<p>With the Moore’s Law era coming to an end, much attention has been given to novel nanoelectronic devices as a key driving force behind technological innovation. Utilizing the inherent device physics of nanoelectronic components, for sensory and computational tasks have proven to be useful in reducing the area and energy requirements of the underlying hardware fabrics. In this work we demonstrate how the intrinsic noise present in nano magnetic devices can pave the pathway for energy efficient neuromorphic hardware. Furthermore, we illustrate how the unique magnetic properties of such devices can be leveraged for accurate estimation of environmental magnetic fields. We focus on spintronic technologies in particular, due to the low current and energy requirements in contrast to traditional CMOS technologies.</p><p>Image segmentation is a crucial pre-processing stage used in many object identification tasks that involves simplifying the representation of an image so it can be conveniently analyzed in the later stages of a problem. This is achieved through partitioning a complicated image into specific groups based on color, intensity or texture of the pixels of that image. Locally Excitatory Globally Inhibitory Oscillator Network or LEGION is one such segmentation algorithm, where synchronization and desynchronization between coupled oscillators are used for segmenting an image. In this work we present an energy efficient and scalable hardware implementation of LEGION using stochastic Magnetic Tunnel Junctions that leverage the fast parallel</p><p> nature of the algorithm. We demonstrate that the proposed hardware is capable of segmenting binary and gray-scale images with multiple objects more efficiently than<br> existing hardware implementations. </p><p>It is understood that the underlying device physics of spin devices can be used for emulating the functionality of a spiking neuron. Stochastic spiking neural networks based on nanoelectronic spin devices can be a possible pathway of achieving brain-like compact and energy-efficient cognitive intelligence. Current computational models attempt to exploit the intrinsic device stochasticity of nanoelectronic synaptic or neural components to perform learning and inference. However, there has been limited analysis on the scaling effect of stochastic spin devices and its impact on the operation of such stochastic networks at the system level. Our work attempts to explore the design space and analyze the performance of nanomagnet based stochastic neuromorphic computing architectures, for magnets with different barrier heights. We illustrate how the underlying network architecture must be modified to account for the random telegraphic switching behavior displayed by magnets as they are scaled into the superparamagnetic regime.<br></p><p>Next we investigate how the magnetic properties of spin devices can be utilized for real world sensory applications. Magnetic Tunnel Junctions can efficiently translate variations in external magnetic fields into variations in electrical resistance. We couple this property of Magnetic Tunnel Junctions with Amperes law to design a non-invasive sensor to measure the current flowing through a wire. We demonstrate how undesirable effects of thermal noise and process variations can be suppressed through novel analog and digital signal conditioning techniques to obtain reliable and accurate current measurements. Our results substantiate that the proposed noninvasive current sensor surpass other state-of-the-art technologies in terms of noise and accuracy.<br></p><br>

Dans le cadre de l'augmentation de la densité de stockage des mémoires magnétorésistives à accès direct (MRAM), les matériaux à anisotropie magnétique perpendiculaire sont particulièrement intéressants car ils possèdent une très forte anisotropie. Cependant, cette augmentation d'anisotropie induit également un accroissement de la consommation d'écriture. Un nouveau concept d'écriture assistée thermiquement a été proposé par le laboratoire SPINTEC. Le principe est de concevoir une structure très stable à température ambiante, mais qui perd son anisotropie lorsqu'elle est chauffée, facilitant ainsi l'écriture. Le but de cette thèse est de valider expérimentalement ce concept. Les premiers chapitres sont consacrés à l'optimisation des matériaux à anisotropie perpendiculaire que sont les multicouches (Co/Pt), (Co/Pd) et (Co/Tb). Leur intégration dans une jonction tunnel magnétique est ensuite présentée. L'évolution de l'anisotropie en température, paramètre crucial au bon fonctionnement de l'assistance thermique, a également été étudiée. Enfin, il est démontré que l'écriture thermiquement assistée est particulièrement efficace : les structures développées présentent une consommation d'écriture réduite par rapport aux structures classiques et une forte stabilité à température ambiante. / In order to increase the storage density of magnetoresistive random access memories (MRAM), magnetic materials with perpendicular anisotropy are very appealing thanks to high anisotropy. However, the enhancement of anisotropy induces an increase of writing consumption as well. A new thermally assisted switching concept has been proposed by SPINTEC laboratory. The principle is to design a highly stable structure at stand-by temperature which loses its anisotropy when heated, making thus the switching easier. The aim of this thesis is to validate experimentally this concept. The first chapters describe the optimisation of out-of-plane magnetic materials such as (Co/Pt), (Co/Pd) and (Co/Tb) multilayers. Their integration in magnetic tunnel junctions is then presented. The evolution of anisotropy with temperature is a critical parameter for thermally assisted writing and has been therefore studied. Finally, the efficiency of this thermally assisted writing is demonstrated: the developed structures present a reduced consumption compared to standard structures and high stability at room temperature.

