Embracing data science in catalysis research

Nature Catalysis, Journal Year: 2024, Volume and Issue: 7(6), P. 624 - 635

Published: April 23, 2024

Language: Английский

---

Integrating artificial intelligence in energy transition: A comprehensive review

Energy Strategy Reviews, Journal Year: 2025, Volume and Issue: 57, P. 101600 - 101600

Published: Jan. 1, 2025

Language: Английский

---

Artificial intelligence in drug development

Nature Medicine, Journal Year: 2025, Volume and Issue: 31(1), P. 45 - 59

Published: Jan. 1, 2025

Language: Английский

---

Autonomous mobile robots for exploratory synthetic chemistry

Nature, Journal Year: 2024, Volume and Issue: 635(8040), P. 890 - 897

Published: Nov. 6, 2024

Abstract Autonomous laboratories can accelerate discoveries in chemical synthesis, but this requires automated measurements coupled with reliable decision-making 1,2 . Most autonomous involve bespoke equipment 3–6 , and reaction outcomes are often assessed using a single, hard-wired characterization technique 7 Any algorithms 8 must then operate narrow range of data 9,10 By contrast, manual experiments tend to draw on wider instruments characterize products, decisions rarely taken based one measurement alone. Here we show that synthesis laboratory be integrated into an by mobile robots 11–13 make human-like way. Our modular workflow combines robots, platform, liquid chromatography–mass spectrometer benchtop nuclear magnetic resonance spectrometer. This allows share existing human researchers without monopolizing it or requiring extensive redesign. A heuristic decision-maker processes the orthogonal data, selecting successful reactions take forward automatically checking reproducibility any screening hits. We exemplify approach three areas structural diversification chemistry, supramolecular host–guest chemistry photochemical synthesis. strategy is particularly suited exploratory yield multiple potential as for assemblies, where also extend method function assay evaluating binding properties.

Language: Английский

---

A dynamic knowledge graph approach to distributed self-driving laboratories

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: Jan. 23, 2024

Abstract The ability to integrate resources and share knowledge across organisations empowers scientists expedite the scientific discovery process. This is especially crucial in addressing emerging global challenges that require solutions. In this work, we develop an architecture for distributed self-driving laboratories within World Avatar project, which seeks create all-encompassing digital twin based on a dynamic graph. We employ ontologies capture data material flows design-make-test-analyse cycles, utilising autonomous agents as executable components carry out experimentation workflow. Data provenance recorded ensure its findability, accessibility, interoperability, reusability. demonstrate practical application of our framework by linking two robots Cambridge Singapore collaborative closed-loop optimisation pharmaceutically-relevant aldol condensation reaction real-time. graph autonomously evolves toward scientist’s research goals, with effectively generating Pareto front cost-yield three days.

Language: Английский

---

An integrated high-throughput robotic platform and active learning approach for accelerated discovery of optimal electrolyte formulations

Nature Communications, Journal Year: 2024, Volume and Issue: 15(1)

Published: March 29, 2024

Abstract Solubility of redox-active molecules is an important determining factor the energy density in redox flow batteries. However, advancement electrolyte materials discovery has been constrained by absence extensive experimental solubility datasets, which are crucial for leveraging data-driven methodologies. In this study, we design and investigate a highly automated workflow that synergizes high-throughput experimentation platform with state-of-the-art active learning algorithm to significantly enhance organic solvents. Our identifies multiple solvents achieve remarkable threshold exceeding 6.20 M archetype molecule, 2,1,3-benzothiadiazole, from comprehensive library more than 2000 potential Significantly, our integrated strategy necessitates assessments fewer 10% these candidates, underscoring efficiency approach. results also show binary solvent mixtures, particularly those incorporating 1,4-dioxane, instrumental boosting 2,1,3-benzothiadiazole. Beyond designing efficient developing high-performance batteries, machine learning-guided robotic presents robust general approach expedited functional materials.

