Ring Polymer Molecular Dynamics Approach to Quantum Dissociative Chemisorption Rates

Zhang, Liang; Zuo, Junxiang; Suleimanov, Yury V.; Guo, Hua

doi:10.1021/acs.jpclett.3c01848

Cited by 6 publications

(3 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…When the accuracy of rate constant prediction is the goal, semiclassical or quasi-classical trajectory (QCT) calculations are typically undertaken using a system-specific potential energy surface (PES) − instead of using a classical MD force field. Trajectory based rate constant calculation packages such as DiNT, VENUS , or internal codes can be used to train these PESs and calculate the rate constants using trajectories. Jasper et al used DiNT along with permutationally invariant polynomial (PIP) expansions to calculate singlet and triplet rate constants for the H + HO 2 reactions.…”

Section: Introductionmentioning

confidence: 99%

Quasi-Classical Trajectory Calculation of Rate Constants Using an Ab Initio Trained Machine Learning Model (aML-MD) with Multifidelity Data

Shi,

Lele,

Jasper

et al. 2024

J. Phys. Chem. A

View full text Add to dashboard Cite

Machine learning (ML) provides a great opportunity for the construction of models with improved accuracy in classical molecular dynamics (MD). However, the accuracy of a ML trained model is limited by the quality and quantity of the training data. Generating large sets of accurate ab initio training data can require significant computational resources. Furthermore, inconsistent or incompatible data with different accuracies obtained using different methods may lead to biased or unreliable ML models that do not accurately represent the underlying physics. Recently, transfer learning showed its potential for avoiding these problems as well as for improving the accuracy, efficiency, and generalization of ML models using multifidelity data. In this work, ab initio trained ML-based MD (aML-MD) models are developed through transfer learning using DFT and multireference data from multiple sources with varying accuracy within the Deep Potential MD framework. The accuracy of the force field is demonstrated by calculating rate constants for the H + HO 2 → H 2 + 3 O 2 reaction using quasi-classical trajectories. We show that the aML-MD model with transfer learning can accurately predict the rate constants while reducing the computational cost by more than five times compared to the use of more expensive quantum chemistry training data sets. Hence, the aML-MD model with transfer learning shows great potential in using multifidelity data to reduce the computational cost involved in generating the training set for these potentials.

show abstract

Section: Introductionmentioning

confidence: 99%

Quasi-Classical Trajectory Calculation of Rate Constants Using an Ab Initio Trained Machine Learning Model (aML-MD) with Multifidelity Data

Shi,

Lele,

Jasper

et al. 2024

J. Phys. Chem. A

View full text Add to dashboard Cite

show abstract

“…16 Methods derived from PI theory such as centroid molecular dynamics 17 and ring-polymer molecular dynamics 18 (RPMD) have early been applied to proton transfer reactions 19 then extensively used to successfully treat tunneling in chemical reactions involving light atoms. [20][21][22][23][24][25][26] Though exact simulation of the quantum dynamics of a reaction remains out of reach even for relatively simple molecular systems, theoretical arguments based on semi-classical theory 27 clarify how PI methods can provide remarkably good estimations of reaction rates even lower than the so-called crossover temperature below which tunneling is expected to play a dominant role. 28,29 In this work we employ the RPMD approach to study the temperature behavior of free energy barriers and the deviations from the transition state theory (TST) by evaluating the so-called recrossing factors.…”

Section: Introductionmentioning

confidence: 99%

Elucidating Heavy Atom Tunneling Kinetics

Angiolari,

Mandelli,

Huppert

et al. 2024

Preprint

View full text Add to dashboard Cite

In this work, we characterize the temperature dependence of kinetic properties in heavy atom tunneling reactions by means of molecular dynamics simulations, including nuclear quantum effects (NQEs) via Path Integral theory. To this end, we consider the prototypical Cope rearrangement of semibullvalene. The reaction was studied in the 25--300~K temperature range observing that the inclusion of NQEs modifies the temperature behavior of both free energy barriers and dynamical recrossing factors with respect to classical dynamics. Notably, while in classical simulations the activation free energy shows a very weak temperature dependence, it becomes strongly dependent on temperature when NQEs are included. This temperature behavior shows a transition from a regime where the quantum effects are limited and can mainly be traced back to zero point energy, to a low temperature regime where tunneling plays a dominant role. In this regime, the free energy curve literally tunnels below the potential energy barrier along the reaction coordinate, allowing much faster reaction rates. Finally, the temperature dependence of the rate constants obtained from molecular dynamics simulations was compared with available experimental data and with semi-classical transition state theory calculations, showing comparable behaviors and similar transition temperatures from thermal to (deep) tunneling regime.

show abstract

“… 56 Furthermore, the RPMD approach has also been used to simulate the DC of H 2 on Cu(111) and D 2 O on Ni(111), where it was observed that RPMD can accurately reproduce wave packet QD. 57 Recently, DC rates of H 2 on Pt(111) and Ag(111) have been (approximately) computed using RPMD rate theory, 58 which yielded a qualitative improvement over classical rate theory. However, in both DC studies, the surface atoms were kept fixed in their ideal positions, neglecting any dynamical effects as a result of surface atom motion, such as barrier height modulation and energy transfer between the molecule and the metal surface.…”

mentioning

confidence: 99%

Accurate Reaction Probabilities for Translational Energies on Both Sides of the Barrier of Dissociative Chemisorption on Metal Surfaces

Gerrits,

Jackson,

Bogaerts

2024

J. Phys. Chem. Lett.

View full text Add to dashboard Cite

Molecular dynamics simulations are essential for a better understanding of dissociative chemisorption on metal surfaces, which is often the rate-controlling step in heterogeneous and plasma catalysis. The workhorse quasi-classical trajectory approach ubiquitous in molecular dynamics is able to accurately predict reactivity only for high translational and low vibrational energies. In contrast, catalytically relevant conditions generally involve low translational and elevated vibrational energies. Existing quantum dynamics approaches are intractable or approximate as a result of the large number of degrees of freedom present in molecule−metal surface reactions. Here, we extend a ring polymer molecular dynamics approach to fully include, for the first time, the degrees of freedom of a moving metal surface. With this approach, experimental sticking probabilities for the dissociative chemisorption of methane on Pt(111) are reproduced for a large range of translational and vibrational energies by including nuclear quantum effects and employing full-dimensional simulations.

show abstract

Ring Polymer Molecular Dynamics Approach to Quantum Dissociative Chemisorption Rates

Cited by 6 publications

References 56 publications

Quasi-Classical Trajectory Calculation of Rate Constants Using an Ab Initio Trained Machine Learning Model (aML-MD) with Multifidelity Data

Quasi-Classical Trajectory Calculation of Rate Constants Using an Ab Initio Trained Machine Learning Model (aML-MD) with Multifidelity Data

Elucidating Heavy Atom Tunneling Kinetics

Accurate Reaction Probabilities for Translational Energies on Both Sides of the Barrier of Dissociative Chemisorption on Metal Surfaces

Contact Info

Product

Resources

About