Data-Driven Review of Thermoelectric Materials: Performance and Resource Considerations

Abstract

In this review, we describe the creation of a large database of thermoelectric materials prepared by abstracting information from over 100 publications. The database has over 18 000 data points from multiple classes of compounds, whose relevant properties have been measured at several temperatures. Appropriate visualization of the data immediately allows certain insights to be gained with regard to the property space of plausible thermoelectric materials. Of particular note is that any candidate material needs to display an electrical resistivity value that is close to 1 mW cm at 300 K, i.e., samples should be significantly more conductive than the Mott minimum metallic conductivity. The Herfindahl-Hirschman index, a commonly accepted measure of market concentration, has been calculated from geological data (known elemental reserves) and geopolitical data (elemental production) for much of the periodic table. The visualization strategy employed here allows rapid sorting of thermoelectric compositions with respect to important issues of elemental scarcity and supply risk.

A Simple Computational Proxy for Screening Magnetocaloric Compounds

“…[132][133][134][135][136][137][138][139] Statistical analysis, interactive visualization, and machine learning allow for new and insightful ways of understanding materials behavior and guiding research efforts. The experimental and computed data in this study, provided in full in the Supporting Information, afford opportunities for this type of analysis.…”

Research Update: Towards designed functionalities in oxide-based electronic materials

“…15,16 However, protocols to identify specific, including unknown, ACS for oxides that have the required technology-enabling properties and functionalities have only recently been offered. This in part relies on establishing structure-property relationships and stability conditions from structural-chemistry data and calculable material properties, [17][18][19][20][21][22][23] which are often established from disparate yet physically motivated models and domain knowledge. Such protocols address the question of how to go beyond calculations of individual materials ("given an ACS, predict its properties," i.e., the "direct approach") and simple exploitation of available materials data to intelligent exploration of those ACS that have a target property ("inverse approach"), including unknown compounds?…”

Perspective: Interactive material property databases through aggregation of literature data

“…Researchers have physically begun to mine data from the literature in order to better understand materials for thermoelectrics, 7,35,36 lithium and lithium-ion batteries, 37 catalysis, 21,38 kinetics, 39 and more. The process, shown in Figure 1 for creating interactive databases, involves gathering appropriate publications, identifying key data in the publications, and then employing a combination of graduate students and/or post-doctoral fellows, undergraduate interns, and sometimes even high school students to work on data extraction.…”

“…The process then involves physically entering numbers into a text file or database, frequently after having digitized plots in the publications, using freeware tools such as . 40 At this stage, metadata such as the crystal structure attributes of the compound being measured, the elemental abundance and availability of the constituents, 35 or the preparation and processing method are entered as well. Finally, the text file is read into web-based visualization suites using software such as , 41 which is freely available for use to academic, not-for-profit entities.…”

“…In our recent work, we have encoded the plots to allow users to hover over specific data points to expose additional information regarding property measurement and metadata (i.e., author, year, synthesis route, processing conditions, comments, temperature, etc.) 35,37 Additionally, if a user clicks on a data point, it will open a new tab and redirect the user to the primary literature source using the document DOI. Finally, users are given the choice for exporting the plots in either raster or vector formats.…”