Supervisors: Dr James Bull, Dr Phil Miller, Prof Alan Spivey

This project will develop a heterocycle ‘vending machine’ that will enable the diversity-oriented-synthesis of heterocycles in a self-optimised fashion. By combining recent advances in light-activated fundamental organic transformations, flow chemistry technology and self-optimisation algorithms, a telescoped multi-step process will be developed for the synthesis of novel heterocyclic compounds. A variety of simple precursors will be combined through a sequence of formation and reaction of diazo compounds and cyclisation. It is envisioned that simple commercially available building blocks will be fed into the machine and reacted in a self-optimised process to produce high value heterocyclic compounds with potential applications in the pharmaceutical sector. This approach, will deliver varied substituted heterocycles on demand, avoiding many repetitive steps by being directly applicable across different substrates. Compounds will be targeted to be appropriate as fragments or lead-like compounds for screening against biological targets. The integration of in-line optical detection in the flow process will facilitate the generation of real-time continuous data that will be fed into a self-optimising algorithm enabling reaction parameters (light, reaction time/flow rate, heat etc.) to be changed ‘on the fly’ in order to map out the optimum reaction conditions for a specific reaction.

Supervisors: Dr Ceri Hammond, Dr Antonio Del Rio Chanona

Quickly transitioning to net zero manufacturing requires researchers to use modern methods for process discovery and development. This is especially true for the development of heterogeneously catalysed processes, which are still typically optimised by painstaking trial-and-error methods. As such, the overall objective of this project is to combine pioneering spectroscopy and reactor technology with machine learning and real-time optimisation to achieve self-optimising reactor technology applicable to the synthesis of important chemical products from biomass feedstock using heterogeneous catalysts.

We recently pioneered development of a novel heterogeneous catalyst reactor equipped with fiber optic technology. This reactor allowed us to follow the conversion of glucose to various bio-based chemicals of industrial interest using heterogeneous zeolite catalysts, at true continuous operational conditions and in real time, by performing operando optical spectroscopy. In particular, unique optical signals related to the active sites of the catalyst and all of the reaction pathways of the chemical process were identified, including the desired reaction pathway and those of undesired side reactions. Moreover, we could relate the intensity of each optical signal to the quantity of products formed, as verified by offline methods (HPLC and 1H-13C HSQC NMR), resulting in real time information of the performance of each catalyst in terms of activity, selectivity, and stability. As the reaction pathway signals are charge transfer bands associated with activation of the substrate by the catalyst (as opposed to chromophores of reactants and products), they also provide direct insight into the transition states of the various reaction pathways.

These breakthroughs are unprecedented in catalysis research, and represent the starting point for this project.This project will use this breakthrough to target the application of machine learning to heterogeneous catalysis and biomass conversion. In particular, we will combine fast data generation through operando optical technology with online learning algorithms to construct data-driven models in real-time through time series modelling. This will facilitate the development of ML algorithms that uniquely account for activity, selectivity and stability, at various operational conditions. We will validate this hypothesis, and then use the Intelligent Optical Reactor (IOR) technology to accelerate bio-based chemical manufacture by developing self-optimising reactor technology. In later stages, we will attempt to use the transition state insights provided by the reactor to generate advanced structureactivity relationships of relevance to future catalyst design.

Supervisors: Dr Yuval Elani, Dr James Wilton-Ely

The synthesis of nanoparticles is of major interest to a wide range of industries, including the pharmaceutical, fine chemicals, vaccine, chemical, food, biomedical and personal care sectors. Current approaches rely on one-by-one synthesis of different particle types based on rational design principles. In this project, we will establish a new paradigm based on high-throughput particle manufacture and discovery. We propose a pioneering microfluidic technology that will enable the automatic generation of vast nanoparticle libraries. This will allow us to synthesise hundreds of thousands of distinct particle types, with varied morphological, compositional and physico-chemical features. In doing so, we will lay the foundations of a new research area based on high-throughput particle discovery, and offer a drastic departure from existing approaches. We will use gold nanoparticles as a testbed to validate our system due to their tremendous potential in applications as varied as photothermal agents, delivery vehicles and platforms for catalysis, sensing and imaging. Building particle libraries and seeing how composition is related to architecture and particle properties will allow us to build a rule book which explains how composition, form, and function are intertwined.

