Vapourtec and Accelerated Materials collaborate to enhance automated chemistry efficiency
In a significant development for the field of flow chemistry, engineering specialists Vapourtec have announced a collaboration with Accelerated Materials to expand the application of machine learning (ML) in chemical optimisation processes. This partnership aims to improve the efficiency and accuracy of predictive modelling in flow chemistry, a technique increasingly utilised in pharmaceutical research and development.
The integration of ML into chemical processes addresses several limitations inherent in conventional optimisation methods. Professor Alexei Lapkin, co-founder and Chief Strategy Officer of Accelerated Materials, explained: “Traditional optimisation methods are based on variations of one-variable-at-a-time (OVAT) methods. The OVAT method frequently fails to capture any interaction effects between variables and is data inefficient.”
While Design of Experiments (DoE) methods have been employed to mitigate these issues, they too face challenges when modelling complex non-linear, dynamic systems. ML solutions aim to bridge this gap, offering enhanced efficiency in data acquisition and improved exploration of optimisation space.
Seamless integration with existing systems
The collaboration centres on integrating Accelerated Materials’ AMLearn™ software with Vapourtec’s R-Series flow chemistry system. Duncan Guthrie, founder of Vapourtec, highlighted the unique position of their system in the market: “Vapourtec’s R-Series is the only commercially available flow chemistry system that provides the opportunity to integrate ML algorithms using high-level commands to specify reaction methods rather than by communication with individual system components.”
This integration utilises Vapourtec’s industry-standard OPC UA interface, allowing for simple configuration and compatibility with various reactor types, including photochemical and electrochemical reactors.
The system’s versatility extends to its analytical capabilities, featuring data retrieval from a range of pre-configured Process Analytical Technology (PAT) devices. This includes Raman, FTIR, and UV detectors, enabling comprehensive real-time analysis of reaction processes.
Implications for pharmaceutical research
For the medical and pharmaceutical research community, this advancement in flow chemistry optimisation holds significant promise. The ability to rapidly analyse vast datasets and identify patterns in experimental conditions could potentially accelerate drug discovery processes and reduce development timelines.
Prof. Lapkin emphasised the broader impact of AI integration in chemistry: “Bringing AI into chemistry accelerates the discovery and optimisation of new compounds, reduces experimental costs and shortens the timelines for development.”
