Extended tight‐binding quantum chemistry methods

Wiley Interdisciplinary Reviews Computational Molecular Science, Journal Year: 2020, Volume and Issue: 11(2)

Published: Aug. 9, 2020

Abstract This review covers a family of atomistic, mostly quantum chemistry (QC) based semiempirical methods for the fast and reasonably accurate description large molecules in gas condensed phase. The theory is derived from density functional (DFT) perturbation expansion electron fluctuation terms to various orders similar original tight binding model. term “eXtended” their name (xTB) emphasizes parameter availability almost entire periodic table elements ( Z ≤ 86) improvements underlying regarding, example, atomic orbital basis set, level multipole approximation treatment important electrostatic dispersion interactions. A common feature most members consistent parameterization on phase theoretical reference data geometries, vibrational frequencies noncovalent interactions, which are primary properties interest typical applications systems composed up few thousand atoms. Further specialized versions were developed electronic spectra corresponding response properties. Besides provided background with some implementation details efficient free xtb program, benchmarks structural thermochemical including (transition‐)metal discussed. completed by recent extensions model force‐field (FF) as well its application solids under boundary conditions. general applicability together excellent cost‐accuracy ratio high robustness make xTB very attractive fields computer‐aided chemical research. article categorized under: Electronic Structure Theory > Ab Initio Methods Semiempirical Software Quantum Chemistry

Language: Английский

---

r2SCAN-3c: A “Swiss army knife” composite electronic-structure method

The Journal of Chemical Physics, Journal Year: 2021, Volume and Issue: 154(6)

Published: Feb. 10, 2021

The recently proposed r2SCAN meta-generalized-gradient approximation (mGGA) of Furness and co-workers is used to construct an efficient composite electronic-structure method termed r2SCAN-3c. To this end, the unaltered functional combined with a tailor-made triple-ζ Gaussian atomic orbital basis set as well refitted D4 geometrical counter-poise corrections for London-dispersion superposition error. performance new evaluated GMTKN55 database covering large parts chemical space about 1500 data points, additional benchmarks non-covalent interactions, organometallic reactions, lattice energies organic molecules ices, adsorption on polar salt non-polar coinage-metal surfaces. These comprehensive tests reveal spectacular robustness r2SCAN-3c: It by far surpasses its predecessor B97-3c at only twice cost provides one best results all semi-local density-functional theory (DFT)/QZ methods ever tested one-tenth cost. Specifically, reaction conformational it outperforms prominent hybrid-DFT/QZ approaches two three orders magnitude lower Perhaps, most relevant remaining issue r2SCAN-3c self-interaction error (SIE), owing mGGA nature. However, SIE slightly reduced compared other (m)GGAs, demonstrated in examples. After all, remarkably robust chosen our group default, replacing previous DFT partially even expensive high-level standard applications systems up several hundreds atoms.

Language: Английский

---

Computational molecular spectroscopy

Nature Reviews Methods Primers, Journal Year: 2021, Volume and Issue: 1(1)

Published: May 27, 2021

Language: Английский

---

Robust and Efficient Implicit Solvation Model for Fast Semiempirical Methods

Journal of Chemical Theory and Computation, Journal Year: 2021, Volume and Issue: 17(7), P. 4250 - 4261

Published: June 29, 2021

We present a robust and efficient method to implicitly account for solvation effects in modern semiempirical quantum mechanics force fields. A computationally yet accurate model based on the analytical linearized Poisson–Boltzmann (ALPB) is parameterized extended tight binding (xTB) density functional (DFTB) methods as well recently proposed GFN-FF general field. The perform over broad range of systems applications, from conformational energies transition-metal complexes large supramolecular association reactions charged species. For hydration free small molecules, GFN1-xTB(ALPB) reaching accuracy sophisticated explicitly solvated approaches, with mean absolute deviation only 1.4 kcal/mol compared experiment. Logarithmic octanol–water partition coefficients (log Kow) are computed about 0.65 using GFN2-xTB(ALPB) experimental values indicating consistent description differential solvent effects. Overall, more than twenty solvents each six tested. They readily available xtb dftb+ programs diverse computational applications.

Language: Английский

---

Extension and evaluation of the D4 London-dispersion model for periodic systems

Physical Chemistry Chemical Physics, Journal Year: 2020, Volume and Issue: 22(16), P. 8499 - 8512

Published: Jan. 1, 2020

We present an extension of the DFT-D4 model [J. Chem. Phys., 2019, 150, 154122] for periodic systems. The main new ingredients are additional reference polarizabilities highly-coordinated group 1-5 elements derived from pseudo-periodic electrostatically-embedded cluster calculations. To illustrate performance updated method, several test cases considered, which we compare D4 to its predecessor D3(BJ), as well a comprehensive set other dispersion-corrected methods. largest improvements observed solid-state 16 inorganic salts, where achieves unprecedented accuracy, surpassing other, computationally much more demanding approaches. For cell volumes and lattice energies two sets chemically diverse molecular crystals, accuracy gain is less pronounced compared already excellently performing D3(BJ) method. challenging adsorption small organic molecules on metallic ionic surfaces, provides values in good agreement with experimental and/or high-level references. These results suggest application proposed physically improved yet efficient dispersion correction standard DFT calculations low-cost approaches like semi-empirical or even force-field models.

