United States

The paper presents the results of numerical modeling and experimental studies of the regularities of the manifestation of the memristor effect of oxide nanodots obtained by the method of local anodic oxidation.

Materials Challenges for Memory

Memristive properties of titanium oxide nanodot structures for resistive memory

probe nanotechnology

nanodots

nanotechnology

rram

memristors

neural networks

The 15th Joint MMM-INTERMAG Conference (2022 Joint) is sponsored jointly by AIP Publishing and the IEEE Magnetics Society. Members of the international scientific and engineering communities interested in recent developments in fundamental and applied magnetism are invited to attend and contribute to the technical sessions. The technical program will include invited and contributed papers in oral and poster sessions, invited symposia, a plenary session, and an evening session, with about 1500 presentations overall. This Conference provides an outstanding opportunity for worldwide participants to meet their colleagues and collaborators and discuss developments in all areas of magnetism research.

The 2022 Joint Conference will offer both in-person and prerecorded on-demand content.

For tickets to the event, please register at the following link: https://magnetism.org/register

Please register for the Joint MMM-Intermag conference to join the event.

2022 Joint MMM-INTERMAG

This Conference provides an outstanding opportunity for worldwide participants to meet their colleagues and collaborators and discuss developments in all areas of magnetism research.

poster


![](https://assets.underline.io/uploads/markdown_image/1/image/2e10c515564e9ec3287ee4c664011c20.png)

The energy cost of storing and moving data is enormous, and is growing at such a rapid pace that within a few years the power demand may overwhelm some developed nations. Non-volatile memory (NVM) and neuromorphic computing are enabling critical advances in this area, but there are many scientific and technological challenges. This conference will cover the materials challenges of different memory technologies, from the fundamental processes and mechanisms, to industry perspectives on materials specifications and manufacturing.

High density, high-speed and low power consuming nonvolatile memories are currently being vigorously explored for use in next-generation computation, particularly due to the performance gap between the logic and memory elements of the current computational architecture.[1–3] Of various explored material systems, electrically switchable spontaneous polarization of ferroelectric materials enables a robust nonvolatile memory solution.[4–6] Using ultrathin films of ferroelectric materials as a tunnel barrier in metal/ferroelectric/metal trilayer structure, so called ferroelectric tunnel junctions (FTJ), is being explored widely as a potential nonvolatile memory elements. Unlike to ferroelectric RAM (FeRAM), FTJ offers nondestructive readout, in addition to low operation energy and high operation speed.[7]
In this work,[8] we have demonstrated FTJs with a very large OFF/ON resistance ratio and relatively low resistance area product (RA) with ~ 1 nm thick Zr doped HfO2 (HZO) ferroelectric tunnel barrier. We stabilized ferroelectricity in ultrathin films of rhombohedral HZO (R-HZO) through the substrate-induced compressive strain. The resistance area product at the bias voltage (~ 300 mV) required for one-half of the zero-bias TER ratio is three orders of magnitude lower than reported value with relatively thick ferroelectric barriers. [9, 10] The OFF/ON resistance ratio of 135 achieved with ~ 1 nm barrier, which is enhanced to 105 with increasing the barrier thickness to 2.5 nm, is the highest reported value for HZO barrier based FTJs. These results set the stage for further exploration of Hafnia-based FTJs for non-volatile memory applications.

References
[1] S. Manipatruni, D. E. Nikonov, I. A. Young, Nat.
Phys. 2018, 14, 338.
[2] H. S. P. Wong, S. Salahuddin, Nat. Nanotechnol.
2015, 10, 191.
[3] B. Prasad, Y. L. Huang, R. V. Chopdekar, Z. Chen,
J. Steffes, S. Das, Q. Li, M. Yang, C. C. Lin, T.
Gosavi, D. E. Nikonov, Z. Q. Qiu, L. W. Martin, B.
D. Huey, I. Young, J. Íñiguez, S. Manipatruni, R.
Ramesh, Adv. Mater. 2020, 32, 1.
[4] S. Manipatruni, D. E. Nikonov, C.-C. Lin, T. A.
Gosavi, H. Liu, B. Prasad, Y.-L. Huang, E.
Bonturim, R. Ramesh, I. A. Young, Nature 2019,
565, 35
[5] S. Oh, H. Hwang, I. K. Yoo, APL Mater. 2019,
091109 7.