Les mémoires non-volatiles magnétiques à effet de couple de transfert de spin - STT-MRAM sont un nouveau type de mémoire pouvant remplacer les mémoires DRAM ou SRAM. Chaque point de mémoire STT-MRAM est une jonction tunnel magnétique sous forme d’un pilier de taille nanométrique, composée de deux couches magnétiques séparées par une barrière d'oxide. L'empilement multicouche doit être élaboré sous ultravide par épitaxie par faisceau moléculaire (M.B.E.) ou par pulvérisation cathodique (P.V.D.). Ces méthodes d’élaboration sont développées par la société Vinci Technologies (finançant ce travail de thèse par une bourse CIFRE). L’amplitude de la magnétorésistance tunnel, utilisée pour lire les informations stockées dans la mémoire, dépend de l'orientation relative des aimantations des deux couches magnétiques. Par ailleurs, l'écriture de l’information dans le dispositif est obtenue grâce à l'effet de couple de transfert de spin, qui permet la manipulation de l’aimantation en utilisant un courant polarisé. Enfin, la stabilité thermique du dispositif est donnée par la barrière en énergie séparant les deux orientations d'aimantation (vers le haut et vers le bas dans le cas d'un dispositif perpendiculaire). Pour que les STT-MRAM soient une technologie compétitive, la tension critique nécessaire au retournement de l’aimantation (tension d'écriture) ainsi que le temps de retournement doivent être réduits, tandis que la stabilité thermique doit rester suffisamment élevée pour assurer la conservation de l'information. Au cours de ma thèse, en collaboration avec Vinci Technologies, les équipements nécessaires à la croissance des couches minces composant les jonctions tunnels (M.B.E. et P.V.D.) ont été optimisées. Grâce à cela, nous avons pu obtenir des couches minces avec une anisotropie perpendiculaire (hors du plan) bien caractérisée. J'ai ensuite concentré mon étude sur les dispositifs STT-MRAM industriels (IBM et STT) présentant une aimantation perpendiculaire pour comprendre le mécanisme de retournement de l’aimantation induite par le courant. J'ai alors pu identifier les paramètres pertinents influençant la valeur de la tension de retournement et proposer des solutions pour l'abaisser tout en préservant la stabilité thermique. Grâce à une étude concernant la probabilité de retournement d'aimantation, comparée à une modélisation macrospin et micromagnétique, j'ai mis en évidence un mécanisme de retournement variable en fonction de la configuration magnétique initiale. En effet, le champ rayonné par une couche magnétique sur une autre et la forme de la jonction tunnel ont un impact important sur la manipulation de l'aimantation / Spin Transfer Torque - Magnetic Random Access Memories – STT-MRAM – are developed as a new type of memory which could replace DRAM or SRAM. In the case of STT- MRAM, each memory point is a nanopillar magnetic tunnel junction composed of two magnetic layers separated by an oxide barrier. The multilayer stack can be grown under ultra-high vacuum using Molecular Beam Epitaxy (MBE) or Physical Vapor Deposition (PVD). Those systems are developed by the company Vinci Technologies (sponsoring this PhD work). The tunnel magnetoresistance signal which depends on the relative orientation of the two magnetizations is used to read the information stored in the device. The writing of the information in the device is realized thanks to the spin transfer torque effect, which allows magnetization manipulation using a spin current. The thermal stability of the device is given by the energy barrier separating the two magnetization orientations (up and down in the case of a perpendicular device). For STT-MRAM to be a competitive technology, the critical voltage needed for magnetization switching (writing voltage) as well as the switching time have to be reduced while the thermal stability remains high enough to ensure the retention of information. During my thesis, in collaboration with Vinci-Technologies several tools to grow thin films have been optimized. With such equipment, we were able to grow thin films with well characterized perpendicular (out-of-plane) anisotropy. I have then focused my study on industrial STT-MRAM devices (from two companies: IBM and STT) with an out-of-plane magnetization direction so as to understand the mechanism of current induced magnetization switching. By doing so, I could identify the relevant parameters influencing the switching voltage value and propose solutions to lower it while preserving thermal stability. Through a probabilistic study of magnetization reversal, coupled with macrospin and micromagnetic modeling studies, I have evidenced different switching mechanisms depending on the initial magnetic configuration. Indeed both the stray field from one magnetic layer to the other and the shape of the nanopillar have a large impact on magnetization manipulation

