Exploiting random phenomena in magnetic materials for development of new generation computing strategies

Random phenomena are ubiquitous in magnetism, including the random orientation of magnetization in assemblies of isotropic magnets, arbitrary domain patterns in multilayers, vortex chirality, or Brownian skyrmion motion. While randomness is usually undesirable when seeking repeatable results, it also creates opportunities in stochastic computing, true random number generation, and unclonable functions for data security.

As a paradigmatic case, we will study arrays of magnetic nanodots, whose interactions can be tuned by external magnetic fields or via voltage-driven magnetoionic control. A non-interacting array of magnetic dots can model a completely random magnetic memory. However, tuning their interactions can allow us partial control of randomness, turning the system into a physically probabilistic system. From this basis, different combinations can yield computing structures capable of fuzzy logic, stochastic computation, p-bits or random number generation.

The thesis will explore these systems through theory, simulation, and experiment. Simulations will clarify the random nature of the system and guide strategies for controlling it, while experiments will focus on fabrication, characterization, and manipulation of nanodots to realize functional computational elements. This combined approach will ensure both fundamental understanding and practical demonstration.

From an energy-saving perspective, magnetic systems are highly efficient compared to mechanical or electrical counterparts, since they involve no moving parts and reduced electric losses. Moreover, one of the major bottlenecks in computation is the energy cost of reading and writing information. Advances in controlling and processing magnetic data with voltage (e.g., magnetoionic systems) could therefore have a strong impact, both by enabling new architectures and by lowering energy consumption. This is the central aim of the proposed thesis project.

The candidate should have a good background in materials science, solid-state physics, and preferably some skills in computer modelling and programming. Practical experience or a strong interest in thin-film deposition, patterning and materials characterization (structural, magnetic, and electrical) will be advantageous, supporting the development of magnetoelectric expertise during the PhD.

The candidate must be curious, adaptable, collaborative, and possess strong problem-solving skills, since the project involves several different methodologies that attacks the problem from several point of views. Proficiency in data analysis, modelling, and scientific communication will be essential for interpreting computing results, experimental results, troubleshooting experiments, and disseminating findings through publications and presentations. Initiative, independent learning, and teamwork are key to success in this interdisciplinary research environment.

Our research focuses on the design/modelling, synthesis, and characterization of advanced materials with tailored properties for cutting-edge engineering applications. By precisely controlling their structure at the nanoscale, we create materials with optimized magnetic performance and enhanced thermal stability.

We study and model a variety of systems, including nanowires, lithographically patterned micro- and nano-objects, thin films, and nanocomposite/glassy alloys. Each offers opportunities to uncover new physical phenomena and functionalities.

Sustainability and energy efficiency are guiding principles of our work, shaping both the development of materials and their envisioned applications. In recent years, we have devoted particular attention to nanomaterials for brain-inspired memory and computing, with the aim of realizing energy-efficient, high-performance devices that bridge materials science and neuromorphic engineering.