Process Development
Turning a route that works into a process that is cheap, safe, reproducible and legal to run at scale.
The first route to a molecule is almost never the route you want to run for ten years. It was designed to prove the compound could be made, usually quickly, usually by someone who was not thinking about effluent load or a reactor occupancy of forty hours.
Process development is the work of replacing it with something better without changing what comes out of the other end.
What we optimise for
In roughly this order of priority, because they trade off against each other and pretending they do not is how programmes go wrong:
- Safety — no step that cannot be controlled thermally at full scale, whatever it costs in yield.
- Reproducibility — narrow variance beats a high average. Consistency is what your filing commits you to.
- Impurity profile — controlled and understood, not just currently passing.
- Cost — raw material selection, step count, solvent recovery, cycle time.
- Environmental load — solvent volumes and effluent that the site can actually treat.
Capabilities
Route scouting
Parallel evaluation of candidate routes against cost, safety, step count and freedom to operate, rather than committing to the first one that works.
Design of experiments
Statistical experimental design to map the process space and find the parameters that genuinely matter instead of changing one factor at a time.
Reaction calorimetry
Heat of reaction, adiabatic temperature rise and thermal stability measured before scale-up, not inferred from literature.
Crystallisation development
Polymorph screening, seeding strategy and particle size control - the steps that decide whether your formulation team has a good week.
Work-up simplification
Reducing extractions, chromatography and isolations, which are the steps that quietly destroy throughput at plant scale.
Cost improvement
Structured cost-down programmes on existing commercial products, with the saving shared rather than pocketed.
A process is finished when removing anything else from it would break it - not when it first produces material that passes.
Speak to someone who knows the answer
Technical, regulatory or commercial - your enquiry reaches a person who works on this, not a shared inbox.