DFG-Abschlussberichte

Permanent URI for this collection

https://oa.tib.eu/renate/handle/123456789/18569

Browse

Browsing DFG-Abschlussberichte by Subject "Device Scalability"

Now showing 1 - 1 of 1
  • Thumbnail Image
    Item
    Collaborative Research Center 917 : Resistively Switching Chalcogenides for Future Electronics : Structure, Kinetics, and Device Scalability : Final report : 2019/2-2020-2021-2022-2023/1
    (Hannover : Technische Informationsbibliothek, 2025-03-25) Wuttig, Matthias; Waser, Rainer; Dronskowski, Richard; Dittmann, Regina; Simon, Ulrich; Mayer, Joachim
    The goal of SFB 917 has been the development of novel nanoswitches that can be reproducibly and reversibly changed between two states on very short time and length scales. Such nanoswitches can enable new storage and memory devices as well as neuro-inspired architectures for information technology. In the third and final funding period of SFB 917 we have witnessed and contributed to three major trends. The exponential growth in the demand for data storage and processing has continued. Hardware improvements are therefore urgently needed to meet the increased demands for data storage and processing as well as the related increase in energy consumption. SFB 917 strives to realize novel storage devices by exploiting the full potential of chalcogenide-based nanoswitches. Yet, it has become increasingly clear in the last few years, that improvements in device performance alone are insufficient to deal with the exponential growth mentioned above. The advance of Large Language Models (LLMs) like ChatGPT and related software has produced a further increase in data processing related energy consumption. This is a major challenge considering the expected further increase of professional and private usage of such machine learning tools. To minimize the related energy consumption, particularly energy-efficient software and hardware developments are mandatory. Within SFB9917, we have this intensified our efforts to work on effects related to a reduction in energy consumption. While energy-efficient devices can help, an improvement in hardware architecture offers significantly more leverage. We have thus explored the potential of chalcogenide based nano-switches in neuro-inspired computer architectures. To this end, several new large-scale research projects have been initiated (NeuroSys and NeuroTec), which extend our research on these devices in increasingly more complex architectures, offering new opportunities to harvest the findings of SFB 917 in new applications. To tailor chalcogenide-based nano-switches, major advances in instrumentation as well as an in-depth understanding of the origin of underlying phenomena and unconventional properties in these materials have been mandatory. Challenges included the characterization of switching in these materials on nanosecond time and nanometer length scales. Sophisticated tools have been built and utilized. Understanding unconventional properties are required building a bridge between concepts of inorganic chemistry and material properties leading to novel treasure maps which help to identify and tailor chalcogenides for specific applications. These successes are described in detail in the present report. To demonstrate that these findings are also relevant for industry, close cooperation with industrial partners has been established to ensure that the findings made within SFB 917 can also be implemented on the shortest possible time scales.