The Racetrack memory is one of the devices that has attracted more attention within the field of spintronics. In this type of memory, the main idea is to move the bits of magnetic information rather than the disc, as it happens in conventional magnetic hard drives. In the racetrack memory, the domains separated by magnetic domain walls (DW), travel pushed by a spin polarised current and the sequence of information can be read at a point by for instance, a magnetic tunnel junction. One of the biggest challenges this memory faces is the reproducibility of the movement and position of the magnetic domain walls. Pinning the DW at a particular point is easy, but reproducibly depinning it from that point, not using a large amount of current, is a big challenge. The scientific community has intensively worked on this problem in the last 20 years. The main approaches have been to increase the polarization of the current via surface effects such as the spin-orbit coupling, to use topology to make the magnetic "bit" more stable, such as in the case of skyrmions and to use different types of pinning points so the positioning and movement of the DW can be more reliable and more energy efficient. In this last approach, we centre the work in the thesis. In this work, we present an alternative method to pin and unpin a DW in a magnetic strip. To do so, we have used the magnetic interaction of the DW with an adjacent layer deposited on top with perpendicular magnetic anisotropy PMA. using a thorough magnetic characterization we have disentangled the magnetic interaction of this PMA multilayer with the different layers of a spin valve. Then using the knowledge of this magnetic interaction of this PMA multilayer with the different layers of a spin valve. Then, using the knowledge of this magnetic characterisation, we have fabricated nanostrips where the PMA multilayer has been used as a pinning point for the DW and the performance of this basic unit of a Racetrack memory has been electrically characterized. We found that this type of defect is much stronger at pinning the DW than we predicted from the magnetic characterisation. But then, its pinning capability can be easily tuned by separating the structure from the DW with an inactive layer. We therefore believe that the results presented in this thesis can pave the road to develop an alternative type of defects in racetrack memories. With the crucial advantage of not posing an additional source of Joule Heating. The development of this work involved learning a wide variety of fabrication and characterization techniques. The deposition of the films was done by sputtering. The magnetic characterisation was done by (Vibrating Sample Magnetometer) and AGFM (Alternating Gradient Force Magnetometer). The electric characterisation was done with transport rings custom-built in our laboratory. The fabrication of devices was done in a clean room involving several techniques, such as e-beam lithography, optical lithography, plasma etching, etc. We have occasionally used other techniques, such as magnetic force microscopy and/or atomic force microscopy. The work began in October 2020, just at the beginning of a series of COVID restrictions that limited access to the laboratories. This was a large barrier during the first year of this thesis, particularly since it was mostly experimental. Despite the problems, the main initial goals of this thesis have been achieved. The main results were published in the Journal of Applied Physics, and the work was made an editorial-featured paper.
© 2001-2026 Fundación Dialnet · Todos los derechos reservados