Search

X-ray absorption spectroscopy (XAS) is a powerful characterization technique for probing the local chemical environment of absorbing atoms. However, analyzing XAS data presents significant challenges, often requiring extensive, computationally intensive simulations, as well as significant domain expertise. These limitations hinder the development of fast, robust XAS analysis pipelines that are essential in high-throughput studies and for autonomous experimentation. We address these challenges with OmniXAS, a framework that contains a suite of transfer learning approaches for XAS prediction, each contributing to improved accuracy and efficiency, as demonstrated on K-edge spectra database covering eight 3d transition metals (Ti-Cu). The OmniXAS framework is built upon three distinct strategies. First, we use M3GNet to derive latent representations of the local chemical environment of absorption sites as input for XAS prediction, achieving up to order-of-magnitude improvements over conventional featurization techniques. Second, we employ a hierarchical transfer learning strategy, training a universal multi-task model across elements before fine-tuning for element-specific predictions. Models based on this cascaded approach after element-wise fine-tuning outperform element-specific models by up to 69%. Third, we implement cross-fidelity transfer learning, adapting a universal model to predict spectra generated by simulation of a different fidelity with a higher computational cost. This approach improves prediction accuracy by up to 11% over models trained on the target fidelity alone. Our approach boosts the throughput of XAS modeling by orders of magnitude versus first-principles simulations and is extendable to XAS prediction for a broader range of elements. This transfer learning framework is generalizable to enhance deep-learning models that target other properties in materials research., Comment: Main manuscript: 22 pages, 11 figures. Supplemental material (12 pages, 6 figures) available as a separate file in arXiv ancillary files (additional downloadable files)

The ability to rapidly develop materials with desired properties has a transformative impact on a broad range of emerging technologies. In this work, we introduce a new framework based on the diffusion model, a recent generative machine learning method to predict 3D structures of disordered materials from a target property. For demonstration, we apply the model to identify the atomic structures of amorphous carbons ($a$-C) as a representative material system from the target X-ray absorption near edge structure (XANES) spectra--a common experimental technique to probe atomic structures of materials. We show that conditional generation guided by XANES spectra reproduces key features of the target structures. Furthermore, we show that our model can steer the generative process to tailor atomic arrangements for a specific XANES spectrum. Finally, our generative model exhibits a remarkable scale-agnostic property, thereby enabling generation of realistic, large-scale structures through learning from a small-scale dataset (i.e., with small unit cells). Our work represents a significant stride in bridging the gap between materials characterization and atomic structure determination; in addition, it can be leveraged for materials discovery in exploring various material properties as targeted.

Heterostructures of topological insulator Bi$_2$Se$_3$ on transition metal dichalcogenides (TMDCs) offer a new materials platform for studying novel quantum states by exploiting the interplay among topological orders, charge orders and magnetic orders. The diverse interface attributes, such as material combination, charge re-arrangement, defect and strain, can be utilized to manipulate the quantum properties of this class of materials. Recent experiments of Bi$_2$Se$_3$/NbSe$_2$ heterostructures show signatures of strong Rashba band splitting due to the presence of a BiSe buffer layer, but the atomic level mechanism is not fully understood. We conduct first-principles studies of the Bi$_2$Se$_3$/BiSe/TMDC heterostructures with five different TMDC substrates (1T phase VSe$_2$, MoSe$_2$, TiSe$_2$, and 2H phase NbSe$_2$, MoSe$_2$). We find significant charge transfer at both BiSe/TMDC and Bi$_2$Se$_3$/BiSe interfaces driven by the work function difference, which stabilizes the BiSe layer as an electron donor and creates interface dipole. The electric field of the interface dipole breaks the inversion symmetry in the Bi$_2$Se$_3$ layer, leading to the giant Rashba band splitting in two quintuple layers and the recovery of the Dirac point in three quintuple layers, with the latter otherwise only occurring in thicker samples with at least six Bi$_2$Se$_3$ quintuple layers. Besides, we find that strain can significantly affect the charge transfer at the interfaces. Our study presents a promising avenue for tuning topological properties in heterostructures of two-dimensional materials, with potential applications in quantum devices., Comment: 10 pages, 6 figures

