This dissertation describes an experimental study on the patterning of thin films and spin valve devices. Initially the change in the magnetisation reversal of ferromagnetic Ni80Fe15Mo5 thin films was investigated as the shape anisotropy was increased using optical lithography to pattern wire arrays. These structures show a progressive increase in coercivity and a transition between single and two-stage reversal with increasing milling depth. A similar patterning technique was applied to unpinned (Ni80Fe20/Cu/Ni80Fe20) pseudo spin valve (PSV) structures in order to enhance the coercivity of one of the ferromagnetic layers. The increased coercivity induced by micropatterning changed the natural similarity of the magnetic layers and the structure exhibited a small spin valve response. These initial measurements were carried out with separate milling and electrical characterisation steps. However, it was decided that it would be ideal to design a technique to do in-situ magnetoresistance measurements during milling. This meant that the samples could be milled and characterised in the same step, leading to a much cleaner and more efficient process. In-situ magnetoresistance measurements were carried out during micropatterning of PSV devices, and the measurements showed the evolution in the electrical response as wire structures were gradually milled through the thickness. Contrary to what was expected, the structures showed a maximum spin valve response when fully milled through. The effect of further increasing the shape anisotropy by reducing the wire width, and changing the material properties in the PSV structure has also been investigated. MR measurements were taken as the temperature was increased from 291K to 493K, and the results show that the patterned PSV structures have a better thermal stability than exchange biased spin valves with an IrMn pinning layer. The experiment was extended to the nanoscale, and the results show that a significant increase in MR is not observed despite the fact that the magnetic configuration tends more towards single domain. This is thought to be due to an increase in the initial resistance of the structures. A small increase in MR was observed as the wire width was decreased from 730 to 470 nm, although the spin valve response is heavily dependent on the gallium dosage density during patterning in the Focused Ion Beam (FIB). Micromagnetic simulations were carried out, which agree with the experimental results and showed the change in the magnetisation reversal from rotation to switching as the dimensions were reduced on the nanoscale.

Previous research on new novel substrates for giant magnoresistance structures has indicated that a net increase in the effect is present. The substrates studied were V-grooved or stepped, however research presented in this thesis used an embossed surface manufactured from alumina oxide which consisted of regular hexagonal arrays with spacing of 110 nm and pore diameter of 60 nm. The physical properties measurements unveiled a net enhancement of the giant magnetoresistance effect thru the whole range of the copper spacer thicknesses deposited with direct current magnetron sputter. The maximum net increase appeared for a spacer thickness of 4.0 nm where the flat silicon substrate yielded a 3 % increase but the embossed surface substrate generated a 12% increase with an overall effect of a 4-fold net enhancement of the effect. Both the aluminum oxide substrates and the thin films structures can be manufactured inexpensively and can be also mass-produced, which are welcoming advantages for the technology sector of magnetic sensing.

The Cu/NiFe/Cu/Co/(CoOx) spin-valves have been prepared by the ion-beam sputtering method. Their GMR ratio and the time stability have been investigated by the magnetoresistance and the MOKE measurements at room temperature. The reproducibility of the preparation of the samples have been studied as well, i.e. two identically configurations of the layers should have the same magnetotransport properties.

