Discovering and understanding materials through computation

Unsupervised representation learning of Kohn-Sham states and consequences for downstream predictions of many-body effects

Representation learning for the electronic structure problem is a major challenge of machine learning in computational condensed matter and materials physics. Within quantum mechanical first principles approaches, density functional theory (DFT) is the preeminent tool for understanding electronic structure, and the high-dimensional DFT wavefunctions serve as building blocks for downstream calculations of correlated many-body excitations and related physical observables. Here, we use variational autoencoders (VAE) for the unsupervised learning of DFT wavefunctions and show that these wavefunctions lie in a low-dimensional manifold within latent space. Our model autonomously determines the optimal representation of the electronic structure, avoiding limitations due to manual feature engineering. To demonstrate the utility of the latent space representation of the DFT wavefunction, we use it for the supervised training of neural networks (NN) for downstream prediction of quasiparticle bandstructures within the GW formalism. The GW prediction achieves a low error of 0.11 eV for a combined test set of two-dimensional metals and semiconductors, suggesting that the latent space representation captures key physical information from the original data. Finally, we explore the generative ability and interpretability of the VAE representation.

Two-Gap Superconductivity and the Decisive Role of Rare-Earth d Electrons in Infinite-Layer Nickelates

We present a theoretical prediction of a phonon-mediated two-gap superconductivity in infinite-layer nickelates Nd_{1-x}Sr_{x}NiO_{2} by performing ab initio GW and GW perturbation theory calculations. Electron GW self-energy effects significantly alter the characters of the two-band Fermi surface and enhance the electron-phonon coupling, compared with results based on density functional theory. Solutions of the fully k-dependent anisotropic Eliashberg equations yield two dominant s-wave superconducting gaps-a large gap on a band of rare-earth Nd d and interstitial orbital characters and a small gap on a band of transition-metal Ni d character. Increasing hole doping induces a non-rigid-band response in the electronic structure, leading to a rapid drop of the superconducting T_{c} in the overdoped regime in agreement with experiments.

Text-to-Microstructure Generation Using Generative Deep Learning

Designing novel materials is greatly dependent on understanding the design principles, physical mechanisms, and modeling methods of material microstructures, requiring experienced designers with expertise and several rounds of trial and error. Although recent advances in deep generative networks have enabled the inverse design of material microstructures, most studies involve property-conditional generation and focus on a specific type of structure, resulting in limited generation diversity and poor human-computer interaction. In this study, a pioneering text-to-microstructure deep generative network (Txt2Microstruct-Net) is proposed that enables the generation of 3D material microstructures directly from text prompts without additional optimization procedures. The Txt2Microstruct-Net model is trained on a large microstructure-caption paired dataset that is extensible using the algorithms provided. Moreover, the model is sufficiently flexible to generate different geometric representations, such as voxels and point clouds. The model's performance is also demonstrated in the inverse design of material microstructures and metamaterials. It has promising potential for interactive microstructure design when associated with large language models and could be a user-friendly tool for material design and discovery.

Modulating the electronic structure of ion phthalocyanine-based molecular catalysts for electrocatalytic nitrogen reduction: a DFT study

The highly localized Fe d orbital in ion phthalocyanine (FePc)-based molecular catalysts significantly hinders their electrocatalytic nitrogen reduction reaction (eNRR) performance. Herein, we theoretically designed a series of FePc-based molecules with adjacent metal phthalocyanine sites to form an asymmetric delocalized electronic structure on Fe centers, promoting the catalytic activity and lowering the overpotential of the eNRR, as well as suppressing the hydrogen evolution reaction (HER) side reaction.

In Silico Chemical Experiments in the Age of AI: From Quantum Chemistry to Machine Learning and Back

Computational chemistry is an indispensable tool for understanding molecules and predicting chemical properties. However, traditional computational methods face significant challenges due to the difficulty of solving the Schrödinger equations and the increasing computational cost with the size of the molecular system. In response, there has been a surge of interest in leveraging artificial intelligence (AI) and machine learning (ML) techniques to in silico experiments. Integrating AI and ML into computational chemistry increases the scalability and speed of the exploration of chemical space. However, challenges remain, particularly regarding the reproducibility and transferability of ML models. This review highlights the evolution of ML in learning from, complementing, or replacing traditional computational chemistry for energy and property predictions. Starting from models trained entirely on numerical data, a journey set forth toward the ideal model incorporating or learning the physical laws of quantum mechanics. This paper also reviews existing computational methods and ML models and their intertwining, outlines a roadmap for future research, and identifies areas for improvement and innovation. Ultimately, the goal is to develop AI architectures capable of predicting accurate and transferable solutions to the Schrödinger equation, thereby revolutionizing in silico experiments within chemistry and materials science.

