A5010887205- Machine Learning in Materials Science
- Innovative Microfluidic and Catalytic Techniques Innovation
- Crystallization and Solubility Studies
- X-ray Diffraction in Crystallography
- Computational Drug Discovery Methods
- Synthesis and Properties of Aromatic Compounds
- Organic Electronics and Photovoltaics
- Conducting polymers and applications
- Advanced Multi-Objective Optimization Algorithms
- Fullerene Chemistry and Applications
- Scientific Computing and Data Management
- Perovskite Materials and Applications
- Advanced Photocatalysis Techniques
- Process Optimization and Integration
- Modular Robots and Swarm Intelligence
- Advanced Memory and Neural Computing
- Advanced Thermoelectric Materials and Devices
- Analytical Chemistry and Chromatography
- Gene expression and cancer classification
- Various Chemistry Research Topics
- Catalytic Processes in Materials Science
- Crystallography and molecular interactions
- Molecular Sensors and Ion Detection
- Advanced Chemical Physics Studies
- Machine Learning and Algorithms
University of Toronto
2019-2021
Vector Institute
2019-2021
Harvard University
2018-2021
Tianjin University
2017-2021
Canadian Institute for Advanced Research
2018-2020
University of Zurich
2016-2018
Nankai University
2017-2018
University of British Columbia
2018
Harvard University Press
2018
University College London
2016
An autonomous laboratory for thin film discovery is used to optimize the doping and annealing of organic semiconductors.
We report Phoenics, a probabilistic global optimization algorithm identifying the set of conditions an experimental or computational procedure which satisfies desired targets. Phoenics combines ideas from Bayesian with concepts kernel density estimation. As such, allows to tackle typical problems in chemistry for objective evaluations are limited, due either budgeted resources time-consuming conditions, including experimentation enduring computations. proposes new based on all previous...
Fundamental advances to increase the efficiency as well stability of organic photovoltaics (OPVs) are achieved by designing ternary blends, which represents a clear trend toward multicomponent active layer blends. The development high-throughput and autonomous experimentation methods is reported for effective optimization polymer blends OPVs. A method automated film formation enabling fabrication up 6048 films per day introduced. Equipping this platform with Bayesian optimization,...
Autonomous process optimization involves the human intervention-free exploration of a range parameters to improve responses such as product yield and selectivity. Utilizing off-the-shelf components, we develop closed-loop system for carrying out parallel autonomous experiments in batch. Upon implementation our stereoselective Suzuki-Miyaura coupling, find that definition set meaningful, broad, unbiased is most critical aspect successful optimization. Importantly, discern phosphine ligand,...
Designing functional molecules and advanced materials requires complex design choices: tuning continuous process parameters such as temperatures or flow rates, while simultaneously selecting catalysts solvents. To date, the development of data-driven experiment planning strategies for autonomous experimentation has largely focused on parameters, despite urge to devise efficient selection categorical variables. Here, we introduce Gryffin, a general-purpose optimization framework variables...
Finding the ideal conditions satisfying multiple pre-defined targets simultaneously is a challenging decision-making process, which impacts science, engineering, and economics. Additional complexity arises for tasks involving experimentation or expensive computations, as number of evaluated must be kept low. We propose Chimera general purpose achievement scalarizing function multi-target optimization where evaluations are limiting factor. combines concepts priori with lexicographic...
The current Edisonian approach to discovery requires up two decades of fundamental and applied research for materials technologies reach the market. Such a slow capital-intensive turnaround calls disruptive strategies expedite innovation. Self-driving laboratories have potential provide means revolutionize experimentation by empowering automation with artificial intelligence enable autonomous discovery. However, lack adequate software solutions significantly impedes development self-driving...
Abstract Research challenges encountered across science, engineering, and economics can frequently be formulated as optimization tasks. In chemistry materials recent growth in laboratory digitization automation has sparked interest optimization-guided autonomous discovery closed-loop experimentation. Experiment planning strategies based on off-the-shelf algorithms employed fully research platforms to achieve desired experimentation goals with the minimum number of trials. However, experiment...
