Module guide · Available

Solution Thermodynamics: how components behave in mixtures

When substances share a phase, pure-component properties are no longer enough. Each component contributes differently than it would alone, and unlike molecules interact differently from like ones. This module describes those contributions and departures, and produces the activity and fugacity quantities that phase and reaction equilibrium calculations need.

Part of Chemical Engineering Thermodynamics. Builds on Pure Component Properties.

The question this module answers

Mix equal volumes of ethanol and water and the result occupies less than the sum of the two; the mixture also warms slightly. Neither fact follows from the pure liquids alone. Solution thermodynamics explains such behaviour by asking what each component contributes to a mixture property and how far the real mixture departs from a simple reference.

The answers matter because equilibrium is decided component by component. Whether ethanol escapes to the vapour, or a reaction goes further in one solvent than another, depends on each species’ chemical potential in the mixture. This module defines those quantities and the models that estimate them; it names the equilibrium criterion but leaves equilibrium solving to Phase Equilibrium and Chemical Equilibrium.

Before you start

From Pure Component Properties: Gibbs energy, residual properties, fugacity and fugacity coefficient, cubic equation-of-state parameters and vapour pressure. These are used here without being derived again.

Two branches, one bridge to equilibrium

There are two established ways to describe non-ideal mixtures, and the module keeps both. One starts from an ideal solution and corrects it with activity coefficients; it is the natural description of liquid solutions. The other extends an equation of state to mixtures and gives fugacity coefficients directly; it suits gases and fluids an equation of state represents well. Neither is derived from the other.

Mixture statetemperature, pressure, composition
Solution non-ideality Partial molar propertieswith chemical potential Properties of mixing Excess properties Activity coefficients γi
Real-fluid mixtures Pure-component equation-of-state parameters Mixture equation of statemixing rules, kij Fugacity coefficients φi
The two branches of solution thermodynamics. Within each branch an arrow marks a quantity built from the box above; the final arrow shows that phase-equilibrium calculations use both branches. The fugacity-coefficient branch is a peer of the activity branch, not a step after it; both feed phase-equilibrium calculations. Dashed boxes are context outside this module’s five topics.

Topic order and scientific dependency

The platform presents the five topics in a teaching order that differs from the dependency map above. Fugacity Coefficients, whose chapter opens with the ideal-gas mixture, comes second; the chain from mixing to activity coefficients follows. Chemical potential, the ideal-gas mixture and the ideal solution are taught inside these topics rather than as separate ones, and binary interaction parameters are model data, not a topic.

Platform topic order: a learning progression, not a scientific dependency

  1. Partial Molar Properties
  2. Fugacity Coefficients
  3. Properties of Mixing
  4. Excess Properties
  5. Activity Coefficients

Scientifically, mixing, excess properties and activity coefficients form one chain built on partial molar properties; fugacity coefficients stand on their own equation-of-state branch.

Partial molar properties and chemical potential

What does each component contribute to a mixture property?

A partial molar property M̄i is the change in a total mixture property when a little of component i is added at constant temperature, pressure and amounts of everything else. The mixture property is recovered by summability, and the contributions cannot change independently because of the Gibbs–Duhem equation:

M = Σ xi M̄i   Σ xi dM̄i = 0  (constant T, P)

The partial molar Gibbs energy is the chemical potential, μi = Ḡi. Equality of each component’s chemical potential across phases is the condition for phase equilibrium. It is introduced here and applied in Phase Equilibrium.

Mixing versus excess: two different comparisons

What changes simply because components are mixed, and how far is the real solution from ideal?

A property of mixing, ΔMmix, compares the mixture with the unmixed pure components at the same temperature and pressure. An excess property, ME, compares the real mixture with an ideal solution of the same composition. They are linked by

ΔMmix = ΔMid + ME

For an ideal solution there is no volume change or heat of mixing, but there is still an entropy of mixing, −R Σ xi ln xi, so the Gibbs energy of mixing is negative. Real behaviour lives entirely in the excess term.

Which properties can be calculated

Excess Gibbs energy is always available from an activity model. Excess enthalpy and entropy need the model’s temperature dependence, so they are offered only where that dependence is defensible: for example library Wilson and NRTL correlations within their validity ranges, and UNIFAC’s temperature-dependent group interactions. Models without it, such as Margules and Van Laar with constant parameters, report excess enthalpy and entropy as unavailable. Excess volume is non-zero only where a genuine excess-volume correlation exists; otherwise the ideal-volume assumption, VE = 0, is stated as such.

