Search for: All records

Abstract Although first principles based anharmonic lattice dynamics is one of the most common methods to obtain phonon properties, such method is impractical for high-throughput search of target thermal materials. We develop an elemental spatial density neural network force field as a bottom-up approach to accurately predict atomic forces of ~80,000 cubic crystals spanning 63 elements. The primary advantage of our indirect machine learning model is the accessibility of phonon transport physics at the same level as first principles, allowing simultaneous prediction of comprehensive phonon properties from a single model. Training on 3182 first principles data and screening 77,091 unexplored structures, we identify 13,461 dynamically stable cubic structures with ultralow lattice thermal conductivity below 1 Wm −1 K −1 , among which 36 structures are validated by first principles calculations. We propose mean square displacement and bonding-antibonding as two low-cost descriptors to ease the demand of expensive first principles calculations for fast screening ultralow thermal conductivity. Our model also quantitatively reveals the correlation between off-diagonal coherence and diagonal populations and identifies the distinct crossover from particle-like to wave-like heat conduction. Our algorithm is promising for accelerating discovery of novel phononic crystals for emerging applications, such as thermoelectrics, superconductivity, and topological phonons for quantum information technology. 

Abstract Existing machine learning potentials for predicting phonon properties of crystals are typically limited on a material-to-material basis, primarily due to the exponential scaling of model complexity with the number of atomic species. We address this bottleneck with the developed Elemental Spatial Density Neural Network Force Field, namely Elemental-SDNNFF. The effectiveness and precision of our Elemental-SDNNFF approach are demonstrated on 11,866 full, half, and quaternary Heusler structures spanning 55 elements in the periodic table by prediction of complete phonon properties. Self-improvement schemes including active learning and data augmentation techniques provide an abundant 9.4 million atomic data for training. Deep insight into predicted ultralow lattice thermal conductivity (<1 Wm −1  K −1 ) of 774 Heusler structures is gained by p–d orbital hybridization analysis. Additionally, a class of two-band charge-2 Weyl points, referred to as “double Weyl points”, are found in 68% and 87% of 1662 half and 1550 quaternary Heuslers, respectively. 

Discovering new materials is a challenging task in materials science crucial to the progress of human society. Conventional approaches based on experiments and simulations are labor-intensive or costly with success heavily depending on experts’ heuristic knowledge. Here, we propose a deep learning based Physics Guided Crystal Generative Model (PGCGM) for efficient crystal material design with high structural diversity and symmetry. Our model increases the generation validity by more than 700% compared to FTCP, one of the latest structure generators and by more than 45% compared to our previous CubicGAN model. Density Functional Theory (DFT) calculations are used to validate the generated structures with 1869 materials out of 2000 are successfully optimized and deposited into the Carolina Materials Databasewww.carolinamatdb.org, of which 39.6% have negative formation energy and 5.3% have energy-above-hull less than 0.25 eV/atom, indicating their thermodynamic stability and potential synthesizability.

 

Despite the machine learning (ML) methods have been largely used recently, the predicted materials properties usually cannot exceed the range of original training data. We deployed a boundless objective-free exploration approach to combine traditional ML and density functional theory (DFT) in searching extreme material properties. This combination not only improves the efficiency for screening large-scale materials with minimal DFT inquiry, but also yields properties beyond original training range. We use Stein novelty to recommend outliers and then verify using DFT. Validated data are then added into the training dataset for next round iteration. We test the loop of training-recommendation-validation in mechanical property space. By screening 85,707 crystal structures, we identify 21 ultrahigh hardness structures and 11 negative Poisson’s ratio structures. The algorithm is very promising for future materials discovery that can push materials properties to the limit with minimal DFT calculations on only ~1% of the structures in the screening pool.