Coherent injection, detection and manipulation of spins in semiconductor nansotructures can herald a new genre of information processing devices that are extremely energy-efficient and non-volatile. For them to work reliably, spin coherence must be maintained across the device by suppressing spin relaxation. Suppression can be accomplished by structural engineering, such as by confining spin carriers to the lowest subband in a semiconductor quantum wire. Accordingly, we have fabricated 50-nm diameter InSb nanowire spin valves capped with Co and Ni nanocontacts in which a single conduction subband is occupied by electrons at room temperature. This extreme quantum confinement has led to a 10-fold increase in the spin relaxation time due to dramatic suppression of the D’yakonov -Perel’ (DP) spin relaxation mechanism. We have observed the spin-valve and Hanle effects at room temperature in these systems. Observing both effects allowed us to estimate the carrier mobility and the spin relaxation length/time and we found that the latter is ~10 times larger than the value reported in bulk InSb despite a four orders of magnitude decrease in the carrier mobility due to surface roughness scattering. We ascribe this dramatic increase in spin relaxation time to the suppression of the DP relaxation mode due to single subband occupancy. Modulation of spin relaxation rate by an external agent can open new possibilities for spintronic devices. Any agent that can excite electrons from the lowest subband to higher subbands will dramatically increase the DP spin relaxation rate. We have shown that the spin relaxation rate in InSb nanowires can be modulated with infrared light. In the dark, almost all the electrons in the nanowires are in the lowest conduction subband, resulting in near-complete absence of DP relaxation and long spin coherence length. This results in a high resistance state in a spin valve whose ferromagnetic contacts have anti-parallel spin polarizations. Under infrared illumination, higher subbands get populated and the DP spin relaxation mechanism is revived, leading to a three-fold decrease in the spin relaxation length. As a result, injected spins flip in the spacer layer of the spin valve and this causes the spin valve resistance to drop. Therefore, this effect can be exploited to implement an infrared detector. We also studied the transport behavior of a single nanowire (~50 nm diameter) captured between two non-magnetic contact pads. The wire was attached between the pads using dielectrophoresis. A giant (∼10,000,000%) negative magnetoresistance at 39 mT field was observed at room temperature in Cu nanowires contacted with Au contact pads. In these nanowires, potential barriers form at the two Cu/Au interfaces because of Cu oxidation that results in an ultrathin copper oxide layer forming between Cu and Au. Current flows when electrons tunnel through, and/or thermionically emit over these barriers. A magnetic field applied transverse to the direction of current flow along the wire deflects electrons toward one edge of the wire because of the Lorentz force, causing electron accumulation at that edge and depletion at the other. This makes the potential barrier at the accumulated edge shorter and at the depleted edge taller. The modulation of the potential barrier height with a magnetic field dramatically alters the tunneling and/or thermionic emission rate causing a giant magnetoresistance. Currently, effort is underway to demonstrate strain sensitive anisotropic magnetoresistance (AMR) in a single Co-Cu-Co nanowire spin valve. AMR is caused by spin-orbit coupling effects which makes the resistance of a ferromagnet depend on the angle between the direction of current flow and the magnetization. The resistance maximizes when the angle is 00 or 1800 and minimizes when the angle is 900. When an external magnetic field is applied in a direction opposite to a ferromagnet’s magnetization, the latter begins to rotate in the direction of the field and hence its resistance continuously changes. This results in a trough in the magnetoresistance of a spin valve structure between the two fields when the magnetization starts to rotate and when the magnetization completes the rotation. We have observed a magnetoresistance peak (instead of trough) in the Co-Cu-Co spin valve, which is due to the normal spin valve effect that overshadows AMR. However, when an intense infrared light source is brought close to the sample, the peak gets overshadowed by a trough, showing that the AMR effect becomes dominant. We attribute this intriguing feature to the fact that the AMR effect is strongly influenced by strain. Heating by the light source generates strain in the Co contacts owing to unequal thermal expansion of Co and the underlying substrate. We also observed that the AMR effect becomes more pronounced as the light source is brought closer to the sample, resulting in increased heating and hence increased strain generation.

This thesis explores the use of 2D materials (graphene and hBN) for spintronics. Interest on these materials in spintronics arose from theoretical predictions of high spin filtering in out-of-plane transport through graphene and hBN sandwiched by ferromagnets. Similarly, 5-layer graphene was forecast to be a perfect spin filter. In the case of in-plane spin transport, graphene was expected to be an excellent material due to its low spin-orbit coupling and low number of defects. Although there already exist experimental works that attempted to explore the aforementioned predictions, they have failed so far to comply with the expected results. Earlier experimental works in graphene and hBN out-of-plane spin transport achieved low spin filtering on the order of a few percent; while spin relaxation parameters in graphene for in-plane spin transport remained one or two orders of magnitude below the predicted values. In the case of vertical devices, the failure to meet the theoretical expectations was attributed to the oxidation of the ferromagnets and the lack of an epitaxial interface between the later and the graphene or hBN. Similarly, the exact mechanisms that lead to high spin relaxation for in-plane spin transport in graphene are not completely understood, in part due to the low-quality of the explored devices. In this thesis we analyze new architectures and procedures that allowed us to fabricate ultraclean and oxidation-free interfaces between ferromagnets and graphene or hBN. In these devices we encountered negative and reversible magnetoresistance, that could not be explained with the previous theoretical models. We propose a new model based on a thorough characterization of the devices and well-known properties of graphene that were not taken into account in the previous model. We also employed a novel type of contact to graphene (1D-contacts) and applied it for the first time to achieve spin-injection in graphene. The main advantage of this type of contact is the full encapsulation of graphene with hBN, which leads to high quality graphene spintronic devices.

