Die Arbeit befasst sich mit der Elektronenspinresonanz (ESR) stark korrelierter Elektronensysteme mit ferromagnetischen Wechselwirkungen. Es wurden dafür Messungen an den Kondogitter-Systemen CeRuPO und CeOsPO, der Dotierungsreihe CeFeAs1-xPxO, den niederdimensionalen frustrierten Quadratgittern AA’VO(PO4)2 sowie in dem schwach ferromagnetischen Metall Nb1-yFe2+y durchgeführt. Alle Verbindungen zeigen entweder eine ferromagnetische Ordnung oder befinden sich in der Nähe einer ferromagnetischen Instabilität, die die Eigenschaften des stark korrelierten Systems beeinflusst.

Le transfert de spin est un moyen de retourner l’aimantation d’une couche dans une jonction tunnel magnétique. Le courant nécessaire à cette tâche dépend des matériaux et dans le contexte actuel consommer moins est devenu un enjeu important. Une solution consiste à utiliser des matériaux ayant une forte polarisation en spin et un faible amortissement magnétique. Ces matériaux sont appelés demi-métaux ferromagnétiques. Du fait de l’existence d’un gap de spin chez les spins minoritaires au niveau de Fermi, ces composés possèdent une polarisation en spin de 100% et un faible amortissement magnétique. En théorie, certains Heusler, tels que Co1.5Fe1.5Ge et Co2MnSi, possèdent ces propriétés s’ils cristallisent dans la bonne phase cristallographique. En pratique, des mesures indirectes semblent confirmer ce comportement mais pourtant aucune preuve directe de cette demi-métallicité n’a été observée jusqu’à présent. C’est dans ce cadre que cette thèse s’inscrit. Après avoir déterminé les conditions de croissance de Co1.5Fe1.5Ge, à l’aide d’une série de mesure et notamment à l’aide de la diffraction anomale, nous avons déterminé l’ordre chimique complet de cet alliage qui est bien celui recherché. Les mesures des propriétés magnétiques donnent des résultats en accord avec la théorie. Mais l’utilisation de ce composé dans des jonctions tunnel magnétiques montre une faible magnétorésistance tunnel. La spectroscopie de photoémission résolue en spin nous a permis d’expliquer ces résultats. Dans le même esprit, nous nous sommes tournés vers le Co2MnSi, un composé qui semble plus prometteur où le gap de spin et de faibles valeurs d’amortissement magnétiques ont été mesurés / Spin transfer is one way of switching the magnetization of a layer in a magnetic tunnel junction. The current needed at this task depends on the materials and in the current context, consume less became an important issue. Materials with a high spin polarization and a low magnetic damping are one solution of this problem. They are called half metal ferromagnets. Because of the existence of a pseudo-gap in the minority spin channel at the Fermi energy, these compounds show a 100% spin polarization and an extremely low magnetic damping. In theory, some Heusler, such as Co1.5Fe1.5Ge and Co2MnSi, possess theses properties if they crystallize in the good crystallographic phase. In practice, there is strong indication of this behavior by mean of indirect techniques. However, no direct evidence of this pseudo-gap has been observed. It is in this context that this thesis is. After having determined growth conditions of Co1.5Fe1.5Ge, by mean of several techniques and especially by anomalous diffraction, we determined the complete chemical order which is the one we were looking for. Magnetic properties measurements show results in agreement with the theory. But the use of this compound in magnetic tunnel junctions shows low tunnel magnetoresistance. Spin resolved photoemission spectroscopy measurements explain very well these results. In the same spirit, we started to study Co2MnSi which seems more promising as this pseudo-gap and low magnetic damping have been observed

