Molecular Machine Learning for Chemical Catalysis: Prospects and Challenges

Accounts of Chemical Research, Journal Year: 2023, Volume and Issue: 56(3), P. 402 - 412

Published: Jan. 30, 2023

ConspectusIn the domain of reaction development, one aims to obtain higher efficacies as measured in terms yield and/or selectivities. During empirical cycles, an admixture outcomes from low high yields/selectivities is expected. While it not easy identify all factors that might impact efficiency, complex and nonlinear dependence on nature reactants, catalysts, solvents, etc. quite likely. Developmental stages newer reactions would typically offer a few hundreds samples with variations participating molecules conditions. These "observations" their "output" can be harnessed valuable labeled data for developing molecular machine learning (ML) models. Once robust ML model built specific under predict outcome any new choice substrates/catalyst seconds/minutes thus expedite identification promising candidates experimental validation. Recent years have witnessed impressive applications world, most them aimed at predicting important chemical or biological properties. We believe integration effective workflows made richly beneficial discovery.As technology, direct adaptation used well-developed domains, such natural language processing (NLP) image recognition, unlikely succeed discovery. Some challenges stem ineffective featurization space, unavailability quality its distribution, making right technically deployment. It shall noted there no universal suitable inherently high-dimensional problem reactions. Given these backgrounds, rendering tools conducive exciting well challenging endeavor same time. With increased availability efficient algorithms, we focused tapping potential small-data discovery (a thousands samples).In this Account, describe both feature engineering approaches applied diverse contemporary interest. Among these, catalytic asymmetric hydrogenation imines/alkenes, β-C(sp3)–H bond functionalization, relay Heck employed approach using quantum-chemically derived physical organic descriptors features─all designed enantioselectivity. The selection features customize interest described, along emphasizing insights could gathered through use features. Feature methods Buchwald–Hartwig cross-coupling, deoxyfluorination alcohols, enantioselectivity N,S-acetal formation are found excellent predictions. propose transfer protocol, wherein trained large number (105–106) fine-tuned library target task reactions, alternative (102–103 reactions). exploitation deep neural network latent space method generative tasks useful substrates demonstrated strategy.

Language: Английский

---

Exploring the Structural, Dynamic, and Functional Properties of Metal‐Organic Frameworks through Molecular Modeling

Advanced Functional Materials, Journal Year: 2023, Volume and Issue: 34(43)

Published: Oct. 17, 2023

Abstract This review spotlights the role of atomic‐level modeling in research on metal‐organic frameworks (MOFs), especially key methodologies density functional theory (DFT), Monte Carlo (MC) simulations, and molecular dynamics (MD) simulations. The discussion focuses how periodic cluster‐based DFT calculations can provide novel insights into MOF properties, with a focus predicting structural transformations, understanding thermodynamic properties catalysis, providing information or that are fed classical simulations such as force field parameters partial charges. Classical simulation methods, highlighting selection, databases MOFs for high‐throughput screening, synergistic nature MC MD described. By equilibrium dynamic these methods offer wide perspective behavior mechanisms. Additionally, incorporation machine learning (ML) techniques quantum is discussed. These enhance accuracy, expedite setup, reduce computational costs, well predict parameters, optimize geometries, estimate stability. charting growth promise field, aim to recommendations facilitate more broadly research.

Language: Английский

---

Computational Evolution Of New Catalysts For The Morita–Baylis–Hillman Reaction**

Angewandte Chemie International Edition, Journal Year: 2023, Volume and Issue: 62(18)

Published: Feb. 14, 2023

We present a de novo discovery of an efficient catalyst the Morita-Baylis-Hillman (MBH) reaction by searching chemical space for molecules that lower estimated barrier rate-determining step using genetic algorithm (GA) starting from randomly selected tertiary amines. identify 435 candidates, virtually all which contain azetidine N as catalytically active site, is discovered GA. Two are further study based on their predicted synthetic accessibility and have barriers than known catalyst. Azetidines not been used catalysts MBH reaction. One suggested successfully synthesized showed eightfold increase in activity over commonly believe this first experimentally verified generative model.

