arXiv subjects

Bjørk Hammer

Publications and source records attributed to Bjørk Hammer.

32 records · Page 2Linked to original sources

arXiv2021

Generating stable molecules using imitation and reinforcement learning

Chemical space is routinely explored by machine learning methods to discover interesting molecules, before time-consuming experimental synthesizing is attempted. However, these methods often rely on a graph representation, ignoring 3D information necessary for determining the stability of the molecules. We propose a reinforcement learning approach for generating molecules in cartesian coordinates allowing for quantum chemical prediction of the stability. To improve sample-efficiency we learn basic chemical rules from imitation learning on the GDB-11 database to create an initial model applicable for all stoichiometries. We then deploy multiple copies of the model conditioned on a specific stoichiometry in a reinforcement learning setting. The models correctly identify low energy molecules in the database and produce novel isomers not found in the training set. Finally, we apply the model to larger molecules to show how reinforcement learning further refines the imitation learning model in domains far from the training data.

physics.chem-ph↗

arXiv2020

Global optimization of atomistic structure enhanced by machine learning

Global Optimization with First-principles Energy Expressions (GOFEE) is an efficient method for identifying low energy structures in computationally expensive energy landscapes such as the ones described by density functional theory (DFT), van der Waals-enabled DFT, or even methods beyond DFT. GOFEE relies on a machine learned surrogate model of energies and forces, trained on-the-fly, to explore configuration space, eliminating the need for expensive relaxations of all candidate structures using first-principles methods. In this paper we elaborate on the importance of the use of a Gaussian kernel with two length scales in the Gaussian Process Regression (GPR) surrogate model. We further explore the role of the use in GOFEE of the lower confidence bound for relaxation and selection of candidate structures. In addition, we present two improvements to the method: 1) the population generation now relies on a clustering of all low-energy structures evaluated with DFT, with the lowest energy member of each cluster making up the population. 2) the very final relaxations in well-sampled basins of the energy landscape, the final exploitation steps, are now performed as continued relaxation paths within the first-principles method, to allow for arbitrarily fine relaxations of the best structures, independently of the predictive resolution of the surrogate model. The versatility of the GOFEE method is demonstrated by applying it to identify the low-energy structures of gas-phase fullerene-type 24-atom carbon clusters and of dome-shaped 18-atom carbon clusters supported on Ir(111).

physics.chem-ph↗

arXiv2020

Atomistic Structure Learning Algorithm with surrogate energy model relaxation

The recently proposed Atomistic Structure Learning Algorithm (ASLA) builds on neural network enabled image recognition and reinforcement learning. It enables fully autonomous structure determination when used in combination with a first-principles total energy calculator, e.g. a density functional theory (DFT) program. To save on the computational requirements, ASLA utilizes the DFT program in a single-point mode, i.e. without allowing for relaxation of the structural candidates according to the force information at the DFT level. In this work, we augment ASLA to establish a surrogate energy model concurrently with its structure search. This enables approximative but computationally cheap relaxation of the structural candidates before the single-point energy evaluation with the computationally expensive DFT program. We demonstrate a significantly increased performance of ASLA for building benzene while utilizing a surrogate energy landscape. Further we apply this model-enhanced ASLA in a thorough investigation of the c(4x8) phase of the Ag(111) surface oxide. ASLA successfully identifies a surface reconstruction which has previously only been guessed on the basis of scanning tunnelling microscopy images.

cond-mat.mtrl-sci↗

arXiv2019

Efficient global structure optimization with a machine learned surrogate model

We propose a scheme for global optimization with first-principles energy expressions (GOFEE) of atomistic structure. While unfolding its search, the method actively learns a surrogate model of the potential energy landscape on which it performs a number of local relaxations (exploitation) and further structural searches (exploration). Assuming Gaussian Processes, an acquisition function is used to decide on which of the resulting structures is the more promising. Subsequently, a single point first-principles energy calculation is conducted for that structure. The method is demonstrated to outperform by two orders of magnitude a well established first-principles based evolutionary algorithm in finding surface reconstructions. Finally, GOFEE is utilized to identify initial stages of the edge oxidation and oxygen intercalation of graphene sheets on the Ir(111) surface.

physics.chem-ph↗

arXiv2019

Atomistic structure learning

One endeavour of modern physical chemistry is to use bottom-up approaches to design materials and drugs with desired properties. Here we introduce an atomistic structure learning algorithm (ASLA) that utilizes a convolutional neural network to build 2D compounds and layered structures atom by atom. The algorithm takes no prior data or knowledge on atomic interactions but inquires a first-principles quantum mechanical program for physical properties. Using reinforcement learning, the algorithm accumulates knowledge of chemical compound space for a given number and type of atoms and stores this in the neural network, ultimately learning the blueprint for the optimal structural arrangement of the atoms for a given target property. ASLA is demonstrated to work on diverse problems, including grain boundaries in graphene sheets, organic compound formation and a surface oxide structure. This approach to structure prediction is a first step toward direct manipulation of atoms with artificially intelligent first principles computer codes.