Computational Discovery of Intermolecular Singlet Fission Materials Using Many-Body Perturbation Theory

Intermolecular singlet fission (SF) is the conversion of a photogenerated singlet exciton into two triplet excitons residing on different molecules. SF has the potential to enhance the conversion efficiency of solar cells by harvesting two charge carriers from one high-energy photon, whose surplus energy would otherwise be lost to heat. The development of commercial SF-augmented modules is hindered by the limited selection of molecular crystals that exhibit intermolecular SF in the solid state. Computational exploration may accelerate the discovery of new SF materials. The GW approximation and Bethe-Salpeter equation (GW+BSE) within the framework of many-body perturbation theory is the current state-of-the-art method for calculating the excited-state properties of molecular crystals with periodic boundary conditions. In this Review, we discuss the usage of GW+BSE to assess candidate SF materials as well as its combination with low-cost physical or machine learned models in materials discovery workflows. We demonstrate three successful strategies for the discovery of new SF materials: (i) functionalization of known materials to tune their properties, (ii) finding potential polymorphs with improved crystal packing, and (iii) exploring new classes of materials. In addition, three new candidate SF materials are proposed here, which have not been published previously.

Insights into the DHQ-BN: mechanical, electronic, and optical properties

Computational materials research is vital in improving our understanding of various class of materials and their properties, contributing valuable information that helps predict innovative structures and complement empirical investigations. In this context, DHQ-graphene recently emerged as a stable two-dimensional carbon allotrope composed of decagonal, hexagonal, and quadrilateral carbon rings. Here, we employ density functional theory calculations to investigate the mechanical, electronic, and optical features of its boron nitride counterpart (DHQ-BN). Our findings reveal an insulating band gap of 5.11 eV at the HSE06 level and good structural stability supported by phonon calculations and ab initio molecular dynamics simulations. Moreover, DHQ-BN exhibits strong ultraviolet (UV) activity, suggesting its potential as a highly efficient UV light absorber. Its mechanical properties, including Young's modulus (230 GPa) and Poisson's ratio (0.7), provide insight into its mechanical resilience and structural stability.

Data-Driven Predetermination of Cu Oxidation State in Copper Nanoparticles: Application to the Synthesis by Laser Ablation in Liquid

Copper-based nanocrystals are reference nanomaterials for integration into emerging green technologies, with laser ablation in liquid (LAL) being a remarkable technique for their synthesis. However, the achievement of a specific type of nanocrystal, among the whole library of nanomaterials available using LAL, has been until now an empirical endeavor based on changing synthesis parameters and characterizing the products. Here, we started from the bibliographic analysis of LAL synthesis of Cu-based nanocrystals to identify the relevant physical and chemical features for the predetermination of copper oxidation state. First, single features and their combinations were screened by linear regression analysis, also using a genetic algorithm, to find the best correlation with experimental output and identify the equation giving the best prediction of the LAL results. Then, machine learning (ML) models were exploited to unravel cross-correlations between features that are hidden in the linear regression analysis. Although the LAL-generated Cu nanocrystals may be present in a range of oxidation states, from metallic copper to cuprous oxide (Cu2O) and cupric oxide (CuO), in addition to the formation of other materials such as Cu2S and CuCN, ML was able to guide the experiments toward the maximization of the compounds in the greatest demand for integration in sustainable processes. This approach is of general applicability to other nanomaterials and can help understand the origin of the chemical pathways of nanocrystals generated by LAL, providing a rational guideline for the conscious predetermination of laser-synthesis parameters toward the desired compounds.

