dxtb—An efficient and fully differentiable framework for extended tight-binding DOI
Marvin Friede, Christian Hölzer, Sebastian Ehlert

и другие.

The Journal of Chemical Physics, Год журнала: 2024, Номер 161(6)

Опубликована: Авг. 9, 2024

Automatic differentiation (AD) emerged as an integral part of machine learning, accelerating model development by enabling gradient-based optimization without explicit analytical derivatives. Recently, the benefits AD and computing arbitrary-order derivatives with respect to any variable were also recognized in field quantum chemistry. In this work, we present dxtb—an open-source, fully differentiable framework for semiempirical extended tight-binding (xTB) methods. Developed entirely Python leveraging PyTorch array operations, dxtb facilitates extensibility rapid prototyping while maintaining computational efficiency. Through comprehensive code vectorization optimization, essentially reach speed compiled xTB programs high-throughput calculations small molecules. The excellent performance scales large systems, batch operability yields additional execution on parallel hardware. particular, energy evaluations are par existing programs, whereas automatically differentiated nuclear is only 2 5 times slower compared their counterparts. We showcase utility calculating various molecular spectroscopic properties, highlighting its capacity enhance simplify such evaluations. Furthermore, streamlines tasks offers seamless integration chemistry paving way physics-inspired end-to-end models. Ultimately, aims further advance capabilities methods, providing extensible foundation future developments hybrid learning applications. accessible at https://github.com/grimme-lab/dxtb.

Язык: Английский

A Guide to In Silico Drug Design DOI Creative Commons
Yiqun Chang, Bryson A. Hawkins, Jonathan J. Du

и другие.

Pharmaceutics, Год журнала: 2022, Номер 15(1), С. 49 - 49

Опубликована: Дек. 23, 2022

The drug discovery process is a rocky path that full of challenges, with the result very few candidates progress from hit compound to commercially available product, often due factors, such as poor binding affinity, off-target effects, or physicochemical properties, solubility stability. This further complicated by high research and development costs time requirements. It thus important optimise every step in order maximise chances success. As recent advancements computer power technology, computer-aided design (CADD) has become an integral part modern guide accelerate process. In this review, we present overview CADD methods applications, silico structure prediction, refinement, modelling target validation, are commonly used area.

Язык: Английский

Процитировано

133
Recent advances, application and prospect in g-C3N4-based S-scheme heterojunction photocatalysts DOI
Pengyu Hao,

Zhouze Chen,

Yujie Yan

и другие.

Separation and Purification Technology, Год журнала: 2023, Номер 330, С. 125302 - 125302

Опубликована: Окт. 12, 2023

Язык: Английский

Процитировано

89
Electric effects reinforce charge carrier behaviour for photocatalysis DOI

Aoqiang Shu,

Chencheng Qin,

Miao Li

и другие.

Energy & Environmental Science, Год журнала: 2024, Номер 17(14), С. 4907 - 4928

Опубликована: Янв. 1, 2024

Recent studies on enhancing charge carrier behavior through electric effects for efficient photocatalysis are summarized, evaluating the in-depth function of these effects. This provides unique perspectives to optimize photocatalytic processes.

Язык: Английский

Процитировано

29
Quantum chemical investigation of choline chloride-based deep eutectic solvents DOI

Zubera Naseem,

Rao Aqil Shehzad,

Sobia Jabeen

и другие.

Chemical Physics, Год журнала: 2023, Номер 571, С. 111936 - 111936

Опубликована: Апрель 10, 2023

Язык: Английский

Процитировано

32
A comparative analysis of the influence of hydrofluoroethers as diluents on solvation structure and electrochemical performance in non-flammable electrolytes DOI Creative Commons
Wessel van Ekeren, Marcelo Albuquerque, Gustav Ek

и другие.

Journal of Materials Chemistry A, Год журнала: 2023, Номер 11(8), С. 4111 - 4125

Опубликована: Янв. 1, 2023

TTE (1,1,2,2-tetrafluoroethyl 2,2,3,3-tetrafluoropropyl ether) shows better performance than BTFE (bis(2,2,2-trifluoroethyl)ether as diluent in a localized highly concentrated electrolyte based on lithium bis(fluorosulfonyl)imide triethylposphate.

