-
Machine learning the electronic structure of matter across temperatures
Authors:
Lenz Fiedler,
Normand A. Modine,
Kyle D. Miller,
Attila Cangi
Abstract:
We introduce machine learning (ML) models that predict the electronic structure of materials across a wide temperature range. Our models employ neural networks and are trained on density functional theory (DFT) data. Unlike most other ML models that use DFT data, our models directly predict the local density of states (LDOS) of the electronic structure. This provides several advantages, including…
▽ More
We introduce machine learning (ML) models that predict the electronic structure of materials across a wide temperature range. Our models employ neural networks and are trained on density functional theory (DFT) data. Unlike most other ML models that use DFT data, our models directly predict the local density of states (LDOS) of the electronic structure. This provides several advantages, including access to multiple observables such as the electronic density and electronic total free energy. Moreover, our models account for both the electronic and ionic temperatures independently, making them ideal for applications like laser-heating of matter. We validate the efficacy of our LDOS-based models on a metallic test system. They accurately capture energetic effects induced by variations in ionic and electronic temperatures over a broad temperature range, even when trained on a subset of these temperatures. These findings open up exciting opportunities for investigating the electronic structure of materials under both ambient and extreme conditions.
△ Less
Submitted 17 September, 2023; v1 submitted 9 June, 2023;
originally announced June 2023.
-
Predicting electronic structures at any length scale with machine learning
Authors:
Lenz Fiedler,
Normand A. Modine,
Steve Schmerler,
Dayton J. Vogel,
Gabriel A. Popoola,
Aidan P. Thompson,
Sivasankaran Rajamanickam,
Attila Cangi
Abstract:
The properties of electrons in matter are of fundamental importance. They give rise to virtually all molecular and material properties and determine the physics at play in objects ranging from semiconductor devices to the interior of giant gas planets. Modeling and simulation of such diverse applications rely primarily on density functional theory (DFT), which has become the principal method for p…
▽ More
The properties of electrons in matter are of fundamental importance. They give rise to virtually all molecular and material properties and determine the physics at play in objects ranging from semiconductor devices to the interior of giant gas planets. Modeling and simulation of such diverse applications rely primarily on density functional theory (DFT), which has become the principal method for predicting the electronic structure of matter. While DFT calculations have proven to be very useful to the point of being recognized with a Nobel prize in 1998, their computational scaling limits them to small systems. We have developed a machine learning framework for predicting the electronic structure on any length scale. It shows up to three orders of magnitude speedup on systems where DFT is tractable and, more importantly, enables predictions on scales where DFT calculations are infeasible. Our work demonstrates how machine learning circumvents a long-standing computational bottleneck and advances science to frontiers intractable with any current solutions. This unprecedented modeling capability opens up an inexhaustible range of applications in astrophysics, novel materials discovery, and energy solutions for a sustainable future.
△ Less
Submitted 8 December, 2022; v1 submitted 20 October, 2022;
originally announced October 2022.
-
Electron dynamics in extended systems within real-time time-dependent density functional theory
Authors:
Alina Kononov,
Cheng-Wei Lee,
Tatiane Pereira dos Santos,
Brian Robinson,
Yifan Yao,
Yi Yao,
Xavier Andrade,
Andrew David Baczewski,
Emil Constantinescu,
Alfredo A. Correa,
Yosuke Kanai,
Normand Modine,
Andre Schleife
Abstract:
Due to a beneficial balance of computational cost and accuracy, real-time time-dependent density functional theory has emerged as a promising first-principles framework to describe electron real-time dynamics. Here we discuss recent implementations around this approach, in particular in the context of complex, extended systems. Results include an analysis of the computational cost associated with…
▽ More
Due to a beneficial balance of computational cost and accuracy, real-time time-dependent density functional theory has emerged as a promising first-principles framework to describe electron real-time dynamics. Here we discuss recent implementations around this approach, in particular in the context of complex, extended systems. Results include an analysis of the computational cost associated with numerical propagation and when using absorbing boundary conditions. We extensively explore the shortcomings for describing electron-electron scattering in real time and compare to many-body perturbation theory. Modern improvements of the description of exchange and correlation are reviewed. In this work, we specifically focus on the Qb@ll code, which we have mainly used for these types of simulations over the last years, and we conclude by pointing to further progress needed going forward.
△ Less
Submitted 9 May, 2022;
originally announced May 2022.