Depuis la découverte de la magnétorésistance (MR) géante en 1988 par le groupe d'Albert Fert (prix Nobel de physique en 2007), le domaine de l'électronique de spin a connu un essor sans précédent, justifié par toutes les applications qu'elle permet d'envisager en électronique.Depuis une vingtaine d'années, il est question d'utiliser le degré de liberté de spin directement dans les matériaux semi-conducteurs avec le gros avantage par rapport aux métaux de pouvoir manipuler électriquement le spin des porteurs. L'électronique de spin dans les matériaux semi-conducteurs utilise pour coder l'information non seulement la charge des porteurs (électrons et trous), mais aussi leur spin. En associant charge et spin, on ajoute de nouvelles fonctionnalités aux dispositifs de micro-électronique traditionnels.Le premier challenge consiste à contrôler l’injection et la détection d’une population de porteurs polarisés en spin dans les semi-conducteurs traditionnels (Si, Ge).Pour cela, nous avons étudié des dispositifs hybrides de type MIS: Métal ferromagnétique/Isolant/Semi-conducteur qui nous permettent d'injecter et de détecter électriquement un courant de spin. La première partie de cette thèse concerne les dispositifs à 3 terminaux sur différents substrats qui utilisent une unique électrode ferromagnétique pour injecter et détecter par effet Hanle l’accumulation de spin dans les semi-conducteurs. Une amplification des signaux de spin extraits expérimentalement par rapport aux valeurs théoriques du modèle diffusif est à l’origine d’une controverse importante. Nous avons alors démontré que l’origine du signal de MR ou de l’amplification ne peut être expliquée par la présence de défauts dans la barrière tunnel. A l’inverse, nous prouvons la présence d’états d’interface qui peuvent expliquer l’amplification du signal de spin. De plus, la réduction de la densité d’états d’interface par une préparation de surface montre des changements significatifs comme la diminution du signal de spin.La deuxième partie de ces travaux concerne la transition vers les vannes de spin latérales sur semi-conducteurs. Dans ces dispositifs utilisant deux électrodes FM, le découplage entre l’injection et la détection de spin permet de s’affranchir des effets de magnétorésistance parasites car seul un pur courant de spin est détecté dans le semi-conducteur. Par une croissance d’une jonction tunnel ferromagnétique épitaxiée, nous avons démontré l’injection de spin dans des substrats de silicium et germanium sur isolant. En particulier nous observons un fort signal de spin jusqu’à température ambiante dans le germanium.Finalement, les prémices de la manipulation de spin par l’étude du couplage spin-orbite ont été étudiées dans les substrats d’arséniure de gallium et de germanium. En effet, nous avons induit par effet Hall de spin (une conséquence du couplage spin-orbite) une accumulation de spin qui a été sondée en utilisant la spectroscopie de muon. On démontre alors, à basse température, la présence de l’accumulation grâce au couplage entre les spins électroniques accumulés et les noyaux de l’arséniure de gallium. / Since the discovery of the giant magnetoresistance in 1988 by the group of Albert Fert (Nobel Prize in 2007), the field of spintronics has been growing very fast due to its potential applications in micro-electronics.For almost 20 years, it has been proposed to introduce the spin degree of freedom directly in the semiconducting materials. Spintronics aims at using not only the charge of carriers (electrons and holes) but also their intrinsic spin degree of freedom. In that case, spins might be manipulated with electric fields. By using both charge and spin, one might add new functionalities to traditional micro-electronic devices.Indeed, the first challenge of semiconductor spintronics is to create and detect a spin polarized carrier population in traditional semiconductors like Si and Ge to further manipulate them.For this purpose, we have used hybrid ferromagnetic metal/insulator/semiconductor devices which allow us to perform electrical spin injection and detection. The first part of this thesis deals with 3 terminal devices grown on different substrates and in which a single ferromagnetic electrode is used to inject and detect spin polarized electrons using the Hanle effect. A spin signal amplification is measured experimentally as compared to the value from the theoretical diffusive model, this raised a controversy concerning 3 terminal measurements. We demonstrate that localized defects in the tunnel barrier cannot be at the origin of the measured MR signal and spin signal amplification. Instead, we show that the presence of interface states is the origin of the spin signal amplification in all the substrates. By using a proper surface preparation and the MBE growth of the magnetic tunnel junctions, we reduce the density of interface states and show a significant modification of the spin signals.In a second part, we present the transition from 3 terminal measurements to lateral spin valves on semiconductors. In the last configuration by using two ferromagnetic electrodes, charge and spin currents are decoupled in order to avoid any spurious magnetoresistance artefacts. Using epitaxially grown magnetic tunnel junctions we can prove the spin injection in silicon and germanium. Especially, we are able to measure non local spin signals in germanium up to room temperature.Finally, we study the spin Hall effect in gallium arsenide and germanium substrates. For this propose we induce spin accumulation using the spin Hall effect (i.e spin-orbit coupling) and probe it using muon spectroscopy. We demonstrate, at low temperature the presence of spin accumulation by the coupling between nuclear spins and the electron spin accumulation.

