Closed loop optimisation for sustainable chemical manufacture

Lead Research Organisation: University of Southampton
Department Name: Sch of Chemistry

Abstract

The overall aim of the proposed research is to enable the development and operation of new, agile, more cost-effective and sustainable chemical manufacturing processes.

The future of sustainable chemicals manufacturing is in flexible, modular and intensive processes. New automated reaction tools and hardware are becoming ubiquitous but optimisation of how they are used and the methods of dealing with the larger amounts of experimental data available are still largely manual processes, and generally only carried out for long duration production runs. A crucial missing component is a fast automated closed-loop methodology for development and running of optimised chemicals manufacturing processes.

This proposal will close this gap by developing an automated system for experimentation that brings together automated hardware for reaction execution, methods for reaction composition data acquisition and analysis, the intelligent selection of future experiments, and the development of process models in real-time. The multi-disciplinary challenge of this topic requires research in a variety of fields, including chemistry, statistics, engineering, chemometrics and computer science. Each of the individual research questions are novel and substantial challenges in their own right; their fusion will allow the automatic optimisation of reaction chemistry for a variety of applications and on a variety of different scales.

Such a system would become a key tool in both academic and industrial chemistry, making feasible the routine manufacture of even small amounts of material via optimised processes, and increasing the efficiency of processes on all scales. Hence, it has the potential to enable new ways of working towards sustainable and green chemistry.

Planned Impact

The main beneficiaries of the research will be academic and industrial chemists manufacturing quantities of material through reaction processes on a variety of scales. They will benefit through the development and demonstration of the proposed automated experimentation and optimisation system.

More specifically, in the Technology Strategy Board's (TSB) document 'A landscape for the future of high value manufacturing in the UK' (Feb 2012) a number of strategic themes were identified that are addressed in this proposal, including:
- Manufacturing systems: 'Increasing the global competitiveness of UK manufacturing technologies by creating more efficient and effective manufacturing systems' which includes redesigning processes to increase their yields and increase operation efficiency, design and manufacture of smaller products such as specialised drugs.
- Manufacturing processes: 'Developing new, agile, more cost-effective manufacturing processes' which includes flexibility of production and manufacturing supporting customised and rapidly reconfigurable manufacturing, adaptive manufacturing including single step, flexible reconfiguration and process technology that can adapt to feedstock of different types and compositions and mass customisation techniques.

Further, the TSB document 'The future UK life sciences manufacturing landscape: opportunities and challenges for high value manufacturing in the pharmaceutical and biopharmaceutical sectors' (Nov. 2012) identified the importance of value adding new products/processes including:
- More flexible production facilities located to support responsive, adaptable manufacturing capability;
- Making products to order in response to individual patient need and to reduce inventory;
- Manufacturing for personalised medicines.

The key technologies and capabilities required included: multifunction equipment with quick turn around - one plant for multiple products; continuous processing across a variety of platforms and unit operations; appropriate process controls and associated software and measurement to allow quality control, flexibility and small batch, complex processing.

The work proposed here is well aligned with these needs and as such will have immediate impact in chemical and life sciences manufacturing industries.

Further industrial and manufacturing impact will be achieved through:
- Provision of fast, low cost multi-objective optimisation and process understanding that allows rapid implementation and scale-up, delivering a substantial competitive advantage to U.K. chemicals manufacture through lower cost routes and reduced time to manufacture.
- Enabling optimum conditions to be used for the manufacture of small batches of chemicals, a key future area as described above. Custom synthesis of compounds is a substantial business in the U.K. but is under pressure from low manpower cost economies.
- Ensuring more sustainable manufacture with less waste though use of optimum conditions. Processes may be optimised for sustainability, rather than cost only.
- The ability to make use of current experimental downtime (evenings, weekends) and the freeing-up the scientist from routine operations during the day. It puts the focus on more highly skilled and valuable positions in the U.K. economy. - Enabling the transition to modular 'where needed' chemical manufacture by allowing processes to be rapidly implemented in a variety of different plants and scales, with variable input materials, to produce the same quality product.
- Improvement of real-time optimisation of large-scale manufacturing, including increase in process flexibility through better process understanding.

Ultimately, the developed methods will enhance U.K. economic competitiveness and quality of life by decreasing the cost and time to market of manufactured products in a variety of areas including pharmaceuticals, agrochemicals and fine chemicals.

Publications

10 25 50
 
Description • Developed and demonstrated an autonomous closed loop reaction optimisation system using a sophisticated commercial flow reactor with in-line hplc monitoring. The use of multiple analytical methods simultaneously has been enabled, but the data generated not yet incorporated in the closed loop system.
• Demonstrated the efficient closed loop optimisation of a batch polymerisation process
• Demonstrated closed loop optimisation of a C-H activation reaction by parameterisation of a mechanistic model for the transformation using in-line Gas Chromatography for analysis.
• Developed new statistical and mathematical methods for efficient close loop optimisation, in particular aimed at providing a process model (empirical or mechanistic) for the transformation, rather than just finding an optimum.
Exploitation Route There has been industrial take-up of the findings, particularly from our key partners. As the work is published we expect the techniques to be adopted by others.
Sectors Chemicals,Manufacturing, including Industrial Biotechology,Pharmaceuticals and Medical Biotechnology

 
Description The findings have been used in continued interactions with industry, for example some of the techniques developed have been implemented at Syngenta, although the full 'closed loop' systems did not reach the level of development where industrial use was viable. A number of additional grants in collaboration with industrial partners have been obtained by several of the investigators to continue the work which will lead to exploitation.
First Year Of Impact 2017
Sector Agriculture, Food and Drink,Chemicals
Impact Types Societal,Economic

 
Description CASE conversion awards
Amount £56,000 (GBP)
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 10/2015 
End 09/2018
 
Description Industrial Studentship
Amount £50,000 (GBP)
Organisation UCB Pharma 
Sector Private
Country United Kingdom
Start 10/2015 
End 09/2018
 
Description Rapid acquisition of process data using gradients in flow
Amount £80,000 (GBP)
Organisation Syngenta International AG 
Sector Private
Country Switzerland
Start 10/2019 
End 03/2023
 
Description Responsive Mode. Combining Chemical Robotics and Statistical Methods to Discover Complex Functional Products (A Lapkin, D Woods, L. Cronin)
Amount £1,227,510 (GBP)
Funding ID EP/R009902/1 
Organisation Engineering and Physical Sciences Research Council (EPSRC) 
Sector Public
Country United Kingdom
Start 02/2018 
End 01/2021