Electrochemical degradation of solid electrolytes is a major roadblock in the development of solid-state batteries, and the formed solid-solid interphase (SSI) plays a key role in the performance of solid-state batteries. In this study, by combining experimental X-ray absorption spectroscopy (XAS) measurements, first-principles simulations, and unsupervised machine learning, we have unraveled the atomic-scale oxidative degradation mechanisms of sulfide electrolytes at the interface using the baseline Li3PS4 (LPS) electrolyte as a model system. The degradation begins with a decrease of Li neighbor affinity to S atoms upon initial delithiation, followed by the formation of S-S bonds as the PS4 tetrahedron deforms. After the first delithiation cycle, the PS4 motifs become strongly distorted and PS3 motifs start to form. Spectral fingerprints of the local structural evolution are identified, which correspond to the main peak broadening and the peak shifting to a higher energy by about 2.5 eV in P K-edge XAS and a new peak emerging at 2473 eV in S K-edge XAS during delithiation. The spectral fingerprints serve as a proxy for the electrochemical stability of phosphorus sulfide solid electrolytes beyond LPS, as demonstrated in argyrodite Li6PS5Cl. We observed that the strong distortion and destruction of PS4 tetrahedra and the formation of S-S bonds are correlated with an increased interfacial impedance. To the best of our knowledge, this study showcases the first atomic-scale insights into the oxidative degradation mechanism of the LPS electrolyte, which can provide guidance for controlling macroscopic reactions through microstructural engineering and, more generally, can advance the rational design of sulfide electrolytes.

Electron energy loss spectroscopy (EELS) and X-ray absorption spectroscopy (XAS) provide detailed information about bonding, distributions and locations of atoms, and their coordination numbers and oxidation states. However, analysis of XAS/EELS data often relies on matching an unknown experimental sample to a series of simulated or experimental standard samples. This limits analysis throughput and the ability to extract quantitative information from a sample. In this work, we have trained a random forest model capable of predicting the oxidation state of copper based on its L-edge spectrum. Our model attains an $R^2$ score of 0.85 and a root mean square valence error of 0.24 on simulated data. It has also successfully predicted experimental L-edge EELS spectra taken in this work and XAS spectra extracted from the literature. We further demonstrate the utility of this model by predicting simulated and experimental spectra of mixed valence samples generated by this work. This model can be integrated into a real time EELS/XAS analysis pipeline on mixtures of copper containing materials of unknown composition and oxidation state. By expanding the training data, this methodology can be extended to data-driven spectral analysis of a broad range of materials.

Fast screening of drug molecules based on the ligand binding affinity is an important step in the drug discovery pipeline. Graph neural fingerprint is a promising method for developing molecular docking surrogates with high throughput and great fidelity. In this study, we built a COVID-19 drug docking dataset of about 300,000 drug candidates on 23 coronavirus protein targets. With this dataset, we trained graph neural fingerprint docking models for high-throughput virtual COVID-19 drug screening. The graph neural fingerprint models yield high prediction accuracy on docking scores with the mean squared error lower than $0.21$ kcal/mol for most of the docking targets, showing significant improvement over conventional circular fingerprint methods. To make the neural fingerprints transferable for unknown targets, we also propose a transferable graph neural fingerprint method trained on multiple targets. With comparable accuracy to target-specific graph neural fingerprint models, the transferable model exhibits superb training and data efficiency. We highlight that the impact of this study extends beyond COVID-19 dataset, as our approach for fast virtual ligand screening can be easily adapted and integrated into a general machine learning-accelerated pipeline to battle future bio-threats., Comment: 8 pages, 5 figures, 2 tables, accepted by ICLMA2023

