Distillation Minimum Stages (Fenske) & Gilliland Calculator

Understanding the Fenske–Underwood–Gilliland distillation calculator — concept and principle

Before a distillation column is simulated rigorously, engineers size it with the shortcut Fenske–Underwood–Gilliland method. For a binary (or light-key/heavy-key) separation it gives the minimum number of theoretical stages at total reflux, the minimum reflux ratio and an estimate of the stages needed at the reflux you actually plan to run.

Less reflux than infinite means more stages: 6.43 at total reflux, 14.1 at R = 1.43.

How it is calculated

Fenske: N_min = ln[(x_D/(1 − x_D)) × ((1 − x_B)/x_B)] ÷ ln α
Underwood (binary, saturated-liquid feed): R_min = [x_D/x_F − α(1 − x_D)/(1 − x_F)] ÷ (α − 1)
Gilliland (Eduljee form): X = (R − R_min)/(R + 1), Y = 0.75[1 − X^0.5668], and N = (Y + N_min)/(1 − Y), since Y = (N − N_min)/(N + 1).

x values are mole fractions of the light key; α is the relative volatility of light to heavy key; R = (R/R_min) × R_min.

The principle

At total reflux each ideal stage multiplies the light/heavy ratio by α, so after N stages the ratio has grown by α^N. Taking logarithms gives the Fenske equation. At the other extreme, the minimum reflux is where the operating line touches the equilibrium curve at the feed point, creating a "pinch" that would need infinitely many stages; Underwood's equation finds it. Gilliland correlated real column designs between these two limits, and Eduljee fitted that curve with a simple formula.

Worked example

Inputs: x_D = 0.95, x_B = 0.05, α = 2.5, x_F = 0.5, R/R_min = 1.3.
Separation factor = (0.95/0.05) × (0.95/0.05) = 361; N_min = ln 361 ÷ ln 2.5 = 5.889 ÷ 0.9163 = 6.427.
R_min = (0.95/0.5 − 2.5 × 0.05/0.5) ÷ 1.5 = (1.9 − 0.25) ÷ 1.5 = 1.1.
R = 1.3 × 1.1 = 1.43; X = 0.33 ÷ 2.43 = 0.1358; Y = 0.75 × (1 − 0.1358^0.5668) = 0.5081.
N = (0.5081 + 6.4269) ÷ (1 − 0.5081) = 14.10 theoretical stages.

Each point is the calculator re-run with only “Relative volatility α” changed; the gold point is the default example. Change the inputs above and this chart follows your numbers.

Relative volatility α	Minimum stages Nmin
0.625	-12.52943082
1.25	26.39053615
1.88	9.368114597
2.5	6.426866226
3.12	5.168247125
3.75	4.455344762
5	3.658965601

How to read the result

Theoretical stages normally count the partial reboiler as one stage. Divide by an overall tray efficiency to get real trays. Reflux ratios of about 1.2–1.5 × R_min are a common economic range: a lower ratio means more stages but less energy, a higher one the opposite. If α is below about 1.2, stage counts climb sharply.

Common mistakes

Using the heavy-key fraction instead of the light-key fraction for x_D or x_B.
Applying the binary Underwood formula to a feed that is not saturated liquid.
Assuming constant α when it varies strongly from top to bottom (use a geometric mean of the top and bottom values).

Keep in mind: Shortcut methods assume constant relative volatility and constant molar overflow. They are for screening; real column design needs rigorous simulation, physical property data and review by qualified engineers.

Frequently asked questions

Does N_min include the reboiler?

Yes, in the usual convention the Fenske count includes the reboiler as an equilibrium stage, so subtract one for trays inside the column.

Where does the feed go?

The Kirkbride equation or a rigorous simulation estimates the optimum feed stage; this calculator does not give it.

Sources: M. R. Fenske, Industrial and Engineering Chemistry 24 (1932) 482 · E. R. Gilliland, Industrial and Engineering Chemistry 32 (1940) 1220 · J. D. Seader, E. J. Henley and D. K. Roper, Separation Process Principles, Wiley

Checked 8 Nov 2026. Standard textbook formulas in SI units, computed in your browser. For learning and estimates — real designs must follow your local codes and be checked by a qualified engineer.

Your numbers