arXiv subjects

Diego Rubi

Publications and source records attributed to Diego Rubi.

11 recordsLinked to original sources

arXiv2026

Device-area selection of memristive transport regimes in epitaxial $Hf_{0.5}Zr_{0.5}O_{2}$-based ferroelectric devices

Ferroelectric memristive devices based on hafnia are promising systems for neuromorphic electronics, yet the interplay between polarization-modulated resistive changes and defect-mediated transport often leads to complex and debated switching mechanisms. Here, we investigate this competition in epitaxial Hf$_{0.5}$Zr$_{0.5}$O$_2$/La$_{0.67}$Sr$_{0.33}$MnO$_3$ heterostructures with Pt top electrodes by combining structural, ferroelectric, and memristive characterization with a statistical analysis across a broad range of device areas spanning three orders of magnitude. We identify two distinct memristive regimes with opposite resistance--voltage chiralities. Small devices exhibit a low-resistance state that scales inversely with area, consistent with area-distributed tunneling transport, while larger devices display an area-independent resistance indicative of localized conductive channels. A statistical nucleation model quantitatively captures this behavior and yields a crossover characteristic area $A^* \approx 10^3~\mu\mathrm{m}^2$. This crossover also correlates with the onset of ferroelectric wake-up in larger devices, linking conductive-channel nucleation and oxygen-vacancy redistribution within a unified physical picture. These results establish lateral device size as a key parameter controlling the dominant transport mechanism in epitaxial hafnia-based devices.

cond-mat.mtrl-sci↗

arXiv2025

Modeling Memristor-Based Neural Networks with Manhattan Update: Trade-offs in Learning Performance and Energy Consumption

We present a systematic study of memristor based neural networks trained with the hardware-friendly Manhattan update rule, focusing on the trade offs between learning performance and energy consumption. Using realistic models of potentiation/depression (P/D) curves, we evaluate the impact of nonlinearity (NLI), conductance range, and number of accessible levels on both a single perceptron (SP) and a deep neural network (DNN) trained on the MNIST dataset. Our results show that SPs tolerate P/D nonlinearity up to NLI $\leq 0.01$, while DNNs require stricter conditions of NLI $\leq$ 0.001 to preserve accuracy. Increasing the number of discrete conductance states improves convergence, effectively acting as a finer learning rate. We further propose a strategy where one memristor of each differential pair is fixed, reducing redundant memristor conductance updates. This approach lowers training energy by nearly 50% in DNN with little to no loss in accuracy. Our findings highlight the importance of device algorithm codesign in enabling scalable, low power neuromorphic hardware for edge AI applications.

cond-mat.mes-hall↗

arXiv2025

Synaptic plasticity in Co/Nb:STO memristive devices: The role of oxygen vacancies

Neuromorphic computing aims to develop energy-efficient devices that mimic biological synapses. One promising approach involves memristive devices that can dynamically adjust their electrical resistance in response to stimuli, similar to synaptic weight changes in the brain. However, a key challenge is understanding and controlling the coexistence of different types of synaptic plasticity, such as short-term and long-term plasticity. In this work, we show that plasticity behaviors in Co/Nb:STO Schottky memristors originate from oxygen vacancy electromigration, which modulates the Schottky barrier and enables both short-term and long-term plasticity. Our experiments reveal that resistance changes follow a power-law during reading (short-term plasticity) and increase stepwise with successive pulses (long-term memory retention). These behaviors are successfully reproduced by our model, which demonstrates the correlation between oxygen vacancy distribution and Schottky barrier modulation. Our findings highlight these memristors as promising candidates for neuromorphic applications.

cond-mat.mes-hall↗

arXiv2025

Oxygen vacancy engineering in pulsed laser deposited BaSnO$_3$ thin films on SrTiO$_3$