Accurate classification of molecular chemical motifs from experimental measurement is an important problem in molecular physics, chemistry and biology. In this work, we present neural network ensemble classifiers for predicting the presence (or lack thereof) of 41 different chemical motifs on small molecules from simulated C, N and O K-edge X-ray absorption near-edge structure (XANES) spectra. Our classifiers not only reach a maximum average class-balanced accuracy of 0.99 but also accurately quantify uncertainty. We also show that including multiple XANES modalities improves predictions notably on average, demonstrating a "multi-modal advantage" over any single modality. In addition to structure refinement, our approach can be generalized for broad applications with molecular design pipelines.

X-ray absorption spectroscopy (XAS) is an element-specific materials characterization technique that is sensitive to structural and electronic properties. First-principles simulated XAS has been widely used as a powerful tool to interpret experimental spectra and draw physical insights. Recently, there has also been growing interest in building computational XAS databases to enable data analytics and machine learning applications. However, there are non-trivial differences among commonly used XAS simulation codes, both in underlying theoretical formalism and in technical implementation. Reliable and reproducible computational XAS databases require systematic benchmark studies. In this work, we benchmarked Ti K-edge XAS simulations of ten representative Ti-O binary compounds, which we refer to as the Ti-O-10 dataset, using three state-of-the-art codes: xspectra, ocean and exciting. We systematically studied the convergence behavior with respect to the input parameters and developed a workflow to automate and standardize the calculations to ensure converged spectra. Our benchmark comparison shows: (1) the two Bethe-Salpeter equation (BSE) codes (ocean and exciting) have excellent agreement in the energy range studied (up to 35 eV above the onset) with an average Spearman's rank correlation score of 0.998; (2) good agreement is obtained between the core-hole potential code (xspectra) and BSE codes (ocean and exciting) with an average Spearman's rank correlation score of 0.990. Our benchmark study provides important standards for first-principles XAS simulations with broad impact in data-driven XAS analysis., Comment: 32 pages, 11 figures and 7 tables

X-ray absorption spectroscopy (XAS) is a premier technique for materials characterization, providing key information about the local chemical environment of the absorber atom. In this work, we develop a database of sulfur K-edge XAS spectra of crystalline and amorphous lithium thiophosphate materials based on the atomic structures reported in Chem. Mater., 34, 6702 (2022). The XAS database is based on simulations using the excited electron and core-hole pseudopotential approach implemented in the Vienna Ab initio Simulation Package. Our database contains 2681 S K-edge XAS spectra for 66 crystalline and glassy structure models, making it the largest collection of first-principles computational XAS spectra for glass/ceramic lithium thiophosphates to date. This database can be used to correlate S spectral features with distinct S species based on their local coordination and short-range ordering in sulfide-based solid electrolytes. The data is openly distributed via the Materials Cloud, allowing researchers to access it for free and use it for further analysis, such as spectral fingerprinting, matching with experiments, and developing machine learning models., Comment: 11 pages, 3 figures, 1 table

A new semi-supervised machine learning method for the discovery of structure-spectrum relationships is developed and demonstrated using the specific example of interpreting X-ray absorption near-edge structure (XANES) spectra. This method constructs a one-to-one mapping between individual structure descriptors and spectral trends. Specifically, an adversarial autoencoder is augmented with a novel rank constraint (RankAAE). The RankAAE methodology produces a continuous and interpretable latent space, where each dimension can track an individual structure descriptor. As a part of this process, the model provides a robust and quantitative measure of the structure-spectrum relationship by decoupling intertwined spectral contributions from multiple structural characteristics. This makes it ideal for spectral interpretation and the discovery of new descriptors. The capability of this procedure is showcased by considering five local structure descriptors and a database of over fifty thousand simulated XANES spectra across eight first-row transition metal oxide families. The resulting structure-spectrum relationships not only reproduce known trends in the literature, but also reveal unintuitive ones that are visually indiscernible in large data sets. The results suggest that the RankAAE methodology has great potential to assist researchers to interpret complex scientific data, test physical hypotheses, and reveal new patterns that extend scientific insight.