Language: Английский

---

Materials genome engineering accelerates the research and development of organic and perovskite photovoltaics

Materials Genome Engineering Advances, Journal Year: 2024, Volume and Issue: 2(1)

Published: March 1, 2024

Abstract The emerging photovoltaic (PV) technologies, such as organic and perovskite PVs, have the characteristics of complex compositions processing, resulting in a large multidimensional parameter space for development optimization technologies. Traditional manual methods are time‐consuming labor‐intensive screening optimizing material properties. Materials genome engineering (MGE) advances an innovative approach that combines efficient experimentation, big database artificial intelligence (AI) algorithms to accelerate materials research development. High‐throughput (HT) platforms perform experimental tasks rapidly, providing amount reliable consistent data creation databases. Therefore, novel combining HT AI can design application, which is beneficial establishing material‐processing‐property relationships overcoming bottlenecks PV This review introduces key technologies involved MGE overviews accelerating role field PVs.

Language: Английский

---

Navigating phase diagram complexity to guide robotic inorganic materials synthesis

Nature Synthesis, Journal Year: 2024, Volume and Issue: 3(5), P. 606 - 614

Published: April 9, 2024

Abstract Efficient synthesis recipes are needed to streamline the manufacturing of complex materials and accelerate realization theoretically predicted materials. Often, solid-state multicomponent oxides is impeded by undesired by-product phases, which can kinetically trap reactions in an incomplete non-equilibrium state. Here we report a thermodynamic strategy navigate high-dimensional phase diagrams search precursors that circumvent low-energy, competing by-products, while maximizing reaction energy drive fast transformation kinetics. Using robotic inorganic laboratory, perform large-scale experimental validation our precursor selection principles. For set 35 target quaternary oxides, with chemistries representative intercalation battery cathodes electrolytes, robot performs 224 spanning 27 elements 28 unique precursors, operated 1 human experimentalist. Our frequently yield higher purity than traditional precursors. Robotic laboratories offer exciting platform for data-driven science, from develop fundamental insights guide both chemists.

Language: Английский

---

Autonomous reaction Pareto-front mapping with a self-driving catalysis laboratory

Nature Chemical Engineering, Journal Year: 2024, Volume and Issue: 1(3), P. 240 - 250

Published: Feb. 27, 2024

Ligands play a crucial role in enabling challenging chemical transformations with transition metal-mediated homogeneous catalysts. Despite their undisputed catalysis, discovery and development of ligands have proven to be resource-intensive undertaking. Here, response, we present self-driving catalysis laboratory, Fast-Cat, for autonomous resource-efficient parameter space navigation Pareto-front mapping high-temperature, high-pressure, gas–liquid reactions. Fast-Cat enables ligand benchmarking multi-objective catalyst performance evaluation minimal human intervention. Specifically, utilize perform rapid identification the hydroformylation reaction between syngas (CO H2) olefin (1-octene) presence rhodium various classes phosphorus-based ligands. By reactor benchmarking, demonstrate Fast-Cat's knowledge scalability, essential fine/specialty industries. We report details modular flow chemistry platform its experiment-selection strategy generation optimized experimental conditions in-house data required supplying machine-learning approaches investigations. A is presented efficient high-throughput screening using rhodium-catalyzed as case study. used Pareto map investigate varying several

Language: Английский

---

Advancing perovskite solar cell commercialization: Bridging materials, vacuum deposition, and AI-assisted automation

Next Materials, Journal Year: 2024, Volume and Issue: 3, P. 100103 - 100103

Published: Jan. 9, 2024

Organic-inorganic metal halide perovskites solar cells (PSCs) have been emerging as a counterpart or supplement of silicon-based cells. They shown various interesting optoelectronic properties and impressive power conversion efficiencies, even outperforming the theoretical limits in tandem configurations. However, challenges such long-term stability scalable manufacturing remain significant obstacles to commercialization. Key factors like material composition crystal quality are essential for reliability performance PSCs. Traditional solution-based processes face scalability reproducibility. This has drawn attention vacuum processes, which successfully employed commercial mass production devices displays. Also, recent innovations automated deposition systems aided by machine learning offer promising solutions. These technological advancements enable rapid optimization combinations facilitating transition from lab-scale prototypes industrial applications. review highlights converging efforts multiple disciplines—materials science, process engineering learning—that experimental validation commercial, sustainable energy In summary, work sets path forward, where collective expertize can address lingering challenges, making clean, accessible, affordable an attainable goal.

Language: Английский

---