How Artificial Intelligence Can Help Design Advanced Materials

As industries search for corrosion-resistant alloys and protective coatings, artificial intelligence is playing an increasingly pivotal role in designing new alloys. Photo by Getty Images.

The Max Planck Society for the Advancement of Science (Munich, Germany) is a formally independent non-governmental and nonprofit association of German research institutes. Founded in 1911 as the Kaiser Wilhelm Society, it was renamed in 1948 in honor of its former president, theoretical physicist Max Planck. 

The society is funded by Germany’s federal and state governments.

In recent months, researchers at the society have introduced a new machine learning (ML) model, which they believe can enhance the predictive accuracy of corrosion-resistant alloy designs by up to 15% when compared to existing frameworks.

Role of Artificial Intelligence

As industries search for corrosion-resistant alloys and protective coatings, artificial intelligence (AI) is playing an increasingly pivotal role in designing new alloys. Yet, the predictive power of AI models in foreseeing corrosion behavior and suggesting optimal alloy formulas has remained elusive, according to the researchers.
To address this, the model developed by a Max Planck Institute for Eisenforschung tean uncovers new, but realistic corrosion-resistant alloy compositions. According to the researchers, its distinct power arises from fusing both numerical and textual data. 

Using Unique Alloy Properties

Initially developed for the critical realm of resisting pitting corrosion in high-strength alloys, they believe this model’s versatility can be extended to all alloy properties. The researchers published their latest results in the journal Science Advances.

“Every alloy has unique properties concerning its corrosion resistance,” says Kasturi Narasimha Sasidhar, lead author of the publication and former postdoctoral researcher at the Max-Planck-Institut für Eisenforschung.

“These properties do not only depend on the alloy composition itself, but also on the alloy’s manufacturing process,” Sasidhar adds. “Current machine learning models are only able to benefit from numerical data. However, processing methodologies and experimental testing protocols, which are mostly documented by textual descriptors, are crucial to explain corrosion.”

(a) Schematic representation of the entire process-aware deep neural network model, and (b) Schematic illustration of the data processing workflow carried out within the natural language processing (NLP) module. LSTM stands for long short-term memory. Image from Science Advances via Max Planck Society.

Future Steps for the Model

The researcher team used language processing methods, akin to ChatGPT, in combination with ML techniques for numerical data. From there, they developed a fully automated natural language processing framework. By involving textual data into the ML framework, the researchers say they can identify enhanced alloy compositions resistant to pitting corrosion. 

“We trained the deep-learning model with intrinsic data that contain information about corrosion properties and composition,” says Michael Rohwerder, co-author of the publication and head of the corrosion group at the Max-Planck-Institut für Eisenforschung. “Now, the model is capable of identifying alloy compositions that are critical for corrosion-resistance, even if the individual elements were not fed initially into the model.”

In the recently devised framework, Sasidhar and his team harnessed manually gathered data as textual descriptors. As of now, their objective lies in automating the process of data mining and seamlessly integrating it into an existing framework. 

Going forward, they believe incorporating microscopic images will mark another milestone by presenting the next generation of AI frameworks with textual, numerical, and image-based data.

Source: Max-Planck-Gesellschaft,

Related Articles