Supervisors: Dr Rob Davies, Prof Jason Hallett

Enzymes are naturally occurring catalysts which possess remarkably high activity and exquisite selectivity. As such they hold much promise as a new generation of industrial catalysts. However, the fragility of enzymes especially at higher temperatures and in non-native solvents can pose a key stumbling block for their successful deployment. In this project enzymes will be immobilized within a matrix (a Metal Organic Framework or MOF), leading to enhanced resistance against high temperatures and solvents, whilst still permitting easy substrate access to the enzyme active site and maintaining high activity. Novel MOF-enzyme constructs will be designed, prepared and characterized before their full evaluation (using high throughput methodology and kinetic profiling) in a number of industrially important processes key for future transition to a bio-based economy.

Supervisors: Dr Qilei Song, Dr Kim Jelfs, Prof Anthony Kucernak

This project will accelerate the materials discovery by computational design, and materials analysis by automated high-throughput reaction platforms

With the rapid development of renewable energy, there is an urgent need for low-cost energy storage technologies. Conventional battery materials are reaching their capacity limit and there are growing concerns about sustainability. Organic redox-active materials have emerged as a new generation of sustainable materials for energy storage in redox flow batteries or Lithium/sodium ion batteries.  Organic electroactive molecules such as quinones, nitroxide radical derivatives, phenazine derivatives show great promise for batteries. However, there are still some scientific challenges, such as solubility, redox reaction stability and air stability, which remain to be improved. The current synthesis approach is still mainly based on trial and error.  Computational high-throughput screening strategies can provide guidance to accelerate the search for redox materials suitable for high-energy-power and low-cost battery systems.

In this proposal, we will combine computational and experimental approach to accelerate the design and synthesis of organic redox molecules for battery applications. In the past decade, a large quantity of data have been generated on organic electrode materials in Li/Na-ion batteries and organic redox flow batteries. The library of  molecules with varied structure and functional groups reported in the literature will be analyzed and key structural features of the molecules and properties will be input into the library to establish the structure-property relationships.  Computational modelling is critically important for advanced the development of organic electrode materials, as it can be used to investigate the intrinsic properties of molecular compounds and corresponding interactions in the electrolyte. Then high-throughput computational screening and machine learning will be applied to accelerate the materials discovery, by studying the thousands of molecules in the chemical database and predict optimized structures and molecules with suitable properties.  Building on the modelling and prediction, automated synthesis and modification of molecules will be performed. The molecules will then be rapidly screened, such as electrochemical reaction properties, kinetics, solubility, and degradation. The data generated from the synthesis and reactions will be used to validate the computational predictions. Finally, the selected molecules will be tested in multichannel battery testing facility and data generated will be used to further guide the automation and optimization of the molecules.

This project will accelerate the materials discovery by computational design, and materials analysis by automated high-throughput reaction platforms, and data analysis, and train chemistry student to develop relevant skills.

Supervisors: Dr Aron Walsh, Prof James Durrant, Dr Reshma Rao

Water oxidation is the step limiting the efficiency of electrocatalytic hydrogen production from water.  The development of sustainable catalysts is limited by the lack of understanding on the catalytic cycle.

Transient Absorption Spectroscopy (TAS) is a powerful tool to characterise the photophysical reactions of semiconducting materials. It measures the change in optical absorption of a material following a perturbation from the ground-state. While it is widely used for solar energy conversion, it is a new probe for electrochemical reactions.  The spectra are challenging to interpret, and it can be difficult to separate intrinsic (bulk behaviour) from the extrinsic (defects, dopants and surfaces) effects.

This project will develop a toolkit for (i) the theoretical prediction of spectroelectrochemical signatures from quantum chemistry and (ii) advanced automated analysis of large experimental datasets that the student will be trained to collect. The project will involve reaction technology engineering over a multidimensional configurational space including materials composition, temperature, and electrochemical conditions. These developments will enable the extend the TAS method from qualitative kinetic analysis to quantitative insights into the in situ operation of state-of-the-art electrochemical systems for clean energy applications.