This thesis discusses spin-transfer torques in MgO-based magnetic tunnel junctions. The voltage-field switching phase diagrams have been experimentally determined for in-plane CoFeB/MgO/CoFeB magnetic tunnel junctions. In order to limit the effect of thermal activation, experiments have been carried out using nanosecond voltage pulses, as well as at low-temperature (4.2 K). The bias-dependence of the two spin-torque terms (Slonczewski-like and field-like) has been determined from thermally-excited ferromagnetic resonance measurements, yielding values which are in good agreement with previous reports. Additionally, material parameters such as the effective magnetisation and the damping factor have also been extracted. Using these values as input, the switching voltages as function of the applied magnetic field have been calculated numerically and analytically by solving the modified Landau-Lifshitz-Gilbert equation. Unlike previous studies, the field-like spin-torque has also been included. Moreover, different configurations have been considered for the magnetic anisotropy directions of the reference and free layer, respectively. / Diese Arbeit befasst sich mit Spin-Transfer-Torque-Effekten in MgO-basierten magnetischen Tunnelstrukturen. Die Phasendiagramme als Funktion von Spannung und Magnetfeld von CoFeB/MgO/CoFeB-Tunnelstrukturen mit Magnetisierung in der Ebene wurden experimentell bestimmt. Um thermische Anregungseffekte zu limitieren, wurden die Experimente einerseits mit nanosekundenlangen Spannungspulsen und andererseits bei niedrigen Temperaturen (4.2 K) durchgeführt. Die Spannungsabhängigkeit der beiden Spin-Torque-Parameter (in-plane und senkrechter Spin-Transfer-Torque) wurde aus Messungen der thermisch angeregten ferromagnetischen Resonanz bestimmt, wobei sich Werte ergaben, die gut mit vorangegangenen Untersuchungen übereinstimmen. Zusätzlich wurden Werte für Materialparameter wie die effektive Magnetisierung und den Dämpfungsparameter gewonnen. Unter Verwendung der erhaltenen Werte wurden die Schaltspannungen als Funktion des angelegten Magnetfeldes analytisch und numerisch berechnet, indem die erweiterte Landau-Lifshitz-Gilbert-Gleichung gelöst wurde. Im Gegensatz zu vorangegangenen Untersuchungen wurde der senkrechte Spin-Transfer-Torque dabei mit einbezogen. Darüber hinaus wurden verschiedene Konfigurationen für die Richtung der magnetischen Anisotropie der freien und fixierten Schicht berücksichtigt.

The properties of bulk 3D materials of metals or semiconductors are manifested with various length scales(e.g., Bohr excitonic radius, magnetic correlation length, mean free path etc.) and are important in controlling their properties. When the size of the material is smaller than these characteristics length scales, the confinement effects operate reflecting changes in their physical behavior. Materials with such confinement effects can be designated as low dimensional materials. There are exceedingly large numbers of low dimensional materials and the last half a century has probably seen the maximum evolution of such materials in terms of synthesis, characterization, understanding and modification of their properties and applications. The field of” nanoscience and nanotechnology”, have become a mature field within the last three decades where, for certain application, synthesis of materials of sizes in the nanometer range can be designed and controlled. Interface plays a very important role in controlling properties of heterogeneous material of every dimensionality. For example, the interface forms in 2D thin films or interface of heterogeneous nanoparticles(0D). In recent times, a large number of remarkable phenomena have triggered understanding and controlling properties arises due to nature of certain interface. In the field of nanoparticles, it is well known that the photoluminescence property depends very strongly on the nature of interface in heterostructured nanoparticles. In the recent time a large variety of heterostructured nanoparticles starting from core-shell to quantum dot-quantum well kind has been synthesized to increase the photoluminescence efficiency up to 80%. Along with improvement of certain properties due to heterostructure formation inside the nanoparticles, the techniques to understand the nature of those interfaces have improved side by side. It has been recently shown that variable energy X-ray Photoemission Spectroscopy (XPS) can be employed to understand the nature of interfaces (internal structure) of such heterostructure nanoparticles in great detail with high accuracy. While most of the previous studies of variable energy XPS, uses photonenergies sensitive to smaller sized particle, we have extended the idea of such nondestructive approach of understanding the nature of buried interfaces to bigger sized nanoparticles by using photon energy as high as 8000eV, easily available in various 3rd generation synchrotron centers. The nature of the interface also plays an important role in multilayer thin films. Major components of various electronic devices, like read head memory devices, field effect transistors etc., rely on interface properties of certain multilayer thin film