| Sumario: | [EN] Since its birth marked by the discovery of the Giant Magnetoresistance (GMR) in 1988, the field of spintronics has evolved rapidly, maturing and giving rise to new subfields. The ultimate goal of spintronics is the development of a new paradigm with potential to overcome the limits imposed by the Moore’s law in conventional electronics. This paradigm is based on the use of the spin degree of freedom, together with the electrical charge. A functional spintronic technology requires the capability to control three fundamental operations: generation, transport and detection of spin angular momentum. Therefore, spin dynamics are focus of an intense research aiming at the enhancement of the performance of these three operations. In particular, spin currents are a fundamental object in the field of spintronics. Furthermore, pure spin currents, which are not accompanied by a charge flow, allow to propagate spin without Joule power dissipation losses. In this regard, pure spin currents carried by collective magnetic excitations (magnons) bring the possibility of insulator-based spintronics using electrically insulating materials with long-range magnetic order (MOIs). This thesis is devoted to the study of thermal spin transport, i.e., the coupling between spin and heat currents, which constitutes the area of research of the spintronics subfield of spin caloritronics or thermal spintronics. This field is envisaged to significantly contribute to the development of a new generation of highly efficient thermoelectric devices; to this end, a deeper understanding on the fundamental mechanisms governing thermal spin transport is still required. Through the work developed in this thesis, the spin Seebeck effect (SSE) in maghemite (γ-Fe2O3) -based low-dimensional nanostructures has been exhaustively investigated. The SSE is one of the most prominent transport phenomenon integrating the field of spin caloritronics, as it enables the direct generation of a magnon spin current upon the application of a thermal gradient in magnetic materials. SSE is observed in FM/NM bilayers, where FM is a magnetically ordered material and NM is a paramagnetic or diamagnetic metal (most often, Pt). In general, strategies aimed at improving the SSE efficiency can be targeted at three levels: (1) the heat-to-spin current conversion in the FM layer; (2) the interface level, comprising both interfacial heat-to-spin current generation and spin transfer from FM to NM; and (3) spin current detection at NM. This last step is usually performed by means of spin-to-charge conversion via the inverse spin Hall effect (ISHE) enabled by spin-orbit coupling (SOC). The first part of this thesis is focused on the development as thin films of materials of interest within this topic. In particular, the preparation of high quality maghemite and iridium(IV) oxide (IrO2) thin films is investigated. Additionally, the fabrication of Y3Fe5O12 (YIG) by an affordable chemical method is also addressed. Maghemite and YIG constitute examples of MOIs holding an already successful history of applications in other fields. Meanwhile, IrO2 is a NM metal with strong SOC and high electrical resistivity making it an appealing candidate for spin current detection via the ISHE. Then, the SSE is extensively investigated in γ-Fe2O3/Pt bilayers of nanometric thickness. A thorough depiction of the SSE is reached, accounting for both contributions to the thermal spin current: the one originated at the FM/NM interface, and the one originated at the thickness of γ-Fe2O3 due to the thermally induced magnon accumulation. To this end, the influence of different transport parameters is discussed. Furthermore, a second method for measuring the SSE in steady heating conditions is implemented. This alternative method is based on the current-induced heating approach, in which the NM material holds a triple role: heater, thermometer, and spin-to-charge converter. The experimental equivalence between both methods is analyzed. Additionally, this current-induced method enables the simultaneous detection of the recently discovered spin Hall magnetoresistance (SMR), which is hence characterized in the γ-Fe2O3/Pt bilayers. Secondly, the ultrafast dynamics of the interfacial SSE is investigated by means of the Terahertz Emission Spectroscopy all-optical technique (TES). In this case, three different FM/NM structures using FM materials with different degree of electrical conductivity are used: insulating maghemite, half-metallic magnetite and metallic iron. The comparison of the photoinduced thermal spin currents in each sample enables the characterization of the time scale at which these spin currents rise and decay. Based on their different dynamics, the SSE and the spin-dependent version of the thermoelectric Seebeck effect (spin-dependent Seebeck effect, SDSD) are differentiated. The SDSE-associated thermal spin current is carried by electrons, in contrast to the magnonic nature of spin currents excited by the SSE, and thus SDSE is operative only in conductive materials. Finally, the third level of the SSE —the spin current detection in the NM layer—is addressed using IrO2 as NM in γ-Fe2O3/IrO2 bilayers of nanometric thickness. To date, a few experimental works studying spin-to-charge conversion in polycrystalline or amorphous IrO2 have been reported. In contrast, IrO2 thin films investigated in this thesis are strongly textured in a preferential direction, and the role of the different SOC mechanisms contributing to ISHE in this kind of samples is analyzed. Besides, the striking differences found between ISHE spin-to-charge conversion in textured samples with respect to previous studies in polycrystalline or amorphous IrO2 are discussed. The results open the door to the interesting possibility of tuning the desired functionality of high-resistance spin-Hall-based devices.
|
|---|