cond-mat.mtrl-sci↗

arXiv2018

Machine learning enhanced global optimization by clustering local environments to enable bundled atomic energies

We show how to speed up global optimization of molecular structures using machine learning methods. To represent the molecular structures we introduce the auto-bag feature vector that combines: i) a local feature vector for each atom, ii) an unsupervised clustering of such feature vectors for many atoms across several structures, and iii) a count for a given structure of how many times each cluster is represented. During subsequent global optimization searches, accumulated structure-energy relations of relaxed structural candidates are used to assign local energies to each atom using supervised learning. Specifically, the local energies follow from assigning energies to each cluster of local feature vectors and demanding the sum of local energies to amount to the structural energies in the least squares sense. The usefulness of the method is demonstrated in basin hopping searches for 19-atom structures described by single- or double-well Lennard- Jones type potentials and for 24-atom carbon structures described by density functional theory (DFT). In all cases, utilizing the local energy information derived on-the-fly enhances the rate at which the global minimum energy structure is found.

physics.comp-ph↗

arXiv2018

Size-dependent phase transitions in MoS2 nanoparticles controlled by a metal substrate

Nanomaterials based on MoS2 are remarkably versatile; MoS2 nanoparticles are proven catalysts for processes such as hydrodesulphurization and the hydrogen evolution reaction, and transition metal dichalcogenides in general have recently emerged as novel 2D components for nanoscale electronics and optoelectronics. The properties of such materials are intimately related to their structure and dimensionality. For example, only the edges exposed by MoS2 nanoparticles (NPs) are catalytically active, and extended MoS2 systems show different character (direct or indirect gap semiconducting, or metallic) depending on their thickness and crystallographic phase. In this work, we show how particle size and interaction with a metal surface affect the stability and properties of different MoS2 NPs and the resulting phase diagrams. By means of calculations based on the Density Functional Theory (DFT), we address how support interactions affect MoS2 nanoparticles of varying size, composition, and structure. We demonstrate that interaction with Au modifies the relative stability of the different nanoparticle types so that edge terminations and crystallographic phases that are metastable for free-standing nanoparticles and monolayers are expressed in the supported system. These support-effects are strongly size-dependent due to the mismatch between Au and MoS2 lattices, which explains experimentally observed transitions in the structural phases for supported MoS2 NPs. Accounting for vdW interactions and the contraction of the Au(111) surface underneath the MoS2 is further found to be necessary for quantitatively reproducing experimental results. This work demonstrates how the properties of nanostructured MoS2 and similar layered systems can be modified by the choice of supporting metal.

cond-mat.mtrl-sci↗

arXiv2017

Contact-Induced Semiconductor-to-Metal Transition in Single-Layer WS$_2$

Low-resistance ohmic contacts are a challenge for electronic devices based on two-dimensional materials. We show that an atomically precise junction between a two-dimensional semiconductor and a metallic contact can lead to a semiconductor-to-metal transition in the two-dimensional material--a finding which points the way to a possible method of achieving low-resistance junctions. Specifically, single-layer WS$_2$ undergoes a semiconductor-to-metal transition when epitaxially grown on Ag(111), while it remains a direct band gap semiconductor on Au(111). The metallicity of the single layer on Ag(111) is established by lineshape analysis of core level photoemission spectra. Angle-resolved photoemission spectroscopy locates the metallic states near the Q point of the WS$_2$ Brillouin zone. Density functional theory calculations show that the metallic states arise from hybridization between Ag bulk bands and the local conduction band minimum of WS$_2$ near the Q point.

cond-mat.mtrl-sci↗

arXiv2016

Crystalline and electronic structure of single-layer TaS$_2$

Single-layer TaS$_2$ is epitaxially grown on Au(111) substrates. The resulting two-dimensional crystals adopt the 1H polymorph. The electronic structure is determined by angle-resolved photoemission spectroscopy and found to be in excellent agreement with density functional theory calculations. The single layer TaS$_2$ is found to be strongly n-doped, with a carrier concentration of 0.3(1) extra electrons per unit cell. No superconducting or charge density wave state is observed by scanning tunneling microscopy at temperatures down to 4.7 K.

cond-mat.str-el↗

arXiv2016

Band gap engineering by Bi intercalation of graphene on Ir(111)