materials. In recent time wide range of unusual phenomenon such as high mobility metallic behavior between two insulating oxide, superconductivity, interface ferroelectricity, unusual magnetism, multiferroicity etc. has been observed at oxide interface making it an interesting field of study. We have shown that variable energy photoemission spectroscopy with high photon energies, can be a useful tool to realize such interfaces and controlling the properties of multilayered devices, as well as to understand the origin of unusual phenomenon exists at several multilayer interfaces. Chapter1 provides a brief description of low dimensional materials, overall perspective of interesting properties in materials with reduced dimensionality. We have emphasized on the importance of determining the internal structure of buried interface of different dimensionalities. We have given a brief overview and importance of different interfaces that we have studied in the subsequent chapters dealing with specific interfaces. Chapter 2 describes experimental and theoretical methods used for the study of interface and self-assembly reported in this thesis. These methods are divided into two categories. The first section deals with different experimental techniques, like, UV-Visible absorption and photoluminescence spectroscopy, X-Photoelectron Spectroscopy(XPS), X-Ray diffraction, Transmission Electron Microscopy(TEM) etc. This section also includes brief overview on synchrotron radiation and methods used for detail analysis of interface structure using variable energy XPS. In the second part of this chapter, we have discussed theoretical methods used in the present study. \ In Chapter 3A we have combined low energy XPS, useful to extract information of the surface of the nanoparticles, with high energy XPS, important to extract bulk information and have characterized the internal structure of nanoparticle system of different heterogeneity. We have chosen two important heterostructure systems namely, inverted core-shell(CdScore-CdSeshell) type nanoparticles and homogeneous alloy(CdSeS)type nanoparticles. Such internal structure study revealed that the actual internal structure of certain nanomaterial can be widely different from the aim of the synthesis and knowledge of internal structure is a prerequisite in understanding their property. We were able to extend the idea of variable energy XPS to higher energy limit. Many speculations have been made about the probable role of interface in controlling properties, like blinking behavior of bigger sized core-shell nanoparticles, but no conclusive support has yet been given about the nature of such interface. After successfully extending the technique to determine the internal structure of heterostructured nanoparticles to very high photon energy region, we took the opportunity to determine the internal structure of nanoparticles of sizes as large as 12nm with high energy photoemission spectroscopy for the first time. In Chapter 3B we emphasize on the importance of interface structure in controlling the behavior of bigger sized nanoparticles systems, the unsettled issues regarding their internal structure, and described the usefulness of high energy XPS in elucidating the internal structure of such big particles with grate accuracy to solve such controversies. The existence of high density storage media relies on the existence of highly sensitive magnetic sensors with large magnetoresistance. Today almost all sensor technologies used in modern hard disk drives rely on tunnel magnetoresistance (TMR) CoFeB-MgO-CoFeB structures. Though device fabrication is refined to meet satisfactory quality assurance demands, fundamental understanding of the refinement in terms of its effect on the nature of the interfaces and the MgO tunnel barrier leading to improved TMR is still missing. Where, the annealing condition required to improve the TMR ratio is itself not confirmatory its effect on the interface structure is highly debatable. In particular, it has been anticipated that under the proposed exotic conditions highly mobile B will move into the MgO barrier and will form boron oxide. In Chapter 4 we are able to shed definite insights to heart of this problem. We have used high energy photoemission to investigate a series of TMR structures and able to provide a systematic understanding of the driving mechanisms of B diffusion in CoFeBTMR structures. We have solved the mix-up of annealing temperature required and have shown that boron diffusion is limited merely to a sub-nanometer thick layer at the interface and does not progress beyond this point under typical conditions required for device fabrication. We have given a brief overview on the evolution of magnetic storage device and have described various concepts relevant for the study of such systems. The interface between two nonmagnetic insulators LaAlO3 and SrTiO3 has shown a variety of interface phenomena in the recent times. In spite of a large number of high profile studies on the interface LaAlO3 