Dense crossbar arrays of non-volatile memory (NVM) devices represent one possible path for implementing massively-parallel and highly energy-efficient neuromorphic computing systems. We first review recent advances in the application of NVM devices to three computing paradigms: spiking neural networks (SNNs), deep neural networks (DNNs), and 'Memcomputing'. In SNNs, NVM synaptic connections are updated by a local learning rule such as spike-timing-dependent-plasticity, a computational approach directly inspired by biology. For DNNs, NVM arrays can represent matrices of synaptic weights, implementing the matrix-vector multiplication needed for algorithms such as backpropagation in an analog yet massively-parallel fashion. This approach could provide significant improvements in power and speed compared to GPU-based DNN training, for applications of commercial significance. We then survey recent research in which different types of NVM devices-including phase change memory, conductive-bridging RAM, filamentary and nonfilamentary RRAM, and other NVMs-have been proposed, either as a synapse or as a neuron, for use within a neuromorphic computing application. The relevant virtues and limitations of these devices are assessed, in terms of properties such as conductance dynamic range, (non)linearity and (a)symmetry of conductance response, retention, endurance, required switching power, and device variability.

No abstract

No abstract

The interaction of water and carbon dioxide with nanostructured epitaxial (Ba,Sr)TiO3(001) thin film and bulk single crystal SrTiO3(001) surfaces was studied using x-ray photoemission spectroscopy (XPS), thermal desorption spectroscopy (TDS), and density functional theory (DFT). On both surfaces, XPS and TDS indicate D2O and CO2 chemisorb at room temperature with broad thermal desorption peaks (423–723 K) and a peak desorption temperature near 573 K. A comparison of thermal desorption Redhead activation energies to adsorption energies calculated using DFT indicates that defect surface sites are important for the observed strong adsorbate-surface reactivity. Numerical calculations of the competetive adsorption/desorption equilibria for H2O and CO2 on SrTiO3(001) surfaces show that for typical atmospheric concentrations of 0.038% carbon dioxide and 0.247% water vapor the surfaces are covered to a large extent with both adsorbates. The high desorption temperature indicates that these adsorbates have the potential to impact measurements of the electronic structure of BaTiO3–SrTiO3(001) surfaces exposed to air, or prepared in high vacuum deposition systems, as well as the electrical properties of thin film ATiO3-based devices.

A method for the experimental determination of surface photoemission core-level shifts for 3d transition metals J. Appl. Phys. 98, 014908 (2005); 10.1063/1.1948508 Surface and interface chemical composition of thin epitaxial SrTiO 3 and BaTiO 3 films: Photoemission investigation