A database of battery materials auto-generated using ChemDataExtractor

Abstract

A database of battery materials is presented which comprises a total of 292,313 data records, with 214,617 unique chemical-property data relations between 17,354 unique chemicals and up to five material properties: capacity, voltage, conductivity, Coulombic efficiency and energy. 117,403 data are multivariate on a property where it is the dependent variable in part of a data series. The database was auto-generated by mining text from 229,061 academic papers using the chemistry-aware natural language processing toolkit, ChemDataExtractor version 1.5, which was modified for the specific domain of batteries. The collected data can be used as a representative overview of battery material information that is contained within text of scientific papers. Public availability of these data will also enable battery materials design and prediction via data-science methods. To the best of our knowledge, this is the first auto-generated database of battery materials extracted from a relatively large number of scientific papers. We also provide a Graphical User Interface (GUI) to aid the use of this database.

What Can Text Mining Tell Us About Lithium‐Ion Battery Researchers’ Habits?

“…As a benchmark in the battery community, Huang et al . constructed a database of battery materials electrochemical properties (such as capacity, conductivity or coulombic efficiency) by mining 229,061 academic papers using the chemistry‐aware natural language processing toolkit, ChemDataExtractor [60] . Furthermore, the potential of extracting information and trends from literature of the emerging all‐solid‐state batteries (ASSBs) field was recently highlighted by Randau et al ., which required the use of approximations and hypotheses to manually analyze a small dataset of about 30 publications [61] .…”

“…[10,[54][55][56][57][58][59] As a benchmark in the battery community, Huang et al constructed a database of battery materials electrochemical properties (such as capacity, conductivity or coulombic efficiency) by mining 229,061 academic papers using the chemistry-aware natural language processing toolkit, ChemDataExtractor. [60] Furthermore, the potential of extracting information and trends from literature of the emerging allsolid-state batteries (ASSBs) field was recently highlighted by Randau et al, which required the use of approximations and hypotheses to manually analyze a small dataset of about 30 publications. [61] Needless to say, such in depth analysis would not be possible for a much more established technology such as LIB, for which thousands of reports must be individually screened: under the hypothesis of reading 200 articles per year, a researcher will need almost 140 years to read all the LIB scientific publications available today!…”

Artificial intelligence and machine learning in glass science and technology: 21 challenges for the 21st century

“…With specific data parsing algorithms such as Snowball, Chemdataextractor has been used to automatically extract large datasets of magnetic materials and their Neel's temperatures 46 and as well as a database of battery material with five leading properties. 45 Several such approaches have already been used in zeolites 47 and inorganic materials. 48 In glass science, this presents a unique challenge as glass compositions have complex representations without any uniformity.…”

“…On the other hand, recent developments in AI can be exploited to automate the data extraction process from journal publications and patents. For example, ChemDataExtractor 44,45 is a tool that can identify chemical species through their symbols in literature. With specific data parsing algorithms such as Snowball, Chemdataextractor has been used to automatically extract large datasets of magnetic materials and their Neel's temperatures 46 and as well as a database of battery material with five leading properties 45 .…”

“…For example, ChemDataExtractor 44,45 is a tool that can identify chemical species through their symbols in literature. With specific data parsing algorithms such as Snowball, Chemdataextractor has been used to automatically extract large datasets of magnetic materials and their Neel's temperatures 46 and as well as a database of battery material with five leading properties 45 . Several such approaches have already been used in zeolites 47 and inorganic materials 48 .…”