and SrTiO3 there is still a raging debate on the nature, origin and the distribution of the two dimensional electron gas that is supposed to be responsible for its exotic physical properties, ranging from unusual transport properties to its diverse ground states, such as metallic, magnetic and superconducting ones, depending on the specific synthesis. The polar discontinuity present across the SrTiO3-LaAlO3 interface is expected to result in half an electron transfer from the top of the LaAlO 3 layer to each TiofSrTiO3 at the interface, but, the extent of localization that can make it behave like delocalized with very high mobility as well as localized with magnetic moments is not yet clear. In Chapter 5 we have given a description of this highly interesting system as well as presented the outcome of our depth resolved XPS investigation on several such samples synthesized under different oxygen pressure. We were able to describe successfully the distribution of charge carriers. While synthesizing and understanding properties of nanoparticles is one issue, using them for device fabrication is another. For example, to make a certain device often requires specific arrangements of nanoparticles in a suitable substrate. Self-assembly formation can be a potential tool in these regards. Just like atom or ions, both nano and colloidal particles also assemble by themselves in ordered or disordered structure under certain conditions, e.g., the drying of a drop of suspension containing the colloid particles over a TEM grid. This phenomenon is known as self-assembly. Though, the process of assembly formation can be a very easy and cost-effective technique to manipulate the properties in the nano region, than the existing ones like lithography but, the lack of systematic study and poor understanding of these phenomena at microscopic level has led to a situation that, there is no precise information available in literature to say about the nature of such assembly. In Chapter 6 we have described experiments that eliminate the dependence of the self-assembly process on many complicating factors like substrate-particle interaction, substrate-solvent interaction etc., making the process of ordering governed by minimum numbers of experimental parameter that can be easily controlled. Under simplified conditions, our experiments unveil an interesting competition between ordering and jamming in drying colloid systems similar to glass transition phenomenon Resulting in the typical phase behavior of the particles. We establish a re-entrant behavior in the order-disorder phase diagram as a function of particle density such that there is an optimal range of particle density to realize the long-range ordering. The results are explained with the help of simulations and phenomenological theory. In summary, we were able to extend the idea of variable energy XPS to higher energy limit advantageous for investigating internal structure of nonmaterial of various dimensionalities and sizes. We were able to comprehend nature of buried interface indicating properties of heterostructures quantum dots and thin films. Our study revealed that depth resolved XPS combined with accessibility of high and variable energies at synchrotron centers can be a very general and effective tool for understanding buried interface. Finally, we have given insight to the mechanism of spontaneous ordering of nanoparticles over a suitable substrate.

Le contexte général de ce travail est le développement et l'intégration de nouveaux matériaux magnétiques ayant des propriétés originales et d'intérêt potentiel pour la spintronique. En tant que matériau ferromagnétique d’aimantation nulle, le composé Sm1-xGdxAl2 (SGA) suscite un intérêt particulier, puisqu’il est capable, dans son état magnétique compensé, de polariser en spin un courant d’électrons. Dans un premier temps, des expériences de photoémission résolues en angle et en spin sur synchrotron ont permis d’effectuer une analyse précise de la structure électronique selon diverses directions de la zone de Brillouin et d’estimer de manière directe la polarisation de spin au niveau de Fermi du composé SGA. Dans un second temps, une attention particulière a été portée aux multicouches [Co/Pt] et aux JTMs à base de [Co/Pt]. Les multicouches [Co/Pt] constituent la seconde électrode des JTMs à base de SGA. Leurs propriétés magnétiques (en particulier l'anisotropie perpendiculaire et l'aimantation à saturation) ont été soigneusement étudiées en fonction de l'épaisseur de Pt et de la nature de la couche tampon (Pt, MgO ou Al2O3), et en liaison avec leurs caractéristiques structurales. Leur intégration dans des JTMs à base de [Co/Pt] a permis ensuite de remonter d’une part à la polarisation tunnel effective des multicouches [Co/Pt] et d’autre part aux configurations magnétiques des différentes électrodes, configurations parfaitement expliquées et reproduites par des simulations micro-magnétiques. Dans un troisième temps, les résultats de magnéto-transport au sein des JTMs