Overview

Pharmaceutical process development often relies on time-consuming experimentation to optimise purification operations such as distillation, solvent swaps, and crystallisation. At PolyModelsHub, we develop mechanistic and hybrid “system models” using our PMH-core engine to accelerate this workflow and provide deeper insight into process–property interactions.

For batch distillation, we employ intentionally simple yet robust modelling frameworks that balance computational efficiency with predictive capability. Our models typically use NRTL as the primary thermodynamic package and are calibrated using minimal targeted experiments designed in collaboration with our customers. This approach enables rapid understanding of VLE behaviour, impurity trajectories, and solvent removal efficiency while significantly reducing development cycle times.

A key feature of our workflow is the integration of liquid–liquid equilibrium (LLE) models with batch distillation and crystallisation models. This allows us to track impurity distribution across phases, evaluate the impact of solvent composition on solubility, and assess crystallisation outcomes after solvent exchange. By linking these unit operations through shared thermodynamics and material balances, our system models capture the broader process context that is often missed when unit operations are studied in isolation.

To further enhance process understanding, we apply global sensitivity analysis (GSA) to quantify how uncertainties in process parameters propagate to critical quality attributes (CQAs). This provides a structured view of parameter importance, guiding both experimental design and process optimisation.

Overall, our integrated modelling framework helps pharmaceutical teams reduce experimentation, accelerate cycle times, and make more confident solvent selection and process development decisions. These system models—when combined with targeted calibration and GSA—offer a powerful route toward smarter, data-efficient process development.

Speaker

Charalampos (Harry) Christodoulou, Co-Founder and Director of Science and Modelling, PolyModels Hub

Harry is Co-Founder and Director of Science and Modelling at PolyModels Hub, specialising in mechanistic and hybrid models for pharmaceutical processes. With prior experience at GSK in systems modelling and simulation, and a PhD in chemical engineering from UCL, Harry’s work focuses on building predictive, efficient process models and end-to-end workflows that accelerate pharma process development.

The material presented at this event has not been peer-reviewed. Any opinions are the presenter's own and do not necessarily represent those of IChemE or the Fluid Separations Special Interest Group. The information is given in good faith but without any liability on the part of IChemE.

Time

13:00–14:00 GMT.

Software

The presentation will be delivered via Microsoft Teams. We recommend downloading the app from the Microsoft website, rather than using the web portal.