Language: Английский

---

Computational Discovery of Stable Metal–Organic Frameworks for Methane-to-Methanol Catalysis

Journal of the American Chemical Society, Journal Year: 2023, Volume and Issue: 145(26), P. 14365 - 14378

Published: June 20, 2023

The challenge of direct partial oxidation methane to methanol has motivated the targeted search metal–organic frameworks (MOFs) as a promising class materials for this transformation because their site-isolated metals with tunable ligand environments. Thousands MOFs have been synthesized, yet relatively few screened promise in conversion. We developed high-throughput virtual screening workflow that identifies from diverse space experimental not studied catalysis, are thermally stable, synthesizable, and unsaturated metal sites C–H activation via terminal metal-oxo species. carried out density functional theory calculations radical rebound mechanism methane-to-methanol conversion on models secondary building units (SBUs) 87 selected MOFs. While we showed oxo formation favorability decreases increasing 3d filling, consistent prior work, previously observed scaling relations between hydrogen atom transfer (HAT) disrupted by greater diversity our MOF set. Accordingly, focused Mn MOFs, which favor intermediates without disfavoring HAT or leading high release energies─a key feature hydroxylation activity. identified three comprising centers bound weak-field carboxylate ligands planar bent geometries kinetics thermodynamics. energetic spans these indicative turnover frequencies warrant further catalytic studies.

Language: Английский

---

Directional multiobjective optimization of metal complexes at the billion-system scale

Nature Computational Science, Journal Year: 2024, Volume and Issue: 4(4), P. 263 - 273

Published: March 29, 2024

Language: Английский

---

LigandDiff: de Novo Ligand Design for 3D Transition Metal Complexes with Diffusion Models

Journal of Chemical Theory and Computation, Journal Year: 2024, Volume and Issue: 20(10), P. 4377 - 4384

Published: May 14, 2024

Transition metal complexes are a class of compounds with varied and versatile properties, making them great technological importance. Their applications cover wide range fields, either as metallodrugs in medicine or materials, catalysts, batteries, solar cells,

Language: Английский

---

Automated Transition Metal Catalysts Discovery and Optimisation with AI and Machine Learning

ChemCatChem, Journal Year: 2024, Volume and Issue: 16(10)

Published: Jan. 5, 2024

Abstract Significant progress has been made in recent years the use of AI and Machine Learning (ML) for catalyst discovery optimisation. The effectiveness ML data science techniques was demonstrated predicting optimising enantioselectivity regioselectivity catalytic reactions through optimisation ligands, counterions reaction conditions. Direct new catalysts/reactions is more difficult requires efficient exploration transition metal chemical space. A range computational descriptor generation, ranging from molecular mechanics to DFT methods, have successfully demonstrated, often conjunction with reduce cost associated TS calculations. Complex aspects reactions, such as solvent, temperature, etc., also incorporated into workflow.

Language: Английский

---

Using Computational Chemistry To Reveal Nature’s Blueprints for Single-Site Catalysis of C–H Activation

ACS Catalysis, Journal Year: 2022, Volume and Issue: 12(15), P. 9281 - 9306

Published: July 15, 2022

The challenge of activating inert C–H bonds motivates a study catalysts that draws from what can be accomplished by natural enzymes and translates these advantageous features into transition-metal complex (TMC) material mimics. Inert bond activation reactivity has been observed in diverse number predominantly iron-containing the heme-P450s to nonheme iron α-ketoglutarate-dependent methane monooxygenases. Computational studies have played key role correlating active-site variables, such as primary coordination sphere, oxidation state, spin reactivity. TMCs, zeolites, metal–organic frameworks (MOFs), single-atom (SACs) are synthetic inorganic materials designed incorporate Fe active sites analogy single enzymes. In systems, computational essential supporting spectroscopic assignments quantifying effects metal-local environment on High-throughput virtual screening tools widely used for bulk metal catalysis do not readily extend single-site where metal–ligand bonding localized d-electrons govern reaction energetics. These also necessitate wave function theory calculations when density functional (DFT) is sufficiently accurate. Where sufficient or experimental data gathered, machine learning helped uncover more general design rules stability. As we continue investigate metalloprotein sites, gain insights enable us stable, active, selective catalysts.

Language: Английский

---

Cyclic iron tetra N-heterocyclic carbenes: synthesis, properties, reactivity, and catalysis

Chemical Society Reviews, Journal Year: 2023, Volume and Issue: 52(6), P. 2238 - 2277

Published: Jan. 1, 2023

Cyclic iron tetracarbenes structurally resemble porphyrins, but the strong equatorial σ-donation results in a different electronic structure and reactivity.

Language: Английский

---

Transition metal oxide complexes as molecular catalysts for selective methane to methanol transformation: any prospects or time to retire?

Physical Chemistry Chemical Physics, Journal Year: 2023, Volume and Issue: 25(7), P. 5313 - 5326

Published: Jan. 1, 2023

The performance of transition metal oxides for converting methane to methanol is assessed and two kinds molecular catalysts are proposed improve their selectivity: with hydrophilic ligands or oxide anionic complexes.

Language: Английский

---