We demonstrate the tunability of oxygen content in pulsed laser deposition (PLD)-grown barium stannate (BaSn$O_3$, BSO) thin films by precisely controlling the background oxygen pressure over a broad range from 0.0004 mbar to 0.13 mbar. The introduction of oxygen vacancies significantly alters the structural properties of BSO films, inducing a monotonic expansion of the out-of-plane lattice parameter and cell volume as the vacancy concentration increases. The progressive formation of oxygen vacancies was spectroscopically tracked using X-ray photoelectron spectroscopy (XPS), providing direct insight into the vacancy evolution. Furthermore, we show that the oxygen stoichiometry in BSO plays a critical role in modulating the sheet resistance of BSO/LaScO$_3$ heterostructures, enabling interface metallic electron conduction. This oxygen content control offers a robust strategy to tailor the electronic properties at the interface, highlighting its potential for oxide electronics and functional interface engineering.

cond-mat.mtrl-sci↗

arXiv2024

Electrically-driven control of nanoscale chemical changes in amorphous complex oxide memristive devices

Our study demonstrates that strong cationic segregation can occur in amorphous complex oxide memristors during electrical operation. With the help of analytic techniques, we observed that switching the electrical stimulation from voltage to current significantly prevents structural changes and cation segregation at the nanoscale, improving also the device cycle-to-cycle variability. These findings could contribute to the design of more reliable oxide-based memristors and underscore the crucial effect that has the type of electrical stimulation applied to the devices on their integrity and reliability.

cond-mat.mtrl-sci↗

arXiv2024

Multi-mem behavior at reduced voltages in La$_{1/2}$Sr$_{1/2}$Mn$_{1/2}$Co$_{1/2}$O$_{3-x}$ perovskite modified with Sm:CeO$_2$

Neuromorphic computing aims to mimic the architecture and the information processing mechanisms of the mammalian brain, appearing as the only avenue that offers significant energy savings compared to the standard digital computers. Memcapacitive devices (which can change their capacitance between different non-volatile states upon the application of electrical stimulation) can significantly reduce the energy consumption of bioinspired circuitry. In the present work, we study the multimem (memristive and memcapacitive) behavior of devices based on thin films of the topotactic redox La$_{1/2}$Sr$_{1/2}$Mn$_{1/2}$Co$_{1/2}$O$_{3-x}$ (LSMCO) perovskite modified with Sm:Ce$O_2$ (SCO), grown on Nb:SrTiO$_{3}$ with (001) and (110) out of plane orientations. Either the self assembling at the nanoscale of both LSMCO and SCO phases or the doping with Ce(Sm) of the LSMCO perovskite were observed for different fabrication conditions and out of plane orientations. The impact of these changes on the device electrical behavior was determined. The optimum devices resulted those with (110) orientation and Ce(Sm) doping the perovskite. These devices displayed a multimem behavior with robust memcapacitance and significantly lower operation voltages (especially the RESET voltage) in comparison with devices based on pristine LSMCO. In addition, they were able to endure electrical cycling (and the concomitant perovskite topotactic redox transition between oxidized and reduced phases) without suffering nanostructural or chemical changes. We link these properties to an enhanced perovskite reducibility upon Ce(Sm) doping. Our work contributes to increase the reliability of LSMCO based multimem systems and to reduce their operating voltages closer to the 1 V threshold, which are key issues for the development of nanodevices for neuromorphic or in memory computing.

physics.app-ph↗

arXiv2023

Effect of memristor\'s potentiation-depression curves peculiarities in the convergence of physical perceptrons

Neuromorphic computing aims to emulate the architecture and information processing mechanisms of the mammalian brain. This includes the implementation by hardware of neural networks. Oxide-based memristor arrays with cross-bar architecture appear as a possible physical implementation of neural networks.In this paper, we obtain experimental potentiation-depression (P-D) curves on different manganite-based memristive systems and simulate the learning process of perceptrons for character recognition. We analyze how the specific characteristics of the P-D curves affect the convergence time -- characterized by the EPOCHs-to-convergence (ETC) parameter -- of the network. Our work shows that ETC is reduced for systems displaying P-D curves with relatively low granularity and non-linear and asymmetric response. In addition, we also show that noise injection during the synaptic weight actualization further reduces the ETC. The results obtained here are expected to contribute to the optimization of hardware neural networks based on memristors cross-bar arrays.

cond-mat.mes-hall↗

arXiv2022

Oxygen vacancy dynamics in Pt/TiOx/TaOy/Pt memristors: exchange with the environment and internal electromigration