Accelerating the Design of High-Energy-Density Hydrocarbon Fuels by Learning from the Data

In the ZINC20 database, with the aid of maximum substructure searches, common substructures were obtained from molecules with high-strain-energy and combustion heat values, and further provided domain knowledge on how to design high-energy-density hydrocarbon (HEDH) fuels. Notably, quadricyclane and syntin could be topologically assembled through these substructures, and the corresponding assembled schemes guided the design of 20 fuel molecules (ZD-1 to ZD-20). The fuel properties of the molecules were evaluated by using group-contribution methods and density functional theory (DFT) calculations, where ZD-6 stood out due to the high volumetric net heat of combustion, high specific impulse, low melting point, and acceptable flash point. Based on the neural network model for evaluating the synthetic complexity (SCScore), the estimated value of ZD-6 was close to that of syntin, indicating that the synthetic complexity of ZD-6 was comparable to that of syntin. This work not only provides ZD-6 as a potential HEDH fuel, but also illustrates the superiority of learning design strategies from the data in increasing the understanding of structure and performance relationships and accelerating the development of novel HEDH fuels.

Block sparsity promoting algorithm for efficient construction of cluster expansion models for multicomponent alloys

Multicomponent alloys are gaining significance as drivers of technological breakthroughs especially in structural and energy storage materials. The vast configuration space of these materials prohibit computational modeling using first-principles based methods alone. The cluster expansion (CE) method is the most widely used tool for modeling configurational disorder in alloys. CE relies on machine learning algorithms to train Hamiltonians and uses first-principles calculated data as training sets. In this paper we present a new compressive sensing-based algorithm for the efficient construction of CE Hamiltonians of multicomponent alloys. Our algorithm constructs highly sparse and physically reasonable models from a carefully selected small training set of alloy structures. Compared to conventional fitting algorithms, the algorithm achieves more than 50% reduction in the training set size. The resultant sparse models can sample the configuration space at least 3 × faster. We demonstrate this algorithm on 4 different alloy systems, namely Ag-Au, Ag-Au-Cu, Ag-Au-Cu-Pd and (Ge,Sn)(S,Se,Te).The sparse CE models for these alloys can rapidly reproduce known ground state orderings and order-disorder transitions. Our method can truly enable high-throughput multicomponent alloy thermodynamics by reducing the cost associated with model construction and configuration sampling.

On the mechanical, electronic, and optical properties of the boron nitride analog for the recently synthesized biphenylene network: a DFT study

CONTEXT

Recently, a new 2D carbon allotrope named biphenylene network (BPN) was experimentally realized. Here, we use density functional theory (DFT) calculations to study its boron nitride analogue sheet's structural, electronic, and optical properties (BN-BPN). Results suggest that BN-BPN has good structural and dynamic stabilities. It also has a direct bandgap of 4.5 eV and significant optical activity in the ultraviolet range. BN-BPN Young's modulus varies between 234.4[Formula: see text]273.2 GPa depending on the strain direction.

METHODS

Density functional theory (DFT) simulations for the electronic and optical properties of BN-BPN were performed using the CASTEP package within the Biovia Materials Studio software. The exchange and correlation functions are treated within the generalized gradient approximation (GGA) as parameterized by Perdew-Burke-Ernzerhof (PBE) and the hybrid functional Heyd-Scuseria-Ernzerhof (HSE06). For convenience, the mechanical properties were carried out using the DFT approach implemented in the SIESTA code, also within the scope of the GGA/PBE method. We used the double-zeta plus polarization (DZP) for the basis set in these cases. Moreover, the norm-conserving Troullier-Martins pseudopotential was employed to describe the core electrons.

Reversible assembly of nanoparticles: theory, strategies and computational simulations

The significant advances in synthesis and functionalization have enabled the preparation of high-quality nanoparticles that have found a plethora of successful applications. The unique physicochemical properties of nanoparticles can be manipulated through the control of size, shape, composition, and surface chemistry, but their technological application possibilities can be further expanded by exploiting the properties that emerge from their assembly. The ability to control the assembly of nanoparticles not only is required for many real technological applications, but allows the combination of the intrinsic properties of nanoparticles and opens the way to the exploitation of their complex interplay, giving access to collective properties. Significant advances and knowledge gained over the past few decades on nanoparticle assembly have made it possible to implement a growing number of strategies for reversible assembly of nanoparticles. In addition to being of interest for basic studies, such advances further broaden the range of applications and the possibility of developing innovative devices using nanoparticles. This review focuses on the reversible assembly of nanoparticles and includes the theoretical aspects related to the concept of reversibility, an up-to-date assessment of the experimental approaches applied to this field and the advanced computational schemes that offer key insights into the assembly mechanisms. We aim to provide readers with a comprehensive guide to address the challenges in assembling reversible nanoparticles and promote their applications.

