Estudo da viabilidade da determinação da dissolução em meio aquoso de compostos óxidos utilizando a ferramenta open source orange data mining

Abstract

Solubility, or the ability of a material to dissolve in a solvent, is a fundamental property of materials, especially for ceramics and glass. The complexity of solute-solvent interactions makes it difficult to accurately calculate solubility, and the lack of precise measurement methods can affect the reliability of the data. The dissolution rate, which is the speed at which a solute dissolve, is crucial in areas such as chemistry and pharmacology. The Noyes-Whitney equation, developed in 1897, describes how the dissolution rate is influenced by various factors. The chemical composition of a glass can significantly influences its properties, including solubility and dissolution rate. Data mining and machine learning can be used to predict these properties, assisting in the discovery and design of new materials. Neural Networks are particularly useful for predicting solubility and dissolution rate due to their ability to model complex relationships. Orange Data Mining is a data mining and machine learning tool that can be used to predict the solubility and dissolution rate of glasses based on their composition. The tool is intuitive and easy to use, allowing researchers to build models without the need for extensive coding. It is also capable of handling large data sets, essential for data mining, proposing a unified approach to predict both solubility and dissolution rate, using only the material composition as input. This simplifies the prediction process, making it more accessible and less dependent on complex data. In this work, through open-source software, a study of the feasibility of using this tool in predicting solubility and dissolution rate in aqueous medium applied to oxide compounds will be presented.