extraction: Using Distribution Data in a Material Balance

Short Answer

A single-stage liquid-liquid extraction is solved by combining total and solute material balances with distribution data: a distribution coefficient for immiscible liquids, or tie lines on a triangular diagram for partially miscible ones. Together, these relations fix the rates and compositions of the two product streams, the raffinate and the extract.

In liquid–liquid extraction, a liquid feed containing a component to be removed—the solute—is contacted with a second, immiscible liquid, the solvent. After mixing and settling, the system separates into two streams: the extract, solvent laden with the transferred solute, and the raffinate, the feed liquid left with less solute. A single-stage contactor is fully described by three groups of relations—mole balances, phase-equilibrium equations, and the process specifications—which is exactly how the standard reference treatment frames it (Lane, “Single-Stage Solvent Extraction,” CRC Press, 2019).

This article answers one focused question: given a feed and a solvent, how do you use distribution data in the material balance to compute the amounts and compositions of the raffinate and extract leaving a single equilibrium stage? The short answer: write the total and solute balances, then close them with equilibrium information—an algebraic distribution coefficient when the two liquids are essentially immiscible, or tie lines on a triangular diagram when they are partially miscible. The sections below define the terms, walk through both calculation routes with a labeled worked example, and note the assumptions and limits.

What a Single Extraction Stage Is

A single stage is the smallest self-contained unit of extraction equipment: a vessel, or one idealized cell of a larger column, where feed and solvent are contacted, allowed to approach phase equilibrium, and then physically separated into two leaving streams. The feed supplies two ingredients—the solute to be removed and a carrier liquid (often water) in which the solute is dissolved. The discussion in this article, like the standard treatments it follows, is limited to three-component mixtures: solute, carrier, and solvent.

The key idealization is the equilibrium stage. It assumes the two leaving streams are in phase equilibrium with each other, so their compositions are linked by the system’s equilibrium (distribution) data regardless of how fast mass transfer actually occurs. How close a real contactor comes to this ideal is an efficiency question, taken up later in the article.

Distribution Data: The Equilibrium Information You Need

Two forms of distribution data are used in practice:

  • Distribution coefficient (K). For a given solute, the ratio of its concentration in the extract phase to its concentration in the raffinate phase at equilibrium: K = y/x. A large K means the solute prefers the solvent phase. K must be expressed on a basis consistent with the material balance (mass or moles, solute-free or total), and it generally varies with concentration and temperature.
  • Tie lines on triangular diagrams. When the solvent and carrier are partially miscible, all three components appear in both phases, and compositions are plotted on a triangular (ternary) diagram. Each tie line is an experimental curve segment connecting the compositions of the two liquid phases that coexist at equilibrium; its endpoints are the extract and raffinate compositions. Experimental phase-equilibrium data and tie-line analysis on triangular diagrams are the standard graphical inputs for these calculations (Lane, 2019; Vega-Mercado, “Separation,” UNESCO–EOLSS).

Where does the data come from? Ultimately from laboratory measurements of two-phase mixtures. For calculation convenience, the raw data are often replaced by fitted representations—polynomial equations fitted to the extract and raffinate equilibrium curves and to the tie lines, or a thermodynamic activity-coefficient model such as the Non-Random Two-Liquid (NRTL) equation (Lane, 2019). These substitutes stand in for the three phase-equilibrium equations that would otherwise be needed for each of the three components.

The Material Balance Around the Stage

Choose a basis (for example, one hour of operation or 100 kg of feed) and denote the entering flows as F (feed) and S (solvent), and the leaving flows as E (extract) and R (raffinate). Two balances always apply:

  • Total balance: F + S = E + R
  • Solute balance: F·xF + S·yS = E·yE + R·xR, where x denotes solute fraction in feed-side streams and y in solvent-side streams (xS and yS describe solute entering with the solvent, often zero for fresh solvent).

