Automated method points way for making complex chemicals

For the first time, researchers at Cambridge University applied the concept of flow chemistry to complex molecules by using an automated feedback system.

‘We don’t physically need to be on site anymore to run these reactions. We can use very simple devices such as webcams to keep track of what’s going on in the lab and remote desktop access for the hardware,’ said Dr Ian Baxendale, who heads the Innovative Technology Centre at Cambridge along with Prof Steven Ley.

Many of the consumer chemicals used in modern society are made using a protracted, multi-step batch process — for example, pharmaceuticals often require at least 10 distinct processes.

Batch production is frequently an inefficient process that demands excess chemicals and solvents, generating large quantities of waste materials. 

’As chemists, we are a very conservative industry, we don’t tend to like technology, we’ve had our batch reactor for 250 years and we tend to stick with what we know,’ Baxendale said.

Employing what is known as a ‘flow assembly’ is, however, a significant challenge for complex molecules because of the need to add sequential reagents at the correct concentration to the product stream. 

In their latest project, the researchers were able to achieve this through a feedback system that uses infrared monitoring to precisely control concentration.

‘We can make very fine adjustments to the process as we need to, in a real-time scenario, without ruining what could be 10 or 15 metric tonnes of material in a batch reactor,’ Baxendale said.

In their initial trials, the group was able to make pyrazoles — which are used in analgesic and anti-inflammatory drugs — at yields comparable to standard methods, but with a significantly reduced amount of methyl hydrazine reagent.

Baxendale explained that flow technology also has advantages in terms of scalability. In the case of batch production, different-sized reactors give different kinetics and purity profiles, so the whole process needs to be re-optimised when scaling-up.

‘We use a small reactor and instead of using size to dictate the amount of product we make we use time as a function. So you use the same reactor under the same conditions for a longer period of time to generate more material — it saves going through this re-iterative process of re-inventing the wheel to scale up,’ Baxendale said.