Neural Network Algorithm for Accurately Estimating Arrhenius Crossover Temperature

Scheme of the machine learning model based on the feedforward artificial neural network. Credit: Materials (2023). DOI: 10.3390/ma16031127

The study of material science has always been a crucial aspect of human civilization, from developing new alloys for aircraft manufacturing to discovering materials that improve the performance of computer chips. The process of discovering new materials has always been a time-consuming and costly process, as it requires a lot of experimentation and testing. The development of new materials requires a deep understanding of the fundamental principles governing the material's properties, such as its melting temperature, glass transition temperature, and brittleness value. In recent years, machine learning has emerged as a promising approach for predicting the properties of new materials. In this article, we will discuss a new algorithm based on neural networks that can accurately predict the Arrhenius crossover temperature, a crucial property for many materials.

The Significance of the Arrhenius Crossover Temperature

Many materials, including metals, plastics, and glass, start their life as viscous liquids that solidify at a certain temperature. The temperature at which a change in the state of aggregation begins is called the Arrhenius temperature. This temperature is essential in determining the properties of the resulting solid material. When approaching the Arrhenius temperature, the atoms of matter begin to move more slowly, indicating the preparation of the liquid for solidification. Understanding the Arrhenius temperature is crucial for optimizing the production process of many materials, including metal alloys. However, estimating the Arrhenius temperature accurately can be a challenging task.

The New Algorithm for Predicting Arrhenius Crossover Temperature

A team of researchers from the Department of Computational Physics and Modeling of Physical Processes and Udmurt Federal Research Center of the Russian Academy of Sciences has developed a new algorithm that can predict the Arrhenius crossover temperature accurately. The algorithm is based on a neural network that was trained on several physical parameters of the material, including its melting temperature, glass transition temperature, and brittleness value. These parameters are used to describe phase transitions and structural changes in liquids during cooling.

Testing the Algorithm for Various Materials

The researchers tested the algorithm for metallic, silicate, borate, and organic glasses. They found that for the created neural network, the melting and glass transition temperatures of the material are significant and sufficient characteristics for estimating the Arrhenius temperature. From these two values, the algorithm determined the Arrhenius temperature for all analyzed liquids with an accuracy of more than 90%. The team further plans to adapt the created algorithm to more complex materials, such as polymers.

Implications of the New Algorithm

The new algorithm can have significant implications for the production of many materials, including metal alloys. Estimating the Arrhenius temperature accurately can help speed up the production process and simplify quality control. The algorithm can also help researchers discover new materials with desirable properties. By using machine learning algorithms like this, researchers can save a lot of time and resources that would otherwise be spent on trial and error.

Journal Information: Bulat N. Galimzyanov et al, Arrhenius Crossover Temperature of Glass-Forming Liquids Predicted by an Artificial Neural Network, Materials (2023). DOI: 10.3390/ma16031127