Four unknowns remain—the flow rates and compositions of E and R—and two balances alone cannot close them. The equilibrium data supply the missing relations, and the process specifications decide which quantities are given and which are sought (Lane, 2019). The route you take depends on miscibility:

FeatureImmiscible (negligibly miscible) liquidsPartially miscible liquids
Equilibrium dataDistribution coefficient KTie lines on a triangular diagram
Calculation modeAlgebraic, closed-formGraphical (with lever rule), or fitted-correlation based
Carrier behaviorCarrier and solvent flows pass through unchangedCarrier partly dissolves in solvent and vice versa
Key resultsRaffinate and extract solute ratios; fraction extracted = E/(1+E)Compositions read from tie-line endpoints; amounts from lever rule

Route 1: Immiscible Liquids — Algebra with a Distribution Coefficient

If the solvent and carrier are mutually insoluble, the carrier simply passes through to the raffinate at rate Fc, and the solvent passes through to the extract at rate S. Working on a solute-free mass-ratio basis (x = kg solute per kg carrier; y = kg solute per kg solvent), the solute balance plus the equilibrium relation yE = K·xR give a direct solution:

  • Solute balance: Fc·xF + S·yS = Fc·xR + S·yE
  • Substituting yE = K·xR and solving: xR = (Fc·xF + S·yS) / (Fc + K·S)

Define the extraction factor Ef = K·S/Fc. For fresh solvent (yS = 0), the fraction of solute recovered into the extract is Ef/(1 + Ef). The two levers are intuitive: a larger distribution coefficient or a larger solvent rate raises recovery—but with diminishing returns, since each added unit of solvent helps less than the last.

Worked Example (Hypothetical)

The following numbers are illustrative, not measured data. A feed of 100 kg/h contains 20 wt% solute in a carrier water stream: 20 kg/h of solute and 80 kg/h of carrier, so xF = 0.25 kg/kg carrier. A pure solvent is added at 100 kg/h, immiscible with water, and the distribution coefficient on a mass-ratio basis is K = 4.

  • Extraction factor: Ef = K·S/Fc = (4 × 100)/80 = 5
  • Fraction extracted: 5/(1 + 5) = 0.833, i.e., about 83%
  • Raffinate solute ratio: xR = (80 × 0.25)/(80 + 400) = 0.0417 kg/kg carrier
  • Check: solute in raffinate = 80 × 0.0417 ≈ 3.3 kg/h; solute extracted = 20 − 3.3 ≈ 16.7 kg/h; yE = 16.7/100 = 0.167, which indeed equals K·xR. The solution satisfies both the balance and the equilibrium relation.

Route 2: Partially Miscible Liquids — Triangular Diagrams and Tie Lines

When solvent and carrier partially dissolve in each other, compositions of both phases must be read from equilibrium data rather than from a single coefficient. The classical graphical procedure on a ternary diagram works as follows (Lane, 2019; Vega-Mercado, EOLSS):

  1. Locate the entering streams. Plot the feed composition F and solvent composition S on the triangular diagram.
  2. Find the mixing point M. The overall composition of the combined F + S mixture lies on the straight segment between F and S, positioned by the lever rule: the distances from M to F and S are inversely proportional to the flow rates.
  3. Apply the equilibrium constraint. Because the leaving streams are in equilibrium, the extract and raffinate compositions lie at the two ends of a tie line, and the overall mixture M = E + R must lie on that tie line between the endpoints. If the exact tie line passing through M is not tabulated, it is found by interpolating among measured tie lines, or by using fitted tie-line correlations instead of raw data.
  4. Back out the stream amounts. With the endpoint compositions known, the lever rule (or the two material balances) gives E and R: each amount is proportional to the distance from M to the opposite endpoint.

The results should always be checked for consistency: the product compositions must satisfy both the total and solute balances and the equilibrium data simultaneously.

Step-by-Step: Working a Single-Stage Calculation

  1. Set a basis (per hour, or a fixed feed mass) and identify what is specified and what is sought.
  2. Assemble equilibrium data at the operating temperature: K for immiscible systems; measured curves and tie lines, polynomial fits, or an NRTL-type model for partially miscible ones (Lane, 2019).
  3. Write the total and solute balances, consistent in units and basis with the equilibrium data.
  4. Close the system with equilibrium: substitute y = K·x algebraically, or impose the tie-line condition graphically.
  5. Solve for the four unknowns (E, R, and their solute contents), then verify: balances close, and the product pair lies on a valid tie line or satisfies K.

Assumptions, Data Quality, and Practical Limits

The equilibrium-stage idealization is the main assumption. Real contacting equipment—mixers, columns, and centrifugal contactors—exchanges solute at a finite rate, and whether equilibrium is actually approached is a mass-transfer and residence-time question. Equipment evaluation, optimization, and scale-up for extraction service are specialized design activities in their own right, as equipment suppliers emphasize in practice-oriented design guidance (Koch Modular, “Designing Liquid-Liquid Extraction Columns,” AIChE presentation, December 2020).