 

Discovering new materials with desired properties has been a dominant and crucial topic of interest in the field of materials science in the past few decades. In this work, novel carbon allotropes and ternary B–C–N structures were generated using the state-of-the-art RG 2 code. All structures were fully optimized using density functional theory with first-principles calculations. Several hundred carbon allotropes and ternary B–C–N structures were identified to be superhard materials. The thermodynamic stability of some randomly selected superhard materials was confirmed by evaluating the full phonon dispersions in the Brillouin zone. The new carbon allotropes and ternary B–C–N structures possess a wide range of mechanical properties generally and Vickers hardness specifically. Through 2D Pearson's correlation map, we first reproduced the well-accepted explanation and relationship of the Vickers hardness of the generated structures with other mechanical properties such as shear modulus, bulk modulus, Pugh's ratio, universal anisotropy, and Poisson's ratio. We then propose two fundamentally new descriptors from the electronic level, namely local potential and electron localization function averaged over a unit cell, both of which exhibit a strong correlation with Vickers hardness. More importantly, these descriptors are easy to access from first-principles calculations (at least two orders of magnitude faster than the traditional calculation of elastic constants), and thus can serve as a fast and accurate approach for screening superhard materials. We also combined these new descriptors with known composition and structural descriptors in the machine learning training process. The new descriptors significantly enhance the performance of the trained machine learning model in predicting the Vickers hardness of unknown materials, which provides strong evidence for local potential and electron localization function to be considered in future high-throughput computation. This work unravels more fundamental but previously unexplored knowledge about superhard materials and the newly proposed electronic level descriptors are expected to accelerate the discovery of new superhard materials. 

The Mg 3 Sb 2− x Bi x family has emerged as the potential candidates for thermoelectric applications due to their ultra-low lattice thermal conductivity ( κ L ) at room temperature (RT) and structural complexity. Here, using ab initio calculations of the electron-phonon averaged (EPA) approximation coupled with Boltzmann transport equation (BTE), we have studied electronic, phonon and thermoelectric properties of Mg 3 Sb 2− x Bi x (x = 0, 1, and 2) monolayers. In violation of common mass-trend expectations, increasing Bi element content with heavier Zintl phase compounds yields an abnormal change in κ L in two-dimensional Mg 3 Sb 2− x Bi x crystals at RT (∼0.51, 1.86, and 0.25 W/mK for Mg 3 Sb 2 , Mg 3 SbBi, and Mg 3 Bi 2 ). The κ L trend was detailedly analyzed via the phonon heat capacity, group velocity and lifetime parameters. Based on quantitative electronic band structures, the electronic bonding through the crystal orbital Hamilton population (COHP) and electron local function analysis we reveal the underlying mechanism for the semiconductor-semimetallic transition of Mg 3 Sb 2-− x Bi x compounds, and these electronic transport properties (Seebeck coefficient, electrical conductivity, and electronic thermal conductivity) were calculated. We demonstrate that the highest dimensionless figure of merit ZT of Mg 3 Sb 2− x Bi x compounds with increasing Bi content can reach ∼1.6, 0.2, and 0.6 at 700 K, respectively. Our results can indicate that replacing heavier anion element in Zintl phase Mg 3 Sb 2− x Bi x materials go beyond common expectations (a heavier atom always lead to a lower κ L from Slack’s theory), which provide a novel insight for regulating thermoelectric performance without restricting conventional heavy atomic mass approach. 

High‐throughput screening has become one of the major strategies for the discovery of novel functional materials. However, its effectiveness is severely limited by the lack of sufficient and diverse materials in current materials repositories such as the open quantum materials database (OQMD). Recent progress in deep learning have enabled generative strategies that learn implicit chemical rules for creating hypothetical materials with new compositions and structures. However, current materials generative models have difficulty in generating structurally diverse, chemically valid, and stable materials. Here we propose CubicGAN, a generative adversarial network (GAN) based deep neural network model for large scale generative design of novel cubic materials. When trained on 375 749 ternary materials from the OQMD database, the authors show that the model is able to not only rediscover most of the currently known cubic materials but also generate hypothetical materials of new structure prototypes. A total of 506 such materials have been verified by phonon dispersion calculation. Considering the importance of cubic materials in wide applications such as solar panels, the GAN model provides a promising approach to significantly expand existing materials repositories, enabling the discovery of new functional materials via screening. The new crystal structures discovered are freely accessible atwww.carolinamatdb.org.