The increasing demand for nearline storage capacity in data centers calls for a continued enhancement in hard disk drive recording density far beyond one terabit per square inch. The thermal stability limit forces the drive manufacturers to develop new concepts in order to achieve this in the long term. Potential solutions are microwave-assisted magnetic recording (MAMR), heat-assisted magnetic recording (HAMR) and bit-patterned media (BPM). A simple example of BPM based on sputter-deposited Co/Pd multilayers and prepatterned substrates at hypothetical recording densities up to one terabit per square inch was studied by magnetic force microscopy (MFM). This system achieved promising results at lower densities, but an actual application for data storage, especially at one terabit per square inch and higher densities, requires elaborate optimizations. For some time now, FePt thin films have attracted much attention as prospective recording layers for high-density magnetic data storage due to their high magnetic anisotropy. The use of FePt films in HAMR is especially promising. This application has been tested successfully by Seagate and its key customers in recent years and is about to be introduced into the nearline hard disk drive market. It requires a tuning of the magnetic properties of FePt, especially of its Curie temperature. The addition of Cu proved to be effective in this regard and can also facilitate the formation of the crucial L10 structure and (001) texture during rapid thermal annealing of sputter-deposited thin films. Such films were prepared as bilayers of Cu and FePt on Si substrates, annealed for 30 s, and analyzed by X-ray diffraction (XRD) and SQUID vibrating sample magnetometry (SQUID-VSM). The influence of large Cu additions on important properties like lattice parameters, mosaicity, magnetic anisotropy and Curie temperature is discussed. The chemical long-range order was calculated from the XRD data, and a dedicated chapter of this thesis covers the most important factors to be considered in such calculations for textured thin films and other samples. The feasibility of creating patterned Fe-Cu-Pt films with perpendicular magnetic anisotropy, as needed for a combination of HAMR and BPM, by deposition through a PMMA mask, a lift-off process and subsequent annealing was investigated as well. The results indicate that the chosen approach might not lead to the required (001) texture when the nanostructures are small enough to compete with today's recording densities, so that either a continuous film might need to be etched after annealing or a seed layer might be required to induce the texture.:1. Motivation: Magnetic Data Storage 2. Experimental Techniques 3. Co/Pd Multilayers on Prepatterned Substrates 4. Fe-Pt and Fe-Cu-Pt Alloys 5. Rapid Thermal Annealing of FePt and FePt/Cu Films 6. Order Parameter Calculation 7. Summary

Manipulation of ferromagnetic (FM) spins (and spin textures) using an antiferromagnet (AFM) as an active element in exchange coupled AFM/FM heterostructures is a promising branch of spintronics. Recent ground-breaking experimental demonstrations, such as electrical manipulation of the interfacial exchange coupling and FM spins, as well as ultrafast control of the interfacial exchange-coupling torque in AFM/FM heterostructures, have paved the way towards ultrafast spintronic devices for data storage and neuromorphic computing device applications.[5,6] To achieve electrical manipulation of FM spins, AFMs offer an efficient alternative to passive heavy metal electrodes (e.g., Pt, Pd, W, and Ta) for converting charge current to pure spin current. However, AFM thin films are often integrated into complex heterostructured thin film architectures resulting in chemical, structural, and magnetic disorder. The structural and magnetic disorder in AFM/FM-based spintronic devices can lead to highly undesirable properties, namely thermal dependence of the AFM anisotropy energy barrier, fluctuations in the magnetoresistance, non-linear operation, interfacial spin memory loss, extrinsic contributions to the effective magnetic damping in the adjacent FM, decrease in the effective spin Hall angle, atypical magnetotransport phenomena and distorted interfacial spin structure. Therefore, controlling the magnetic order down to the nanoscale in exchange coupled AFM/FM-based heterostructures is of fundamental importance. However, the impact of fractional variation in the magnetic order at the nanoscale on the magnetization reversal, magnetization dynamics, interfacial spin transport, and the interfacial domain structure of AFM/FM-based heterostructures remains a critical barrier. To address the aforementioned challenges, we conduct a comprehensive experimental investigation of chemical, structural, magnetization reversal (integral and element-specific), magnetization dynamics, and magnetotransport properties, combined with high-resolution magnetic imaging of the exchange coupled Ni3Fe/IrMn3-based heterostructures. Initially, we study the chemical, structural, electrical, and magnetic properties of epitaxially textured MgO(001)/IrMn3(0-35 nm)/Ni3Fe(15 nm)/Al2O3(2.0 nm) heterostructures. We reveal the impact of magnetic field annealing on the interdiffusion at the IrMn3/Ni3Fe interface, electrical resistivity, and magnetic properties of the heterostructures. We further present an AFM IrMn3 film thickness dependence of the exchange bias field, coercive field, magnetization reversal, and magnetization dynamics of the exchange coupled heterostructures. These experiments reveal a strong correlation between the chemical, structural and magnetic properties of the IrMn3-based heterostructures. We find a significant decrease in the spin-mixing conductance of the chemically-disordered IrMn3/Ni3Fe interface compared to the chemically-ordered counterpart. Independent of the AFM film thickness, we unveil that thermally disordered AFM grains exist in all the samples (measured up to 35-nm-thick IrMn3 films). We develop an iterative magnetic field cooling procedure to systematically manipulate the orientation of the thermally disordered and reversible AFM moments and thus, achieve tunable magnetic, and magnetotransport properties of exchange coupled AFM-based heterostructures. Subsequently, we investigate the impact of fractional variation in the AFM order on the magnetization reversal and magnetotransport properties of the epitaxially textured ɣ-phase IrMn3/Ni3Fe, Ni3Fe/IrMn3/Ni3Fe, and Ni3Fe/IrMn3/Ni3Fe/CoO heterostructures. We probe the element-specific (FM: Ni and Co, and AFM: Mn) magnetization reversal properties of the exchange coupled Ni3Fe/IrMn3/Ni3Fe/Co/CoO heterostructures in various magnetic field cooled states. We present a detailed procedure for separating the spin and orbital moment contributions for magnetic elements using the XMCD sum rule. We address whether Mauri-type domain walls can develop at the (polycrystalline) exchange coupled Ni3Fe/IrMn3/Ni3Fe interfaces. We further study the impact of magnetic field cooling on the AFM Mn (near L2,3-edges) X-ray absorption spectra. Finally, we employ a combination of in-field high-resolution magnetic force microscopy, magnetooptical Kerr effect magnetometry with micro-focused beam, and micromagnetic simulations to study the magnetic vortex structures in exchange coupled FM/AFM and AFM/FM/AFM disk structures. We examine the magnetic vortex annihilation mechanism mediated by the emergence and subsequent annihilation of the vortex-antivortex (V-AV) pairs in simple FM and exchange coupled FM/AFM as well as AFM/FM/AFM disk structures. We image the distorted magnetic vortex structures in exchange coupled FM/AFM disks proposed by Gilbert and coworkers. We further emphasize crucial magnetic vortex properties, such as handedness, effective vortex core radius, core displacement at remanence, nucleation field, annihilation field, and exchange bias field. Our experimental inquiry offers profound insight into the interfacial exchange interaction, magnetization reversal, magnetization dynamics, and interfacial spin transport of the AFM/FM-based heterostructures. Moreover, our results pave the way towards nanoscale control of the magnetic properties in AFM-based heterostructures and point towards future opportunities in the field of AFM spintronic devices.:1. Introduction 2. Magnetic Interactions and Exchange Bias Effect 3. Materials 4. Experimental Methods 5. Structural, Electrical, and Magnetization Reversal Properties of Epitaxially Textured ɣ-IrMn3/ Ni3Fe Heterostructures 6. Magnetization Dynamics of MgO(001)/IrMn3/Ni3Fe Heterostructures in the Frequency Domain 7. Tunable Magnetic and Magnetotransport Properties of MgO(001)/Ni3Fe/IrMn3/Ni3Fe/ CoO/Pt Heterostructures 8. Element-Specific XMCD Study of the Exchange Couple Ni3Fe/IrMn3/Ni3Fe/Co/CoO Heterostructures 9. Distorted Vortex Structure and Magnetic Vortex Reversal Processes in Exchange Coupled Ni3Fe/IrMn3 Disk Structures 10. Conclusions and Outlook Addendum Acronyms Symbols Publication List Author Information Acknowledgments Statement of Authorship