Hf0.5Zr0.5O2 ferroelectric tunnel barriers for nonvolatile memory applications

Phase change alloys have seen widespread use from rewritable optical discs to the present day interest in their use in emerging neuromorphic computing architectures. In spite of this enormous commercial interest, the physics of carriers in these materials is still not fully understood. In particular, the role of non-thermal process in the switching physics of PCM is still debated. Here, we describe the time and space dependence of the coupling between photoexcited carriers and the lattice in both the amorphous and crystalline states of one phase change material, GeTe. Our work reveals a clear evolution of the electron-phonon coupling during the phase transformation as the spectra of photoexcited acoustic phonons in the amorphous (a-GeTe) and crystalline (α-GeTe) phases are different. In particular and surprisingly, our analysis of the photoinduced acoustic pulse duration in crystalline GeTe suggests that a part of the energy deposited during the photoexcitation process takes place over a distance that clearly exceeds that defined by the pump light skin depth. In the opposite, the lattice photoexcitation process remains localized within that skin depth in the amorphous state. We then demonstrate that this is due to supersonic diffusion of photoexcited electron-hole plasma in the crystalline state. Consequently these findings prove the existence of a non-thermal transport of energy which is much faster than lattice heat diffusion.

Non-thermal transport of energy driven by photoexcited carriers in switchable solid states of GeTe

Resistive switching devices based on metal-insulator-metal (MIM) structures have substantial potentials in both non-volatile memories (NVM) and neuromorphic computing circuits. Most of the MIM devices are highly insulating at the pristine state and requires a large bias (electroforming step) to set the devices to the operable resistive switching situation. Electroforming step may cause the stochastic characteristics of the conductive filament formation, corresponding to poor synaptic characteristics.
In the present work, we employed a WOx/ZrOx stack with Ta/Pt electrodes to achieve an electroforming-free Ta/WOx/ZrOx/Pt resistive switching device. This device is also self-rectifying because the ZrOx/Pt interface creates a Schottky barrier, which provides an opportunity to solve the sneak-path-current problem in memristive crossbar array. Furthermore, Ta/WOx/ZrOx/Pt devices show analog-switching behavior, which allow us to demonstrate the synaptic plasticity via pulse stimulation with different amplitude or interval pulse. Then, synaptic functions of potentiation/depression cycles, paired-pulse facilitation (PPF) and spike-timing plasticity (STDP) have been successfully simulated. The excellent linearity and repeatability potentiation/depression cycles in the Ta/WOx/ZrOx/Pt device were applied to model the synaptic weight update in neural network (NN) simulation platform for pattern recognition accuracy using the MINST dataset of hand-written digits. The resulting NN inference accuracy of the Ta/WOx/ZrOx/Pt synaptic devices reaches 97.7% after 100 epochs, indicating that this device make it very suitable for on-chip training of artificial neural network.

Electroforming-free resistive switching of WOx/ZrOx stack for neuromorphic computing systems

Currently, there is a high demand for developing shape memory alloys that can be used at high temperatures such as aerospace and Ti50Pd50-XRuX is found to be one of the promising alloy. This is due to their shape memory effect and super elasticity. Previous studies revealed that the alloy is mechanically stable above 25 at. % Ru with a positive C^' at 0 K. So, an understanding of its transformation temperature behaviour is required, which is currently not clear. The self-consistent-charge density functional tight binding (SCC-DFTB) approach was used to develop parameters for Ti50Pd50-XRuX employing the parameterization technique. The structural and electronic properties were determined to check stability on derived potentials using DFTB+ code. As part of the validation for the developed set, the results are in good agreement to within 5 % as compared with Dmol3 findings. Furthermore, the SCC sets of parameters were also used to determine the lattice expansions to check the transformation temperature of ternary Ti50Pd50-XRuX. It was observed that the addition of Ru reduces the transformation temperature. The findings imply that the SCC-DFTB approach was able to predict the stability and transformation behaviour of Ti50Pd50-XRuX alloys.

Parameterization and transformation temperature of Ti50Pd50-XRuX using SCC-DFTB approach

With rapid development of information technology and artificial intelligence, optically controlled ferroelectric junctions are becoming promising candidates for data storage and logics [1]. We have recently reported that the resistance state of a BaTiO3 ferroelectric tunnel barrier can the switched from ON to OFF states by illumination with a suitable wavelength (blue laser, λ = 405 nm) if an imprint field (EIMP ≈ -2.8MV/cm) exists that unbalances the equilibrium polarization up/down domain configuration. The observation that smaller energy photons (red laser, λ = 648 nm) does not produce such effect strongly suggests that photocarriers play major role on polarization and resistance switching. Here we explore in greater detail the possible role of thermal effects and potential thermally induced suppression of polarization. A number of experiments, including the time-dependent response, its sensitivity to illumination power and recovery processes, have been performed. Data shows the prominence of electronic process (photocarriers and polarization screening) as the ultimate reason for light-induced resistive switching [2].