SGA/MgO/[Co/Pt] sont présentés et discutés / The general context of this work is the development and integration of new magnetic materials with original properties of potential interest for spintronic applications. In this field, the Sm1-xGdxAl2 (SGA) compound drives a particular attention, as a zero-magnetization ferromagnet that can exhibit a spin polarization in its magnetic compensated state. In a first step, synchrotron-based angle and spin resolved photoemission spectroscopy experiments have permitted to perform an accurate analysis of the electronic structure along various directions of the Brillouin Zone and to get a direct estimation of the spin polarization at the Fermi level. In a second step, a special attention has been the paid to [Co/Pt] multilayers and to [Co/Pt]-based MTJs. The [Co/Pt] multilayers would constitute the second electrode in SGA-based MTJs. Their magnetic properties (especially the perpendicular anisotropy and the saturation magnetization) have been carefully investigated as a function of Pt thickness and nature of the buffer layer (Pt, MgO or Al2O3), and in close connection with structural characteristics. Their integration in [Co/Pt]-based MTJs has permitted to determine the [Co/Pt] effective tunnel polarization and to unravel the magnetic configurations of both electrodes which are perfectly explained and reproduced by micromagnetic simulations. In a third step, the results concerning the magneto-transport experiments in SGA/MgO/[Co/Pt] MTJs are presented and discussed

Magnetický tunelový spoj (MTJ) je spintronická součástka komerčně používaná ve vysoce citlivých čtecích hlavách pevných disků. Počínaje rokem 2007 přispěla k udržení exponenciálního nárůstu hustoty magnetického zápisu. Kromě toho se také stala stavebním kamenem rychlé, odolné, úsporné a nevolatilní magnetické paměti s přímým přístupem (MRAM). Tento nový typ polovodičové paměti, stejně jako je tomu u čtecích hlav disků, využívá tunelové spoje založené na krystalickém oxidu hořečnatém (MgO) spolu s 3d kovovými magnetickými prvky (Fe a Co). Pro zmenšení MTJ a současné udržení dlouhodobé stability paměti proti tepelným fluktuacím je zapotřebí silná magnetická anizotropie ve směru kolmém na rozhraní kov|MgO. V této práci proto nejdříve provedeme analýzu magnetokrystalické anizotropie (MCA) kubického prostorově centrovaného Fe, Co a Ni na MgO pomocí ab initio simulací. Dále bude vyvinut program pro výpočet tvarové anizotropie, která je kromě MCA velmi podstatná, neboť v součtu dávají efektivní anizotropii. Na závěr implementujeme program pro výpočet MCA na základě poruchové teorie druhého řádu. To nám umožní dát pozorované anizotropní vlastnosti do souvislosti přímo s elektronickou strukturou systému (pásovou strukturou a hustotou stavů).

Die vorliegende Arbeit befasst sich mit magnetischen Tunnelkontakten (magnetic tunnel junctions, MTJs) auf Basis des Oxids Zinkferrit (ZnxFe3-xO4). Dabei soll das Potential dieses Materials durch die Demonstration des Tunnelmagnetowiderstandes (tunnel magnetoresistance, TMR) in zinkferritbasierten Tunnelkontakten gezeigt werden. Dazu wurde ein Probendesign für MTJs auf Basis der „pseudo spin valve“-Geometrie entwickelt. Die Basis für dieseStrukturen ist ein Dünnfilmstapel aus MgO (Substrat) / TiN / ZnxFe3-xO4 / MgO / Co. Dieser ist mittels gepulster Laserabscheidung (pulsed laser deposition, PLD) hergestellt. Im Rahmen dieser Arbeit wurden die strukturellen, elektrischen und magnetischen Eigenschaften der Dünnfilme untersucht. Des weiteren wurden die fertig prozessierten MTJ-Bauelemente an einem im Rahmen dieser Arbeit entwickeltem und aufgebautem TMR-Messplatz vermessen. Dabei ist es gelungen einen TMR-Effekt von 0.5% in ZnxFe3-xO4-basierten MTJs nachzuweisen. Das erste Kapitel der Arbeit gibt eine Einführung in die spintronischen Effekte Riesenmagnetowiderstand (giant magnetoresistance, GMR) und Tunnelmagnetowiderstand (TMR). Deren technologische Anwendungen sowie die grundlegenden physikalischen Effekte und Modelle werden diskutiert. Das zweite Kapitel gibt eine Übersicht über die Materialklasse der spinellartigen Ferrite. Der Fokus liegt auf den Materialien Magnetit (Fe3O4) sowie Zinkferrit (ZnxFe3-xO4). Die physikalischen Modelle zur Beschreibung der strukturellen, magnetischen und elektrischen Eigenschaften dieser Materialien werden dargelegt sowie ein Literaturüberblick über experimentelle und theoretische Arbeiten gegeben. Im dritten Kapitel werden die im Rahmen dieser Arbeit verwendeten Probenpräparations- und Charakterisierungsmethoden vorgestellt und technische Details sowie physikalische Grundlagen erläutert. Die Entwicklung eines neuen Probendesigns zum Nachweis des TMR-Effekts in ZnxFe3-xO4-basierten MTJs ist Gegenstand des vierten Kapitels. Die Entwicklung des Probenaufbaus sowie die daraus resultierende Probenprozessierung werden beschrieben. Die beiden letzten Kapitel befassen sich mit der strukturellen, elektrischen und magnetischen Charakterisierung der mittels PLD abgeschiedenen Dünnfilme sowie der Tunnelkontaktstrukturen.