First-principles computational spectroscopy is a critical tool for interpreting experiment, performing structure refinement, and developing new physical understanding. Systematically setting up input files for different simulation codes and a diverse class of materials is a challenging task with a very high barrier-to-entry, given the complexities and nuances of each individual simulation package. This task is non-trivial even for experts in the electronic structure field and nearly formidable for non-expert researchers. Lightshow solves this problem by providing a uniform abstraction for writing computational x-ray spectroscopy input files for multiple popular codes, including FEFF, VASP, OCEAN, EXCITING and XSPECTRA. Its extendable framework will also allow the community to easily add new functions and to incorporate new simulation codes., Comment: 3 pages, 1 figure, software can be found open source under the BSD-3-clause license at https://github.com/AI-multimodal/Lightshow

As machine learning (ML) methods continue to be applied to a broad scope of problems in the physical sciences, uncertainty quantification is becoming correspondingly more important for their robust application. Uncertainty aware machine learning methods have been used in select applications, but largely for scalar properties. In this work, we showcase an exemplary study in which neural network ensembles are used to predict the X-ray absorption spectra of small molecules, as well as their point-wise uncertainty, from local atomic environments. The performance of the resulting surrogate clearly demonstrates quantitative correlation between errors relative to ground truth and the predicted uncertainty estimates. Significantly, the model provides an upper bound on the expected error. Specifically, an important quality of this uncertainty-aware model is that it can indicate when the model is predicting on out-of-sample data. This allows for its integration with large scale sampling of structures together with active learning or other techniques for structure refinement. Additionally, our models can be generalized to larger molecules than those used for training, and also successfully track uncertainty due to random distortions in test molecules. While we demonstrate this workflow on a specific example, ensemble learning is completely general. We believe it could have significant impact on ML-enabled forward modeling of a broad array of molecular and materials properties., Comment: 24 pages, 16 figures

Existing quantum algorithms for quantum chemistry work well near the equilibrium geometry of molecules, but the results can become unstable when the chemical bonds are broken at large atomic distances. For any adiabatic approach, this usually leads to serious problems, such as level crossing and/or energy gap closing along the adiabatic evolution path. In this work, we propose a quantum algorithm based on adiabatic evolution to obtain molecular eigenstates and eigenenergies in quantum chemistry, which exploits a smooth geometric deformation by changing bond lengths and bond angles. Even with a simple uniform stretching of chemical bonds, this algorithm performs more stably and achieves better accuracy than our previous adiabatic method [Phys. Rev. Research 3, 013104 (2021)]. It solves the problems related to energy gap closing and level crossing along the adiabatic evolution path at large atomic distances. We demonstrate its utility in several examples, including H${}_2$O, CH${}_2$, and a chemical reaction of H${}_2$+D${}_2\rightarrow$ 2HD. Furthermore, our fidelity analysis demonstrates that even with finite bond length changes, our algorithm still achieves high fidelity with the ground state.

We employ variational autoencoders to extract physical insight from a dataset of one-particle Anderson impurity model spectral functions. Autoencoders are trained to find a low-dimensional, latent space representation that faithfully characterizes each element of the training set, as measured by a reconstruction error. Variational autoencoders, a probabilistic generalization of standard autoencoders, further condition the learned latent space to promote highly interpretable features. In our study, we find that the learned latent variables strongly correlate with well known, but nontrivial, parameters that characterize emergent behaviors in the Anderson impurity model. In particular, one latent variable correlates with particle-hole asymmetry, while another is in near one-to-one correspondence with the Kondo temperature, a dynamically generated low-energy scale in the impurity model. Using symbolic regression, we model this variable as a function of the known bare physical input parameters and "rediscover" the non-perturbative formula for the Kondo temperature. The machine learning pipeline we develop suggests a general purpose approach which opens opportunities to discover new domain knowledge in other physical systems., Comment: Update to match PRB publication + typo fixes + minor edits

