Variational Quantum Circuit-Based Quantum Machine Learning Approach for Predicting Corrosion Inhibition Efficiency of Expired Pharmaceuticals

Aug 5, 2026

Upload by Sage

This study investigates how quantum machine learning (QML) can predict corrosion inhibition capacity for expired pharmaceutical compounds. A QSPR workflow is used, where features derived from density functional theory (DFT) calculations feed a variational quantum circuit (VQC) model, while corrosion inhibition efficiency (CIE) from experimental measurements serves as the target. Model performance varies with encoding and ansatz design, yielding RMSE 6.15, MAE 5.63, and MAD 5.50. Findings emphasize the importance of larger datasets to improve predictive accuracy and highlight QML potential for anti-corrosion material exploration.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages11

Reading Time28 min

AccessFree to download

Abstract

Introduction

Corrosion and corrosion inhibitors

Materials informatics and QSPR-based ML

Prior studies using ML for corrosion inhibition

Motivation for quantum machine learning

Variational Quantum Circuit-Based Quantum Machine Learning Approach for Predicting Corrosion Inhibition Efficiency of Expired Pharmaceuticals MuhamadAkrom1, Muhammad Reesa Rosyid2, Lubna Mawaddah2, Akbar Priyo Santosa2 1Research Center for Quantum Computing and Materials Informatics, Faculty of Computer Science, Universitas Dian Nuswantoro, Indonesia 2Faculty of Computer Science, Universitas Dian Nuswantoro, Indonesia Article Info ABSTRACT | Article history:
Received April 23, 2024 Revised November 15, 2024 Accepted November 15, 2024 Published April 01, 2025 | This study examines the potential of quantum machine learning (QML) to predict the corrosion inhibition capacity of expired pharmaceutical compounds. The investigation employs a QSPR model, using features generated from density functional theory (DFT) calculations as input. Atthe same time, corrosion inhibition efficiency (CIE) values obtained from experimental data serve as the target output. The VQC model demonstrates varied performance across evaluation metrics, especially with encoding and ansatz design. The model achieves fine scores in evaluation metrics, with root mean square error (RMSE) of 6.15, mean absolute error (MAE) of 5.63, and mean absolute deviation (MAD) of 5.50. The research underscores the significance of larger datasets for enhancing predictive accuracy and points to QML's potential in exploring anti-corrosion materials. Although there are some limitations, this study provides a foundational framework for using QML to predict anti-corrosive properties. | | --- | --- | | Keywords:
Ansatz Designs
Corrosion Inhibitors
Drug Compounds Quantum Machine Learning Variational Quantum Circuit | | Corresponding Author: MuhamadAkrom, Research Center for Quantum Computing and Materials Informatics, Faculty of Computer Science, Universitas Dian Nuswantoro, Indonesia [Email: m.akrom@dsn.dinus.ac.id](Email: m.akrom@dsn.dinus.ac.id) 1. INTRODUCTION Corrosion is a process that causes damage to metals due to the electrochemical interaction between the metal surface and its corrosive environment [1], [2] . Gradually, corrosion reduces the lifespan of materials, potentially shortening their expected service life. The global damage caused by corrosion reached at least $2.5 trillion, equivalent to 3.4% of the worldwide GDP in 2013, making mitigating corrosion effects a critical priority. It is estimated that 15-35% of these losses could be prevented by applying corrosion inhibitors [3] . Utilizing corrosion inhibition is highly effective in preventing corrosion. Corrosion inhibitors typically consist of organic compounds containing heteroatoms such as nitrogen (N), phosphorus (P), sulfur (S), arsenic (As), or oxygen (O) within their molecular structure because these compounds possess free electrons or π electrons on aromatic rings or double bonds, allowing for strong interactions between metal atoms and organic molecules to form a protective layer on the metal surface. This layer is absorbed at the interface of the corrosive solution due to structural similarities, leading to numerous compounds being tested as corrosion inhibitors [4],[5],[6] . Currently, materials informatics is gaining popularity due to the widespread development of technologies such as Machine Learning (ML), particularly in the design and development of new materials [7], [8], [9] . The ML approach based on the Quantitative Structure-Property Relationship (QSPR) model is frequently employed to evaluate the performance of compounds because molecular attributes can be measured and related to the chemical structure of these compounds [3], [10],[11] . This approach replaces costly, time-consuming experimental procedures that require significant resources [12],[13], [14]. The ML approach to QSPR targets the Corrosion Inhibition Efficiency (CIE) derived from experimental results and utilizes Quantum Chemical Properties (QCP) obtained from Density Functional Theory (DFT)

FAQs

What is the main goal of this study?

To predict the corrosion inhibition efficiency of expired pharmaceutical drug compounds using a variational quantum circuit-based quantum machine learning approach.

How are input features and targets defined in the model?

DFT-derived quantum/chemical features are used as inputs, while experimentally obtained corrosion inhibition efficiency (CIE) values are used as the target output.

Which factors affect the VQC model’s performance?

The study reports that encoding and ansatz design choices lead to varied performance across evaluation metrics.

Aug 5, 2026

Upload by Sage