Best‐Practice DFT Protocols for Basic Molecular Computational Chemistry**

Angewandte Chemie International Edition, Journal Year: 2022, Volume and Issue: 61(42)

Published: Sept. 14, 2022

Nowadays, many chemical investigations are supported by routine calculations of molecular structures, reaction energies, barrier heights, and spectroscopic properties. The lion's share these quantum-chemical applies density functional theory (DFT) evaluated in atomic-orbital basis sets. This work provides best-practice guidance on the numerous methodological technical aspects DFT three parts: Firstly, we set stage introduce a step-by-step decision tree to choose computational protocol that models experiment as closely possible. Secondly, present recommendation matrix guide choice depending task at hand. A particular focus is achieving an optimal balance between accuracy, robustness, efficiency through multi-level approaches. Finally, discuss selected representative examples illustrate recommended protocols effect choices.

Language: Английский

---

Computational Methods in Heterogeneous Catalysis

Chemical Reviews, Journal Year: 2020, Volume and Issue: 121(2), P. 1007 - 1048

Published: Dec. 22, 2020

The unprecedented ability of computations to probe atomic-level details catalytic systems holds immense promise for the fundamentals-based bottom-up design novel heterogeneous catalysts, which are at heart chemical and energy sectors industry. Here, we critically analyze recent advances in computational catalysis. First, will survey progress electronic structure methods atomistic catalyst models employed, have enabled catalysis community build increasingly intricate, realistic, accurate active sites supported transition-metal catalysts. We then review developments microkinetic modeling, specifically mean-field kinetic Monte Carlo simulations, bridge gap between nanoscale insights macroscale experimental kinetics data with increasing fidelity. finally advancements theoretical accelerating discovery. Throughout review, provide ample examples applications, discuss remaining challenges, our outlook near future.

Language: Английский

---

Improving the Accuracy of Atomistic Simulations of the Electrochemical Interface

Chemical Reviews, Journal Year: 2022, Volume and Issue: 122(12), P. 10651 - 10674

Published: May 6, 2022

Atomistic simulation of the electrochemical double layer is an ambitious undertaking, requiring quantum mechanical description electrons, phase space sampling liquid electrolytes, and equilibration electrolytes over nanosecond time scales. All models electrochemistry make different trade-offs in approximation electrons atomic configurations, from extremes classical molecular dynamics a complete interface with point-charge atoms to correlated electronic structure methods single electrode configuration no or electrolyte. Here, we review spectrum techniques suitable for electrochemistry, focusing on key approximations accuracy considerations each technique. We discuss promising approaches, such as enhanced configurations computationally efficient beyond density functional theory (DFT) methods, that will push simulations present frontier.

Language: Английский

---

Autonomous Reaction Network Exploration in Homogeneous and Heterogeneous Catalysis

Topics in Catalysis, Journal Year: 2022, Volume and Issue: 65(1-4), P. 6 - 39

Published: Jan. 13, 2022

Autonomous computations that rely on automated reaction network elucidation algorithms may pave the way to make computational catalysis a par with experimental research in field. Several advantages of this approach are key catalysis: (i) Automation allows one consider orders magnitude more structures systematic and open-ended fashion than what would be accessible by manual inspection. Eventually, full resolution terms structural varieties conformations as well respect type number potentially important elementary steps (including decomposition reactions determine turnover numbers) achieved. (ii) Fast electronic structure methods uncertainty quantification warrant high efficiency reliability order not only deliver results quickly, but also allow for predictive work. (iii) A degree autonomy reduces amount human work, processing errors, bias. Although being inherently unbiased, it is still steerable specific regions an emerging addition new reactant species. This fidelity formalization some catalytic process surprising silico discoveries. In we first review state art embed autonomous explorations into general field from which draws its ingredients. We then elaborate conceptual issues arise context procedures, discuss at example system.

Language: Английский

---

A human-machine interface for automatic exploration of chemical reaction networks

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

Published: May 1, 2024

Abstract Autonomous reaction network exploration algorithms offer a systematic approach to explore mechanisms of complex chemical processes. However, the resulting networks are so vast that an all potentially accessible intermediates is computationally too demanding. This renders brute-force explorations unfeasible, while with completely pre-defined or hard-wired constraints, such as element-specific coordination numbers, not flexible enough for systems. Here, we introduce STEERING WHEEL guide otherwise unbiased automated exploration. The algorithm intuitive, generally applicable, and enables one focus on specific regions emerging network. It also allows guiding data generation in context mechanism exploration, catalyst design, other optimization challenges. demonstrated elucidation transition metal catalysts. We highlight how catalytic cycles reproducible way. objectives fully adjustable, allowing harness both structure-specific (accurate) calculations well broad high-throughput screening possible intermediates.