Correlated electron materials have been at the forefront of condensed matter research in the past couple of decades. Correlation in materials, especially, with open d and f electronic shells often lead to very exciting and intriguing phenomenon like high temperature superconductivity, Mott metal-insulator transition, colossal magnetoresistance (CMR). This thesis focuses on triple-layered Sr4Ru3O10, Sr substituted double layered (Ca1-- xAx)3Ru2O7 (A = Ba, Sr) and 5d system Sr2IrO4 and Sr3Ir2O7. Triple-layered Sr4Ru3O10 displays interesting phenomena ranging from quantum oscillations, tunneling magnetoresistance, unusual low temperature specific heat, strong spin-lattice coupling to switching behavior. The central feature, however, is the unique borderline magnetism: along the c-axis. Sr4Ru3O10 shows spontaneous ferromagnetism, indicating a strong Coulomb exchange interaction, U and a large density of states at the Fermi surface, g(EF ), hence Ug(EF ) ≥ 1 (Stoner criterion). But within the ab-plane it features a pronounced peak in magnetization and a first-order metamagnetic transition. The coexistence of the interlayer ferromagnetism and the intralayer metamagnetism makes Sr4Ru3O10 a really unique system. Also, in this thesis the spin-valve behavior exhibited by impurity doping at the Ca site by Ba and Sr in the double layered Ca3Ru2O7 is reported. Spin valve effect is a phenomenon only realized in multilayer thin films. Here, spin valve is observed in bulk single crystals of impurity dopedCa3Ru2O7, Ca3(Ru1-xCrx)2O7 and (Ca1- xAx)3Ru2O7 (A = Ba, Sr). 5d Iridates are expected to be more metallic and less magnetic than their 3d and 4f counterparts because of the extended 5d orbitals. In marked contrast, many iridates are magnetic insulators with exotic properties. The focus in this thesis is on Sr2IrO4 which diplays a novel Jeff = 1/2 Mott state. Magnetic, electrical, and thermal measurements on single-crystals of Sr2IrO4, reveal a novel giant magneto-electric effect (GME) arising from a frustrated magnetic/ferroelectric state. The GME and electric polarization hinge on a spin-orbit gapping of 5d-bands, rather than the magnitude and spatial dependence of magnetization, as traditionally accepted.