-
Accelerating Finite-temperature Kohn-Sham Density Functional Theory with Deep Neural Networks
Authors:
J. Austin Ellis,
Lenz Fiedler,
Gabriel A. Popoola,
Normand A. Modine,
J. Adam Stephens,
Aidan P. Thompson,
Attila Cangi,
Sivasankaran Rajamanickam
Abstract:
We present a numerical modeling workflow based on machine learning (ML) which reproduces the the total energies produced by Kohn-Sham density functional theory (DFT) at finite electronic temperature to within chemical accuracy at negligible computational cost. Based on deep neural networks, our workflow yields the local density of states (LDOS) for a given atomic configuration. From the LDOS, spat…
▽ More
We present a numerical modeling workflow based on machine learning (ML) which reproduces the the total energies produced by Kohn-Sham density functional theory (DFT) at finite electronic temperature to within chemical accuracy at negligible computational cost. Based on deep neural networks, our workflow yields the local density of states (LDOS) for a given atomic configuration. From the LDOS, spatially-resolved, energy-resolved, and integrated quantities can be calculated, including the DFT total free energy, which serves as the Born-Oppenheimer potential energy surface for the atoms. We demonstrate the efficacy of this approach for both solid and liquid metals and compare results between independent and unified machine-learning models for solid and liquid aluminum. Our machine-learning density functional theory framework opens up the path towards multiscale materials modeling for matter under ambient and extreme conditions at a computational scale and cost that is unattainable with current algorithms.
△ Less
Submitted 9 July, 2021; v1 submitted 10 October, 2020;
originally announced October 2020.
-
Kinetics-Controlled Degradation Reactions at Crystalline LiPON/Li(x)CoO(2) and Crystalline LiPON/Li-metal Interfaces
Authors:
Kevin Leung,
Alexander J. Pearse,
A. Alec Talin,
Elliot J. Fuller,
Gary W. Rubloff,
Normand A. Modine
Abstract:
Detailed understanding of solid-solid interface structure-function relations is critical for the improvement and wide deployment of all solid-state batteries. The interfaces between lithium phosphorous oxynitride ("LiPON") solid electrolyte material and lithium metal anode, between LiPON and Li(x)CoO(2) cathode surfaces, have been reported to generate solid electrolyte interphase ("SEI")-like prod…
▽ More
Detailed understanding of solid-solid interface structure-function relations is critical for the improvement and wide deployment of all solid-state batteries. The interfaces between lithium phosphorous oxynitride ("LiPON") solid electrolyte material and lithium metal anode, between LiPON and Li(x)CoO(2) cathode surfaces, have been reported to generate solid electrolyte interphase ("SEI")-like products and/or disordered regions. Using electronic structure calculations and crystalline LiPON models with atomic-layer-deposition-like stoichiometry, we predict LiPON models with purely P-N-P backbones are kinetically inert towards lithium at room temperature. In contrast, transfer of oxygen atoms from low-energy Li(x)CoO(2) (104) surfaces to LiPON is much faster under ambient conditions. The mechanisms of the primary reaction steps, LiPON motifs that readily react with lithium metal, experimental results on amorphous LiPON to partially corroborate these predictions, and possible mitigation strategies to reduce degradations are discussed. LiPON interfaces are found to be useful case studies for highlighting the importance of kinetics-controlled processes during battery assembly at moderate processing temperatures.
△ Less
Submitted 23 June, 2018;
originally announced June 2018.
-
Graphene/MoS2 van der Waals Bilayer as the Anode Material for Next Generation Li-ion Battery: A First-Principles Investigation
Authors:
Towfiq Ahmed,
N. A. Modine,
Jian-Xin Zhu
Abstract:
We performed density functional theory (DFT) calculations for a bi-layered heterostructure combining a graphene layer with a MoS2 layer with and without intercalated Li atoms. Our calculations demonstrate the importance of the van der Waals (vdW) interaction, which is crucial for forming stable bonding between the layers. Our DFT calculation correctly reproduces the linear dispersion, or Dirac con…
▽ More
We performed density functional theory (DFT) calculations for a bi-layered heterostructure combining a graphene layer with a MoS2 layer with and without intercalated Li atoms. Our calculations demonstrate the importance of the van der Waals (vdW) interaction, which is crucial for forming stable bonding between the layers. Our DFT calculation correctly reproduces the linear dispersion, or Dirac cone, feature at the Fermi energy for the isolated graphene monolayer and the band gap for the MoS2 monolayer. For the combined graphene/MoS2 bi-layer, we observe interesting electronic structure and density of states (DOS) characteristics near the Fermi energy, showing both the gap like features of the MoS2 layer and in-gap states with linear dispersion contributed mostly by the graphene layer. Our calculated total density of states (DOS) in this vdW heterostructure reveals that the graphene layer significantly contributes to pinning the Fermi energy at the center of the band gap of MoS2. We also find that intercalating Li ions in between the layers of the graphene/MoS2 heterostructure enhances the binding energy through orbital hybridizations between cations (Li adatoms) and anions (graphene and MoS2 monolayers). Moreover, we calculate the dielectric function of the Li intercalated graphene/MoS2 heterostructure, the imaginary component of which can be directly compared with experimental measurements of optical conductivity in order to validate our theoretical prediction. We observe sharp features in the imaginary component of the dielectric function, which shows the presence of a Drude peak in the optical conductivity, and therefore metallicity in the lithiated graphene/MoS2 heterostructure.