We report on the structural and electronic properties of a single bismuth layer intercalated underneath a graphene layer grown on an Ir(111) single crystal. Scanning tunneling microscopy (STM) reveals a hexagonal surface structure and a dislocation network upon Bi intercalation, which we attribute to a $\sqrt{3}\times\sqrt{3}R30°$ Bi structure on the underlying Ir(111) surface. Ab-initio calculations show that this Bi structure is the most energetically favorable, and also illustrate that STM measurements are most sensitive to C atoms in close proximity to intercalated Bi atoms. Additionally, Bi intercalation induces a band gap ($E_g=0.42\,$eV) at the Dirac point of graphene and an overall n-doping ($\sim 0.39\,$eV), as seen in angular-resolved photoemission spectroscopy. We attribute the emergence of the band gap to the dislocation network which forms favorably along certain parts of the moiré structure induced by the graphene/Ir(111) interface.

cond-mat.mtrl-sci↗

arXiv2016

Single-layer MoS$_2$ on Au(111): band gap renormalization and substrate interaction

The electronic structure of epitaxial single-layer MoS$_2$ on Au(111) is investigated by angle-resolved photoemission spectroscopy, scanning tunnelling spectroscopy, and first principles calculations. While the band dispersion of the supported single-layer is close to a free-standing layer in the vicinity of the valence band maximum at $\bar{K}$ and the calculated electronic band gap on Au(111) is similar to that calculated for the free-standing layer, significant modifications to the band structure are observed at other points of the two-dimensional Brillouin zone: At $\barΓ$, the valence band maximum has a significantly higher binding energy than in the free MoS$_2$ layer and the expected spin-degeneracy of the uppermost valence band at the $\bar{M}$ point cannot be observed. These band structure changes are reproduced by the calculations and can be explained by the detailed interaction of the out-of-plane MoS$_2$ orbitals with the substrate.

cond-mat.str-el↗

arXiv2015

In Situ Detection of Active Edge Sites in Single-Layer MoS$_2$ Catalysts

MoS2 nanoparticles are proven catalysts for processes such as hydrodesulphurization and hydrogen evolution, but unravelling their atomic-scale structure under catalytic working conditions has remained significantly challenging. Ambient pressure X-ray Photoelectron Spectroscopy (AP-XPS) allows us to follow in-situ the formation of the catalytically relevant MoS2 edge sites in their active state. The XPS fingerprint is described by independent contributions to the Mo3d core level spectrum whose relative intensity is sensitive to the thermodynamic conditions. Density Functional Theory (DFT) is used to model the triangular MoS2 particles on Au(111) and identify the particular sulphidation state of the edge sites. A consistent picture emerges in which the core level shifts for the edge Mo atoms evolve counter-intuitively towards higher binding energies when the active edges are reduced. The shift is explained by a surprising alteration in the metallic character of the edge sites, which is a distinct spectroscopic signature of the MoS2 edges under working conditions.

cond-mat.mtrl-sci↗

arXiv2014

High Crystallinity and Decoupling of Graphene on a Metal: Reduced Coulomb Screening and Tunable pn-Junctions

High quality epitaxial graphene films can be applied as templates for tailoring graphene-substrate interfaces that allow for precise control of the charge carrier behavior in graphene through doping and many-body effects. By combining scanning tunneling microscopy, angle-resolved photoemission spectroscopy and density functional theory we demonstrate that oxygen intercalated epitaxial graphene on Ir(111) has high structural quality, is quasi free-standing, and shows signatures of many-body interactions. Using this system as a template, we show that tunable pn-junctions can be patterned by adsorption and intercalation of rubidium, and that the n-doped graphene regions exhibit a reduced Coulomb screening via enhanced electron-plasmon coupling. These findings are central for understanding and tailoring the properties of graphene-metal contacts e.g. for realizing quantum tunneling devices.

cond-mat.mtrl-sci↗

arXiv2011

Steps on Rutile TiO2(110): Active Sites for Water and Methanol Dissociation

We present a detailed investigation of the structure and activity of extended defects namely monoatomic steps on (1x1)-TiO2(110). Specifically, the two most stable <001> and <1-11> step edges are considered. Employing an automated genetic algorithm that samples a large number of candidates for each step edge, more stable, reconstructed structures were found for the <1-11> step edge, while the bulk truncated structures were recovered for the <001> step edge. We demonstrate how oxygen vacancies along these defects have lower formation energies than on flat terraces and how water and methanol molecules adsorb dissociatively on reduced <1-11> step edges. Our findings are in agreement with earlier experimental results and indicate an important contribution from step edges to the reactivity of the TiO2(110) surface.

cond-mat.mtrl-sci↗