Dr Dean Thomas FHEA

Autonomous chemical robots have the potential to revolutionise molecular discovery, yet their power also exposes chemistry to new forms of risk. Without intrinsic containment, these systems could execute reactions beyond human oversight, posing hazards both within the laboratory and across the digital landscape. Here we present the Control of Substances Hazardous to Health chemputer (CoSHHputer), a secure autonomous synthesis platform that integrates physical protection, digital encryption, and virtual replication to perform and verify hazardous reactions safely at gram scale

The Turing Machine is the fundamental abstraction behind all modern computers, and any modern computer can simulate one, an equivalence known as Turing completeness. A system with this property can, in theory, perform any algorithmically defined task through discrete unit operations. Here, we extend this concept to robotic platforms that execute physical operations for chemical synthesis via a chemically-aware programming language, XDL

This work introduces an automated chemistry platform that incorporates microwave heating, addressing previous limitations around safety, scalability, and standardisation. The system uses two different microwave modules for temperature control and demonstrates its flexibility by executing a range of chemical reactions. The results highlight the platform’s reliability and versatility, expanding the capabilities of automated synthesis and paving the way for wider use in programmable chemistry

This work introduces a methodology for automated reaction monitoring based on spectral similarity metrics (Jaccard index and cross-correlation) applied to diverse spectroscopic modalities (NMR, UV/Vis, IR, EPR)

The assembly of molecular nanomachines using atomically precise manipulations promises to enable molecular nanotechnology with unprecedented functional properties. However, this vision is limited by the ability to autonomously manufacture nanomachines, with current synthetic efforts being heavily labour intensive. What is needed is a system that can program the assembly of matter under digital control unifying molecular nanotechnology and macroscale chemical processes.

The ChemPU platform streamlines chemistry by automating kinetic measurements, bridging synthesis and analysis. It’s versatile, adaptable, and facilitates precise data encoding of reactions via the Chemical Description Language (XDL). This innovation has the potential to enhance efficiency and create a valuable database of kinetic data for broader applications.

We report the synthesis and operation of a molecular energy ratchet that transports a crown ether from solution onto a thread, along the axle, over a fluorophore, and off the other end of the thread back into bulk solution, all in response to a single pulse of a chemical fuel (CCl3CO2H). The system provides a potential alternative signaling approach for artificial molecular machines that read symbols from sequence-encoded molecular tapes.

Nature has evolved highly efficient biological machines that use chemical fuels to drive directional molecular motion and complex reactions with exceptional precision. Mimicking this control at the molecular level could revolutionise fields such as materials science, surface chemistry, and drug delivery. This thesis presents a series of artificial molecular machines that harness external stimuli to achieve controlled, directional motion for targeted applications.

Polymer beads have been used as the core of magnetic particles for around twenty years. Here we report studies to attach polymetallic complexes to polymer beads for the first time, producing beads of around 115 microns diameter that are attached to 10¹⁴ hybrid inorganic-organic [2]rotaxanes.

The sorption of species from a solution into and onto solids underpins the sequestering of waste and pollutants, precious metal recovery, heterogeneous catalysis, analysis and separation science, and other technologies. Non-equilibrium sorption by immobilised artificial molecular machines enables the transduction of energy from chemical fuels for the use, storage and release of energy and information.

We report on catalysis by a fuel-induced transient state of a synthetic molecular machine. Dissipative catalysis by synthetic molecular machines has implications for the future design of networks that feature communication and signaling between the components

Tetrahydroquinolines feature in molecules with wide ranging properties, including those with biological activity and materials applications. This report demonstrates that widely accessible N-alkyl arylamines can be directly coupled with new classes of alkenes to synthesise tetrahydroquinolines. The reaction is facilitated by a commercially available electron-deficient borane catalyst, and enables the creation of novel tetrahydroquinoline scaffolds.

A fundamental redesign of scientific instrumentation is needed to keep pace with the rapid advances in artificial intelligence (AI), says a coalition of more than 25 prominent researchers and national lab scientists from the U.S., U.K., Canada, and Japan. A call to action published today by Renaissance Philanthropy, “On the Need for Autonomous Science Instruments”, identifies a critical bottleneck in the booming field of autonomous experimentation threatening the future of AI-driven scientific discovery.

The article explores how AI is transforming scientific research, raising the question of whether machine-generated discoveries can be described as "beautiful" in the same way as human work. As AI becomes increasingly central to fields like chemistry, with advancements in retrosynthesis, reaction planning, and Nobel Prize-winning applications, it blurs the line between human and machine creativity. The evolution of AI systems challenges us to rethink the appreciation in science as more discoveries are driven by autonomous processes.

The article explores the pitfalls of the "teach, test, forget" cycle in STEM education, where short-term memorisation often outweighs long-term understanding. It calls for a shift towards more interactive, experiential learning and smaller class sizes to promote critical thinking and knowledge retention

The article shares the story behind the 2022 paper on artificial molecular pumps, exploring the development process and challenges faced. These pumps mimic biological motors, using chemical fuel to drive cargo between phases, offering potential applications in environmental and material sciences. The article highlights the design evolution and the significance of these molecular machines in future technologies.

The article reflects on the challenges faced by researchers during the COVID-19 lockdowns, where limited lab time disrupted crucial work. It stresses the importance of accepting that lost time cannot be fully recovered and cautions against overworking in an attempt to make up for it. Instead, advocating for a healthier work-life balance, encouraging researchers to reset expectations and focus on sustainable progress.