The pursuit of more advanced electronics, finding solutions to energy needs,and tackling a wealth of social issues often hinges upon the discovery andoptimization of new functional materials that enable disruptive technologies orapplications. However, the discovery rate of these materials is alarmingly low.Much of the information that could drive this rate higher is scattered acrosstens of thousands of papers in the extant literature published over severaldecades, and almost all of it is not collated and thus cannot be used in itsentirety. Many of these limitations can be circumvented if the experimentalisthas access to systematized collections of prior experimental procedures andresults that can be analyzed and built upon. Here, we investigate theproperty-processing relationship during growth of oxide films by pulsed laserdeposition. To do so, we develop an enabling software tool to (1) mine theliterature of relevant papers for synthesis parameters and functionalproperties of previously studied materials, (2) enhance the accuracy of thismining through crowd sourcing approaches, (3) create a searchable repositorythat will be a community-wide resource enabling material scientists to leveragethis information, and (4) provide through the Jupyter notebook platform, simplemachine-learning-based analysis to learn the complex interactions betweengrowth parameters and functional properties (all data and codes available onhttps://github.com/ORNL-DataMatls). The results allow visualization of growthwindows, trends and outliers, and which can serve as a template for analyzingthe distribution of growth conditions, provide starting points for relatedcompounds and act as feedback for first-principles calculations. Such toolswill comprise an integral part of the materials design schema in the comingdecade.
展开▼