Supervisors: Prof George Britovsek, Dr Maxie Roessler, Dr Mark Chadwick

The oligomerization of ethylene to higher alpha olefins is an essential part of chemical manufacturing. This route to value-added longer chain olefins and hydrocarbons from ethylene, potentially bioderived through ethanol dehydration, will only become more important as we move away from fossil fuels. Chromium-based catalysts are some of the most established for this transformation, but these have two issues. Firstly, the catalysts lack selectivity, and indeed sometimes polymerise the olefin. Secondly, the active catalyst species are highly challenging to investigate, due to the paramagnetic chromium centre. In this project new ligand systems, based on the well-known PNP chromium catalysts will be developed for this important transformation. To understand how the catalyst needs to be tuned, advanced electron paramagnetic resonance spectroscopic techniques will be used giving a hitherto unprecedented understanding of the active species’ structure in catalysis. This data will establish structure/activity relationships and allow for the design of a selective, highly-active, catalyst for olefin oligomerization.

Supervisors: Dr Charles Romain, Prof Kim Jelfs

New technologies are urgently needed to develop truly sustainable materials, i.e. made from renewables, featuring useful properties while having end-of-life options (e.g. chemical recycling) that will contribute towards building our circular economy. Identifying and making such sustainable materials is challenging and it requires innovative strategies to access such “advanced” polymers. Whereas computational strategies to predict (co)polymers properties (e.g. glass transition temperature, T_g) are rapidly developing, the prediction of polymer “degradation” remains underexplored.

In this project, a combined experimental-computational approach making use of machine learning will be used to predict the chemical recyclability of polymers via estimation of thermodynamic parameters and by exploiting the ring-opening polymerisation (ROP) of cyclic monomers (e.g. lactones, cyclic carbonates, etc…). Ultimately, this project will contribute to guide towards the design of new monomers to enable polymer chemical recycling.

Supervisors: Prof Matthew Fuchter, Prof Nick Long, Dr Becky Greenaway

Photoswitches are molecules that are capable of being reversibly interconverted between (at least) two states by light irradiation. The incorporation of photoswitches into multifunctional systems has huge relevance for next-generation light-addressable applications that span catalysis, biology and materials science. The Fuchter group have pioneered the development of heteroaromatic azo switches (J. Am. Chem. Soc. 2014, 136, 11878; J. Am. Chem. Soc. 2017, 139, 1261) and so far employed them in photopharmacology (J. Am. Chem. Soc. 2020, 142, 17457) and energy storage (J. Am. Chem. Soc. 2020, 142, 8688) applications. Despite the promise of existing switches however, there remains a number of challenges to be solved in the area of switch design – from completeness of switching to red-light switching, which are often determined by the application in question and therefore critical to address. It is non-intuitive to identify a new switch structure that meets all the performance requirements for a given application, with many established photoswitches falling short of what is needed. Coupled to this, there is a potentially huge design space to navigate when it comes to homing in on improved azo switch designs. This project will harness the power of both computational and experimental high throughout discovery to develop tailored high performance heteroaromatic photoswitches. We will then seek to showcase the switches developed in two light-addressable applications: lactide polymerisation catalysis and incorporation into switchable porous materials. This will demonstrate the power of high throughput approaches in this exciting area of research.

Supervisors: Dr Maxie Roessler, Dr Alex Ganose, Prof Klaus Hellgardt, Dr David Sale (Syngenta)

Film-electrochemical electron paramagnetic resonance spectroscopy (FE-EPR) is a transformative new experimental technique that enables detailed insight into catalytic reactions by detecting and interrogating unpaired electrons. In this project, we will (1) develop and design software and hence optimally harness information from FE-EPR, (2) engineer a user-friendly and widely applicable set-up, (3) design and screen novel working electrode materials and (4) apply FE-EPR to industrially relevant catalytic reactions, focussing on TEMPO-catalysed alcohol oxidation, Cu-catalysed radical rearrangements and organic radical intermediate formation, stability and characterisation.

