Integrating Data\nMining and Natural Language Processing\nto Construct a Melting Point Database for Organometallic Compounds

As\nsemiconductor devices are miniaturized, the importance of atomic\nlayer deposition (ALD) technology is growing. When designing ALD precursors,\nit is important to consider the melting point, because the precursors\nshould have melting points lower than the process temperature. However,\nobtaining melting point data is challenging due to experimental sensitivity\nand high computational costs. As a result, a comprehensive and well-organized\n database for the melting point of the OMCs has not been fully reported\nyet. Therefore, in this study, we constructed a database of melting\npoints for 1,845 OMCs, including 58 metal and 6 metalloid elements.\nThe database contains CAS numbers, molecular formulas, and structural\ninformation and was constructed through automatic extraction and systematic\ncuration. The melting point information was extracted using two methods:\n1) 1,434 materials from 11 chemical vendor databases and 2) 411 materials\nidentified through natural language processing (NLP) techniques with\nan accuracy of 86.3%, based on 2,096 scientific papers published over\nthe past 29 years. In our database, the OMCs contain up to around\n250 atoms and have melting points that range from −170 to\n1610 °C. The main source is the Chemsrc database, accounting\nfor 607 materials (32.9%), and Fe is the most common central metal\nor metalloid element (15.0%), followed by Si (11.6%) and B (6.7%).\nTo validate the utilization of the constructed database, a multimodal\nneural network model was developed integrating graph-based and feature-based\ninformation as descriptors to predict the melting points of the OMCs\nbut moderate performance. We believe the current approach reduces\nthe time and cost associated with hand-operated data collection and\nprocessing, contributing to effective screening of potentially promising\nALD precursors and providing crucial information for the advancement\nof the semiconductor industry.