This thesis discusses spin-transfer torques in MgO-based magnetic tunnel junctions. The voltage-field switching phase diagrams have been experimentally determined for in-plane CoFeB/MgO/CoFeB magnetic tunnel junctions. In order to limit the effect of thermal activation, experiments have been carried out using nanosecond voltage pulses, as well as at low-temperature (4.2 K). The bias-dependence of the two spin-torque terms (Slonczewski-like and field-like) has been determined from thermally-excited ferromagnetic resonance measurements, yielding values which are in good agreement with previous reports. Additionally, material parameters such as the effective magnetisation and the damping factor have also been extracted. Using these values as input, the switching voltages as function of the applied magnetic field have been calculated numerically and analytically by solving the modified Landau-Lifshitz-Gilbert equation. Unlike previous studies, the field-like spin-torque has also been included. Moreover, different configurations have been considered for the magnetic anisotropy directions of the reference and free layer, respectively.:1 Introduction 2 Fundamentals 2.1 Magnetoresistance 2.1.1 Giant magnetoresistance 2.1.2 Tunnel magnetoresistance 2.2 Spin-transfer torque effect 2.2.1 Physical picture of the STT 2.2.2 In-plane and perpendicular STT 2.3 Equation of motion for the magnetisation 2.3.1 The Landau-Lifshitz-Gilbert equation 2.3.2 Extension including spin-transfer-torque (LLGS) 2.4 Applications of MR and spin-transfer torque 2.4.1 Read heads in hard disk drives 2.4.2 Spin-transfer torque magnetic random access memory 2.5 STT effects in magnetic tunnel junctions 2.5.1 Current-induced switching 2.5.2 Magnetisation precession 2.5.3 Bias-dependence of STT 2.5.4 Back-hopping 3 Experimental 3.1 Samples 3.1.1 Stack composition 3.1.2 Properties of samples used in this work 3.2 Experimental setup 3.2.1 Overview of equipment for the different measurement techniques 3.2.2 Electromagnet and Kepco power supply 3.2.3 Contacting of the sample 3.2.4 Principle specifications of equipment 3.3 Experimental techniques 3.3.1 Measurement of DC R-H and R-I loops 3.3.2 Measurement of phase diagrams: off and on-pulse 3.3.3 Thermally-excited ferromagnetic resonance 4 Results and discussion 4.1 Switching phase diagrams of MTJs 4.1.1 Theory: Calculating the phase diagram 4.1.2 Experimental phase diagrams 4.2 Thermally excited ferromagnetic resonance 4.2.1 Smoothing and fitting of raw data 4.2.2 Determination of Ms 4.2.3 Signal evolution with bias voltage 4.2.4 Analysis of peak position: perpendicular STT 4.2.5 Analysis of peak linewidth 5 Summary and outlook A Appendix List of figures List of tables Bibliography / Diese Arbeit befasst sich mit Spin-Transfer-Torque-Effekten in MgO-basierten magnetischen Tunnelstrukturen. Die Phasendiagramme als Funktion von Spannung und Magnetfeld von CoFeB/MgO/CoFeB-Tunnelstrukturen mit Magnetisierung in der Ebene wurden experimentell bestimmt. Um thermische Anregungseffekte zu limitieren, wurden die Experimente einerseits mit nanosekundenlangen Spannungspulsen und andererseits bei niedrigen Temperaturen (4.2 K) durchgeführt. Die Spannungsabhängigkeit der beiden Spin-Torque-Parameter (in-plane und senkrechter Spin-Transfer-Torque) wurde aus Messungen der thermisch angeregten ferromagnetischen Resonanz bestimmt, wobei sich Werte ergaben, die gut mit vorangegangenen Untersuchungen übereinstimmen. Zusätzlich wurden Werte für Materialparameter wie die effektive Magnetisierung und den Dämpfungsparameter gewonnen. Unter Verwendung der erhaltenen Werte wurden die Schaltspannungen als Funktion des angelegten Magnetfeldes analytisch und numerisch berechnet, indem die erweiterte Landau-Lifshitz-Gilbert-Gleichung gelöst wurde. Im Gegensatz zu vorangegangenen Untersuchungen wurde der senkrechte Spin-Transfer-Torque dabei mit einbezogen. Darüber hinaus wurden verschiedene Konfigurationen für die Richtung der magnetischen Anisotropie der freien und fixierten Schicht berücksichtigt.:1 Introduction 2 Fundamentals 2.1 Magnetoresistance 2.1.1 Giant magnetoresistance 2.1.2 Tunnel magnetoresistance 2.2 Spin-transfer torque effect 2.2.1 Physical picture of the STT 2.2.2 In-plane and perpendicular STT 2.3 Equation of motion for the magnetisation 2.3.1 The Landau-Lifshitz-Gilbert equation 2.3.2 Extension including spin-transfer-torque (LLGS) 2.4 Applications of MR and spin-transfer torque 2.4.1 Read heads in hard disk drives 2.4.2 Spin-transfer torque magnetic random access memory 2.5 STT effects in magnetic tunnel junctions 2.5.1 Current-induced switching 2.5.2 Magnetisation precession 2.5.3 Bias-dependence