The Anderson Impurity Model (AIM) is a canonical model of quantum many-body physics. Here we investigate whether machine learning models, both neural networks (NN) and kernel ridge regression (KRR), can accurately predict the AIM spectral function in all of its regimes, from empty orbital, to mixed valence, to Kondo. To tackle this question, we construct two large spectral databases containing approximately 410k and 600k spectral functions of the single-channel impurity problem. We show that the NN models can accurately predict the AIM spectral function in all of its regimes, with point-wise mean absolute errors down to 0.003 in normalized units. We find that the trained NN models outperform models based on KRR and enjoy a speedup on the order of $10^5$ over traditional AIM solvers. The required size of the training set of our model can be significantly reduced using furthest point sampling in the AIM parameter space, which is important for generalizing our method to more complicated multi-channel impurity problems of relevance to predicting the properties of real materials., Comment: 31 pages, 27 figures, 5 tables

Disentangling the primary order parameter from secondary order parameters in phase transitions is critical to the interpretation of the transition mechanisms in strongly correlated systems and quantum materials. Here we present a study of structural phase transition pathways in superionic Cu2S nanocrystals that exhibit intriguing properties. Utilizing ultrafast electron diffraction techniques sensitive in both momentum-space and the time-domain, we distinguish the dynamics of crystal symmetry breaking and lattice expansion in this system. We are able to follow the transient states along the transition pathway and so observe the dynamics of both the primary and secondary order parameters. Based on these observations we argue that the mechanism of the structural phase transition in Cu2S is dominated by the electron-phonon coupling. This mechanism advances the understanding from previous results where the focus was solely on dynamic observations of the lattice expansion.

Path-integral ab initio molecular dynamics (PI-AIMD) calculations have been employed to probe the nature of chloride ion solvation in aqueous solution. Nuclear quantum effects (NQEs) are shown to weaken hydrogen bonding between the chloride anion and the solvation shell of water molecules. As a consequence, the disruptive effect of the anion on the solvent water structure is significantly reduced compared to what is found in the absence of NQEs. The chloride hydration structure obtained from PI-AIMD agrees well with information extracted from neutron scattering data. Inparticular, the observed satellite peak in the hydrogen-chloride-hydrogen triple angular distribution serves as a clear signature of NQEs. The present results suggest that NQEs are likely to play acrucial role in determining the structure of saline solutions., Comment: 6 pages, 4figures

The advent of massive data repositories has propelled machine learning techniques to the front lines of many scientific fields, and exploring new frontiers by leveraging the predictive power of machine learning will greatly accelerate big data-assisted discovery. In this work, we show that graph-based neural networks can be used to predict the near edge x-ray absorption structure spectra of molecules with exceptional accuracy. The predicted spectra reproduce nearly all the prominent peaks, with 90% of the predicted peak locations within 1 eV of the ground truth. Our study demonstrates that machine learning models can achieve practically the same accuracy as first-principles calculations in predicting complex physical quantities, such as spectral functions, but at a fraction of the cost., Comment: 6 pages, 3 figures and 1 table

We carry out X-ray absorption spectroscopy experiment at oxygen K-edge in croconic acid (C5H2O5) crystal as a prototype of ferroelectric organic molecular solid, whose electric polarization is generated by proton transfer. The experimental spectrum is well reproduced by the electron-hole excitation theory simulations from configuration generated by ab initio molecular dynamics simulation. When inversion symmetry is broken in ferroelectric state, the hydrogen bonding environment on the two bonded molecules become inequivalent. Such a difference is sensitively probed by the bound excitation in the pre-edge, which are strongly localized on the excited molecules. Our analysis shows that a satellite peak in the pre-edge will emerge at higher excitation energy which serves as a clear signature of ferroelectricity in the material., Comment: 6 pages, 3 figures

