Advances in Synthesis and Integration Methods for Enhanced Properties, and Applications in Emerging Nanomaterials: 2D and Neuromorphic Materials
Sponsored by: TMS: Nanomaterials Committee
Program Organizers: Chang-Yong Nam, Brookhaven National Laboratory; Jung-Kun Lee, University of Pittsburgh; Zubaer Hossain, University of Delaware
Monday 2:00 PM
November 2, 2020
Room: Virtual Meeting Room 25
Location: MS&T Virtual
Session Chair: Jeehwan Kim, Massachusetts Institute of Technology; Deep Jariwala, University of Pennsylvania
2:00 PM Invited
Electrochemical 2D Synapses for Neuromorphic Computing Applications: Feng Xiong1; 1University of Pittsburgh
Inspired by the human brain, neuromorphic computing has recently attracted much research attention. However, majority of the current research efforts towards developing artificial synapses are based on the binary SRAMs, making it impractical to scale up the system to the level of complexity we need. We present a novel approach to build electrochemically-tunable, two-dimensional (2D) synapses with excellent controllability, good energy efficiency, symmetric resistance response, and a rare combination of low-power programming and good retention. In our 2D synapses, the channel conductance (synaptic weight) can be modulated by controlling the concentration of ions between layers of 2D materials through a process called electrochemical intercalation. The major advantage is that we can achieve reversible and precise programming of the 2D device’s conductance to mimic synaptic plasticity with low power consumption. This work can lead to the low-power hardware implementation of neural networks for neuromorphic computing.
2:30 PM Invited
Iontronic Devices for Energy Efficient Electronics and Neuromorphic Computing: Ke Xu1; Zhongmou Chao1; Susan Fullerton-Shirey1; 1University of Pittsburgh
The continuous miniaturization of electronics is nearing an end due to physical constraints. Novel materials systems and device concepts are needed to create next generation electronics that are more powerful, but requires less energy. Two-dimensional (2D) materials are molecularly thin, layered materials that have great potential in energy-efficient electronics and photonics. Iontronics is a newly emerging, interdisciplinary concept that bridges electronics, ionics, solid-state physics and biological science. An iontronic device has electronic properties or functions controlled by ionic motion and arrangement. I will describe some of our work in the Nanoionics and Electronics Lab using ions to control transport in 2D materials for applications such as neuromorphic computing (e.g. iontronic artificial synapses). I will also discuss some results on the direct-writing of silver nanofilaments in ionic liquid-filled solid polymer electrolytes to achieve multiple distinguishable resistance states, which could be useful for artificial neural networks.
3:00 PM Invited
Impact of Processing Parameters on Metal Oxide Resistive Random Access Memory (RRAM) Performance and Implications for non-von Neumann Computing Approaches: Nathaniel Cady1; 1SUNY Polytechnic Institute
Resistive Random Access Memory (ReRAM) devices are a novel form of non-volatile memory expected to replace a variety of current memory technologies and enable the design of new, non-von Neumann circuit architectures. A variety of challenges persist, however, for integrating memristors with CMOS, as well as for tuning device electrical performance. We have found that tuning the processing parameters used in RRAM device integration greatly affects device performance. This includes both deposition conditions (typically atomic layer deposition) and reactive ion etching conditions. We are particularly focused on fabrication strategies that reduce stochastic switching behavior during both binary and analog RRAM device switching. This is a key metric for neuromorphic applications, as variability in device conductance state directly influences the ultimate number of levels (weights) that can be implemented per synapse.
3:30 PM Invited
Efficient Neuromorphic Computing Enabled by Spin-Transfer Torque: Devices, Circuits and Systems: Abhronil Sengupta1; 1Penn State University
While research in designing brain-inspired algorithms have attained a stage where such Artificial Intelligence platforms are being able to outperform humans at several cognitive tasks, an often-unnoticed cost is the huge computational expenses required for running these algorithms in hardware. Bridging the computational efficiency gap necessitates the exploration of devices, circuits and architectures that provide a better match to the computational primitives of biological processing. Recent experiments in spintronic technologies are revealing immense possibilities of implementing a plethora of neural and synaptic functionalities by single spintronic device structures that can be operated at very low terminal voltages. Leveraging insights from such experiments, I will present a multi-disciplinary perspective across the entire stack of devices, circuits and systems to envision the design of an "All-Spin" neuromorphic processor enabled with on-chip learning functionalities that can potentially achieve two to three orders of magnitude energy improvement in comparison to state-of-the-art CMOS implementations.
4:00 PM
Introduction to Research Capabilities at Center for Functional Nanomaterials, a DOE National User Facility: Chang-Yong Nam1; 1Brookhaven National Laboratory
The Center for Functional Nanomaterials (CFN) at Brookhaven National Laboratory is one of the Nanoscale Science Research Centers (NSRCs), a system of five coordinated Nano Centers located in U.S. Dept. of Energy (DOE) national laboratories across the United States. Each Center contains laboratories for synthesis and nanofabrication, one-of-a-kind signature instruments, a suite of supporting instrumentation, and theory, modeling, and simulation expertise, which are open to external users for free and can be accessed via user proposal processes. In this talk, I will briefly introduce the research activities and associated user facilities for materials research available at CFN, which include but are not limited to advanced surface and X-ray characterization capabilities (LEEM, XPEEM, μ-ARPES, μ-XPS), optical spectroscopy, electron microscopy, and automated 2D heterostructure fabrication facility (QPress) that is under development.