Public defence in Engineering Physics, M.Sc. Sreeveni Das

Title of the thesis: Engineering magnetic properties of garnet films and designing CoFeB-based magneto-ionic synapses for spin-based computing

Thesis defender: Sreeveni Das
Opponent: Dr. Stefania Pizzini, Institut Néel - Grenoble, France
Custos: Prof. Sebastiaan van Dijken, Aalto University School of Science

Modern computing systems are driving breakthroughs in artificial intelligence, communications, and data processing. However, as these technologies evolve, their growing energy demands raise concerns about sustainability and scalability. One promising alternative is spintronics, a field that utilizes the spin of electrons to process and store information in a more energy-efficient way. Realizing the full potential of spintronic technologies requires the development of advanced magnetic materials and device architectures. This thesis explores how such materials and platforms can be engineered to support future computing systems with reduced power consumption and enhanced functionality.

The first part of the study investigates how the magnetic properties of garnet thin films can be precisely tuned by optimizing growth conditions and employing ion irradiation techniques. These insulating magnetic oxides exhibit exceptionally low magnetic damping, which makes them strong candidates to replace conventional metallic ferromagnets in applications requiring minimal energy loss during signal transmission. The study demonstrates nearly functional magnetic films with promising properties for spin-based computing technologies. 

The second part of the thesis introduces a device concept designed to emulate the synaptic behavior of the human brain. These artificial synapses form the basis of neuromorphic computing, an emerging field that aims to replicate how the brain processes information. A key feature of the synaptic device is the voltage-driven modulation of magnetic domains, which enables the control of magnetic states without the need for electric current. This mechanism mimics the way biological synapses work: it can store information and adjust its response based on past activity, offering a pathway toward low-power, adaptive computing systems capable of learning and memory functions.

Overall, this research offers a new understanding of how magnetic properties can be controlled through materials engineering and device design at the nanoscale. The findings contribute to ongoing efforts in developing energy-efficient computing technologies and may be of interest to researchers working in spintronics, neuromorphic hardware, and functional materials.

Keywords: Magnetism, Spintronics, Magnetic skyrmions, Neuromorphic computing, Magneto-ionics, Perpendicular magnetic anisotropy, Magnetic garnets

Thesis available for public display 10 days prior to the defence at Aaltodoc.