Self-Optimising Multiphasic Flow Reactor & Work-up Processing
Lead Research Organisation:
University of Leeds
Department Name: Chemical and Process Engineering
Abstract
Three key elements, each with its own significant academic research challenges & questions to be answered: i) Evaluate systems for continuous separation of liquid from immiscible liquids & solids to allow the reactor to be coupled with at-line HPLC & benchtop MS & NMR, whose use might also allow real-time spectroscopic calibration with IR & UV-Vis. The separation methods will be developed to allow sequential at-line measurement (ie through semi-continuous or automated preparation of samples). This will build on existing & proven physical principles for separation including the use of extended surfaces of differing surface energies to promote separation [6]. The development of appropriate separation systems for in-line & at-line monitoring will depend on the reaction system (see iii) & will be carried out in parallel with these developments ii) Establish thermal & barometric control of the multiphasic flow reactor to allow a wider range of chemistries. This will be carried out in close collaboration with the planned PhD student to start in Jan 2016 iii) Exemplar reaction systems are highlighted below & through discussion with AZ a feasible work programme will be established to develop a hierarchical approach to developing multiphasic flow systems. There are a wide range of conditions that can be evaluated, both in terms of the chemical reactions, but also the physical forms (multiphasic) that such reactions require a) Optimisation of a variety of liquid-liquid reactions will be demonstrated. This could include exemplar reactions such as oxidation with liquid oxidants, nitration of organic molecules & biphasic cyclopropanation reactions. Rapid experimentation using automated systems will enable development of mechanistic & statistical models for facile & scalable process development. Characterisation of mass transfer coefficients will be performed for a variety of reactor configurations including both plug flow & CSTR & combinations of active & static mixers. b) A variety of exemplar liquid-liquid work-up separation systems will be evaluated, though a common aspect each requires reliable separation & sampling from both phases. This could be achieved using stacked hydrophilic/hydrophobic plates to drive separation - by using stacks of alternating laser cut components it may be able to make units that have a similar footprint to the freactor & avoid the cost associated with machining channels into layers. In addition, such devices would also allow for counter-current liquid-liquid extraction at a small scale. Excitingly, optimisation will be targeted to deliver high quality product subsequent to biphasic reaction systems. Systems will be critically evaluated & benchmarked for their operability & scalability. c) Liquid-solid systems are perhaps most common & an exemplar is the continuous crystallisation currently being evaluated in an existing AZ supported project. This has already been tested successfully in the iPRD Freactor & involves co-feeding a racemic acid & chiral base to give diastereomeric solid. In this case we would self-optimise the yield by varying solvent, stoichiometry, concentration & residence time by separating the solid with an in-line filter, monitoring first the enantiomer concentrations in the mother liquors, then by diverting a solvent flow dissolve the crystals off the filter & then to the HPLC, MS or ORD detector. The system would require flexibly programmed diversions of solvent flow for sequences of sampling & washing. An ambitious extension of this might be analysis & optimisation of particle size. Cooling crystallisations could also be evaluated. For insoluble solids, designs that can be tested are reverse washing or flows directed through exchangeable multi-parallel filters. Exemplars could include hydrogen transfer with Pd/C, Suzuki cross-coupling or Grignard reactions. d) Exemplar tri-phasic systems
Activities (i)-(iii) will be carried out over years 1-3 as appropriate.
Activities (i)-(iii) will be carried out over years 1-3 as appropriate.
Publications
Chapman M
(2017)
Simple and Versatile Laboratory Scale CSTR for Multiphasic Continuous-Flow Chemistry and Long Residence Times
in Organic Process Research & Development
Clayton A
(2019)
Algorithms for the self-optimisation of chemical reactions
in Reaction Chemistry & Engineering
Clayton A
(2020)
Self-optimising reactive extractions: towards the efficient development of multi-step continuous flow processes
in Journal of Flow Chemistry
Clayton A
(2020)
Automated self-optimisation of multi-step reaction and separation processes using machine learning
in Chemical Engineering Journal
Doherty S
(2018)
Highly efficient aqueous phase reduction of nitroarenes catalyzed by phosphine-decorated polymer immobilized ionic liquid stabilized PdNPs
in Catalysis Science & Technology
Manson J
(2019)
A Hybridised Optimisation of an Automated Photochemical Continuous Flow Reactor
in CHIMIA
Martínez-Carrión A
(2019)
Kinetic Treatments for Catalyst Activation and Deactivation Processes based on Variable Time Normalization Analysis
in Angewandte Chemie
Martínez-Carrión A
(2019)
Kinetic Treatments for Catalyst Activation and Deactivation Processes based on Variable Time Normalization Analysis.
in Angewandte Chemie (International ed. in English)
Schweidtmann A
(2018)
Machine learning meets continuous flow chemistry: Automated optimization towards the Pareto front of multiple objectives
in Chemical Engineering Journal
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
EP/N509681/1 | 30/09/2016 | 29/09/2021 | |||
1803783 | Studentship | EP/N509681/1 | 30/09/2016 | 29/03/2020 | Adam Clayton |
Description | A new multi-objective algorithm has been developed and successfully applied to the self-optimisation of chemical systems. Also, a simulator has been designed to test the performance of multi-objective optimisation algorithms on real chemical examples. Complex systems were the target of this investigation, including multi-step flow processes. This led to the first demonstration of automated optimisation including non-reactive unit operations. Additionally, a new laboratory-scale reactor has been developed, characterised and assessed for a wide variety of chemistries, including multiphasic and photochemical reactions. |
Exploitation Route | The multi-objective optimisation approach can be used by chemists to automatically optimise any process in which two or more relevant performance criteria are conflicting. The associated simulator can also be used by computer scientists to test the performance of their new multi-objective optimisation algorithms against other state-of-the-art algorithms. The developed flow reactor can be utilised by chemists as an initial test for their reactions in flow, and its simple design is particularly suited for chemists new to flow chemistry. |
Sectors | Chemicals Manufacturing including Industrial Biotechology Pharmaceuticals and Medical Biotechnology |
Description | The developed approaches in this project have been adopted by pharmaceutical companies including AstraZeneca and Dr. Reddy's Laboratories. In 2018, AstraZeneca utilised multi-objective self-optimisation to optimise the first step in the synthesis of AZD3293 which was later published in Chem. Eng. J. In 2020, Dr. Reddy's Laboratories set-up a self-optimising flow reactor with multi-objective optimisation capabilities as a result of this work. |
First Year Of Impact | 2018 |
Sector | Chemicals,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology |
Impact Types | Societal Economic |