RESUMEN

Nowadays, biodegradable metals and alloys, as well as their corrosion behavior, are of particular interest. The corrosion process of metals and alloys under various harsh conditions can be studied via the investigation of corrosion atom adsorption on metal surfaces. This can be performed using density functional theory-based simulations. Importantly, comprehensive analytical data obtained in simulations including parameters such as adsorption energy, the amount of charge transferred, atomic coordinates, etc., can be utilized in machine learning models to predict corrosion behavior, adsorption ability, catalytic activity, etc., of metals and alloys. In this work, data on the corrosion indicators of Zn surfaces in Cl-, S-, and O-rich harsh environments are collected. A dataset containing adsorption height, adsorption energy, partial density of states, work function values, and electronic charges of individual atoms is presented. In addition, based on these corrosion descriptors, it is found that a Cl-rich environment is less harmful for different Zn surfaces compared to an O-rich environment, and more harmful compared to a S-rich environment.