Cryoanalytics Laboratory Observations: "The Lab allows us to research material properties at extremely low temperatures"
Interview with Prof. Dr. rer. nat. habil. Martin Ziegler, Ilmenau University of Technology (Technische Universität Ilmenau, TU), Department of Electrical Engineering and Information Technology, Micro- and Nanoelectronic Systems (MNES)
The objective of the cryoanalytics laboratory at the TU Ilmenau is to develop a deeper "understanding" of materials by freezing physical processes to subsequently develop the next generation of hardware. Professor Martin Ziegler and his team are studying new methods to support digitization for the upcoming decades. Check out how in this interview!
Professor Ziegler, the Ilmenau University of Technology now has a cryo lab. What is your research focus?
Martin Ziegler: We explore biologically inspired electronic systems and study how they work. Digitization is prompting ever new challenges in information processing, especially as it pertains to artificial intelligence or machine learning. This technology is rapidly improving when you think of mobile phones with voice recognition (or speech recognition), for example. But it also involves an increase in energy demand. IT hardware is not designed to handle these types of technologies, meaning these types of information processing also require different types of hardware. What makes this an urgent issue? Currently, information technology already accounts for nearly 20 percent of global electricity consumption. At this rate, today’s energy generation will not be enough to meet the rising power requirements of IT hardware in 15 to 20 years.
There are two ways to solve this problem. Option one: We put an end to the digital revolution. Obviously, that’s impossible because we cannot go back to the way things were. Option two: We make information processing more energy efficient, which means we need different hardware to accommodate the digitization process. In doing so, we also bear in mind that computers are not necessarily superior to the human brain. Modern computers can do so much, but they are still vastly inferior to biology in some areas. And this is exactly the realm that interests us. The research interest of our field lies in the transfer of biological information processing and information storage into electronic systems. The technical term is "neuromorphic electronics". We are funded by the Federal Ministry of Education and Research, which launched a program in 2018 that promotes research infrastructure projects at German universities, while specifically targeting the field of microelectronics. The Technische Universität Ilmenau focuses on neuromorphic electronics.
You use the foundations of biology – the synapses in the human brain, for example – to come up with ways of faster data processing. How do the extremely low temperatures in your research come into play in this setting?
Ziegler: We don’t necessarily need the cryo lab for biological reasons. Our goal is to build a bridge between new materials for electronic components and how they can be used, and the storage and processing of information. This requires an understanding of fundamental physics concepts, that is to say, an understanding of materials we use in electronics. The cryo lab is essential in this setting. It allows us to research material properties at extremely low temperatures. Think of it like this: A material involves a multitude of atoms and physical processes. If you subject it to extremely low temperatures, you freeze these processes in time so to speak. You get a frozen picture of a dynamic particle system with multiple fluctuations. The temperature allows you to turn distinct physical processes on or off and examine them. Basically, the cryo lab features high-resolution microscopes that enable us to detect single atoms, for example. We are entering the realm of quantum mechanics, which is subject to different principles. We make use of these quantum mechanics principles to develop innovative electronic components.
View into the cryogenic scanning tunneling microscope.
We need innovative materials because we aim to take advantage of other physical processes for our bioinspired circuits. Let me give you a brief explanation: a computer does not work like the human brain. A computer has a CPU, which is the program. That means it has a defined processing cycle and sequence of events that is very similar to a recipe. The human brain works differently by using decentralized information processing. This means that the 100 billion neurons of a brain connect on average to 1,000 other neurons. These connections are flexible and are called synapses. In other words, the brain does not have a CPU. Instead, it relies on local, decentralized information processing. If you compare these two systems, they make different demands on the components and the systems we plan to develop. We can partially work with the existing structures, but we must come up with innovative approaches if we truly want to create efficient systems. The cryo lab is part of the overall concept.
When do you expect results that lead to industrial applications?
Ziegler: Energy consumption dictates our timeframe, making it the top project for the next ten years. Otherwise – as I have mentioned earlier - electrical energy will no longer cover the power requirements of the necessary IT hardware. That being said, we are a university and do not develop products but come up with innovative ideas. When it comes to basic research, we are typically about 20 years ahead of any commercial launch.
What future applications do you see when it comes to medical technology?
Ziegler: Potential applications include hearing assistive technologies. Let’s take a conversation in a cafeteria or a room with multiple sound sources. If our ears are healthy, we can focus our attention on what a person is saying over any background noises. We can filter out the pertinent signals from distracting noise sources and understand what was said. That is not always the case with conventional hearing aids. We are currently working on creating a cochlea replica, designed to enable adaptive sound localization. This could be a very exciting application for medical technology.
The interview was conducted by Simone Ernst and translated from german by Elena O'Meara. COMPAMED-tradefair.com