△ Less
Submitted 25 February, 2015;
originally announced February 2015.
-
HARES: an efficient method for first-principles electronic structure calculations of complex systems
Authors:
U. V. Waghmare,
Hanchul Kim,
I. J. Park,
Normand Modine,
P. Maragakis,
Efthimios Kaxiras
Abstract:
We discuss our new implementation of the Real-space Electronic Structure method for studying the atomic and electronic structure of infinite periodic as well as finite systems, based on density functional theory. This improved version which we call HARES (for High-performance-fortran Adaptive grid Real-space Electronic Structure) aims at making the method widely applicable and efficient, using h…
▽ More
We discuss our new implementation of the Real-space Electronic Structure method for studying the atomic and electronic structure of infinite periodic as well as finite systems, based on density functional theory. This improved version which we call HARES (for High-performance-fortran Adaptive grid Real-space Electronic Structure) aims at making the method widely applicable and efficient, using high performance Fortran on parallel architectures. The scaling of various parts of a HARES calculation is analyzed and compared to that of plane-wave based methods. The new developments that lead to enhanced performance, and their parallel implementation, are presented in detail. We illustrate the application of HARES to the study of elemental crystalline solids, molecules and complex crystalline materials, such as blue bronze and zeolites.
△ Less
Submitted 22 December, 2000; v1 submitted 10 June, 2000;
originally announced June 2000.
-
Variational Hilbert space truncation approach to quantum Heisenberg antiferromagnets on frustrated clusters
Authors:
N. A. Modine,
Efthimios Kaxiras
Abstract:
We study the spin-$\frac{1}{2}$ Heisenberg antiferromagnet on a series of finite-size clusters with features inspired by the fullerenes. Frustration due to the presence of pentagonal rings makes such structures challenging in the context of quantum Monte-Carlo methods. We use an exact diagonalization approach combined with a truncation method in which only the most important basis states of the…
▽ More
We study the spin-$\frac{1}{2}$ Heisenberg antiferromagnet on a series of finite-size clusters with features inspired by the fullerenes. Frustration due to the presence of pentagonal rings makes such structures challenging in the context of quantum Monte-Carlo methods. We use an exact diagonalization approach combined with a truncation method in which only the most important basis states of the Hilbert space are retained. We describe an efficient variational method for finding an optimal truncation of a given size which minimizes the error in the ground state energy. Ground state energies and spin-spin correlations are obtained for clusters with up to thirty-two sites without the need to restrict the symmetry of the structures. The results are compared to full-space calculations and to unfrustrated structures based on the honeycomb lattice.
△ Less
Submitted 17 July, 1995;
originally announced July 1995.
-
Adaptive coordinate, real-space electronic structure calculations on parallel computers
Authors:
Gil Zumbach,
N. A. Modine,
Efthimios Kaxiras
Abstract:
We present a method for electronic structure calculations that retains all of the advantages of real space and addresses the inherent inefficiency of a regular grid, which has equal precision everywhere. The computations are carried out on a {\em regular} mesh in {\em curvilinear space}, which allows natural and efficient decomposition on parallel computers, and effective use of iterative numeri…
▽ More
We present a method for electronic structure calculations that retains all of the advantages of real space and addresses the inherent inefficiency of a regular grid, which has equal precision everywhere. The computations are carried out on a {\em regular} mesh in {\em curvilinear space}, which allows natural and efficient decomposition on parallel computers, and effective use of iterative numerical methods. A novel feature is the use of error analysis to optimize the curvilinear grid for highly inhomogeneous electronic distributions. We report accurate all-electron calculations for H$_2$, O, and O$_{2}$.
△ Less
Submitted 19 June, 1995;
originally announced June 1995.
