Patryk Krzysteczko

The Memristive Magnetic Tunnel Junction as a Nanoscopic Synapse-Neuron System

Memristors cover a gap in the capabilities of basic electronic components by remembering the history of the applied electric potentials, and are considered to bring neuromorphic computers closer by imitating the performance of synapses.[1–3] We used memristive magnetic tunnel junctions[4,5] based on MgO to demonstrate that the synaptic functionality is complemented by neuron-like behavior in these nanoscopic devices. The synaptic functionality originates in a resistance change caused by a voltage-driven oxygen vacancy motion[6] within the MgO layer. The additional functionality provided by magnetic electrodes enabled a current-driven resistance modulation due to spin-transfer torque.[7] We showed that a phenomenon known as back-hopping[8–11] leads to repeated switching between two resistance levels accompanied by current spiking, which emulates neuronal behavior. As a result, this remarkably simple system, which is composed of two magnetic layers separated by an insulator, provides a sufficient basis for the fabrication of a complete neural network. According to a simple model, the human brain consists of a network of neurons, axons and synapses. Neurons exchange electrical impulses (i.e., spikes) along their axons, and synapses are the functionalized junctions between axons and neurons. The structure of the network is defined by the spatial arrangement of neurons and axons and by the strength of each synaptic connection. In this framework, one can equate the current configuration of the network with “knowledge” and its modulation with “learning”. In the conventional (Hebbian) concept of learning,[12] synaptic strength is modified by the coincident activity of presynaptic and postsynaptic neurons. In the past few years, a series of experiments has revealed a new picture. These studies suggest that instead of mere coincidence, the precise timing of presynaptic and postsynaptic spikes (i.e., spike-timing-dependent plasticity, STDP) plays a decisive role in determining the type of synaptic modification.[13,14] This modification can lead to a persistent increase or decrease in synaptic strength, commonly referred to as long-term potentiation (LTP) or long-term depression (LTD). Software inspired by neuronal networks is widely used for specific tasks, such as pattern recognition. In addition, hardware

Excitation of microwaveguide modes by a stripe antenna

We have studied experimentally the excitation of propagating spin-wave modes of a microscopic Permalloy-film waveguide by a stripe antenna. We show that due to the strong quantization of the spin-wave spectrum, the excitation of particular modes has essentially different frequency dependencies leading to a nonmonotonous variation of the modulation depth of the resulting spin-wave beam as a function of the excitation frequency. In addition, we address the effect of nonreciprocity of spin-wave excitation and found that for the case of Permalloy microwaveguides this effect is much weaker pronounced than for waveguides made from dielectric magnetic films with low saturation magnetization.

Nano-optics with spin waves at microwave frequencies

With the recent development in nanoscale patterning techniques, the potential of practical applications of nanometer-size structures for signal processing has been growing continuously. Experimental findings on the manipulation of optical signals in nanostructures have recently given rise to a widely addressed scientific area—subwavelength nano-optics. Here, we demonstrate that spin waves in microscopic ferromagnetic film structures represent a superb object for realization of the principles of nano-optics in the microwave frequency range. We show experimentally that by using the unique properties of spin waves, one can easily channelize, split, and manipulate submicrometer-width spin-wave beams propagating in microscopic magnetic-film waveguides.

Memristive switching of MgO based magnetic tunnel junctions

Here we demonstrate that both, tunnel magnetoresistance (TMR) and resistive switching (RS), can be observed simultaneously in nanoscale magnetic tunnel junctions. The devices show bipolar RS of 6% and TMR ratios of about 100%. For each magnetic state, multiple resistive states are created depending on the bias history, which provides a method for multibit data storage and logic. The electronic transport measurements are discussed in the framework of a memristive system. Differently prepared MgO barriers are compared to gain insight into the switching mechanism.

Self-focusing of spin waves in Permalloy microstripes

Excitation and propagation of spin waves in Permalloy microstripes magnetized in their plane perpendicularly to the axis have been investigated by means of microfocus Brillouin light scattering spectroscopy with high spatial resolution. We show that the spatial profile of the spin-wave beam demonstrates a focusing at a certain distance from the excitation source depending on the stripe width. A model connecting the observed phenomenon with an interference of different spin-wave modes existing in the stripe due to the finite-size effect is proposed.

Mapping of localized spin-wave excitations by near-field Brillouin light scattering

We report on the experimental study of the spatial characteristics of high-frequency spin-wave modes localized at the edges of micrometer-size in-plane magnetized permalloy ellipses. Using a near-field Brillouin light scattering technique, we have mapped the modes with the spatial resolution of few tens of nanometers. We show that the width of the localization area strongly depends on the applied magnetic field and reduces to about 85 nm for high fields. We also demonstrate that the existing theoretical models do not appropriately describe spatial characteristics of the modes.

Anisotropic Magnetoresistance State Space of Permalloy Nanowires with Domain Wall Pinning Geometry

The domain wall-related change in the anisotropic magnetoresistance in L-shaped permalloy nanowires is measured as a function of the magnitude and orientation of the applied magnetic field. The magnetoresistance curves, compiled into so-called domain wall magnetoresistance state space maps, are used to identify highly reproducible transitions between domain states. Magnetic force microscopy and micromagnetic modelling are correlated with the transport measurements of the devices in order to identify different magnetization states. Analysis allows to determine the optimal working parameters for specific devices, such as the minimal field required to switch magnetization or the most appropriate angle for maximal separation of the pinning/depinning fields. Moreover, the complete state space maps can be used to predict evolution of nanodevices in magnetic field without a need of additional electrical measurements and for repayable initialization of magnetic sensors into a well-specified state.

Current induced resistance change of magnetic tunnel junctions with ultra-thin MgO tunnel barriers

Ultra-thin magnetic tunnel junctions with low resistive MgO tunnel barriers are prepared to examine their stability under large current stress. The devices show magnetoresistance ratios of up to 110% and an area resistance product of down to 4.4 Omega mu m(2). If a large current is applied, a reversible resistance change is observed, which can be attributed to two different processes during stressing and one relaxation process afterwards. Here, we analyze the time dependence of the resistance and use a simple model to explain the observed behavior. The explanation is further supported by numerical fits to the data in order to quantify the timescales of the involved phenomena

Tunnel junction based memristors as artificial synapses.

We prepared magnesia, tantalum oxide, and barium titanate based tunnel junction structures and investigated their memristive properties. The low amplitudes of the resistance change in these types of junctions are the major obstacle for their use. Here, we increased the amplitude of the resistance change from 10% up to 100%. Utilizing the memristive properties, we looked into the use of the junction structures as artificial synapses. We observed analogs of long-term potentiation, long-term depression and spike-time dependent plasticity in these simple two terminal devices. Finally, we suggest a possible pathway of these devices toward their integration in neuromorphic systems for storing analog synaptic weights and supporting the implementation of biologically plausible learning mechanisms.

Tunneling magneto thermocurrent in CoFeB/MgO/CoFeB based magnetic tunnel junctions

We study the tunneling magneto thermopower and tunneling magneto thermocurrent of CoFeB/MgO/CoFeB magnetic tunnel junctions (MTJ). The devices show a clear change of the thermoelectric properties upon reversal of the magnetisation of the CoFeB layers from parallel to the antiparallel orientation. When switching from parallel to antiparallel the thermopower increases by up to 55% whereas the thermocurrent drops by 45%. These observations can be well explained by the Onsager relations taking into account the tunneling magneto resistance of the MTJ. These findings contrast previous studies on Al2O3 based MTJ systems, revealing tunneling magneto thermo power but no tunneling magneto thermocurrent.