of STT 2.5.4 Back-hopping 3 Experimental 3.1 Samples 3.1.1 Stack composition 3.1.2 Properties of samples used in this work 3.2 Experimental setup 3.2.1 Overview of equipment for the different measurement techniques 3.2.2 Electromagnet and Kepco power supply 3.2.3 Contacting of the sample 3.2.4 Principle specifications of equipment 3.3 Experimental techniques 3.3.1 Measurement of DC R-H and R-I loops 3.3.2 Measurement of phase diagrams: off and on-pulse 3.3.3 Thermally-excited ferromagnetic resonance 4 Results and discussion 4.1 Switching phase diagrams of MTJs 4.1.1 Theory: Calculating the phase diagram 4.1.2 Experimental phase diagrams 4.2 Thermally excited ferromagnetic resonance 4.2.1 Smoothing and fitting of raw data 4.2.2 Determination of Ms 4.2.3 Signal evolution with bias voltage 4.2.4 Analysis of peak position: perpendicular STT 4.2.5 Analysis of peak linewidth 5 Summary and outlook A Appendix List of figures List of tables Bibliography

<p>Deep ‘Analog Artificial Neural Networks’ (AANNs) perform complex classification problems with high accuracy. However, they rely on humongous amount of power to perform the calculations, veiling the accuracy benefits. The biological brain on the other hand is significantly more powerful than such networks and consumes orders of magnitude less power, indicating some conceptual mismatch. Given that the biological neurons are locally connected, communicate using energy efficient trains of spikes, and the behavior is non-deterministic, incorporating these effects in Artificial Neural Networks (ANNs) may drive us few steps towards a more realistic neural networks. </p> <p> </p> <p>Emerging devices can offer a plethora of benefits including power efficiency, faster operation, low area in a vast array of applications. For example, memristors and Magnetic Tunnel Junctions (MTJs) are suitable for high density, non-volatile Random Access Memories when compared with CMOS implementations. In this work, we analyze the possibility of harnessing the characteristics of such emerging devices, to achieve neuro-inspired solutions to intricate problems.</p> <p> </p> <p>We propose how the inherent stochasticity of nano-scale resistive devices can be utilized to realize the functionality of spiking neurons and synapses that can be incorporated in deep stochastic Spiking Neural Networks (SNN) for image classification problems. While ANNs mainly dwell in the aforementioned classification problem solving domain, they can be adapted for a variety of other applications. One such neuro-inspired solution is the Cellular Neural Network (CNN) based Boolean satisfiability solver. Boolean satisfiability (k-SAT) is an NP-complete (k≥3) problem that constitute one of the hardest classes of constraint satisfaction problems. We provide a proof of concept hardware based analog k-SAT solver that is built using MTJs. The inherent physics of MTJs, enhanced by device level modifications, is harnessed here to emulate the intricate dynamics of an analog, CNN based, satisfiability (SAT) solver. </p> <p> </p> <p>Furthermore, in the effort of reaching human level performance in terms of accuracy, increasing the complexity and size of ANNs is crucial. Efficient algorithms for evaluating neural network performance is of significant importance to improve the scalability of networks, in addition to designing hardware accelerators. We propose a scalable approach for evaluating Liquid State Machines: a bio-inspired computing model where the inputs are sparsely connected to a randomly interlinked reservoir (or liquid). It has been shown that biological neurons are more likely to be connected to other neurons in the close proximity, and tend to be disconnected as the neurons are spatially far apart. Inspired by this, we propose a group of locally connected neuron reservoirs, or an ensemble of liquids approach, for LSMs. We analyze how the segmentation of a single large liquid to create an ensemble of multiple smaller liquids affects the latency and accuracy of an LSM. In our analysis, we quantify the ability of the proposed ensemble approach to provide an improved representation of the input using the Separation Property (SP) and Approximation Property (AP). Our results illustrate that the ensemble approach enhances class discrimination (quantified as the ratio between the SP and AP), leading to improved accuracy in speech and image recognition tasks, when compared to a single large liquid. Furthermore, we obtain performance benefits in terms of improved inference time and reduced memory requirements, due to lower number of connections and the freedom to parallelize the liquid evaluation process.</p>