Mechanistic Modeling
Mechanistic chromatography models predict separation behavior from physical equations rather than empirical correlations. By describing the underlying physics: convection, diffusion, and adsorption, into mathematical differential equations, these models can extrapolate beyond the conditions used to calibrate them.
What is a Mechanistic Chromatography Model?
A mechanistic model describes a chromatography process using mass-balance partial differential equations (PDEs) that account for each physical mechanism occurring inside the column. Unlike empirical plate-count or regression-based approaches, mechanistic models resolve how solute concentrations evolve in both space and time inside the column, making them powerful tools for process development and optimization.
Why use Mechanistic Modeling?
- Predictive power: Once calibrated, the model can predict chromatograms under new operating conditions like different flow rates, load concentrations, gradient shapes, without additional experiments.
- Physical insight: Model parameters map directly to physical properties (e.g. porosity, diffusion coefficients, binding constants), providing deeper understanding of the separation.
- In-silico optimization: Thousands of virtual experiments can be run to screen process parameters and optimize yield, purity, or productivity computationally.
How Efflux Implements It
Efflux assembles a simulation from modular unit operations (inlets, tubings, mixers, columns, and detectors) connected in a flow graph. Each column is governed by a column model that describes solute transport, paired with a binding model that captures adsorption kinetics. The resulting system of differential-algebraic equations is then solved by the simulation engine.
To see how this works in practice, visit the chromatography simulator page.

