Artificial intelligence today consumes a huge amount of energy. One way to more energy-efficient calculations may be spin waves – magnetic phenomena that scientists are investigating as the basis for a new generation of chips and quantum technologies. Jan Klíma, a doctoral student at CEITEC, Brno University of Technology, is looking for ways to better understand and use their behavior. For his research, he received the prestigious Ministry of Education, Youth and Sports award for excellent graduates and also support from Brno Ph.D. Talent.
Modern electronics are based on controlling the flow of electric charge. However, this principle brings energy losses and heat generation. One way to overcome this limit may be spin waves – a special form of magnetic waves that allow information to be transmitted without the classical flow of electric current. It is their properties that Jan Klíma from Michal Urbánek's research group at CEITEC BUT is investigating.
"Our goal is not only to describe how spin waves behave, but above all to find out how these waves can be controlled. The ability to control these processes will be crucial for their future use in magnonic and quantum technologies," explains Jan Klíma .
Spin waves, also known as magnons, are created by the magnetic properties of materials and propagate in them in a similar way to other types of waves. However, unlike conventional electronics, they do not transfer electric charge and therefore do not generate significant heating. This makes them an interesting candidate for future low-energy technologies – from microprocessors to specialized computing systems.
Magnonics is a modern branch of physics that studies the behavior of spin waves and their quasiparticles, magnons, in nanostructured materials for the purpose of transmitting and processing information with minimal energy consumption.
One area where magnonics is attracting increasing attention is artificial intelligence. Neural networks need nonlinear elements to function – systems whose output is not simply a simple proportion of the input signal. While these properties are created artificially in current computer architectures, spin waves have them naturally.
"It is precisely the nonlinear behavior of spin waves that is interesting for applications in the field of machine learning and neural networks. There are already experiments today that show that they could be used, for example, for signal recognition," says Klíma.
However, his doctoral project focuses on another significant challenge – connecting magnonics with quantum technologies . Spin waves can interact with other physical systems, such as electrical or optical signals, and could thus serve as an intermediary for the transmission of information in quantum devices in the future.
A key problem Klíma is tackling is magnetic noise . This can disrupt sensitive quantum processes and cause information loss. The researcher is therefore trying to find out how to better describe the behavior of spin waves and how to modify the properties of materials to reduce this noise.
"If we could suppress magnetic noise even at higher temperatures, it could lead to more efficient and affordable quantum technologies in the future that would not require as much cooling," he adds.
Magnonics is one of the rapidly developing areas of modern physics. Although it is still a young field, it has experienced significant growth in recent decades and is being studied by leading research centers, especially in Germany, France, Austria, the USA, China and Japan. Michal Urbánek has been working on magnonics since 2015. Over the past decade, his research group has become a respected member of the international community focused on spin wave research.
Jan Klíma has been working in the group since his bachelor's studies, where he also completed his diploma thesis under the supervision of Michal Urbánek. Last year, he received the Ministry of Education, Youth and Sports Award for excellent graduates and support from the Brno Ph.D. Talent program, which helps develop talented doctoral students.
"I was very pleased with the Brno Ph.D. Talent award . The fact that my colleague Dominik Pavelka from the same research group also received support shows that BUT is also successfully developing research in this promising area," concludes Klíma.
Source: Science and Research
