Taylor D. Sparks

Performance and resource considerations of Li-ion battery electrode materials

In this work we present a data-driven approach to the rational design of battery materials based on both resource and performance considerations. A large database of Li-ion battery material has been created by abstracting information from over 200 publications. The database consists of over 16 000 data points from various classes of materials. In addition to reference information, key parameters and variables determining the performance of batteries were collected. This work also includes resource considerations such as crustal abundance and the Herfindahl–Hirschman index, a commonly used measure of market concentration. The data is organized into a free web-based resource where battery researchers can employ a unique visualization method to plot database parameters against one another. This contribution is concerned with cathode and anode electrode materials. Cathode materials are mostly based on an intercalation mechanism, while anode materials are primarily based on conversion and alloying. Results indicate that cathode materials follow a common trend consistent with their crystal structure. On the other hand anode materials display similar behavior, based on elemental composition. Of particular interest is that high energy cathodes are scarcer than high power materials and high performance anode materials are less available. More sustainable materials for both electrodes based on alternative compositions are identified.

Perspective: Web-based machine learning models for real-time screening of thermoelectric materials properties

Chemistries Michael W. Gaultois, a) Anton O. Oliynyk, Arthur Mar, Taylor D. Sparks, Gregory J. Mulholland, and Bryce Meredig b) Materials Research Laboratory and the Department of Chemistry and Biochemistry, University of California, Santa Barbara, California, 93106, USA Department of Chemistry, University of Alberta, Edmonton, Alberta, T6G 2G2, Canada Department of Materials Science and Engineering, University of Utah, Salt Lake City, Utah, 84112, USA Citrine Informatics, Redwood City, California, 94061, USAThe experimental search for new thermoelectric materials remains largely confined to a limited set of successful chemical and structural families, such as chalcogenides, skutterudites, and Zintl phases. In principle, computational tools such as density functional theory (DFT) offer the possibility of rationally guiding experimental synthesis efforts toward very different chemistries. However, in practice, predicting thermoelectric properties from first principles remains a challenging endeavor [J. Carrete et al., Phys. Rev. X 4, 011019 (2014)], and experimental researchers generally do not directly use computation to drive their own synthesis efforts. To bridge this practical gap between experimental needs and computational tools, we report an open machine learning-based recommendation engine (http://thermoelectrics.citrination.com) for materials researchers that suggests promising new thermoelectric compositions based on pre-screening about 25 000 known materials and also evaluates the feasibility of use...

Perspective: Interactive material property databases through aggregation of literature data

Searchable, interactive, databases of material properties, particularly those relating to functional materials (magnetics, thermoelectrics, photovoltaics, etc.) are curiously missing from discussions of machine-learning and other data-driven methods for advancing new materials discovery. Here we discuss the manual aggregation of experimental data from the published literature for the creation of interactive databases that allow the original experimental data as well additional metadata to be visualized in an interactive manner. The databases described involve materials for thermoelectric energy conversion, and for the electrodes of Li-ion batteries. The data can be subject to machine-learning, accelerating the discovery of new materials.

How much improvement in thermoelectric performance can come from reducing thermal conductivity

Large improvements in the performance of thermoelectric materials have come from designing materials with reduced thermal conductivity. Yet as the thermal conductivity of some materials now approaches their amorphous limit, it is unclear if microstructure engineering can further improve thermoelectric performance in these cases. In this contribution, we use large data sets to examine 300 compositions in 11 families of thermoelectric materials and present a type of plot that quickly reveals the maximum possible zT that can be achieved by reducing the thermal conductivity. This plot allows researchers to quickly distinguish materials where the thermal conductivity has been optimized from those where improvement can be made. Moreover, through these large data sets we examine structure-property relationships to identify methods that decrease thermal conductivity and improve thermoelectric performance. We validate, with the data, that increasing (i) the volume of a unit cell and/or (ii) the number of atoms in ...

Enhancement of thermoelectric properties in the Nb–Co–Sn half-Heusler/Heusler system through spontaneous inclusion of a coherent second phase

Half-Heusler XYZ compounds with an 18 valence electron count are promising thermoelectric materials, being thermally and chemically stable, deriving from relatively earth-abundant components, and possessing appropriate electrical transport properties. The typical drawback with this family of compounds is their high thermal conductivity. A strategy for reducing thermal conductivity is through the inclusion of secondary phases designed to minimize negative impact on other properties. Here, we achieve this through the addition of excess Co to half-Heusler NbCoSn, which introduces precipitates of a semi-coherent NbCo2Sn Heusler phase. A series of NbCo1+xSn materials are characterized here using X-ray and neutron diffraction studies and electron microscopy. Electrical and thermal transport measurements and electronic structure calculations are used to understand property evolution. We find that annealing has an important role to play in determining antisite ordering and properties. Antisite disorder in the as-...