The screening of molecular targets benefits from design criteria, which can identify the most promising candidates. We demonstrate that π-depleted polyaromatic molecules present superior π-stacking ability. This realization is quantified using a computational criterion, LOLIPOP, detects ideal π-conjugated frameworks. utility LOLIPOP illustrated by identifying tailored chemosensors.
Abstract Solubility is a ubiquitous phenomenon in many aspects of material science. While solubility can be determined by considering the cohesive forces liquid via Hansen parameters (HSP), quantitative structure–property relationship models are often used for prediction, notably due to their low computational cost. Here, gpHSP, an interpretable and versatile probabilistic approach determining HSP, reported. Our model based on Gaussian processes, Bayesian machine learning that provides...
Abstract A multipurpose interconnection layer based on poly(3,4‐ethylenedioxythiophene) doped with poly(styrene sulfonate) (PEDOT:PSS), and d ‐sorbitol for monolithic perovskite/silicon tandem solar cells is introduced. The of independently processed silicon perovskite subcells a simple add‐on lamination step, alleviating common fabrication complexities devices. It demonstrated experimentally theoretically that PEDOT:PSS an ideal building block manipulating the mechanical electrical...
Abstract Pentaindenocorannulene (C 50 H 20 , 1 ), a deep bowl polynuclear aromatic hydrocarbon, accepts 4 electrons, crystallizes in columnar bowl‐in‐bowl assemblies and forms nested C 60 @ 2 complex. Spectra, structures computations are presented.
Clay minerals are ubiquitous in nature, and the manner which they interact with their surroundings has important industrial environmental implications. Consequently, a molecular-level understanding of adsorption molecules on clay surfaces is crucial. In this regard computer simulations play an role, yet accuracy widely used empirical force fields (FF) density functional theory (DFT) exchange-correlation functionals often unclear systems dominated by weak interactions. Herein we present...
Abstract Pentaindenocorannulene (C 50 H 20 , 1 ), a deep bowl polynuclear aromatic hydrocarbon, accepts 4 electrons, crystallizes in columnar bowl‐in‐bowl assemblies and forms nested C 60 @ 2 complex. Spectra, structures computations are presented.
Accelerating R&D is essential to address some of the challenges humanity currently facing, such as achieving global sustainability goals. Today’s Edisonian approach trial-and-error still prevalent in labs takes up two decades fundamental and applied research for new materials reach market. Turning around this situation calls strategies upgrade expedite innovation. By conducting smart experiment planning that data-driven guided by AI/ML, researchers can more efficiently search through...
SeMOpt uses meta-/few-shot learning to enable knowledge transfer from previous experiments accelerate Bayesian optimization of chemical reactions.
Ocular symptoms developed in four out of 58 patients during clomiphene citrate administration. Ophthalmologic examination revealed scotomata development which disappeared after drug was discontinued one subject. It recurred with readministration but not a placebo. The ocular three subjects were reversed by discontinuance citrate. while the still being administered to fourth patient.
Regeneration of dye sensitizer molecules by reducing species contained in the electrolyte is a key mechanism liquid dye-sensitized solar cells because it competes kinetically with detrimental charge recombination process. Kinetics reduction iodide ions oxidized states (S+) two RuII complex dyes and four organic π-conjugated bridged donor–acceptor sensitizers were examined as function concentration. Results show that different cases can be distinguished. A sublinear behavior regeneration rate...
A general optimization procedure towards the development and implementation of a new family minimal parameter spin-component-scaled double-hybrid (mSD) density functional theory (DFT) is presented. The nature proposed exchange-correlation establishes methodology with empiricism. This (DH) functionals demonstrated using PBEPBE functional, illustrating to mSD-PBEPBE method, performance characteristics shown for set non-covalent complexes covering broad regime weak interactions. With only two...
Optimization strategies based on machine learning (ML), such as Bayesian optimization, show promise across the experimental sciences a superior alternative to traditional design of experiment. Deploying ML optimization tools in R\&D operations increases productivity and efficiency, while reducing time cost necessary identify new molecules, materials, process parameters with desired target properties. Additional benefits can be captured when combining these algorithms automated laboratory...