Theoretical Progress of 2D Six-Membered-Ring Inorganic Materials as Anodes for Non-Lithium-Ion Batteries

The use and storage of renewable and clean energy has become an important trend due to resource depletion, environmental pollution, and the rising price of refined fossil fuels. Confined by the limited resource and uneven distribution of lithium, non-lithium-ion batteries have become a new focus for energy storage. The six-membered-ring (SMR) is a common structural unit for numerous material systems. 2D SMR inorganic materials have unique advantages in the field of non-lithium energy storage, such as fast electrochemical reactions, abundant active sites and adjustable band gap. First-principles calculations based on density functional theory (DFT) can provide a basic understanding of materials at the atomic-level and establish the relationship between SMR structural units and electrochemical energy storage. In this review, the theoretical progress of 2D SMR inorganic materials in the field of non-lithium-ion batteries in recent years is discussed to summarize the common relationship among 2D SMR non-lithium energy storage anodes. Finally, the existing challenges are analyzed and potential solutions are proposed.

Multi-scale computer-aided design and photo-controlled macromolecular synthesis boosting uranium harvesting from seawater

By integrating multi-scale computational simulation with photo-regulated macromolecular synthesis, this study presents a new paradigm for smart design while customizing polymeric adsorbents for uranium harvesting from seawater. A dissipative particle dynamics (DPD) approach, combined with a molecular dynamics (MD) study, is performed to simulate the conformational dynamics and adsorption process of a model uranium grabber, i.e., PAOm-b-PPEGMAn, suggesting that the maximum adsorption capacity with atomic economy can be achieved with a preferred block ratio of 0.18. The designed polymers are synthesized using the PET-RAFT polymerization in a microfluidic platform, exhibiting a record high adsorption capacity of uranium (11.4 ± 1.2 mg/g) in real seawater within 28 days. This study offers an integrated perspective to quantitatively assess adsorption phenomena of polymers, bridging metal-ligand interactions at the molecular level with their spatial conformations at the mesoscopic level. The established protocol is generally adaptable for target-oriented development of more advanced polymers for broadened applications.

Designing 3d metal oxides: selecting optimal density functionals for strongly correlated materials

Transition metal oxides (TMOs) have remarkable physicochemical properties, are non-toxic, and have low cost and high annual production, thus they are commonly studied for various technological applications. Density functional theory (DFT) can help to optimize TMO materials by providing insights into their electronic, optical and thermodynamic properties, and hence into their structure-performance relationships, over a wide range of solid-state structures and compositions. However, this is underpinned by the choice of the exchange-correlation (XC) functional, which is critical to accurately describe the highly localized and correlated 3d-electrons of the transition metals in TMOs. This tutorial review presents a benchmark study of density functionals (DFs), ranging from generalized gradient approximation (GGA) to range-separated hybrids (RSH), with the all-electron def2-TZVP basis set, comparing magneto-electro-optical properties of 3d TMOs against experimental observations. The performance of the DFs is assessed by analyzing the band structure, density of states, magnetic moment, structural static and dynamic parameters, optical properties, spin contamination and computational cost. The results disclose the strengths and weaknesses of the XC functionals, in terms of accuracy, and computational efficiency, suggesting the unprecedented PBE0-1/5 as the best candidate. The findings of this work contribute to necessary developments of XC functionals for periodic systems, and materials science modelling studies, particularly informing how to select the optimal XC functional to obtain the most trustworthy description of the ground-state electron structure of 3d TMOs.

Lattice dynamical properties of antiferromagnetic oxides calculated using self-consistent extended Hubbard functional method

We study the lattice dynamics of antiferromagnetic transition-metal oxides by using self-consistent Hubbard functionals. We calculate the ground states of the oxides with the on-site and intersite Hubbard interactions determined self-consistently within the framework of density functional theory. The on-site and intersite Hubbard terms fix the errors associated with the electron self-interaction in the local and semilocal functionals. Inclusion of the intersite Hubbard terms in addition to the on-site Hubbard terms produces accurate phonon dispersion of the transition-metal oxides. Calculated Born effective charges and high-frequency dielectric constants are in good agreement with experiment. Our study provides a computationally inexpensive and accurate set of first-principles calculations for strongly-correlated materials and related phenomena.