X-ray absorption spectroscopy is a premier element-specific technique for materials characterization. Specifically, the x-ray absorption near-edge structure (XANES) encodes important information about the local chemical environment of an absorbing atom, including coordination number, symmetry, and oxidation state. Interpreting XANES spectra is a key step towards understanding the structural and electronic properties of materials, and as such, extracting structural and electronic descriptors from XANES spectra is akin to solving a challenging inverse problem. Existing methods rely on empirical fingerprints, which are often qualitative or semiquantitative and not transferable. In this paper, we present a machine learning-based approach, which is capable of classifying the local coordination environments of the absorbing atom from simulated K-edge XANES spectra. The machine learning classifiers can learn important spectral features in a broad energy range without human bias and once trained, can make predictions on the fly. The robustness and fidelity of the machine learning method are demonstrated by an average 86% accuracy across the wide chemical space of oxides in eight 3d transition-metal families. We found that spectral features beyond the preedge region play an important role in the local structure classification problem especially for the late 3d transition-metal elements., Comment: 12 numbered pages and 6 figures

Machine learning methods are being explored in many areas of science, with the aim of finding solution to problems that evade traditional scientific approaches due to their complexity. In general, an order parameter capable of identifying two different phases of matter separated by a correspond- ing phase transition is constructed based on symmetry arguments. This parameter measures the degree of order as the phase transition proceeds. However, when the two distinct phases are highly disordered it is not trivial to identify broken symmetries with which to find an order parameter. This poses an excellent problem to be addressed using machine learning procedures. Room tem- perature liquid water is hypothesized to be a supercritical liquid, with fluctuations of two different molecular orders associated to two parent liquid phases, one with high density and another one with low density. The validity of this hypothesis is linked to the existence of an order parameter capable of identifying the two distinct liquid phases and their fluctuations. In this work we show how two different machine learning procedures are capable of recognizing local order in liquid water. We argue that when in order to learn relevant features from this complexity, an initial, physically motivated preparation of the available data is as important as the quality of the data set, and that machine learning can become a successful analysis tool only when coupled to high level physical information., Comment: 14 pages, 12 figures, 2 appendices

We propose a fully ab initio theory to compute the electron density response under the perturbation in the local field. This method is based on our recently developed local dielectric response theory [Phys. Rev. B 92, 241107(R), 2015], which provides a rigorous theoretical framework to treat local electronic excitations in extended systems beyond the commonly employed dipole approximation. We have applied this method to study the electronic part of the molecular polarizability of water in ice Ih and liquid water. Our results reveal that the crystal field of the hydrogen-bond network has strong anisotropic effects, which significantly enhance the out-of-plane component and suppress the in-plane component perpendicular to the bisector direction. The contribution from the charge transfer is equally important, which increases the isotropic molecular polarizability by 5-6%. Our study provides new insights into the dielectric properties of water, which form the basis to understand the electronic excitations in water and to develop accurate polarizable force fields of water., Comment: 7 figures

We present a local representation of the electronic dielectric response function, based on a spatial partition of the dielectric response into contributions from each Wannier function using a generalized density functional perturbation theory. This procedure is fully ab initio, and therefore allows us to rigorously define local metrics, such as "bond polarizability," on Wannier centers. We show that the locality of the response function is determined by the locality of three quantities: Wannier functions of the occupied manifold, the density matrix, and the Hamiltonian matrix. In systems with a gap, the bare dielectric response is exponentially localized, which supports the physical picture of the dielectric response function as a collection of interacting local response that can be captured by a tight-binding model., Comment: 5 pages, 1 table and 3 figures in main text, 3 pages and 4 figures in supplemental material