Language: Английский

---

Beyond Continuum Solvent Models in Computational Homogeneous Catalysis

Topics in Catalysis, Journal Year: 2021, Volume and Issue: 65(1-4), P. 118 - 140

Published: Nov. 16, 2021

Abstract In homogeneous catalysis solvent is an inherent part of the catalytic system. As such, it must be considered in computational modeling. The most common approach to include effects quantum mechanical calculations by means continuum models. When they are properly used, average efficiently captured, mainly those related with polarity. However, neglecting atomistic description molecules has its limitations, and models all alone cannot applied whatever situation. many cases, inclusion explicit system mandatory. purpose this article highlight through selected examples what reasons that urge go beyond employment micro-solvated (cluster-continuum) fully models, way setting limits catalysis. These showcase calculation not only can improve already known mechanisms but yield new mechanistic views a reaction. With aim systematizing use after discussing success limitations issues coordination dynamics, reactions involving small, charged species, as well protic solvents role reagent itself successively considered.

Language: Английский

---

Self-Parametrizing System-Focused Atomistic Models

Journal of Chemical Theory and Computation, Journal Year: 2020, Volume and Issue: 16(3), P. 1646 - 1665

Published: Jan. 17, 2020

Computational studies of chemical reactions in complex environments such as proteins, nanostructures, or on surfaces require accurate and efficient atomistic models applicable to the nanometer scale. In general, an parametrization entities will not be available for arbitrary system classes but demands a fast, automated, system-focused procedure quickly applicable, reliable, flexible, reproducible. Here, we develop combine automatically parametrizable quantum chemically derived molecular mechanics model with machine-learned corrections under autonomous uncertainty quantification refinement. Our approach first generates accurate, physically motivated from minimum energy structure its corresponding Hessian matrix by partial fitting force constants. This is then starting point generate large number configurations which additional off-minimum reference data can evaluated fly. A Δ-machine learning trained these provide correction energies forces including estimates. During procedure, flexibility machine tailored amount training data. The systems enabled fragmentation approach. Due their modular nature, all construction steps allow improvement rolling fashion. may also employed generation electrostatic embedding quantum-mechanical/molecular-mechanical hybrid structures at nanoscale.

Language: Английский

---

Towards a converged strategy for including microsolvation in reaction mechanism calculations

Journal of Computer-Aided Molecular Design, Journal Year: 2021, Volume and Issue: 35(4), P. 473 - 492

Published: Jan. 9, 2021

Language: Английский

---

Computational Organometallic Catalysis: Where We Are, Where We Are Going

European Journal of Inorganic Chemistry, Journal Year: 2021, Volume and Issue: 2021(26), P. 2547 - 2555

Published: June 17, 2021

Abstract This essay gives my personal perspective of the current stage computational methods applied to modeling organometallic catalysis, as well new directions field is taking. The first part deals with what I consider state‐of‐the‐art build up energy profiles, regarding both chemical and models. With a proper choice model methods, quantum mechanical calculations are nowadays able provide accurate profiles reactions in solution involving closed‐shell species. However, most cases they still used “predict past”, providing after‐the‐fact explanations missing out full potential contemporary simulation techniques. Simulations mature enough be incorporated at design guide experimental exploration. taking, incorporating automated exploration combined extensive data analysis machine learning algorithms, approach holy grail catalyst discovering.

Language: Английский

---

Computational mechanistic study in organometallic catalysis: Why prediction is still a challenge

Wiley Interdisciplinary Reviews Computational Molecular Science, Journal Year: 2021, Volume and Issue: 12(4)

Published: Nov. 23, 2021

Abstract Although computational contributions to the understanding of organometallic homogeneous catalysts have become fairly routine, a step‐change in application methods would be achieve efficient, robust, and reliable prediction outcome catalytic transformations. While we concur that there been number recent promising advances interactions between experimental mechanistic studies, mapping reactivity space remains incomplete large‐scale studies make limiting assumptions which restrict their transferability. Close synergies characterization analysis techniques are integrated with data, along data capture, curation, exploitation, vital develop our all aspects pathways (including activation deactivation) allow continual refinement understanding, challenged by testing predictions experimentally. Here review examples formulate protocol for such interactions. This article is categorized under: Electronic Structure Theory > Ab Initio Methods Mechanism Reaction Mechanisms Catalysis Data Science Artificial Intelligence/Machine Learning Density Functional

Language: Английский

---