Photoinduced Desorption Dynamics of CO from Pd(111): A Neural Network Approach.

Bin Jiang,Ivor Lončarić,Juraj Ovčar,Alberto P Sánchez Muzas,Maite Alducin,Yaolong Zhang,Alfredo Serrano Jiménez,J Iñaki Juaristi

doi:10.1021/acs.jctc.1c00347

Abstract

Modeling the ultrafast photoinduced dynamics and reactivity of adsorbates on metals requires including the effect of the laser-excited electrons and, in many cases, also the effect of the highly excited surface lattice. Although the recent ab initio molecular dynamics with electronic friction and thermostats, (Te,Tl)-AIMDEF [AlducinM.;Phys. Rev. Lett.2019, 123, 246802]31922860, enables such complex modeling, its computational cost may limit its applicability. Here, we use the new embedded atom neural network (EANN) method [ZhangY.;J. Phys. Chem. Lett.2019, 10, 496231397157] to develop an accurate and extremely complex potential energy surface (PES) that allows us a detailed and reliable description of the photoinduced desorption of CO from the Pd(111) surface with a coverage of 0.75 monolayer. Molecular dynamics simulations performed on this EANN-PES reproduce the (Te,Tl)-AIMDEF results with a remarkable level of accuracy. This demonstrates the outstanding performance of the obtained EANN-PES that is able to reproduce available density functional theory (DFT) data for an extensive range of surface temperatures (90–1000 K); a large number of degrees of freedom, those corresponding to six CO adsorbates and 24 moving surface atoms; and the varying CO coverage caused by the abundant desorption events.

Highlights

The use of intense (∼1 mJ/cm2) femtosecond laser pulses in the ultraviolet, visible, and near-infrared regime has been shown to be a very efficient way to promote reactions at adsorbate-covered metal surfaces.[1−3] At these wavelengths, a large fraction of the light is absorbed by the metal giving rise to electronic excitations
The desorption of CO from the Pd(111) surface induced by femtosecond laser pulses is simulated with molecular dynamics with electronic friction and thermostat [(Te,Tl)-MDEF] calculations performed in our developed embedded atom neural network (EANN)-potential energy surface (PES)
To rule out these sources of error and further address the accuracy of our developed EANN-PES and additive density generator function (DGF), we have simulated the desorption of CO from 0.75 ML-CO/

Summary

INTRODUCTION

The use of intense (∼1 mJ/cm2) femtosecond (fs) laser pulses in the ultraviolet, visible, and near-infrared regime has been shown to be a very efficient way to promote reactions at adsorbate-covered metal surfaces.[1−3] At these wavelengths, a large fraction of the light is absorbed by the metal giving rise to electronic excitations. The modeling of these experiments requires performing molecular dynamics simulations in an excited environment.[10−17] First, the excitation generated by the laser pulse in the substrate is described in terms of timedependent electronic (Te) and phononic (Tl) temperatures that are obtained using the two-temperature model (2TM).[18] Subsequently, the motion of the adsorbates is determined by solving Langevin equations of motion in the ground-state potential energy surface (PES).

EANN-PES GENERATION AND QUALITY CHECK

PHOTOINDUCED MOLECULAR DYNAMICS ON SURFACES

RESULTS

CONCLUSIONS

■ ACKNOWLEDGMENTS

■ REFERENCES