References
[1] Uchino, Kenji. Ferroelectric devices. CRC press, 2018.
[2] Long, Xiao, et al. "Non-volatile optical switch of resistance in photoferroelectric tunnel junctions." Nature communications 12.1 (2021): 1-9.


Disentangling electronic and thermal contributions to the observed light-induced resistance switching in BaTiO3 ferroelectric tunnel junction

The atomic-scale interfaces of 3D transition-metals oxides are the fascinating playground for several atomic-scale interactions of spins. Novel physical phenomena such as magnetoelectric coupling, unconventional exchange bias (EB) coupling, etc. which originate at such interfaces have huge potential for next-generation nanoscale spintronic devices. Among them, EB is the most studied due to its importance for device applications such as high-density memory and sensor. Here, we demonstrate EB coupling (EB shift, HEB~40 Oe at 2 K) at the interface of epitaxially grown ferromagnetic (FM) La0.67Sr0.33MnO3 (LSMO) – ferroelectric (FE) BaTiO3 (BTO) bilayer thin films on SrTiO3 (STO) single crystal substrates in absence of any conventional antiferromagnetic (AFM) material. Such EB is only observed when the thickness of LSMO is only a few unit cells (<10nm). Importantly, we find that the EB coupling follows training effect, bias field dependency, and temperature dependency. By X-ray magnetic circular dichroism (XMCD) measurement, it is observed that in such epitaxial heterostructure Ti of BTO poses magnetisation at 2K and can reversibly switch between two distinct magnetization states by switching external applied magnetic field. A remanent magnetization (MR) in Ti is also observed. The magnetization of Ti as well the EB vanishes at high temperature (300K) which confirms the EB coupling between Ti of BTO and Mn of LSMO at low temperature. This is an important step towards creating EB coupling between FM and FE materials without conventional AFM materials, which may enable a new class of multi-state memory devices by switching magnetization, FE polarisation and EB coupling in the same device.

Atomic-scale control of exchange bias at BaTiO3 - La0.67Sr0.33MnO3 interface for new multi-state memory

The last years have witnessed major advancements in all-electrical doping control on cuprates. In the vast majority of cases, the tuning of charge carrier density has been achieved via electric field effect by means of either a ferroelectric polarization or by using a dielectric or electrolyte gating. Unfortunately, these approaches are constrained to rather thin superconducting layers and require large electric fields in order to ensure sizable carrier modulations1,2. In this work, we focus on the investigation of oxygen doping in an extended region through current-stimulated oxygen migration in YBa2Cu3O7−δ superconducting bridges. The underlying methodology is rather simple and avoiding sophisticated nanofabrication process steps and complex electronics. A patterned multiterminal transport bridge configuration allows us to electrically assess the directional counterflow of oxygen atoms and vacancies. Importantly, the emerging propagating front of current-dependent doping δ is probed in situ by optical microscopy and scanning electron microscopy. The resulting imaging techniques, together with photo-induced conductivity and Raman scattering investigations reveal an inhomogeneous oxygen vacancy distribution with a controllable propagation speed permitting us to estimate the oxygen diffusivity. These findings provide direct evidence that the microscopic mechanism at play in electrical doping of cuprates involves diffusion of oxygen atoms with the applied current. The resulting fine control of the oxygen content would permit a systematic study of complex phase diagrams and the design of non-volatile memories.

