Machine Learning–Assisted Risk Assessment of Pitting Corrosion Susceptibility of AA1050 in Ethanol‐Containing Fuels

Aug 3, 2026

Risk assessment of corrosion for metallic structures in complex environments is critical for automotive safety and durability. This study investigates ethanol-blended, non-aqueous alcoholate pitting corrosion of AA1050 under ethanol, fuel, water, and chloride contamination using supervised machine learning. Dedicated experiments vary ethanol–fuel–water ratios, temperatures, and surface preparations to build the feature space. A balanced-accuracy score of 0.87 demonstrates strong predictive capability and supports decision making for planning follow-up experiments.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages10

Reading Time25 min

AccessFree to download

Introduction

Background on corrosion challenges in aluminum

Ethanol-based pitting corrosion mechanism and experimental requirements

Physics-based modeling and prior statistical approaches

Supervised machine learning in corrosion science

Materials and Corrosion ARTICLE Machine Learning–Assisted Risk Assessment of Pitting Corrosion Susceptibility of AA1050 in Ethanol‐Containing Fuels Lukas C. Jarren1 | Eugen Gazenbiller1 | Visheet Arya2 | Rüdiger Reitz2 | Matthias Oechsner2 | Christian Feiler1 | Mikhail L. Zheludkevich1,3 | Daniel Höche1 1Institute of Surface Science, Helmholtz‐Zentrum Hereon, Geesthacht, Germany | 2Institute for Materials Technology (IfW), Technische Universität Darmstadt, Darmstadt, Germany | 3Faculty of Engineering, Kiel University, Kiel, Germany Correspondence: Lukas C. Jarren ([lukas.jarren@hereon.de](lukas.jarren@hereon.de)) Received: 12 August 2024 | Revised: 28 August 2024 | Accepted: 12 September 2024 Funding: Financial support was provided by the Deutsche Forschungsgemeinschaft (DFG HO4478/6‐1, OE558/20‐1, DFG HO4478/6‐2, OE558/20‐2) . Keywords: aluminum | corrosion risk | decision support | pitting | supervised machine learning ABSTRACT The ability to assess the risk of corrosion of metallic structures in particular environments holds considerable significance in the field of automotive industry. In recent years, machine learning has evolved into a crucial tool to evaluate the complex and multidimensional corrosion phenomena. In this paper, the special case of non‐aqueous alcoholate pitting corrosion of AA1050 in ethanol‐blended fuels with water and chloride contamination is examined via supervised machine learning techniques in order to distinguish between safe and unsafe conditions. The data space was created by conducting dedicated experiments with varying ethanol–fuel–water ratios, temperatures, and surface preparations. The classifier's performance rating of 0.87 (balanced accuracy) indicates an outstanding predictive ability and highlights the model's usefulness as decision support for subsequent experiments. 1 | Introduction Corrosion challenges in aluminum‐based engineering are known across various sectors, especially automotive. Ensuring the safety and durability of construction components is crucial, yet it often remains a challenging task. Observed corrosion phenomena (e.g., pitting) involve an interplay between coupled chemical, electrochemical, and solid‐state reactions and their transport kinetics, and thus are considered to be highly complex [1–3]. This has triggered and continues to trigger the development of smart corrosion protection concepts and systematic anticorrosive design. Nowadays, with the transition to advanced materials concepts and also to alternative fuels, corrosion engineering faces even new tasks to be addressed. In particular, localized corrosion of aluminum alloys can bea significant issue not only in aqueous environments but also in organic environments like fuel–ethanol mixtures. This ethanol‐based, chemical pitting corrosion mechanism poses a challenge in the automotive industry, where biogenic ethanol is mixed with gasoline. The initiation of ethanol‐based pitting corrosion is temperature‐induced, requiring conditions above ethanol's boiling point for most alloys and surface treatments. For a comprehensive analysis, use of an autoclave reaction vessel is essential. Ethanol is considered to catalyze the breakdown of aluminum's passive layer [4, 5], facilitating the transition from γ ‐AlOOH (boehmite) to γ ‐Al2O3 through temperature‐induced dehydration, accompanied by mechanical stresses [6] . This process reduces the need for aggressive halide anions, which are typically responsible for the passive layer breakdown in aqueous solutions during pit initiation [7] . In the absence of water, the passive layer fails to repassivate, leading to pit propagation as described in This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2024 The Author(s) . Materials and Corrosion published by Wiley‐VCH GmbH. Materials and Corro

FAQs

What corrosion case and material does the study focus on?

The study focuses on ethanol-containing, non-aqueous alcoholate pitting corrosion of AA1050 in ethanol-blended fuels with water and chloride contamination.

How is the dataset created for machine learning?

Dedicated experiments vary ethanol–fuel–water ratios, temperatures, and surface preparations to systematically build and preprocess the data space for model training.

What does the reported classifier performance indicate?

The classifier achieves a balanced accuracy of 0.87, indicating outstanding predictive ability to distinguish safe versus unsafe conditions and to guide subsequent experiments.

Aug 3, 2026

Upload by River Wang