Single-step preparation and consolidation of reduced early-transition-metal oxide/metal n-type thermoelectric composites

Reduced early transition metal oxides/metal composites have been identified here as interesting thermoelectric materials. Numerous compositions in the Nb-rich portion of the WO3–Nb2O5 system have been studied, in composite formulations with elemental W. Spark plasma sintering (SPS) has been employed to achieve rapid preparation and consolidation of composite materials containing W metal precipitates with characteristic length scales that range from under 20 nm to a few microns, that exhibit thermal conductivities that are constant from 300 K to 1000 K, approximately 2.5 W m−1 K−1. Thermoelectric properties of these n-type materials were measured, and the highest-performing compositions were found to reach figure of merit zT values close to 0.1 at 950 K. The measurements point to higher zT values at yet-higher temperatures.

Stable, Heat-Conducting Phosphor Composites for High-Power Laser Lighting

Solid-state lighting using laser diodes is an exciting new development that requires new phosphor geometries to handle the greater light fluxes involved. The greater flux from the source results in more conversion and therefore more conversion loss in the phosphor, which generates self-heating, surpassing the stability of current encapsulation strategies used for light-emitting diodes, usually based on silicones. Here, we present a rapid method using spark plasma sintering (SPS) for preparing ceramic phosphor composites of the canonical yellow-emitting phosphor Ce-doped yttrium aluminum garnet (Ce:YAG) combined with a chemically compatible and thermally stable oxide, α-Al2O3. SPS allows for compositional modulation, and phase fraction, microstructure, and luminescent properties of ceramic composites with varying compositions are studied here in detail. The relationship between density, thermal conductivity, and temperature rise during laser-driven phosphor conversion is elucidated, showing that only modest densities are required to mitigate thermal quenching in phosphor composites. Additionally, the scattering nature of the ceramic composites makes them ideal candidates for laser-driven white lighting in reflection mode, where Lambertian scattering of blue light offers great color uniformity, and a luminous flux >1000 lm is generated using a single commercial laser diode coupled to a single phosphor element.

Balancing Mechanical Properties and Sustainability in the Search for Superhard Materials

The development of superhard materials is focused on two very different classes of compounds. The first contains only light, inexpensive main group elements and requires high pressures and temperatures for preparation whereas the second class combines a transition metal with light main group elements and in general tends to only need high reaction temperatures. Although the preparation conditions are simpler, the second class of compounds suffers from the transition metals used being expensive and exceedingly scarce. Thus, in the search for novel superhard compounds, synthetic accessibility, resource considerations, and material response must be balanced. The research presented here develops high-information density plots drawn from high-throughput first-principle calculations and data mining to reveal the optimal composition space to synthesize new materials. This contribution includes analysis of the experimentally known Vickers hardness for materials as well as screening over 1100 compounds from first-principle calculations to predict their intrinsic hardness. Both data sets are analyzed not only for their mechanical performance but also the compositional scarcity, and Herfindahl-Hirschman index is calculated. Following this methodology, it is possible to ensure targeted materials are not only sustainable and accessible but that they will also have superb mechanical response.

Machine Learning Prediction of Heat Capacity for Solid Inorganics

Many thermodynamic calculations and engineering applications require the temperature-dependent heat capacity (Cp) of a material to be known a priori. First-principle calculations of heat capacities can stand in place of experimental information, but these calculations are costly and expensive. Here, we report on our creation of a high-throughput supervised machine learning-based tool to predict temperature-dependent heat capacity. We demonstrate that material heat capacity can be correlated to a number of elemental and atomic properties. The machine learning method predicts heat capacity for thousands of compounds in seconds, suggesting facile implementation into integrated computational materials engineering (ICME) processes. In this context, we consider its use to replace Neumann-Kopp predictions as a high-throughput screening tool to help identify new materials as candidates for engineering processes. Also promising is the enhanced speed and performance compared to cation/anion contribution methods at elevated temperatures as well as the ability to improve future predictions as more data are made available. This machine learning method only requires formula inputs when calculating heat capacity and can be completely automated. This is an improvement to common best-practice methods such as cation/anion contributions or mixed-oxide approaches which are limited in application to specific materials and require case-by-case considerations.

Machine learning directed search for ultraincompressible, superhard materials

In the pursuit of materials with exceptional mechanical properties, a machine-learning model is developed to direct the synthetic efforts toward compounds with high hardness by predicting the elastic moduli as a proxy. This approach screens 118 287 compounds compiled in crystal structure databases for the materials with the highest bulk and shear moduli determined by support vector machine regression. Following these models, a ternary rhenium tungsten carbide and a quaternary molybdenum tungsten borocarbide are selected and synthesized at ambient pressure. High-pressure diamond anvil cell measurements corroborate the machine-learning prediction of the bulk modulus with less than 10% error, as well as confirm the ultraincompressible nature of both compounds. Subsequent Vickers microhardness measurements reveal that each compound also has an extremely high hardness exceeding the superhard threshold of 40 GPa at low loads (0.49 N). These results show the effectiveness of materials development through state-of-the-art machine-learning techniques by identifying functional inorganic materials.