We investigate the performance of on-the-fly ab initio (OTF-AI) semiclassical dynamics combined with the thawed Gaussian approximation (TGA) for computing vibrationally resolved absorption and photoelectron spectra. Ammonia is used as a prototype of floppy molecules, whose potential energy surfaces display strong anharmonicity. We show that despite complications due to the presence of large amplitude motion, the main features of the spectra are captured by the OTF-AI-TGA, which—by definition—does not require any a priori knowledge of the potential energy surface. Moreover, the computed spectra are significantly better than those based on the popular global harmonic approximation. Finally, we probe the limit of the TGA to describe higher-resolution spectra, where long time dynamics is required.
Download full-text PDF |
Source |
|---|---|
| http://dx.doi.org/10.1021/acs.jpca.5b03907 | DOI Listing |
Publication Analysis
Top Keywords
Similar Publications
Nonadiabatic chemical reaction dynamics for the photoisomerization reaction of 3,5-dimethylisoxazole the S electronic state.
Phys Chem Chem Phys
December 2024
Faculty of Science and Technology, Sophia University, Chiyoda, Tokyo 102-8554, Japan.
Nonadiabatic molecular dynamics simulations were performed to explore the photoisomerization pathway from isoxazole (iso-OXA) to oxazole (OXA), considering four electronic states. The XMS-CASPT2 and SA4-CASSCF theories were employed to describe these electronic structures, which were caused by 12 electrons in 11 orbitals with the cc-pVDZ + sp diffuse basis set; the Gaussian s- and p-type diffuse functions were extracted from Dunning's aug-cc-pVDZ function. The potential energy and its gradient at each time step were computed on-the-fly at these levels in the time evolution of the classical trajectory.
View Article and Find Full Text PDFA machine learning potential construction based on radial distribution function sampling.
J Comput Chem
December 2024
Center for Computational Sciences, University of Tsukuba, Tsukuba, Japan.
Article Synopsis
- - Sampling reference data is essential for building accurate machine learning potentials (MLPs) in molecular dynamics simulations, as poor data coverage can lead to unphysical results.
- - This study introduces a radial distribution function (RDF)-based sampling method that captures unusual molecular structures during MLP-MD simulations by analyzing the shapes of RDFs, enhancing the reference data for better accuracy.
- - By applying this method to a water molecule cluster, researchers found that the resulting MLP-MD simulations displayed realistic behavior, comparable to advanced ab initio simulations, proving the RDF-based sampling's effectiveness in bridging molecular clusters to bulk systems.
Machine learning-driven investigation of the structure and dynamics of the BMIM-BF room temperature ionic liquid.
Faraday Discuss
October 2024
Institute for Computational Physics, University of Stuttgart, 70569 Stuttgart, Germany.
Room-temperature ionic liquids are an exciting group of materials with the potential to revolutionize energy storage. Due to their chemical structure and means of interaction, they are challenging to study computationally. Classical descriptions of their inter- and intra-molecular interactions require time intensive parametrization of force-fields which is prone to assumptions.
View Article and Find Full Text PDFAdvancing Force Fields Parameterization: A Directed Graph Attention Networks Approach.
J Chem Theory Comput
July 2024
Sorbonne Université, CNRS, Université Paris Cité, Laboratoire Jacques-Louis Lions (LJLL), UMR 7598 CNRS, 75005 Paris, France.
Force fields (FFs) are an established tool for simulating large and complex molecular systems. However, parametrizing FFs is a challenging and time-consuming task that relies on empirical heuristics, experimental data, and computational data. Recent efforts aim to automate the assignment of FF parameters using pre-existing databases and on-the-fly data.
View Article and Find Full Text PDFMLatom Software Ecosystem for Surface Hopping Dynamics in Python with Quantum Mechanical and Machine Learning Methods.
J Chem Theory Comput
June 2024
College of Chemistry and Chemical Engineering, Xiamen University, Xiamen, Fujian 361005, China.
We present an open-source MLatom@XACS software ecosystem for on-the-fly surface hopping nonadiabatic dynamics based on the Landau-Zener-Belyaev-Lebedev algorithm. The dynamics can be performed via Python API with a wide range of quantum mechanical (QM) and machine learning (ML) methods, including ab initio QM (CASSCF and ADC(2)), semiempirical QM methods (e.g.
View Article and Find Full Text PDFWant AI Summaries of new PubMed Abstracts delivered to your In-box?
Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!