Data quality matters as much as the arithmetic. Distribution coefficients and tie-line positions vary with concentration and temperature; equilibrium data should match the operating temperature and the composition range of interest. Solvent choice largely determines K and the degree of mutual miscibility, which is why solvent selection is treated as a distinct design step in standard references (Vega-Mercado, EOLSS). Finally, a single stage has a hard physical ceiling on recovery: for the immiscible case, the fraction extracted Ef/(1 + Ef) always falls short of 100%, and no solvent rate removes the last traces of solute economically.

Broad coverage of extraction media, techniques, and practical context for both analytical and preparative liquid-phase extraction is collected in the Handbooks in Separation Science volume on Liquid-Phase Extraction (Poole, ed., Elsevier, 2019).

Where the Single Stage Fits: Crosscurrent and Countercurrent Cascades

A single stage rarely achieves deep solute removal, so real processes usually string stages together. In crosscurrent extraction, fresh solvent is added to the raffinate at each successive stage; in countercurrent multistage extraction, solvent and feed flow in opposite directions through the cascade. Standard treatments analyze these configurations as chains of single equilibrium stages—the same balances and phase-equilibrium relations applied stage by stage (Vega-Mercado, EOLSS). Mastering the single-stage calculation is therefore the prerequisite for all multistage extraction design.

Practical Takeaways

  • A single extraction stage is defined by mole balances plus phase-equilibrium equations plus specifications; distribution data supply the equilibrium part.
  • For immiscible liquids, one distribution coefficient K closes the problem algebraically, and the recovery ceiling is Ef/(1 + Ef) with extraction factor Ef = K·S/Fc.
  • For partially miscible liquids, use a triangular diagram: mixing point from the F–S lever rule, tie line through it (interpolated or fitted), and lever rule for stream amounts.
  • Always verify that the solution satisfies both the balances and the equilibrium data, and remember that equilibrium is an idealization—real stages need efficiency and mass-transfer considerations.

FAQ

What is a distribution coefficient and how is it used in the balance?

It is the ratio of a solute's concentration in the extract phase to its concentration in the raffinate phase at equilibrium (K = y/x). In a single-stage calculation it provides the equilibrium relation y = K·x that, combined with the total and solute balances, closes the problem and lets you solve for the leaving stream compositions and flows.

When can I use algebra alone instead of a triangular diagram?

When the solvent and carrier are essentially immiscible, a single distribution coefficient on a consistent basis is sufficient, and the balances reduce to a closed-form equation. When solvent and carrier are partially miscible, all three components distribute between the phases, and you need tie lines on a triangular diagram (or fitted equilibrium correlations such as NRTL) to fix the leaving compositions.

How much solute can one extraction stage remove?

For an immiscible single stage with fresh solvent, the fraction recovered is Ef/(1 + Ef), where the extraction factor Ef = K·S/F (carrier basis). Recovery can approach—but never reach—100% no matter how much solvent is used, which is why multistage crosscurrent or countercurrent cascades are used when deep removal is required.

How does a single stage relate to crosscurrent and countercurrent extraction?

They are built from the same unit: a cascade of equilibrium stages, each described by the same material balances and phase-equilibrium relations. In crosscurrent operation, fresh solvent is added at each stage to the raffinate; in countercurrent operation, solvent and feed flow through the cascade in opposite directions.

References

  1. Lane, Alan M. "Single-Stage Solvent Extraction." CRC Press eBooks, published November 7, 2019. https://doi.org/10.1201/b22271-21
  2. Vega-Mercado, Humberto. "Separation." UNESCO–EOLSS Sample Chapters, Food Engineering, Vol. III. https://www.eolss.net/sample-chapters/c10/E5-10-04-07.pdf
  3. Poole, Colin F. (ed.). Liquid-Phase Extraction. Handbooks in Separation Science, Elsevier, 2019. https://www.eng.uc.edu/~beaucag/Classes/ChEThermoBeaucage/Books/(HandbooksinSeparationScience)ColinF.Poole(editor)-Liquid-phaseExtraction-ElsevierScienceLtd(2019).pdf
  4. Glatz, Don, and Brendan Cross (Koch Modular). "Designing Liquid-Liquid Extraction Columns: An Introduction." AIChE East Tennessee Local Section presentation, December 8, 2020. https://www.aiche.org/sites/default/files/community/160956/aiche-community-site-files/1720551/liquid-liquidextraction-koch-dec2020.pdf

Related Terms

Leave a Reply

Your email address will not be published. Required fields are marked *