Epit and Epass descriptors of 316L stainless steel estimated by Machine Learning

This database comprises 5 datasets of pitting/passivity descriptors estimated from Potentiodynamic Polarisation (PP) curves. Each dataset consists of 1 CSV file comprising the following features (columns): Epit_x, Epit_y, Epass_x, Epass_y. The “Maps” indexes (rows/data samples) correspond to the PP tests numbering.
Epit_x and Epass_x, Epit_y and Epass_y correspond to Epit and Epass (V), log(jpit), and log(jpass) (µA/cm²).

This descriptors database was derived from the 5 datasets of log(j) Vs E curves obtained in high throughput fashion with the SECCM on 316L stainless steel (5 different combinations of [NaCl] and scan rates).
The descriptors datasets present the same amount of data samples as the source (log(j) Vs E) datasets (287, 377, 119, 125 and 47) available at:
Bertolucci Coelho, Leonardo; Ustarroz, Jon (2023), “Micro-scale potentiodynamic polarisation (log(j)) curves of 316L stainless steel”, Mendeley Data, V1, doi: 10.17632/7j6b6y48jw.1

This descriptors database was deployed as described in the following scientific article, accepted for publication in npj Materials Degradation journal on 25 September 2023:
“Estimating pitting descriptors of 316L stainless steel by machine learning and statistical analysis”.
Leonardo Bertolucci Coelho1,2,∗, Daniel Torres1, Vincent Vangrunderbeek2, Miguel Bernal1, Gian Marco Paldino3, Gianluca Bontempi3, Jon Ustarroz 1,2
1 ChemSIN – Chemistry of Surfaces, Interfaces and Nanomaterials, Université libre de Bruxelles (ULB), Brussels, Belgium
2 Research Group Electrochemical and Surface Engineering (SURF), Vrije Universiteit Brussel, Brussels, Belgium
3 Machine Learning Group (MLG), Université libre de Bruxelles (ULB), Brussels, Belgium
*[e-mail]

In “Estimating pitting descriptors of 316L stainless steel by machine learning and statistical analysis”, we provide a methodology for estimating Epass (passive potential) and Epit (pitting potential) from: 1. typical log(j) Vs E curves with a straightforward passivity breakdown (using an algorithm based on linear regression (LR)); 2. PP curves with more unique profiles mainly due to metastable events (using Artificial Neural Networks (ANN) trained on the LR estimates).

For further details on the acquisition of the PP curves, please refer to:
Bertolucci Coelho, Leonardo (2023), “Micro-scale potentiodynamic polarisation curves of 316L stainless steel ”, Mendeley Data, V3, doi: 10.17632/78rz8vw46x.3
Coelho, L. B. et al. Probing the randomness of the local current distributions of 316 L stainless steel corrosion in NaCl solution. Corros. Sci. 217, 111104 (2023).