Predicting Metal Shape for Better Material Design Using Advanced Computer Models
We recently spoke with Cara-Lena Nies from the Tyndall National Institute to learn more about her pioneering work in materials science. This research leverages advanced computational models to predict the formation and structure of thin metal layers on different surfaces at the atomic level. This is important for creating better materials for electronics and catalysts, as the shape of these metal layers can impact their performance. By predicting these structures before physical testing, the research speeds up material development and saves time and resources. This approach helps scientists design more effective materials for real-world applications.
Title
The Research
This project, led by Cara-Lena Nies from the Tyndall National Institute, utilises advanced, atomic-scale computational models to predict how thin metal layers form on different surfaces. The shape of these metal layers (smooth and spread out or clustered into islands and large particles) is critical for their performance in applications such as electronics and catalysis. For instance, electronics require smooth, uniform metal layers to function efficiently and conduct electricity, while catalysts often benefit from metal clusters that improve chemical reactions by providing a large surface area with multiple reaction sites. By predicting metal shapes before laboratory production, this research accelerates the design of materials, conserving both time and resources, and leads to more effective results through atomic-scale studies of growth mechanisms that would be extremely challenging in a laboratory setting.
Title
The Technology
The project relies on two key technological approaches:
Density Functional Theory (DFT): DFT is employed to predict the shapes of thin metal layers, enabling a detailed understanding of metal atom behaviour on material surfaces. This insight is essential for tailoring material properties to produce the desired metal shapes.
- High-Performance Computing (HPC): The complex models generated through DFT require significant computational power, provided by ICHEC’s high-performance computing resources. These advanced computing capabilities are crucial for processing the large models and vast amounts of data involved in accurately predicting metal growth patterns.
HPC resources:
This project has been allocated 102,400,000 hours, first on Kay, ICHEC's former supercomputer, and more recently on MeluXina, ICHEC's interim platform.
Title
The Impact
A major achievement of this project has been the development of a streamlined model workflow capable of accurately predicting if metals will form smooth or clustered structures on various surfaces. This breakthrough is not only valuable for the project itself but is now being applied to other research initiatives, broadening its scope and significance.
The research also delved into the complexities of metal growth, using high-performance computing to simulate the stability and aggregation of metal atoms. These simulations, which would be unfeasible on standard computing systems, were made possible by ICHEC’s HPC resources. The findings have far-reaching implications for material design, particularly in fields like electronics and catalysis. The ability to predict metal shapes with precision leads to better, more tailored materials, enhancing their performance in real-world applications.
Title
Introduction to the Research Team
Cara-Lena Nies is an expert in atomistic simulations of complex materials using Density Functional Theory. Her work focuses on developing advanced materials for sustainable development and (opto)electronic devices. Currently, she is researching the new catalyst materials for conversion of methane, as well as predicting the morphology of metals on 2D materials. She has previously studied the effect of boron doping on the optical and electronic properties of III-nitride alloys and modelled the growth of copper on modified tantalum nitride (TaN) surfaces, contributing to the advancement of materials science for a range of possible applications.
If you would like to view Cara-Lena's published work, you can explore it here
