The potential advantages displayed by biocatalytic processes for organic synthesis (such as exquisite selectivity under mild operating conditions), have prompted the increasing number of processes running on a commercial scale. However, biocatalysis is still a fairly underutilised technology. As a relatively new technology biocatalytic processes often do not immediately fulfil the required process metrics that are key for an economically and/or environmentally competitive process at an industrial scale (high concentration, high reaction yield, high space-time-yield and high biocatalyst yield). These process metrics can often be attained by improvements in the reaction chemistry, the biocatalyst, and/or by process engineering, which often requires a complex process development strategy. Interestingly this complexity, which arises from the need for integration of biological and process technologies, is also the source of the greatest opportunities. Indeed, recombinant DNA technology offers a superb complement to process technologies. Potentially this is one of the biggest advantages of biocatalysis when compared with conventional chemical catalysis, where all the reaction boundaries are fixed by the physical and thermodynamic properties of the reaction compounds. Therefore, the main avenue that still remains to be explored by process engineers is how to promote process development in a systematic way rather than on a case-by-case basis, as is frequently the case today. One of the main challenges in process development is selecting between different process alternatives. The development effort for a novel process is considerable and thus, an increasing number of conceptual process design methods are now applied in chemical industries. Since the natural environment of the biocatalyst is often very different from the operating conditions suitable for a viable process (high substrate and product concentrations, unnatural substrates, presence of organic solvents, etc.), process development strategies are particularly relevant for biocatalytic processes. However, state-of-the-art methodologies for process development applied to biocatalysis often prove to be unsuccessful. At the early development stage the biocatalysts are usually still under development and many of the reactions have not yet achieved their full potential, many of the process technologies are not yet well described and their relationship with the overall process is not clear.The work described in this thesis presents a methodological approach for early stage development of biocatalytic processes, understanding and dealing with the reaction, biocatalyst and process constraints. When applied, this methodology has a decisive role in helping to identify many of the process bottlenecks up-front and in a straightforward way, whilst indicating development targets, allowing a better use of resources and shortening development time. The methodology is illustrated through three different case studies: H-caprolactam production by a multi-enzymatic process, chiral amine production using Z-transaminase and finally long-chain chiral aliphatic Abstract ii alcohol production by a bi-enzymatic system. For each case study presented, a different tool is used to guide development and evaluate the process when different levels of underlying process knowledge are available. The first case study presents a rational approach for defining a development strategy for multi-enzymatic processes. The proposed methodology requires a profound and structured knowledge of the multi-enzyme systems, integrating chemistry, biological and process engineering. In order to suggest a reduced number of feasible process design options, cofactor and interaction matrices are used, identifying the challenges and addressing them by selecting appropriate process configurations. Based on this information, feasible flowsheets and mass and energy balances are identified. By applying evaluation tools, the number of options can be much reduced and the current process bottlenecks identified. By applying a priori this methodology, the Laboratory experts are better able to understand the most favourable operating conditions at fullscale and thus be able to collect information at these relevant conditions. In the second case study, windows of operation are used to quantify and visualise process performance and feasibility when interactions between process Technologies and biocatalyst performance (or reaction) are significant. The methodology constitutes a useful tool that provides easy interpretable results to enable rational design choices of different available process technologies. In the particular case of the asymmetric synthesis of chiral amines, the reaction constraints (thermodynamic equilibrium) must be solved prior to implementation and these fix the hard boundaries of the operating space. Improvements in the biocatalyst specific activity are also required for a successful full-scale implementation. In the third case study a methodology for bottleneck analysis is presented, incorporating process modelling and engineering evaluation tools. The benefit of such models, when integrated with evaluation tools, is that they can be used to predict the process performance and identify bottlenecks, without requiring experimental examination thereby reducing the resources and time for process development. The use of this methodology in the context of reaction engineering is to propose new operating conditions at which the process performance is improved, while identifying the remaining bottlenecks and suggesting further research efforts. Although the proposed methodology is still in its infancy when compared with other established process development methodologies, it provides a good overview of the whole reaction system and process. The proposed methodological approach establishes a systematic evaluation of different process options and indicates required fundamental data collection and development efforts for further development stages. This methodology could be greatly enhanced by the implementation and integration of in-silico tools for property and thermodynamic data as well as process mechanistic models to assist in the selection of process technologies.
Main Research Area:
Tufvesson, Pär, Woodley, John
Technical University of Denmark, Department of Chemical and Biochemical Engineering, 2013