<h1 class="pgc-h-arrow-right" data-track="1" > This article is from the WeChat public account: X-MOLNews</h1>
For more than half a century, the integrated circuit industry has been driven by transistor microfilm ("Moore's Law"). However, as the difficulty of shrinking the size of transistors continues to increase, the space for performance improvement continues to shrink, and the development of integrated circuits has gradually entered the "post-Moore" era. In the "post-Moore" era, new devices and circuit functions will be the main drivers.
Neuromorphic hardware based on resistive devices is a representative of new technologies in the post-Moore era, with the unique advantages of neural network-like functions, memory-computing integration and large-scale parallel computing, which is expected to break through the inherent bottleneck of the von Neumann architecture. For a long time, the resistance device is facing the contradiction of "current-retention time", that is, limited by the inherent characteristics of the metal conductive filament, the retention time of the state of the resistive device is positively correlated with the operating current, which is contrary to its actual use needs (long-term nonvolatile storage function needs to be implemented at a lower current to ensure energy consumption advantage, while short-term volatile strobe function requires a larger open current to ensure read margin).
In view of the above problems, the brain-like computing team of the Department of Precision Instrumentation of Tsinghua University proposed a new resistance device technology for the conductive filament of the basic elemental semiconductor tellurium (Te), which realizes the high/low conductivity switching of the device through the growth/fracture of the conductive filament of the Te semiconductor.
Figure 1. Resistance characteristic curve of Te conductive filament drag devices and micromicring images of Te filament transmission electrons
The design takes advantage of the electrochemical activity, low melting point, low thermal conductivity and low conductivity of Te semiconductor materials to invert the dependence of the retention time of the state of the resistive device and the operating current. In micron-scale devices, the operating current of long-term nonvolatile storage mode can be as low as several microamperes (in nanoscale devices, it can be as low as tens of picoamps), while the operating current of short-term volatile strobe mode can reach several milliamps. The inversion of the "current-retention time" relationship is inseparable from the use of Te conductive filaments: the electrochemically generated Te conductive filaments have a greater resistance than the metal conductive filaments, and the device current is lower at the same voltage (low conductivity); and at larger currents, the Te conductive filaments tend to accumulate Joule heat (low thermal conductivity), so that the local maximum temperature exceeds the Te melting point (low melting point), so that the conductive filaments are fused again, forming short-term volatile characteristics. This performance successfully solves the "current-retention time" contradiction commonly faced in traditional drag-retaining devices, implements both memory and gating functions that meet the needs of the application on the same single device, and demonstrates the possibility of homogeneous integration of "memory-gating" units.
Due to the unique combination of electrical-thermal properties of Te, the resistance device using Te conductive filament has a large and unique potential optimization space, such as the optimization of the thermal conductivity of the dielectric, the function of the protective electrode, and the optimization of electronegativity.
Figure 2. Homogeneous integrated 1S1R unit based on Te conductive filament and its resistance performance curve
The findings were recently published in Nature Communications. Yang Yifei and Xu Mingkun, 2018 doctoral students from the Department of Precision Instruments of Tsinghua University, and Shujing Jia, postdoctoral fellow of the Institute of Cutting-edge Technology of Chips and Systems of Fudan University, are the co-first authors of this article. Associate Professor Li Huanglong, Associate Professor Pei Jing, Associate Professor of the Department of Precision Instruments of Tsinghua University, Associate Professor Liu Dameng of the Department of Mechanical Engineering, and Duan Wenrui, Associate Researcher of Beijing University of Electronic and Information Technology, are the co-corresponding authors of the paper, and the research has been funded by the National Natural Science Foundation of China, the Young Talents Entrustment Project of the China Association for Science and Technology, the Beijing Brain Science and Brain-like Research Center, the Key Project of the Ministry of Science and Technology, and the Beijing Municipal Major Project.
A new opportunity for the emerging tellurium semiconductor: making resistive switching devices
Yifei Yang, Mingkun Xu, Shujing Jia, Bolun Wang, Lujie Xu, Xinxin Wang, Huan Liu, Yuanshuang Liu, Yuzheng Guo, Lidan Wang, Shukai Duan, Kai Liu, Min Zhu, Jing Pei, Wenrui Duan, Dameng Liu & Huanglong Li
Common Nat., 2021, 12, 6081, DOI: 10.1038/s41467-021-26399-1
News source: Tsinghua University- Tsinghua News Network
https://www.tsinghua.edu.cn/info/1181/87963.htm
Research group recruitment of postdocs:
Recruitment Direction:
1. Experiments: non-volatile memory, memristor, neuromorphic device, new device integration process and circuit design, two-dimensional electronic devices
2. Theory: simulation of transport characteristics of nanoelectronic devices, theoretical calculation of functional electronic materials, research on new devices and new materials (magnetism, ferroelectricity, topology, etc.)
Planned annual recruitment number (perennial): postdocs: 2-3
Recruitment requirements:
Proficient in the experimental tools required for the research direction of the applicant, strong communication, comprehension, English writing skills, the past 2-3 years as the first author has published high-impact articles in a large field, can independently carry out the project.
How to apply:
Send resume to:
Indicate "Postdoctoral Application".
Introduction of the research group:
The research group conducts experimental and theoretical research on new semiconductor information devices and materials. In the past 3 years, the stage results have been achieved: neuromorphic transistor (experiment, Adv. Mater. 2019), Ultra steep sub-threshold swing transistors (Theory, Adv. Mater. 2020), 3D Storage Gating Switch Materials (Theory, Nat. Commun. 2020), memory-computing integrated resistance device (Experiment, Nat Commun. 2021) etc. Its own research conditions include DFT calculation software CASTEP, VASP, ATK, computing server Tsinghua supercomputing, Wuxi supercomputing, UK supercomputing, thin film growth equipment magnetron sputtering instrument, pulsed laser deposition system, device test platform semiconductor parameter analyzer, probe table, material characterization equipment atomic force microscope. The school has a public platform for micro-nano processing. The RESEARCH Group PI is Associate Professor Li Huanglong, relying on the Tsinghua University Brain-like Computing Research Center, Beijing Brain Science and Brain-like Research Center and other platforms to carry out research.