Negative differential resistance behavior in oxide memristors, especially those using NbO2, is gaining renewed interest because of its potential utility in neuromorphic computing. However, there has been a decade-long controversy over whether the negative differential resistance is caused by a relatively low-temperature non-linear transport mechanism or a high-temperature Mott transition. Resolving this issue will enable consistent and robust predictive modeling of this phenomenon for different applications. Here we examine NbO2 memristors that exhibit both a current-controlled and a temperature-controlled negative differential resistance. Through thermal and chemical spectromicroscopy and numerical simulations, we confirm that the former is caused by a ~400 K non-linear-transport-driven instability and the latter is caused by the ~1000 K Mott metal-insulator transition, for which the thermal conductance counter-intuitively decreases in the metallic state relative to the insulating state.

Transition metal oxide memristors, or resistive random-access memory (RRAM) switches, are under intense development for storage-class memory because of their favorable operating power, endurance, speed, and density. Their commercial deployment critically depends on predictive compact models based on understanding nanoscale physicochemical forces, which remains elusive and controversial owing to the difficulties in directly observing atomic motions during resistive switching, Here, using scanning transmission synchrotron x-ray spectromicroscopy to study in-situ switching of hafnium oxide memristors, we directly observed the formation of a localized oxygen-deficiency-derived conductive channel surrounded by a low-conductivity ring of excess oxygen. Subsequent thermal annealing homogenized the segregated oxygen, resetting the cells towards their as-grown resistance state. We show that the formation and dissolution of the conduction channel are successfully modeled by radial thermophoresis and Fick diffusion of oxygen atoms driven by Joule heating. This confirmation and quantification of two opposing nanoscale radial forces that affect bipolar memristor switching are important components for any future physics-based compact model for the electronic switching of these devices.Keywords: Memristors, thermophoresis, operating mechanism, oxygen migration, filament.The recent surge in technological and commercial interest in transition-metal-oxide memristors, especially those utilizing hafnium oxide as the switching material, is accompanied by urgent efforts to formulate a compact predictive model of their behavior in large-scale integrated circuits. 1-9 Several efforts in this direction include first-principles and analytical modeling, 8,10,11 materials characterization, 2,12,13 and circuit characterization and modeling. 14,15 The resultant models are incomplete and controversial owing to a lack of understanding of the nanoscale physico-chemical forces that determine atomic motions during switching, particularly with regard to the presence and sign of temperaturegradient-driven thermophoresis of oxygen atoms, and quantification of the concentration-gradient-driven Fick diffusion. 7,8,11,16,17 Direct in-situ and in-operando studies of localized atomic motion during memristor switching can resolve these issues and improve our modeling, but such observations face steep experimental challenges due to the extremely high resolution and sensitivity required to detect atomic motions inside a functioning cell. 2,4,18,19 In order to non-destructively study the chemical and position changes associated with oxygen atoms during memristor operation, we utilized a synchrotron-based scanning transmission x-ray microscopy (STXM) system tuned to the O K-edge with a spatial resolution of <31 nm and a spectral resolution of ~70 meV. 20 We analyzed a prototype device that had only one oxide layer to permit an unambiguous analysis of the results. To enable x-ray transmission experiments, operational memristor cells for this study wer...

Metal-oxide memristors, or resistive random access memory (RRAM) switches, in particular utilizing HfO x as the resistive switching material, have seen significant interest recently for nonvolatile memory and computation applications. [1][2][3][4][5] There has been particular interest in understanding the role of migration of oxygen atoms in determining the operation of memristors. 6-11 Similar recent advances in understanding the localized nanoscale physico-chemical changes underlying resistance switching 4,12-15 have opened up fresh interests into studying the effect of atomic movements on extended device operation and the nanoscale material behavior during eventual failure and possible techniques to mitigate such failure. [16][17][18] To enable scanning transmission x-ray microscopy (STXM) measurements, each device was built on a 200 nm low-stress Si 3 N 4 film suspended over 50 µm x 50 µm holes etched through a silicon substrate. 13 We fabricated crosspoint HfO x devices with an active area of 2 µm x 2 µm ( Figure 1a) by depositing a bottom electrode (15 nm Pt), a blanket layer of 6 nm HfO 2 , followed by the top electrode (10 nm TiN and 15 nm Pt). Typical currentvoltage plots of these devices (Figure 1b) exhibited the well-recognized resistance switching behavior, or pinched hysteresis loop, that characterizes a memristor. 19 During operation, high and low non-volatile resistance states (also called OFF and ON, respectively) were repeatedly accessed using bipolar voltage pulses. STXM experiments were performed using resonantly tuned x-ray beams mostly in the O K-edge region, with spectral resolution of ~70 meV and a beam diameter <30 nm. 20 The device was electrically connected inside the chamber of the system to enable in-situ operation and ON/OFF cycling to emulate ageing of the memristor. The x-ray absorption spectrum of the material stack within a device crosspoint before its operation (Figure 2a) revealed oxygen bonds to both Hf and Ti, suggesting oxidation of Ti upon sputter deposition of TiN onto HfO 2 and a resulting mixture of Ti and Hf oxides. We used the absorption of the pre O K-edge at 522 eV to monitor total thickness and other structural modulations (especially electrode distortions), the intensity of the 531 eV peak (the lowest conduction band of the stack) as an indicator of the relative conductivity within the crosspoint, 1,21 and the post O K-edge at 570 eV to determine the local oxygen concentration in the film.

Electrical actuation of liquid droplets at the microscale offers promising applications in the fields of microfluidics and lab-on-chip devices. Much prior research has targeted the electrical actuation of electrically conducting liquid droplets using DC voltages (classical electrowetting). Electrical actuation of conducting droplets using AC voltages and the actuation of insulating droplets (using DC or AC voltages) has remained relatively unexplored. This paper utilizes an energy-minimization-based analytical framework to study the electrical actuation of a liquid droplet (electrically conducting or insulating) under AC actuation. It is shown that the electromechanical regimes of classical electrowetting, electrowetting under AC actuation, and insulating droplet actuation can be extracted from the generic electromechanical actuation framework, depending on the electrical properties of the droplet, the underlying dielectric layer and the frequency of the actuation voltage. The paper also presents experiments which quantify the influence of the AC frequency and the electrical properties of the droplet on its velocity under electrical actuation. Velocities of droplets moving between two parallel plates under AC actuation are experimentally measured; these velocities are then related to the actuation force on the droplet which is predicted by the electromechanical model developed in this work. It is seen that the droplet velocities are strongly dependent on the frequency of the AC actuation voltage; the cut-off AC frequency, above which the droplet fails to actuate, is experimentally determined and related to the electrical conductivity of the liquid. This paper then analyzes and directly compares the various electromechanical regimes for actuation of droplets in microfluidic applications.

Reduction of resource consumption in data centers is becoming a growing concern for data center designers, operators and users. Accordingly, interest in the use of renewable energy to provide some portion of a data center's overall energy usage is also growing. One key concern is that the amount of renewable energy necessary to satisfy a typical data center's power consumption can lead to prohibitively high capital costs for the power generation and delivery infrastructure, particularly if on-site renewables are used. In this paper, we introduce a method to operate a data center with renewable energy that minimizes dependence on grid power while minimizing capital cost. We achieve this by integrating data center demand with the availability of resource supplies during operation. We discuss results from the deployment of our method in a production data center.