Language: Английский

---

A Comprehensive Discovery Platform for Organophosphorus Ligands for Catalysis

Journal of the American Chemical Society, Journal Year: 2022, Volume and Issue: 144(3), P. 1205 - 1217

Published: Jan. 12, 2022

The design of molecular catalysts typically involves reconciling multiple conflicting property requirements, largely relying on human intuition and local structural searches. However, the vast number potential requires pruning candidate space by efficient prediction with quantitative structure–property relationships. Data-driven workflows embedded in a library can be used to build predictive models for catalyst performance serve as blueprint novel designs. Herein we introduce kraken, discovery platform covering monodentate organophosphorus(III) ligands providing comprehensive physicochemical descriptors based representative conformer ensembles. Using quantum-mechanical methods, calculated 1558 ligands, including commercially available examples, trained machine learning predict properties over 300000 new ligands. We demonstrate application kraken systematically explore organophosphorus how existing data sets catalysis accelerate ligand selection during reaction optimization.

Language: Английский

---

Computational Discovery of Transition-metal Complexes: From High-throughput Screening to Machine Learning

Chemical Reviews, Journal Year: 2021, Volume and Issue: 121(16), P. 9927 - 10000

Published: July 14, 2021

Transition-metal complexes are attractive targets for the design of catalysts and functional materials. The behavior metal-organic bond, while very tunable achieving target properties, is challenging to predict necessitates searching a wide complex space identify needles in haystacks applications. This review will focus on techniques that make high-throughput search transition-metal chemical feasible discovery with desirable properties. cover development, promise, limitations "traditional" computational chemistry (i.e., force field, semiempirical, density theory methods) as it pertains data generation inorganic molecular discovery. also discuss opportunities leveraging experimental sources. We how advances statistical modeling, artificial intelligence, multiobjective optimization, automation accelerate lead compounds rules. overall objective this showcase bringing together from diverse areas computer science have enabled rapid uncovering structure-property relationships chemistry. aim highlight unique considerations motifs bonding (e.g., variable spin oxidation state, strength/nature) set them their apart more commonly considered organic molecules. uncertainty relative scarcity motivate specific developments machine learning representations, model training, Finally, we conclude an outlook opportunity accelerated complexes.

Language: Английский

---

Late-stage diversification of indole skeletons through nitrogen atom insertion

Science, Journal Year: 2022, Volume and Issue: 377(6610), P. 1104 - 1109

Published: Sept. 1, 2022

Compared with peripheral late-stage transformations mainly focusing on carbon-hydrogen functionalizations, reliable strategies to directly edit the core skeleton of pharmaceutical lead compounds still remain scarce despite recent flurry activity in this area. Herein, we report skeletal editing indoles through nitrogen atom insertion, accessing corresponding quinazoline or quinoxaline bioisosteres by trapping an electrophilic nitrene species generated from ammonium carbamate and hypervalent iodine. This reactivity relies strategic use a silyl group as labile protecting that can facilitate subsequent product release. The utility highly functional group-compatible methodology context several commercial drugs is demonstrated.

Language: Английский

---

Efficient Quantum Chemical Calculation of Structure Ensembles and Free Energies for Nonrigid Molecules

The Journal of Physical Chemistry A, Journal Year: 2021, Volume and Issue: 125(19), P. 4039 - 4054

Published: March 10, 2021

The application of quantum chemical, automatic multilevel modeling workflows for the determination thermodynamic (e.g., conformation equilibria, partition coefficients, pKa values) and spectroscopic properties relatively large, nonrigid molecules in solution is described. Key points are computation rather complete structure (conformer) ensembles with extremely fast but still reasonable GFN2-xTB or GFN-FF semiempirical methods CREST searching approach subsequent refinement at a recently developed, accurate r2SCAN-3c DFT composite level. Solvation effects included all steps by continuum solvation models (ALPB, (D)COSMO-RS). Consistent inclusion thermostatistical contributions framework modified rigid-rotor-harmonic-oscillator approximation (mRRHO) based on xTB/FF computed PES also recommended.

Language: Английский

---

GEOM, energy-annotated molecular conformations for property prediction and molecular generation

Scientific Data, Journal Year: 2022, Volume and Issue: 9(1)

Published: April 21, 2022

Abstract Machine learning (ML) outperforms traditional approaches in many molecular design tasks. ML models usually predict properties from a 2D chemical graph or single 3D structure, but neither of these representations accounts for the ensemble conformers that are accessible to molecule. Property prediction could be improved by using conformer ensembles as input, there is no large-scale dataset contains graphs annotated with accurate and experimental data. Here we use advanced sampling semi-empirical density functional theory (DFT) generate 37 million conformations over 450,000 molecules. The Geometric Ensemble Of Molecules (GEOM) 133,000 species QM9, 317,000 data related biophysics, physiology, physical chemistry. Ensembles 1,511 BACE-1 inhibition also labeled high-quality DFT free energies an implicit water solvent, 534 further optimized DFT. GEOM will assist development ensembles, generative sample conformations.

Language: Английский

---