Язык: Английский

Процитировано

28
Extension of the D3 and D4 London Dispersion Corrections to the full Actinides Series DOI Creative Commons
Lukas Wittmann, Igor Gordiy, Marvin Friede

и другие.

Physical Chemistry Chemical Physics, Год журнала: 2024, Номер 26(32), С. 21379 - 21394

Опубликована: Янв. 1, 2024

Efficient dispersion corrections are an indispensable component of modern density functional theory, semi-empirical quantum mechanical, and even force field methods. In this work, we extend the well established D3 D4 London to full actinides series, francium, radium. To keep consistency with existing versions, original parameterization strategy model was only slightly modified. This includes improved reference Hirshfeld atomic partial charges at ωB97M-V/ma-def-TZVP level fit required electronegativity equilibration charge (EEQ) model. context, developed a new actinide data set called AcQM, which covers most common molecular compound space. Furthermore, efficient calculation dynamic polarizabilities that needed construct

Язык: Английский

Процитировано

15
Functional monomer design for synthetically accessible polymers DOI Creative Commons
Seonghwan Kim, Charles M. Schroeder, Nicholas E. Jackson

и другие.

Chemical Science, Год журнала: 2025, Номер unknown

Опубликована: Янв. 1, 2025

Active learning combined with quantum chemistry reveals the nature of functional monomer design across a diverse chemical space 12M synthetically accessible polymers.

Язык: Английский

Процитировано

1
Do Optimally Tuned Range-Separated Hybrid Functionals Require a Reparametrization of the Dispersion Correction? It Depends DOI
Marvin Friede, Sebastian Ehlert, Stefan Grimme

и другие.

Journal of Chemical Theory and Computation, Год журнала: 2023, Номер 19(22), С. 8097 - 8107

Опубликована: Ноя. 13, 2023

For ground- and excited-state studies of large molecules, it is the state art to combine (time-dependent) DFT with dispersion-corrected range-separated hybrid functionals (RSHs), which ensures an asymptotically correct description exchange effects London dispersion. Specifically for studying excited states, common practice tune range-separation parameter ω (optimal tuning), can further improve accuracy. However, since optimal tuning essentially changes functional, unclear if how much parameters used dispersion correction depend on chosen value. To answer this question, we explore interdependency by refitting DFT-D4 model six established RSHs over a wide range values (0.05-0.45 a0-1) using set noncovalently bound molecular complexes. The results reveal some surprising differences among investigated functionals: While PBE-based ωB97M-D4 generally exhibit weak robust performance values, B88-based RSHs, specifically LC-BLYP, are strongly affected. these, even minor reduction from default value manifests in strong systematic overbinding poor typical optimally tuned values. Finally, discuss strategies mitigate these issues reflect context employed D4 optimization algorithm fit set, outlining future improvements.

Язык: Английский

Процитировано

19
Computational Chemistry as Applied in Environmental Research: Opportunities and Challenges DOI
Christian Sandoval‐Pauker, Sheng Yin, Alexandria Castillo

и другие.

ACS ES&T Engineering, Год журнала: 2023, Номер 4(1), С. 66 - 95

Опубликована: Окт. 12, 2023

The constant development of computer systems and infrastructure has allowed computational chemistry to become an important component environmental research. In the past decade, application quantum classical mechanical calculations model understand increased exponentially. this review, we highlight various applications techniques in areas research (e.g., wastewater/air treatment, sensing, biodegradation). We briefly describe each approach, starting with principle methods followed by molecular mechanics (MM), dynamics (MD), hybrid QM/MM methods. recent introduction artificial intelligence machine learning their potential disrupt field are also discussed. Challenges current future directions address them presented.

Язык: Английский

Процитировано

18
In Defense of (Certain) Pople-Type Basis Sets DOI Creative Commons
Montgomery Gray, Paige E. Bowling, John M. Herbert

и другие.

Опубликована: Янв. 22, 2024

A recent study suggests that Gaussian basis sets in the 6-311G family are inappropriate for thermochemical calculations based on density functional theory, emphasizing need polarization functions but omitting tests of Pople containing a full complement thereof. Here, we point out certain category yield error statistics with respect to benchmark comparable def2-TZVP, at about half cost. More elaborate can rival accuracy def2-QZVPD 5-10% We also clarify role integral thresholds achieving robust convergence presence diffuse functions.

Язык: Английский

Процитировано

9