Orbital degrees of freedom play vital role in prompting novel phenomena in ruthenium based Ruddlesden-Popper compounds through coupling of orbits to spin and lattice. Physical properties are then particularly susceptible to small perturbations by external magnetic fields and/or slight structural changes. Current study pertains to the impact when a more-extended 4d Ruthenium ion is replaced by a less-extended 3d Chromium ion. Perovskite CaRuO3 (n=∞) is characterized by borderline magnetism and non- Fermi liquid behavior – common occurrences in quantum critical compounds. Remarkably, Cr substitution as low as x=0.05 abruptly drives CaRu1−xCrxO3 from a paramagnetic state to an itinerant ferromagnetic state (MS~0.4μB/f.u.), where TC=123K for x=0.22. The Cr-driven magnetism is highly anisotropic suggesting an important role of spin-orbit coupling. Unlike other chemical substitutions in the compound, Cr does not induce any Metal-Insulator transition that is expected to accompany the magnetic transition. The results indicate a coupling of Ru-4d and Cr-3d electrons that is unexpectedly favorable for itinerant ferromagnetism, which often exists delicately in the ruthenates. Bilayered Ca3Ru2O7 (n=2), an abode of huge anisotropy, exhibits a wide range of physical properties – Colossal Magnetoresistance occurring only when the spin polarized state is avoided, Antiferromagnetic-Metallic (AFM-M) state, Quantum Oscillations (periodic in 1/B and in B) that are highly angular dependent, to mention a few. Experimental results obtained so far provide a coherent picture illustrating that orbital order and its coupling to lattice and spin degrees of freedom drive the exotic electronic and magnetic properties in this Mott-like system. Transport and thermodynamic studies on Ca3(Ru1-xCrx)2O7 (0 ≤ x ≤ 0.20) reveal that AFM-M region is broadened with x that ultimately reaches 70K for x=0.20 (~8K for x=0). In this region, electron transport is enhanced and inhibited when B is applied along crystal’s respective axes, confirming an intrinsic half-metallic behavior. Moreover, the difference in coercivities of Ru and Cr magnetic ions pave way for the first-ever observation of a strong spin-valve effect in bulk material, a quantum phenomenon so far realized only in multilayer thin films or heterostructures. This discovery opens new avenues to understand the underlying physics of spin-valves and fully realize its potential in practical devices.

Spin-dependent electron transport in submicron/nano sized magnetic thin film devices fabricated using the optical lithography, e-beam lithography and focused ion beam (FIB) was investigated with the primary aim to find the ballistic magnetoresistance (BMR) in thin film nanoconstrictions. All experimental results were analysed in combination with micromagnetic simulations. The magnetisation reversal processes were investigated in a submicron half-pinned NiFe stripe with a microconstriction. An asymmetric MR curve was observed, and micromagnetic simulations verified it was due to the exchange-bias on the left side, which changed the magnetic switching mechanism. The effects of different pinning sites on the magnetisation switching and domain wall displacement were studied in NiFe film and spin-valve based nanodevices. A sign of domain wall MR was seen on the transversal MR curve of the NiFe nanodevice due to the domain wall induced electron scattering. The size effect on the magnetisation switching and interlayer magnetostatic coupling was demonstrated and characterised in synthetic antiferromagnet (SAF)-pinned spin-valve nanorings. It has been clarified by micromagnetic simulations that these nanorings exhibit a double or single magnetisation switching process, which is determined by the magnetostatic coupling as a function of the ring diameter. The interlayer magnetostatic coupling was efficiently reduced in large SAF-pinned nanorings, resulting in a small shift of the minor MR curve, which is beneficial to the magnetic memory applications. In-situ MR measurements and the investigation of domain wall properties have been carried out in FIB patterned NiFe film nanoconstrictions. Spin-valve like sharp transitions were observed on the MR curves in the 80 nm/130 nm wide nanoconstriction devices. However, our analysis of the results by micromagnetic simulations and domain observations with scanning electron microscopy with polarisation analysis (SEMPA) concluded that these sharp MR transitions originated from the anisotropic magnetoresistance (AMR) effect, due to the fast magnetisation rotation in the nanoconstriction, and not from BMR. The numerical investigation has proved that a further reduction of the constriction width/length is necessary for large MR values.