Memristors are expected to be one of the key building blocks for the development of new bio-inspired nanoelectronics. Memristive effects in transition metal oxides are usually linked to the electromigration at the nanoscale of charged oxygen vacancies (OV). In this paper we address, for Pt/TiOx/TaOy/Pt devices, the exchange of OV between the device and the environment upon the application of electrical stress. From a combination of experiments and theoretical simulations we determine that both TiOx and TaOy layers oxidize, via environmental oxygen uptake, during the electroforming process. Once the memristive effect is stabilized (post-forming behavior) our results suggest that oxygen exchange with the environment is suppressed and the OV dynamics that drives the memristive behavior is restricted to an internal electromigration between TiOx and TaOy layers. Our work provides relevant information for the design of reliable binary oxide memristive devices.

cond-mat.mes-hall↗

arXiv2018

Oxygen vacancies dynamics in redox-based interfaces: Tailoring the memristive response

Redox-based memristive devices are among the alternatives for the next generation of non volatile memories, but also candidates to emulate the behavior of synapses in neuromorphic computing devices. It is nowadays well established that the motion of oxygen vacancies (OV) at the nanoscale is the key mechanism to reversibly switch metal/insulator/metal structures from insulating to conducting, i.e. to accomplish the resistive switching effect. The control of OV dynamics has a direct effect on the resistance changes, and therefore on different figures of memristive devices, such as switching speed, retention, endurance or energy consumption. Advances in this direction demand not only experimental techniques that allow for measurements of OV dynamics, but also of theoretical studies that shed light on the involved mechanisms. Along this goal, we analize the OV dynamics in redox interfaces formed when an oxidizable metallic electrode is in contact with the insulating oxide. We show how the transfer of OV can be manipulated by using different electrical stimuli protocols to optimize device figures such as the ON/OFF ratio or the energy dissipation linked to the writing process. Analytical expressions for attained resistance values, including the high and low resistance states are derived in terms of total transferred OV in a nanoscale region of the interface. Our predictions are validated with experiments performed in Ti/La$_{1/3}$Ca$_{2/3}$MnO$_{3}$ redox memristive devices.

cond-mat.mes-hall↗

arXiv2018

Comment on "Evidence for absence of metallic surface states in BiO2-terminated BaBiO3 thin films"

In a recent work by Ji Seop Oh et al., BaBiO3(001) thin films were grown on SrTiO3 by Pulsed Laser Deposition. It was argued that the films are BiO2-terminated from the modelling of angle-resolved photoemission spectroscopy experiments. The authors claim, in opposition to previous theoretical predictions, that there are no metallic surface states on their films. In this short comment we question that the authors have enough evidence to make such a claim, as we consider that the large mismatch between SrTiO3 and BaBiO3 and the lack of control of their fabrication process with reflection high energy electron difraction could unlikely give high quality films with a single BiO2- termination, which is one of the requisites for the stabilization of these surface metallic states.

cond-mat.mtrl-sci↗

arXiv2016

Tuning the electronic properties at the surface of BaBiO3 thin films

The presence of 2D electron gases at surfaces or interfaces in oxide thin films remains a hot topic in condensed matter physics. In particular, BaBiO3 appears as a very interesting system as it was theoretically proposed that its (001) surface should become metallic if a Bi-termination is achieved (Vildosola et al., PRL 110, 206805 (2013)). Here we report on the preparation by pulsed laser deposition and characterization of BaBiO3 thin films on silicon. We show that the texture of the films can be tuned by controlling the growth conditions, being possible to stabilize strongly (100)-textured films. We find significant differences on the spectroscopic and transport properties between (100)-textured and non-textured films. We rationalize these experimental results by performing first principles calculations, which indicate the existence of electron doping at the (100) surface. This stabilizes Bi ions in a 3+ state, shortens Bi-O bonds and reduces the electronic band gap, increasing the surface conductivity. Our results emphasize the importance of surface effects on the electronic properties of perovskites, and provide strategies to design novel oxide heterostructures with potential interface-related 2D electron gases.

cond-mat.mtrl-sci↗