Supervisors: Dr Ali Salehi-Reyhani, Prof Klaus Hellgardt

There is a need to develop new experimental methods to better monitor, understand and optimize reactions and chemical systems, and to do so rapidly and scalably. The aim of this project is to develop a novel microfluidic platform that is able to accelerate the exploration and mapping of high dimensional chemical space for the optimization of reaction parameters. Traditionally, design of experiments is used to minimize the total number of experiments needed to find optima within any given reaction space. Liquid handling robots aided by online analytics are able to explore chemical space guided by ever sophisticated algorithms. Artificial intelligence based models show promise but have arisen in part due to our limited ability to efficiently map high dimensional chemical space at high resolution within reasonable timescales. Nevertheless, training datasets upon which machine learning relies are typically reported only for optimal synthesis reactions. This inevitably leads to inherent bias in machine-guided predictions. Therefore, robust datasets that span the entirety of a chemical space are necessary. Ultimately then, searching for new molecules is an inherently practical endeavour. Autonomous robotic platforms have been developed to help the modern chemist but tend to be batch reaction systems. They are intrinsically reliant on their starting conditions and this limits their capability to explore novel regions of parameter space. Flow based systems and online methods of reaction monitoring can overcome this and are capable of self guided exploration. However, they are fundamentally limited as single coordinate, point-based sampling approaches thus rely heavily on competent search strategies to efficiently explore parameter space. This limits flow chemistry to simple or low dimensional reaction schemes. The miniaturization of chemical reactors through microfluidic technology offers a promising opportunity for chemical synthesis. Low instantanteous volumes mean that reaction conditions can be screened rapidly in comparison with their robotic or conventional flow based counterparts. Yet, despite their capability, have not extended chemical synthesis in any fundamental way. In this project, we propose to develop a novel approach to chemical space mapping by spatially encoding planes of chemical parameter space onto the spatial plane of a microfluidic reaction chamber. By doing so billions of reaction conditions can be simultaneously established in parallel. It could exponentially speed up chemical data acquisition and pave the way to full and autonomous reaction space mapping on-chip and ultra efficient reaction optimization strategies.

Supervisors: Dr Claire Adjiman, Prof Galindo Amparo, Prof Alan Armstrong

In liquid phase reactions and other liquid phase processes such as crystallization or liquid-liquid separation, solvation properties play a critical role as the choice of solvent can have an impact on reaction rate, affecting free energy barriers as well as the maximum concentration (solubility) that can be achieved, on regioselectivity or enantiomeric excess, and on partition coefficients, amongst many other factors. During the development of pharmaceutical or agrochemical processes, the identification of an appropriate solvent or solvent mixture can be a painstaking process and can require significant amounts of material to perform experiments in different media and at different conditions. The aim of this project is to effect a substantial reduction in the amount of material needed and in the time required to develop predictive models of solvation applicable to a particular substrate or reaction. We will achieve this by developing a novel model-based methodology to design high-throughput experiments (e.g., in a few solvents at different temperatures) with maximum information content so that the model can be developed successfully. We will apply this methodology to the prediction of solubility and selectivity. We will verify the predictive capabilities of the proposed models by conducting experimental campaigns in a larger set of solvents and we will refine the methodology on this basis.

Supervisors: Dr Chris Rowlands, Prof Chris Braddock

This project combines a new automated microfluidic reaction platform with relay cross metathesis (‘ReXM’), a cutting-edge but sensitive and unoptimized synthetic route to biochemically-useful terpenoids, which are a class of compounds accounting for ~60% of all known natural products. The automated reaction platform consists of a computer-controlled valve and heater array, and by actuating these, the chip can be reconfigured to mimic any other microreactor, allowing the reaction conditions (temperature, reaction time, stoichiometry, etc.) to be altered as desired. In this project, the chip will be used to automatically optimize ReXM, producing a powerful method for
producing arbitrary turpenoids. Ultimately this combination of automated optimization and synthetic utility will yield
a ‘dial-a-terpenoid’ system, where the desired reaction is optimized on-the-fly to yield the desired product.

