An official website of the United States government
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.
more »
« less
Unlocking phonon properties of a large and diverse set of cubic crystals by indirect bottom-up machine learning approach
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.
more »
« less
- PAR ID:
- 10465913
- Date Published:
- Journal Name:
- Communications Materials
- Volume:
- 4
- Issue:
- 1
- ISSN:
- 2662-4443
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Lattice thermal conductivity is important for many applications, but experimental measurements or first principles calculations including three-phonon and four-phonon scattering are expensive or even unaffordable. Machine learning approaches that can achieve similar accuracy have been a long-standing open question. Despite recent progress, machine learning models using structural information as descriptors fall short of experimental or first principles accuracy. This study presents a machine learning approach that predicts phonon scattering rates and thermal conductivity with experimental and first principles accuracy. The success of our approach is enabled by mitigating computational challenges associated with the high skewness of phonon scattering rates and their complex contributions to the total thermal resistance. Transfer learning between different orders of phonon scattering can further improve the model performance. Our surrogates offer up to two orders of magnitude acceleration compared to first principles calculations and would enable large-scale thermal transport informatics.more » « less
-
Full Heusler compounds have long been discovered as exceptional n-type thermoelectric materials. However, no p-type compounds could match the high n-type figure of merit ( ZT ). In this work, based on first-principles transport theory, we predict the unprecedentedly high p-type ZT = 2.2 at 300 K and 5.3 at 800 K in full Heusler CsK 2 Bi and CsK 2 Sb, respectively. By incorporating the higher-order phonon scattering, we find that the high ZT value primarily stems from the ultralow lattice thermal conductivity ( κ L ) of less than 0.2 W mK −1 at room temperature, decreased by 40% compared to the calculation only considering three-phonon scattering. Such ultralow κ L is rooted in the enhanced phonon anharmonicity and scattering channels stemming from the coexistence of antibonding-induced anharmonic rattling of Cs atoms and low-lying optical branches. Moreover, the flat and heavy nature of valence band edges leads to a high Seebeck coefficient and moderate power factor at optimal hole concentration, while the dispersive and light conduction band edges yield much larger electrical conductivity and electronic thermal conductivity ( κ e ), and the predominant role of κ e suppresses the n-type ZT . This study offers a deeper insight into the thermal and electronic transport properties in full Heusler compounds with strong phonon anharmonicity and excellent thermoelectric performance.more » « less
-
null (Ed.)Abstract Crystalline solids exhibiting glass-like thermal conductivity have attracted substantial attention both for fundamental interest and applications such as thermoelectrics. In most crystals, the competition of phonon scattering by anharmonic interactions and crystalline imperfections leads to a non-monotonic trend of thermal conductivity with temperature. Defect-free crystals that exhibit the glassy trend of low thermal conductivity with a monotonic increase with temperature are desirable because they are intrinsically thermally insulating while retaining useful properties of perfect crystals. However, this behavior is rare, and its microscopic origin remains unclear. Here, we report the observation of ultralow and glass-like thermal conductivity in a hexagonal perovskite chalcogenide single crystal, BaTiS 3 , despite its highly symmetric and simple primitive cell. Elastic and inelastic scattering measurements reveal the quantum mechanical origin of this unusual trend. A two-level atomic tunneling system exists in a shallow double-well potential of the Ti atom and is of sufficiently high frequency to scatter heat-carrying phonons up to room temperature. While atomic tunneling has been invoked to explain the low-temperature thermal conductivity of solids for decades, our study establishes the presence of sub-THz frequency tunneling systems even in high-quality, electrically insulating single crystals, leading to anomalous transport properties well above cryogenic temperatures.more » « less
-
Fluorographene, a fluorinated graphene-derivative, is expected to feature both high thermal conductivity and electrical insulation simultaneously, making it an emerging material for thermal management in electronic devices. In this paper, we investigated the lattice thermal conductivity and phonon transport properties of monolayer fluorographene using first-principles calculation. The solution of the fully linearized phonon Boltzmann transport equation gives the lattice thermal conductivity of monolayer fluorographene as 145.2 W m−1 K−1 at 300 K, which is about 20 times smaller than that of monolayer graphene. We systematically compared the phonon transport properties of all phonon modes in graphene and fluorographene in terms of phonon polarization. The significantly reduced thermal conductivity of fluorographene can be attributed to the lowering of both the lifetime of the flexural acoustic phonons and the group velocities of all acoustic phonons. We concluded that the broken in-plane mirror symmetry and the weaker in-plane chemical bonds induced by fluorination led to the suppression of the lattice thermal conductivity of fluorographene. Finally, we investigated the anomalously large contribution of optical phonons to the thermal transport process in fluorographene, where the large group velocities of selected optical phonons were derived from the in-plane acoustic modes of graphene. Our work provides a new approach to studying the influence of chemical functionalization on the phonon structure and exploring graphene-derived thermal management materials.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1038/s43246-023-00390-3