Direct Visualization of Current-Stimulated Oxygen Migration in YBa 2 Cu 3 O 7−δ Thin Films

High temperature shape memory alloys (HTSMAs) has attracted a great deal of scientific interest for their ability to operate at a higher temperature above 800 K, for applications as actuators for aerospace and automotive components due to the reversible martensitic transformation. TiPt amongst other binary systems undergoes a reversible martensitic transformation from B2↔B19 at elevated temperatures. Previous studies showed that the TiPt alloy is mechanically unstable with the negative C′ (–32) and soft modes in the negative frequency of the phonon dispersion curves along the gamma region at 0 K. The supercell approach was used to substitute Ti with Hf on TiPt structure to evaluate their mechanical stability from elastic properties and the phonon dispersions curves. The elastic properties suggest that an increase in Hf concentration enhances the mechanical stability of ternary structures. The C' becomes positive and larger at 25 < x < 50, which suggests a reduced martensitic transformation at x ≥ 43.75. The Ti50–xHfxPt50 ternary becomes more ductile with the increase in Hf concentration, which suggests that Hf stabilizes the system at a higher concentration. The analyses of the vibrational properties of Ti50–xHfxPt50 structures with respect to phonon dispersion are also discussed.

First-Principle Study of TiPtHf High Temperature Shape Memory Alloys Applications

Despite being predominantly known for their extraordinary performance in photovoltaic application, hybrid organic-inorganic perovskites (OIPs) are also widely studied in terms of neuromorphic computing and artificial synapses. 
The possibility to mimic the behaviour of neurons with OIP-based memristive devices is attributed mainly to the creation and rupture of metallic and/or vacancy defect conducting filaments under external electric field and modulation of the Schottky barrier at the OIP/metal interface. It has already been shown, that OIPs can emulate simple synaptic learning rules, such as spike-timing dependent plasticity, spike-rate dependent plasticity, metaplasticity and memory consolidation.1,2 However, the unique optoelectronic properties of perovskites and their strong interaction with light create an opportunity to induce the neuromorphic effects with light as well.
Here, we present a memristive photoactive device based on a thin OIP layer sandwiched between Au and ITO electrodes that exhibits voltage- and light-induced synaptic effects. To induce the photo-synaptic effects, we have illuminated the sample with short (200 ms) light pulses. With no charge trapping layer, the device exhibits a gradual increase of photocurrent amplitude (short-term facilitation). However, upon the incorporation of charge trapping carbon nitride nanoparticles (CN-NPs) into the device, we have observed a much more complex photoresponse, with steep learning and slow photocurrent decrease (short-term depression). Such photocurrent generation dynamics suggest complex charge-trapping mechanisms at the Au/OIP and CN-NPs/OIP interfaces.
Our preliminary results show that the photocurrent generation dynamics can be controlled with the switching voltage as well, indicating that the incorporation of charge-trapping species might be useful in tailoring the photo-synaptic effects of light-sensitive OIP-based memristive devices.

Light-induced synaptic effects controlled by incorporation of charge-trapping layer into hybrid perovskite volatile memristor

Spintronics, based on spin dynamics and often intimately connected to effects driven by spin-orbit coupling (SOC), is relevant in technological applications over the last decades. Creating and detecting spin currents is of crucial importance within this field. One route is the direct and inverse spin Hall effect [1, 2]. While theoretically it is the spin Hall conductivity which is routinely calculated, experimentally it is often the spin accumulation at interfaces including all imperfections which is determining the observation [3]. Understanding this spin accumulation is critical and has been studied in theoretically and experimentally to varying degrees [4, 5]. In order to explore this complicated physics, evaluating the spin galvanic and spin Hall effect [6], we utilise the semi-classical Boltzmann equation based on a first-principle, relativistic Korringa-Kohn-Rostoker Green’s function method. Varying the host and impurity material in thin metallic films will allow us to establish trends across a wide range of materials. A particular focus will be on the influence of substitutional impurities on the spin accumulation and their positional dependence. As impurities will break space inversion symmetry this will give access to both symmetric and asymmetric contributions to the spin accumulation. Finally, as interfaces are of high relevance for practical applications we will investigate the key differences arising between surface or adatom impurities.
References
[1] J. E. Hirsch, Phys. Rev. Lett. 83, 1834 (1999).
[2] T. Kimura et al., Phys. Rev. Lett. 98, 156601 (2007).
[3] S. Takahashi and S. Maekawa, Sci. Technol. Adv. Mater. 9, 014105 (2008).
[4] G. Géranton et al., Phys. Rev. B 93, 224420 (2016).
[5] O. Gueckstock et al., Advanced Materials 123, 2006281 (2021).
[6] F. Hellman et al., Rev. Mod. Phys. 89, 025006 (2017).

Premium content

Next from Materials Challenges for Memory

Hf0.5Zr0.5O2 ferroelectric tunnel barriers for nonvolatile memory applications

Similar lecture

Magnetization Dynamics of Single and Trilayer Permalloy Nanodots