Supervisors: Prof Klaus Hellgardt, Prof Mimi Hii

As we switch from petrochemical (mainly hydrocarbons) to bio-renewable (mainly oxygenated) feedstocks, safer and efficient methods to convert C-O bonds to C-H bonds will be an important part of the strategy towards a more sustainable future. While a number of electrochemical reactors have become commercially available in recent years, they are generally built for synthesis/reaction discovery, and are not designed for obtaining reaction data and understanding. In a previous project, we have constructed an electrochemical flow reactor that can be used to reduced amide bonds - one of the most stable C=O bonds found in nature.

In this project, we will aim to extend this to the reduction of other types of C=O compounds (aldehydes, ketones, acids and esters) using a data-guided approach.

Supervisors: Prof Klaus Hellgardt, Dr Antonio Del Rio Chanona, Dr Bernd Schaefer

Photochemistry is capable of disruptively changing value and supply chains by considering alternative, preferentially renewable and green feedstock. Taking these feedstocks rather than oil and natural gas will change the chemical industry as we know it and as it has developed since the onset of industrialization:

Commodity chemicals are produced on large scale in multiple steps often starting from steam cracking. In many cases supply chains are integrated on a single site (“Verbund”-site) for improved heat integration logistics, and safety in pipes. Therefore, changing the initial feedstock from natural gas to bio-based methanol for example will have major implications on the supply and value chains. While this change is widely accepted on our way to carbon neutrality, the technical and economical impact on the chemical industry needs to be considered to maintain commercial viability and competitiveness. Photochemistry is a prime example of technology that has the potential to utilize biogenic feedstock for chemical conversions.

In this project we plan to pave the way toward implementing photochemistry in the large scale chemical industry by investigating the technoeconomical aspects and the changes of value and supply chains. We will consider changing boundary conditions, including the shift to renewable feedstock and CO2-taxation. We will pick up examples investigated within BASF’s photochemistry cluster at Imperial, including the fabrication of di-acids with CO2 as a reagent to be used in the polyester and polyamide fabrication and the conversion of methanol to mono-ethyleneglycol (MEG). The project will be holistic along the R&D workflow and along the value and supply chains and aim at benchmarking and optimizing future processing options.

This project will require an interest in holistic challenges of chemical engineering. It will consider machine learning algorithms and theoretical assessments rather than experimental work.

Applications are invited for the above project, and the successful applicant will undertake the EPSRC CDT (rEaCt) programme, as part of Cohort 5 (Intake 2023).

Please note that the BASF studentship projects follow the same programme as the normal CDT studentships. The difference is that project funding comes from BASF and not EPSRC.

The project is one of a number of BASF funded projects in the area of photochemistry within the CDT, which, in turn, forms part of a wider suite of activities that BASF is supporting within the CDT, all of which follow the principles detailed below.

Objective:  BASF, the world’s leading chemical company, wishes to explore the application of flow chemistry in its R&D workflow for the synthesis of novel agrochemicals, and commodities. In this project, it will assemble a multidisciplinary team of scientists and engineers to explore the value of synergies between a variety of cutting-edge technologies, to develop more effective and efficient chemical production processes.

Setup:  This project will bring together a group of highly motivated graduate students to work on two representative challenges from industrial process development, where photochemistry in combination with flow chemistry methodologies will be integrated holistically with in-line screening and analytics, and chemical process engineering complemented by machine learning and modelling. The members of this interdisciplinary project team will be exposed to a broad range of competencies and interact closely with each other in working on a disruptive change of the R&D workflow in a large chemical enterprise. 

Training:  Industrial R&D scientists based at BASF’s site in Ludwigshafen, the largest chemical “Verbund”site in Europe, will be closely involved in the programme, combining a real-world industrial experience with cutting-edge academic research. Each of the students will be advised by an internationally renowned professor and a BASF scientist and will be part of a centre for doctoral training, bringing together fellow graduate students in closely related fields of research and building a deep understanding of the area of specialization. 

This project forms part of a wider suite of activities that BASF is supporting within the CDT. The successful candidate will be able to interact with, and leverage the benefits of, this wider activity. There will be opportunities for placement(s) (total of up to one year duration) within the BASF organisation during the studentship period.