ACS Publications Establishes Data-Driven AI Design Framework for Mg-Sr-X Ternary Anodes in High-Performance Mg-Air Batteries

August 17, 2026

ACS Publications – Industrial & Engineering Chemistry Research USA

Overview

A paper in ACS Publications establishes an AI-driven alloy design framework for data-driven discovery of ternary anodes for high-performance Mg-air batteries. Integrating machine learning, explainable AI, and operations research, the framework trains an XGBoost model on a curated dataset of 819 experimental records, enhancing prediction accuracy for discharge performance metrics. This accelerates the development of next-generation metal-air batteries.

In Depth

Key Findings

Research published in ACS Publications’ Industrial & Engineering Chemistry Research journal has established an AI-driven alloy design framework for the data-driven discovery of ternary anodes (Mg-Sr-X alloys) for high-performance magnesium-air (Mg-air) batteries. This framework successfully integrates machine learning, explainable AI (XAI), and operations research techniques, utilizing an XGBoost model trained on a curated dataset of 819 experimental records to significantly enhance the prediction accuracy of discharge performance metrics.

Technical / Clinical Details

This AI-driven framework operates through the following steps:

Data Curation: Constructing a large and high-quality dataset consisting of 819 experimental records for Mg-air batteries (including alloy composition, manufacturing conditions, and discharge performance data).
Machine Learning Model Construction: Training an XGBoost (eXtreme Gradient Boosting) model using this dataset. XGBoost is an ensemble learning method known for its high predictive accuracy and processing speed.
Discharge Performance Prediction: The trained XGBoost model accurately predicts the discharge performance (e.g., discharge voltage, current density, energy density, cycle stability) of a new Mg-Sr-X alloy when its composition is provided.
Explainable AI (XAI): Incorporating XAI tools to interpret the model’s predictions. This helps understand which alloy compositional elements or process conditions have the greatest impact on performance, assisting material scientists in more intuitively refining designs.
Operations Research Techniques: Based on the prediction model and XAI results, efficiently identifying the most promising alloy compositions to be experimentally synthesized and tested, thereby optimizing experimental design.

This integrated approach dramatically shortens the discovery cycle for high-performance Mg-air battery anodes and reduces development costs compared to conventional trial-and-error methods.

Background & Context

Metal-air batteries are garnering significant attention as next-generation battery technologies for electric vehicles and large-scale energy storage due to their high theoretical energy density. Mg-air batteries, in particular, offer advantages over lithium, being abundant in resources and safer. However, the performance of anode materials (especially discharge stability and efficiency) has been a major hurdle to practical implementation. Traditional alloy design processes, heavily reliant on experimentation, are time-consuming and costly, making it difficult to find optimal compositions. AI and data-driven approaches offer powerful solutions to overcome this bottleneck, enabling more efficient material exploration and optimization.

Strategic Significance & Outlook

The established AI-driven alloy design framework will accelerate the performance enhancement of Mg-air batteries, marking a critical step towards their commercialization. The combination of a highly accurate XGBoost model and explainable AI will enable material scientists to gain deeper insights and rapidly design unprecedented high-performance anodes. This technology is applicable not only to Mg-air batteries but also to other metal-air battery systems and the development of a wide range of energy storage materials, holding the potential to contribute to the integration of renewable energy and the realization of sustainable transportation systems. This is a powerful example of AI as an indispensable element in shaping the future of clean energy technologies.

Source: https://pubs.acs.org/iecred/article/doi/10.1021/acs.iecr.6c01891/5253646/Data-Driven-Discovery-of-Mg-Sr-X-Ternary-Anodes

Get our weekly technology intelligence — free

Receive an infographic that lets you judge at a glance whether each field’s analysis report is worth reading.

Subscribe Free — Weekly Tech Intelligence

By subscribing, you’ll receive Troy-Technical’s weekly technology intelligence newsletter.