Activity and activity coefficients

How strongly does each liquid species depart from ideal-solution behaviour?

The activity coefficient γi measures that departure for one component, and the excess Gibbs energy is their composition-weighted sum:

GE/RT = Σ xi ln γi

On the usual Lewis–Randall basis, γi → 1 as the component becomes pure. Its limit at infinite dilution, γi∞, shows how unlike a solute is from its solvent. For a dilute dissolved gas, a Henry’s-law reference is often more natural; the two conventions describe the same physics with different reference points.

The models offered are Margules, Van Laar, Wilson and NRTL, which are fitted to data through binary interaction parameters, and original UNIFAC, which predicts activity coefficients from the functional groups in each molecule. A fitted model is usually more reliable within the conditions of its data; a predictive model is useful where no fitted parameters exist, at the cost of larger uncertainty.

The NRTL activity-coefficient worked example follows one ethanol–water calculation from Library parameters to γ, including the pure-liquid reference and the ordering of directional parameters.

Activity versus mixture fugacity

How non-ideal is each species in a real gas or equation-of-state mixture?

Both branches end at the same physical quantity, the fugacity of a species in the mixture, f̂i, but reach it differently:

  • Activity route: f̂i = xi γi fi, a pure-component reference fugacity corrected for solution non-ideality. Typical for liquids at low to moderate pressure, including polar and associating mixtures.
  • Equation-of-state route: f̂i = yi φi P for a vapour (with liquid composition for a liquid), where φi comes from a cubic equation of state (Soave–Redlich–Kwong or Peng–Robinson) with van der Waals one-fluid mixing rules and a binary interaction parameter kij. Typical for gas mixtures and hydrocarbon-like systems, including higher pressures.

The ideal-gas mixture is the reference for the second route: there φi = 1 and f̂i = yiP, the partial pressure. Choosing between the routes is a modelling judgement about the system, and Phase Equilibrium lets the learner see its consequences.

Where the parameters come from

A fitted activity model is only as good as its binary interaction parameters, so their origin is part of the result. Three sources can be chosen explicitly, and the platform never switches between them silently:

Library
stored parameters with their source and the temperature range of the data behind them.
UNIFAC-estimated
parameters for a fitted model obtained by regression to UNIFAC predictions; labelled as estimates.
Custom
values entered by the learner, for example from a textbook problem.

Native UNIFAC needs group data rather than binary parameters. For equations of state, a library kij is used where one exists; otherwise kij = 0 is applied and stated as an approximation. Missing or ambiguous data make a calculation unavailable instead of falling back to ideal behaviour.

Common misunderstandings

  • “Activity coefficients correct for gas non-ideality.” They describe the solution, usually a liquid. Vapour non-ideality is described by fugacity coefficients.
  • “The Gibbs energy of mixing measures non-ideality.” An ideal solution already has a negative ΔGmix. Non-ideality is measured by the excess Gibbs energy, which can be positive or negative.
  • “A good fit of γ also gives a good heat of mixing.” Excess enthalpy depends on how the parameters change with temperature; parameters fitted at one temperature say little about it.
  • “Fugacity coefficients lie downstream of activity coefficients.” They belong to a separate equation-of-state description of the same mixture.

Scope and limits

  • No equilibrium solving. Bubble and dew points, flash calculations and phase diagrams are in Phase Equilibrium.
  • Classical mixing rules only. Equation-of-state/excess-Gibbs-energy mixing rules (such as Wong–Sandler or MHV) and electrolyte models (such as Pitzer or electrolyte NRTL) are not offered.
  • Data-limited properties. Excess enthalpy, entropy and volume depend on the model’s temperature and volume information, as described above.
  • Model estimates. Results are for learning; parameters outside their stated ranges are extrapolations.

Sources and next steps

  • J. M. Smith, H. C. Van Ness, M. M. Abbott and M. T. Swihart, Introduction to Chemical Engineering Thermodynamics, 9th ed.: the solution-thermodynamics sequence of partial molar properties, mixture fugacity, mixing and excess properties, and activity coefficients.
  • This guide summarises ChETL’s documented module scope as of 9 November 2026; editorial responsibility rests with the ChETL Scientific Editor. See sources, review and limits; report errors through Contact.

Launch ChETL

Launch ChETL