VLizard documentation

Overview of thermodynamic consistency tests

VLizard offers a wide range of tests for checking the thermodynamic consistency of your experimental VLE data (both isobaric and isothermal). This is not to be confused with generic statistical analysis, which is provided by many other software tools, while thermodynamic consistency examines compliance of experimental data with fundamental laws of thermodynamics. Most of these procedures are based on the Gibbs-Duhem equation (G-D), with different approaches to its simplification or approximation (the partial differential equation is inconvenient for direct application). That means there can be no single test to guarantee the correctness of your data, but a combination of several tests, each with different scope and limitations, can help you identify systematic errors in your data.

This document covers only practical aspects of individual tests. Theory is briefly explained in Theoretical background document. Alternatively, see all literary sources for VLizard.

Which test to choose?

Fredenslund test

The most versatile test, it should always be performed.

van Ness test

While all other tests examine only VLE dataset & vapor pressure models to see if they internally comply with G-D,
the van Ness test also needs a fitted thermodynamic model. It then examines if G-D is valid between data & model.

Gamma offset test

A novel procedure based on activity coefficient definition rather than G-D, which detects if the VLE data aligns with vapor pressure models (a necessary condition for thermodynamic consistency).

Redlich-Kister test

One of the oldest and most well-known tests based on G-D in an integral form.

Herington test

An empirical extension of the Redlich-Kister test.

Slope test