Válvulas de Spin têm sido utilizadas na fabricação de sensores magnéticos e memórias de acesso randômico, sendo muito importantes do ponto de vista tecnológico. Neste trabalho, foram exploradas as análises das curvas de reversão de primeira ordem da magnetorresistência (MR-FORC), bem como ajustes das curvas de histereses da magnetização e magnetorresistência, para estudar o fenômeno de exchange-bias, anisotropia magnética e propriedades magnetorresistivas. As válvulas de spin estudadas foram do tipo NiFe/Cu/NiFe/IrMn, tendo camadas semente e de cobertura de Ta, preparadas por sputtering. Um modelo fenomenológico de parede de domínios no material antiferromagnético (AFM) foi utilizado, levando em conta as anisotropias magnéticas e interações entre as camadas. Também foram consideradas certas dispersões da anisotropia dos grãos ferromagnéticos (FM) e antiferromagnéticos (com distribuições Gaussianas) em torno dos respectivos eixos de anisotropia uniaxiais. Para o ajuste da magnetização para algumas amostras, foi necessário utilizar uma rotação no plano de um ângulo nos eixos de anisotropia uniaxiais do FM e AFM, em relação à direção do campo magnético aplicado durante a deposição dos filmes. Bons ajustes das curvas de histereses das magnetizações foram obtidos nas direções medidas do campo magnético aplicado. Um método baseado em medidas de variações angulares da magnetorresistência em campos constantes foi proposto para extrair este ângulo para cada amostra. Foram obtidas razoáveis concordâncias entre estes ângulos e os correspondentes extraídos dos ajustes das curvas de magnetização. Através da análise dos diagramas da MR-FORC e de simulações indicados dos resultados dos ajustes das histereses da magnetização, foi encontrada uma relação direta entre os campos de interação (e suas incertezas) com os campos de exchange-bias (HEB) dos grãos da distribuição (extraídos das simulações, usando a largura da distribuição obtida do ajuste). Resumindo, esta análise mostrou que esta técnica permite extrair informações comparativas sobre a dispersão dos eixos de anisotropia dos grãos FM e AFM em torno do eixo de anisotropia uniaxial, o que pode ser importante na caracterização dos sensores magnetorresistivos. Além disso, análise dos diagramas MR-FORC indicaram início da presença de descontinuidade na camada de NiFe presa em 27, com um aumento acentuado (acima do previsto) para a amostra com 25. Este aumento acima do previsto corrobora com nossa hipótese. As simulações das curvas de histerese da magnetorresistência não foram muito bons, indicando que melhorias devem ser introduzidas no modelo utilizado para a simulação da histerese da magnetorresistência, obtidos a partir dos ângulos das camadas ferromagnéticas livre e presa. A questão referente a presença em algumas das amostras de um desalinhamento entre os eixos fácéis do FM e do AFM ainda é uma questão em aberta, mas neste trabalho foi encontrado que este ângulo é igual a 2. / Spin Valves have been employed as magnetic sensors and used in random access memories, showing they are very important in terms of technological point of view. In this work, analyses of the magnetoresistance first order reversal curves (MR-FORC) have been used, as well as fittings of the magnetization and magnetoresistance hysteresis, to study the exchange-bias phenomena, magnetic anisotropies and magnetoresistance in spin valves. Sputtering has been used to the deposition of NiFe/Cu/NiFe/IrMn, and Ta has been deposited as seed and buffer layers. A domain wall model (in the antiferromagnetic layer) taking into account the magnetic anisotropies and the interactions between the layers has been employed to fit the magnetization hysteresis. Some textures have been also introduced to take into account the ferromagnetic (FM) and antiferromagnetic (AFM) grains dispersion (with Gaussian distributions) centered around the respective uniaxial anisotropy axes. However, to obtain good fits for some samples, it has been necessary to include an in-plane rotation of an angle of the both FM and AFM easy axes in relation to the field direction applied during the growing of the films. Good fits of the magnetization hysteresis have been obtained for all measured directions of the applied field. A new method based on the angular variation of the magnetoresistance to constant fields has been proposed to extract directly these angles. Reasonable agreements have been obtained between these angles and the corresponding ones extracted from the fits of the magnetization loops. Through the analyses of the MR-FORC and from the simulations indicated by the parameters (obtained from the fittings of magnetization loops), a direct relation between the interaction fields (and its uncertainties) and the exchange-bias fields of the grains of the distribution (extracted from the simulations, using the width of the distribution obtained from the magnetization fittings) has been identified. In summary, this analysis has showed that this technique allows to extract comparative information about the dispersion of the anisotropy axes of the FM and AFM grains around the uniaxial axis, which can be very import to the characterization of spin-valve based sensors. Besides, MR-FORC analyses have also indicated the presence of a threshold of discontinuity of the pinned NiFe layer at 27, showing a huge increase (above of the expected) to the sample at 25, and this unexpected increasing has corroborated with our hypothese. Simulations of the magnetoresistance loops have not been good, indicating that improvements should be included in the model employed to simulate these curves, obtained from the pinned and free angles of the NiFe layers. Concerning the case of the presence of misalignments of FM and AFM for some samples, it is still an open question, but in this work, we have found that this angle () is equal to 2.