Continuous-flow process automation suffers from a lack of software tools to design and inspect flow networks prior to runtime. Here, we present MechWolf, an open-source Python framework that allows efficient description, analysis, and implementation of automated flow-chemistry processes via an Internet of Things (IoT) approach as a solution to this problem. In this framework, we delineate three logically distinct parts of flow-chemistry networks: individual components (<i>i.e.</i> pumps, valves, and sensors), apparatuses (unique configurations of these elemental components), and protocols (lists of procedures defined for apparatuses). For each layer of abstraction, we utilized a novel object-oriented approach that readily allows for introspection, via either generated natural-language descriptions or visualizations. MechWolf introduces a completely novel protocol compiler to convert concise protocol descriptions written in Python into an unambiguous list of protocols for each component that features thorough error checking and intelligent automated inferences. Compiled protocols can then be wirelessly communicated to active components over the internet and executed in unison with millisecond accuracy. Automated data collection is supported, meaning that experiments can be designed, analyzed, and optimized entirely automatically. MechWolf is designed as a decentralized framework based on mature technologies with no single point of failure in order to ensure longevity in a lab environment. Finally, the entire package is extensively documented and user extensible, with automated tests to guarantee compatibility of future developments with published code. We also present the application of this package to specific, common flow-chemistry setups.

MechWolf: Programmatic framework for continuous-flow chemical process description, analysis, and automation

256th National Meeting (2018)

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.