How to Optimize Yield and Purity

Once a calibrated model is in place, simulation can replace lab experiments for exploring operating conditions. This guide covers how to use in-silico optimization to maximize yield and productivity whilst retaining purity requirements.

Overview

In preparative chromatography, yield and purity are often in conflict — collecting a wider fraction window increases yield but decreases purity, and vice versa. Traditionally, finding the right balance requires running many experiments at different conditions. With a calibrated model, these experiments can be replaced by simulations.

The optimization workflow typically involves:

1. Defining what you want to optimize — yield at a target purity, maximum productivity, or a weighted combination.
2. Choosing which process parameters to vary — gradient slope, flow rate, load factor, fraction cut points, or a combination of these.
3. Have the optimizer run thousands of simulations, converging to one or multiple final process conditions.
4. Analyzing the robustness of the final process.

This approach lets you map the full yield-purity design space from a single calibrated model, rather than running dozens of lab experiments. It also makes it straightforward to revisit the optimization when requirements change — for example, if the target purity specification is tightened.

Defining the Objective

Choosing Parameters to Vary

Running Simulations

Analyzing the Pareto Front

Selecting an Operating Point