technology is everywhere. Currently, it is the Von Neumann's architecture that is applied to the electronic computing systems in which the different elements (memory, processor, and controller) are separated. [1] Nearly all the circuits within the memory and the processor are composed of complementary metal-oxidesemiconductor (CMOS) devices, which can be a problem once this technology cannot be further miniaturized without compromising its performance. [2] A practical example could be that in order to increase the operating frequency and the device density, which is necessary when downscaling, the power supply and the operation temperature would also increase, which would obviously degrade the system performance. Moreover, the Von Neumann system is not suited to solve real world problems where inputs and outputs are sometimes not specified [3] or to execute adaptive learning algorithms as it would be necessary in tasks, such as classification of unstructured data or pattern recognition. [4] To overcome the Von Neumann's bottleneck, the development of artificial intelligence technologies and new computer architecture designs are necessary. One of such novel approaches is neuromorphic computation, which operates with extremely low power consumption. It also maintains the massive parallelism found in the human brain [5] where neurons communicate information by electrical or chemical input signals passing through synapses. Synapses present a very important behavior called plasticity which consists in changing their strength (synaptic weight), either facilitating or inhibiting the connection between two neurons, through potentiation and depression, respectively. [6] One of the solutions to emulate biological synapses is the resistive switching (RS) device, or memristor. A memristor is basically a nonlinear two-terminal device whose conductance can be altered by external inputs and depends on the history of current that has flowed through the device. Application of both weight programming and weight processing signals are through its input terminals thus mimicking a synapse. In fact, the memristor can simulate the synapses' plasticity by continuously adapting its resistance into excitatory and inhibitory weights upon application of electrical Amorphous indium-gallium-zinc-oxide (a-IGZO) based memristive devices with molybdenum contacts as both top and bottom electrodes are presented aiming to be used in neuromorphic applications. Devices down to 4 µm 2 are fabricated using conventional photolithography processes, with an extraordinary yield of 100%. X-ray photoelectron spectroscopy and transmission electron microscopy performed on the developed structures confirm the presence of a thin intermixed oxide layer (4-5 nm) containing Mo 6+ oxidation state at the interface with the bottom contact. This results in Schottky diodelike characteristics at the pristine state with a rectification ratio of 3 orders of magnitude. The devices have electroforming-free and area-dependent analog resistive switching properties. Temperature...

Solution-processed metal oxides have been investigated as an alternative to vacuum-based oxides to implement low-cost, high-performance electronic devices on flexible transparent substrates. However, their electrical properties need to be enhanced to apply at industrial scale. Amorphous indium-gallium-zinc oxide (a-IGZO) is the most-used transparent semiconductor metal oxide as an active channel layer in thin-film transistors (TFTs), due to its superior electrical properties. The present work evaluates the influence of composition, thickness and ageing on the electrical properties of solution a-IGZO TFTs, using solution combustion synthesis method, with urea as fuel. After optimizing the semiconductor properties, low-voltage TFTs were obtained by implementing a back-surface passivated 3-layer In:Ga:Zn 3:1:1 with a solution-processed high-к dielectric; AlOx. The devices show saturation mobility of 3.2 cm2 V−1 s−1, IOn/IOff of 106, SS of 73 mV dec−1 and VOn of 0.18 V, thus demonstrating promising features for low-cost circuit applications.

of less destructive switching events, since the device reliability is independent of the reproducibility of the filament formation. [6] Frequently, an analog switching is observed, which is applicable for neuromorphic systems. [7][8][9][10] Reports exist in the literature, which explore area-dependent switching involving the homogeneous migration of defects or ions through the thickness of the switching matrix. [11,12] This can be referred to as 3D RS. In a device with asymmetric electrodes, both 1D RS and 3D RS result in counter eightwise (c8w) switching loops, when the bias is applied to the current blocking electrode. [13] Another type of area-dependent RS exists when ions are exchanged (or trapping occurs at the interface) between the switching matrix and the electrode. If the current transport through this interface is modulated as a consequence of the ion exchange/trapping, then this can be referred to as 2D RS. The direction of the switching loops is opposite to 1D RS and 3D RS, that is, eightwise (8w) when the bias is applied to the current blocking electrode. Frequently, 2D volatile switching has been observed as a secondary process, affecting the high resistive state (HRS) of an otherwise 1D RS device. [14][15][16][17] In addition, the choice between digital and analog operation modes by different device initialization schemes has been reported for NiO-based devices. [18] However, the resistance window for the analog mode was below one order of magnitude and the mechanism was discussed as 3D RS.Both in 3D RS and 2D RS, the active interface represents a barrier to the current flow, for example, a Schottky junction. [19] In the case of Ti/Pr 0.7 Ca 0.3 MnO 3 (PCMO) devices the underlying mechanism is based on a redox reaction between the titanium oxide interlayer and the PCMO. [20,21] In oxide-based devices with a platinum Schottky barrier bottom electrode, electronic trapping at interface states was proposed as a mechanism. [15] To allow a sizeable data retention, structural stabilization of the trapped charge needs to occur. [22] To achieve a high ratio between HRS and low resistance state (LRS) in 3D and 2D RS, a highly mobile Fermi level, that is, a highly variable charge carrier concentration is desirable. In thin-film transistors (TFT), AOS are known to yield tremendously high on/off ratios. [23] Diodes of AOS have rectification ratios of 10 6 . [24] These applications show how the material is able to change from an insulator to a metallic conductor by A room-temperature-processed resistive switching Schottky diode that can be operated in two distinct modes, depending solely on the choice of device initialization mode, is presented. Electroforming in the diode's reverse polarity leads to an abrupt filamentary switching with inherently long data retention at the expense of rectification. After this electroforming process, the devices may work in either a bipolar or unipolar manner with a resistance window of at least two orders of magnitude. Device initialization in the forward direction shows a smoo...

Neuromorphic computation based on resistive switching devices represents a relevant hardware alternative for artificial deep neural networks. For the highest accuracies on pattern recognition tasks, an analog, linear, and symmetric synaptic weight is essential. Moreover, the resistive switching devices should be integrated with the supporting electronics, such as thin-film transistors (TFTs), to solve crosstalk issues on the crossbar arrays. Here, an a-Indium-gallium-zinc-oxide (IGZO) memristor is proposed, with Mo and Ti/Mo as bottom and top contacts, with forming-free analog switching ability for an upcoming integration on crossbar arrays with a-IGZO TFTs for neuromorphic hardware systems. The development of a TFT compatible fabrication process is accomplished, which results in an a-IGZO memristor with a high stability and low cycle-to-cycle variability. The synaptic behavior through potentiation and depression tests using an identical spiking scheme is presented, and the modulation of the plasticity characteristics by applying non-identical spiking schemes is also demonstrated. The pattern recognition accuracy, using MNIST handwritten digits dataset, reveals a maximum of 91.82% accuracy, which is a promising result for crossbar implementation. The results displayed here reveal the potential of Mo/a-IGZO/Ti/Mo memristors for neuromorphic hardware.

Solution-based memristors are emergent devices, due to their potential in electrical performance for neuromorphic computing combined with simple and cheap fabrication processes. However, to reach a practical application in crossbar...