resumo
The machine learning framework reported herein can greatly accelerate the development of more effective and sustainable corrosion inhibitors for aluminum alloys, which still rely mostly on the experience of corrosion scientists, and trial and error laboratory testing. It can be used to design inhibitors for specific applications, which can be immobilized into nanocontainers or included directly into coatings in the search for less hazardous corrosion protective technologies. Therefore, a machine learning (ML) classification model that allows to identify promising compounds ( > 70% inhibitor efficiency) among less promising ones, and an online application (https://datacor.shinyapps.io/datacortech/) were developed for the virtual screen (simulation) of potential inhibitors for aluminum alloys, capable of considering the molecular structure and the influence of pH as an input.
palavras-chave
PARTITION-COEFFICIENTS; QUANTUM-CHEMISTRY; PREDICTION; DESIGN; MODELS
categoria
Materials Science
autores
Galvao, TLP; Ferreira, I; Maia, F; Gomes, JRB; Tedim, J
nossos autores
João Tedim
Professor associado
José R. B. Gomes
Investigador Principal
Tiago Luís Pereira Galvão
Professor AuxiliarGrupos
G3 - Materiais Eletroquímicos, Interfaces e Revestimentos
G6 - Materiais Virtuais e Inteligência Artificial
Projectos
CICECO - Aveiro Institute of Materials (UIDB/50011/2020)
CICECO - Aveiro Institute of Materials (UIDP/50011/2020)
Associated Laboratory CICECO-Aveiro Institute of Materials (LA/P/0006/2020)
Dados Inteligentes para Desenhar Inibidores de Corrosão (DataCor)
Collaboratory for Emerging Technologies, CoLab (EMERGING TECHNOLOGIES)
agradecimentos
This work was developed within the scope of the project CICECO-Aveiro Institute of Materials, UIDB/50011/2020, UIDP/50011/2020 & LA/P/0006/2020, financed by national funds through the FCT/MCTES (PIDDAC). It was also financed in the framework of the project DataCor (refs. POCI-01-0145-FEDER-030256, PTDC/QUI-QFI/30256/2017 and https://datacorproject.wixsite.com/datacor) and has received funding from the European Union's Horizon 2020 research and innovation program under the Marie Sklodowska-Curie grant agreement ID 101007430 (COAT4LIFE). TLPG thanks the Portuguese Foundation for Science and Technology (FCT) for the grant in the scope of the Individual Call to Scientific Employment Stimulus - 5th Edition, Refs. 2022.08205.CEECIND.