Information and Communication Technologies (ICTs) are a broader term for Information Technology (IT), which refers to all communication technologies, including the internet, wireless networks, cell phones, computers, software, middleware, video conferencing, social networking, and other media applications. In addition, services enabling users to access, retrieve, store, transmit, and manipulate information in a digital form are included. Technically, ICT deals with the transmission and reception of information of some kind. This concept has become ubiquitous in everyday life, largely magnified by the ever-growing potential of autonomy in various applications. Unfortunately, the pace of such developments is limited by the existing platforms in ICT. In particular, the intricacy of these novel applications require capability exceeding that of the often used complementary-metal-oxide-semiconductor (CMOS) devices and require alternative information processing paradigms. Consequently, neuromorphic computing has evolved, with a particular regard to computer science research focusing on the development of a computer built to mimic the human brain’s neural networks. In fact, in order to realize such bio-inspired neuromorphic systems in hardware, numerous efforts have been made to realize scalable and power efficient artificial synapses. Thereto, the development of non-CMOS based neuronal architectures using memristors is an active area of research.
Recent advances in nanotechnology revealed the huge potential of low-dimensional nanostructures such as semiconductor nanowires as attractive building blocks of memristors. This is related to their enhanced surface-to-volume ratio and quantum confinement in significantly altering their electrical properties with respect to bulk materials. Moreover, such nanowires possess excellent physical advantages that make them of particular interest in the semiconductor industry. Nonetheless, the actual realization of memory devices based on the memristive effect is sluggishly progressing, due to processing requirements and the need for exotic materials which are not compatible with today’s CMOS technology. To address this, a consortium of researchers from the Institute of Solid State Electronics at the Vienna University of Technology in Austria: Mr. Raphael Böckle, Dr. Masiar Sistani, Dr. Philipp Staudinger and Dr. Alois Lugstein, in collaboration with Dr. Michael Stanislaus Seifner at the Centre for Analysis and Synthesis and Dr. Sven Barth at Goethe University Frankfurt, conduct experimental study on a Ge quantum wire device featuring distinct signatures of memristive behavior favorable for integration in CMOS platform technology. Their work is currently published in the research journal, Nanotechnology.
In their approach, they focused on demonstrating a memristive device, utilizing quantized charge carriers in ultra-scaled Ge NWs modulated by surface trap assisted electrostatic gating. To realize this, 18 nm thin single crystalline Ge nanowires were integrated in a back-gated field effect transistor (FET) architecture. The developed system was subjected to various tests for analysis and characterization at room-temperature and cryogenic temperatures.
The authors found that by embedding the quasi-1D Ge quantum wire into an electrostatically modulated back-gated FET, individual current transport channels could be addressed directly by controlling the surface trap assisted electrostatic gating. More so, it was seen that the resulting quantization of the current represented the ultimate limit of memristors with practically zero off-state current and low footprint. Most importantly, the proposed device was reported to possess the advantage of non-destructive successive reading cycles capability.
In summary, the study demonstrated a low footprint CMOS compatible Ge quantum wire device with memristive behavior and practically zero off-state current. The resistive switching behavior of the Ge quantum wire device was elaborated by conducting electrical measurements at room-temperature and cryogenic measurements. Thereof, the different modes of a classical memory system could be demonstrated, thus providing a platform which may pave the way for fully CMOS compatible neuromorphic computing. In a statement to Advances in Engineering, the authors explained their findings offer a framework towards fully CMOS compatible ultra-scaled Ge based memristors.
R Böckle, M Sistani, P Staudinger, M S Seifner, S Barth, A Lugstein. Ge quantum wire memristor. Nanotechnology 31 (2020) 445204 (6pp).