Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles

[Image: see text] Highly energetic electron–hole pairs (hot carriers) formed from plasmon decay in metallic nanostructures promise sustainable pathways for energy-harvesting devices. However, efficient collection before thermalization remains an obstacle for realization of their full energy generati...

Descripción completa

Detalles Bibliográficos
Autores principales: Habib, Adela, Lubbers, Nicholas, Tretiak, Sergei, Nebgen, Benjamin
Formato: Online Artículo Texto
Lenguaje:English
Publicado: American Chemical Society 2023
Acceso en línea:https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10165650/
https://www.ncbi.nlm.nih.gov/pubmed/37078657
http://dx.doi.org/10.1021/acs.jpca.2c08757
_version_ 1785038303304613888
author Habib, Adela
Lubbers, Nicholas
Tretiak, Sergei
Nebgen, Benjamin
author_facet Habib, Adela
Lubbers, Nicholas
Tretiak, Sergei
Nebgen, Benjamin
author_sort Habib, Adela
collection PubMed
description [Image: see text] Highly energetic electron–hole pairs (hot carriers) formed from plasmon decay in metallic nanostructures promise sustainable pathways for energy-harvesting devices. However, efficient collection before thermalization remains an obstacle for realization of their full energy generating potential. Addressing this challenge requires detailed understanding of physical processes from plasmon excitation in the metal to their collection in a molecule or a semiconductor, where atomistic theoretical investigation may be particularly beneficial. Unfortunately, first-principles theoretical modeling of these processes is extremely costly, preventing a detailed analysis over a large number of potential nanostructures and limiting the analysis to systems with a few 100s of atoms. Recent advances in machine learned interatomic potentials suggest that dynamics can be accelerated with surrogate models which replace the full solution of the Schrödinger Equation. Here, we modify an existing neural network, Hierarchically Interacting Particle Neural Network (HIP-NN), to predict plasmon dynamics in Ag nanoparticles. The model takes as a minimum as three time steps of the reference real-time time-dependent density functional theory (rt-TDDFT) calculated charges as history and predicts trajectories for 5 fs in great agreement with the reference simulation. Further, we show that a multistep training approach in which the loss function includes errors from future time-step predictions can stabilize the model predictions for the entire simulated trajectory (∼25 fs). This extends the model’s capability to accurately predict plasmon dynamics in large nanoparticles of up to 561 atoms, not present in the training data set. More importantly, with machine learning models on GPUs, we gain a speed-up factor of ∼10(3) as compared with the rt-TDDFT calculations when predicting important physical quantities such as dynamic dipole moments in Ag(55) and a factor of ∼10(4) for extended nanoparticles that are 10 times larger. This underscores the promise of future machine learning accelerated electron/nuclear dynamics simulations for understanding fundamental properties of plasmon-driven hot carrier devices.
format Online
Article
Text
id pubmed-10165650
institution National Center for Biotechnology Information
language English
publishDate 2023
publisher American Chemical Society
record_format MEDLINE/PubMed
spelling pubmed-101656502023-05-09 Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles Habib, Adela Lubbers, Nicholas Tretiak, Sergei Nebgen, Benjamin J Phys Chem A [Image: see text] Highly energetic electron–hole pairs (hot carriers) formed from plasmon decay in metallic nanostructures promise sustainable pathways for energy-harvesting devices. However, efficient collection before thermalization remains an obstacle for realization of their full energy generating potential. Addressing this challenge requires detailed understanding of physical processes from plasmon excitation in the metal to their collection in a molecule or a semiconductor, where atomistic theoretical investigation may be particularly beneficial. Unfortunately, first-principles theoretical modeling of these processes is extremely costly, preventing a detailed analysis over a large number of potential nanostructures and limiting the analysis to systems with a few 100s of atoms. Recent advances in machine learned interatomic potentials suggest that dynamics can be accelerated with surrogate models which replace the full solution of the Schrödinger Equation. Here, we modify an existing neural network, Hierarchically Interacting Particle Neural Network (HIP-NN), to predict plasmon dynamics in Ag nanoparticles. The model takes as a minimum as three time steps of the reference real-time time-dependent density functional theory (rt-TDDFT) calculated charges as history and predicts trajectories for 5 fs in great agreement with the reference simulation. Further, we show that a multistep training approach in which the loss function includes errors from future time-step predictions can stabilize the model predictions for the entire simulated trajectory (∼25 fs). This extends the model’s capability to accurately predict plasmon dynamics in large nanoparticles of up to 561 atoms, not present in the training data set. More importantly, with machine learning models on GPUs, we gain a speed-up factor of ∼10(3) as compared with the rt-TDDFT calculations when predicting important physical quantities such as dynamic dipole moments in Ag(55) and a factor of ∼10(4) for extended nanoparticles that are 10 times larger. This underscores the promise of future machine learning accelerated electron/nuclear dynamics simulations for understanding fundamental properties of plasmon-driven hot carrier devices. American Chemical Society 2023-04-20 /pmc/articles/PMC10165650/ /pubmed/37078657 http://dx.doi.org/10.1021/acs.jpca.2c08757 Text en © 2023 The Authors. Published by American Chemical Society https://creativecommons.org/licenses/by-nc-nd/4.0/Permits non-commercial access and re-use, provided that author attribution and integrity are maintained; but does not permit creation of adaptations or other derivative works (https://creativecommons.org/licenses/by-nc-nd/4.0/).
spellingShingle Habib, Adela
Lubbers, Nicholas
Tretiak, Sergei
Nebgen, Benjamin
Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles
title Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles
title_full Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles
title_fullStr Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles
title_full_unstemmed Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles
title_short Machine Learning Models Capture Plasmon Dynamics in Ag Nanoparticles
title_sort machine learning models capture plasmon dynamics in ag nanoparticles
url https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10165650/
https://www.ncbi.nlm.nih.gov/pubmed/37078657
http://dx.doi.org/10.1021/acs.jpca.2c08757
work_keys_str_mv AT habibadela machinelearningmodelscaptureplasmondynamicsinagnanoparticles
AT lubbersnicholas machinelearningmodelscaptureplasmondynamicsinagnanoparticles
AT tretiaksergei machinelearningmodelscaptureplasmondynamicsinagnanoparticles
AT nebgenbenjamin machinelearningmodelscaptureplasmondynamicsinagnanoparticles