# Data-driven optimisation of variation of residual concentration to estimate the hydrogen diffusion coefficient and uptake via MATLAB

## TL;DR

This paper describes a method to estimate hydrogen diffusion and uptake in metals using experimental data and MATLAB optimization.

## Contribution

A novel data-driven MATLAB approach is introduced to calculate hydrogen diffusion parameters from experimental solubility tests.

## Key findings

- A two-step electrochemical process was used to charge and discharge hydrogen in metallic samples.
- MATLAB optimization successfully estimated hydrogen diffusion coefficients and uptake parameters.
- The method can be adapted to different geometries through simulation.

## Abstract

This article presents the data collected during experimental solubility tests in University of Bergamo lab and the developed code to estimate the principal parameters for hydrogen uptake. The experimental approach proposed involves in two-step process: the first phase consists of electrochemical charging to saturate the metallic sample via cathodic polarization at potentials lower than the equilibrium potential for hydrogen evolution, followed by a second phase of discharging under anodic polarization at potentials higher than the equilibrium potential. During the discharging phase, the time-dependent anodic current is influenced by the flux of diffusible hydrogen exiting the sample. This flux is governed by the initial concentration of diffusible hydrogen in the material, the diffusion coefficient, and the time elapsed between the end of the charging step and the beginning of the discharging phase. Via a data-driven optimisation it is possible to obtain the characteristic parameter of hydrogen diffusion in the material (apparent diffusion coefficient, hydrogen uptake and waiting time) using the MATLAB code. The data permit to elaborate the cylindrical geometry, however it is possible to modify the target curve via a simulation on a specific geometry (not included in the data) and use the same method to elaborate the experimental data of a specific geometry.