2D material based memristors for neuromorphic computing and sensing

Two-terminal memristors have emerged as a basic circuit element for neuromorphic computing and non-Boolean logic. Recently, two-dimensional (2D) materials such as MoS2 have shown promise for memristor technology, in both lateral and vertical geometries (see Figure) [1][2]. The memristive behavior of these devices has been correlated with field-driven grain boundary motion in 2D materials, although a detailed understanding of this process remains elusive. Alternative realizations may rely on piezoresistive or ferroelectric effects, as well as phase change processes in 2D materials.

Schematic drawings of a (left) lateral  and (middle) vertical memristor based on grain boundaries in MoS 2. (right) IV characteristic of a metal/MoS 2 /metal junction, showing memristive behavior [2].

© Thomas Müller

Schematic drawings of a (a) lateral and (b) vertical memristor based on grain boundaries in MoS2. (c) IV characteristic of a metal/MoS2/metal junction, showing memristive behavior.


In this project we will study different concepts and architectures of 2D material-based memristive devices for neuromorphic computing and sensing. Microscopic studies in combination with theoretical modeling will allow to gain insight in the device physics. Devices will be integrated in a crossbar architecture and hardware implementation of 2D material-based  neuromorphic computing and sensing networks will be demonstrated.


2D semiconductors will be grown by chemical vapor deposition and the dependence of the memristive behavior on material properties (grain-size), device structure (geometry) and material composition (2D materials and contact metals) will be investigated. Device geometries will be optimized for achieving reliable control of multi-level weights. Memristors will be integrated in a crossbar, with non-linearities based on conventional 2D semiconductor circuits, to realize an artificial neural network. Microscopic studies of defects and grain boundary modification under strong electric fields/local heating will be performed to gain insight in the device physics. The results will be compared with macroscopic device models and microscopic simulations of defect modification under strong fields/local heating.


The student working on this project will work in close collaboration with the Diebold group to study defect and grain boundary modifications in 2D material memristors in UHV. Microscopic modelling will be performed in collaboration with the groups of Libisch (electronic properties of defects/grain boundaries) and Madsen (structural properties). Grasser will model device operation on a macroscopic scale to optimize the performance of memristors using the insights gained from the microscopic studies. Lendl will provide spatially-resolved current maps for memristors. Devices and networks will be fabricated and tested in the Mueller group. The 2D semiconductors used in this project will be grown in-house by the Mueller group, and also provided as source material for the Eder, Holzer (chemical functionalization), Filipovic (gas sensing), and Marchetti-Deschmann (laser desorption of biosamples) groups.


Thomas Mueller's research focuses on fabricating and  characterizing (opto)electronic devices based on 2D materials,  such as graphene, monolayer semiconductors, and 2D heterostructures. The group has a strong record in 2D materials  transfer and nano\-device fabrication. Internationally  recognized contributions on the electronic and optical properties of 2D materials include  the first ultrafast graphene-based photodetectors, nanolectronic microchips [3], van der Waals heterostructures in photovoltaics and photoconductivity studies of 2D semiconductors [4].


Group of Prof. Müller, opens an external URL in a new window


  1. V. K. Sangwan, D. Jariwala, I. S. Kim, K.-S. Chen, T. J. Marks, L. J. Lauhon, and M. C. Hersam. Gate-tunable memristive phenomena mediated by grain boundaries in single-layer MoS2. Nature Nanotechnology 10, 403–406 (2015). DOI: 10.1038/nnano.2015.56.
  2. R. Ge, X. Wu, M. Kim, J. Shi, S. Sonde, L. Tao, Y. Zhang, J. C. Lee, and D. Akinwande. Atomristor: Nonvolatile resistance switching in atomic sheets of transition metal dichalcogenides. Nano Letters 18, 434–441 (2017). DOI: 10.1021/acs.nanolett.7b04342.
  3. S. Wachter, D. K. Polyushkin, O. Bethge, and T. Mueller. A microprocessor based on a two-dimensional semiconductor. Nature Comm. 8, 14948 (2017). DOI: 10.1038/ncomms14948.
  4. M. M. Furchi, D. K. Polyushkin, A. Pospischil, and T. Mueller. Mechanisms of photoconductivity in atomically thin MoS2. Nano Letters 14, 6165–6170 (2014). DOI: 10.1021/nl502339q.