Materials Data toward Machine Learning: Advances and Challenges

Machine learning (ML) is believed to have enabled a paradigm shift in materials research, and in practice, ML has demonstrated its power in speeding up the cost-efficient discovery of new materials and autonomizing materials laboratories. In this Perspective, current research progress in materials data which are the backbones of ML are reviewed, focusing on high-throughput data generation, standardized data storage, and data representation. More importantly, the challenging issues in materials data that should be overcome to unlock the full potential of ML in materials research and development, including classic 5V (volume, velocity, variety, veracity, and value) issues, 3M (multicomponent, multiscale, and multistage) challenges, co-mining of experimental and computational data, and materials data toward transferable/explainable ML or causal ML, are discussed.

Understanding Contact Electrification at Water/Polymer Interface

Contact electrification (CE) involves a complex interplay of physical interactions in realistic material systems. For this reason, scientific consensus on the qualitative and quantitative importance of different physical mechanisms on CE remains a formidable task. The CE mechanism at a water/polymer interface is a crucial challenge owing to the poor understanding of charge transfer at the atomic level. First-principle density functional theory (DFT), used in the present work, proposes a new paradigm to address CE. Our results indicate that CE follows the same trend as the gap between the highest occupied and lowest unoccupied molecular orbitals (HOMO and LUMO) of polymers. Electron transfer occurs at the outmost atomic layer of the water/polymer interface and is closely linked to the functional groups and atom locations. When the polymer chains are parallel to the water layer, most electrons are transferred; conversely, if they are perpendicular to each other, the transfer of charges can be ignored. We demonstrate that a decrease in the interface distance between water and the polymer chains leads to CE in quantitative agreement with the electron cloud overlap model. We finally use DFT calculations to predict the properties of CE materials and their potential for triboelectric nanogenerator energy harvesting devices.

Simple Structural Descriptor Obtained from Symbolic Classification for Predicting the Oxygen Vacancy Defect Formation of Perovskites

Symbolic classification is an approach of interpretable machine learning for building mathematical formulas that fit certain data sets. In this work, symbolic classification is used to establish the relationship between oxygen vacancy defect formation energy and structural features. We find a structural descriptor n_a(r_a/E_na - r_b), where n_a is the valence of the a-site ion, r_a is the radius of the a-site ion, E_na is the electronegativity of the a-site ion, and r_b is the radius of the b-site ion. It accelerates the screening of defect-free oxide perovskites in advance of density functional theory (DFT) calculations and experimental characterization. Our results demonstrate the potential of symbolic classification for accelerating the data-driven design and discovery of materials with improved properties.

Gaussian Process Regression for Materials and Molecules

We provide an introduction to Gaussian process regression (GPR) machine-learning methods in computational materials science and chemistry. The focus of the present review is on the regression of atomistic properties: in particular, on the construction of interatomic potentials, or force fields, in the Gaussian Approximation Potential (GAP) framework; beyond this, we also discuss the fitting of arbitrary scalar, vectorial, and tensorial quantities. Methodological aspects of reference data generation, representation, and regression, as well as the question of how a data-driven model may be validated, are reviewed and critically discussed. A survey of applications to a variety of research questions in chemistry and materials science illustrates the rapid growth in the field. A vision is outlined for the development of the methodology in the years to come.

Ab Initio Quantum-Mechanical Predictions of Semiconducting Photocathode Materials

Ab initio Quantum-Mechanical methods are well-established tools for material characterization and discovery in many technological areas. Recently, state-of-the-art approaches based on density-functional theory and many-body perturbation theory were successfully applied to semiconducting alkali antimonides and tellurides, which are currently employed as photocathodes in particle accelerator facilities. The results of these studies have unveiled the potential of ab initio methods to complement experimental and technical efforts for the development of new, more efficient materials for vacuum electron sources. Concomitantly, these findings have revealed the need for theory to go beyond the status quo in order to face the challenges of modeling such complex systems and their properties in operando conditions. In this review, we summarize recent progress in the application of ab initio many-body methods to investigate photocathode materials, analyzing the merits and the limitations of the standard approaches with respect to the confronted scientific questions. In particular, we emphasize the necessary trade-off between computational accuracy and feasibility that is intrinsic to these studies, and propose possible routes to optimize it. We finally discuss novel schemes for computationally-aided material discovery that are suitable for the development of ultra-bright electron sources toward the incoming era of artificial intelligence.