Moore's law, which has described the increase in transistor density in integrated circuits for over 50 years, had a silent partner: Dennard scaling. This principle stated that as transistors got smaller, their power density remained constant, ensuring an overall performance improvement. However, Dennard scaling did not account for leakage currents, which can lead to significant chip heating. Since this became important around 2006, processor clock frequencies have been practically limited to approximately 4 GHz.
This fact, sometimes known as the thermal bottleneck, has made thermal management one of the most important research fields in condensed matter. Maintaining a low thermal budget is perhaps the most pressing issue in modern electronics. This PhD project takes up the challenge of heat control in solids, with the main goal of advancing knowledge regarding anisotropic thermal transport. It will tackle two classes of anisotropy: natural, built-in anisotropy resulting from crystal structure or morphology, and artificial anisotropy induced by tailor-made nanostructuring. In the latter case, superlattices (SLs) will be the main case study, known to foster phonon coherence, as well as thin films with a controlled thickness gradient.
The student will receive training in state-of-the-art methodologies for studying thermal transport in nanoscale systems, including theoretical and experimental techniques. The theoretical activity will focus on calculating phonon dispersion and lattice thermal conductivity of functional oxides and nanostructured semiconductors, primarily using density-functional theory (DFT). These calculations will require substantial computational resources in high-performance infrastructures like the Barcelona Supercomputing Center (BSC) and the Supercomputing Center of Galicia (CESGA).
The experimental activity will revolve around a recently developed in-house experimental setup suitable for studying steady-state thermal transport with enhanced sensitivity to in-plane heat flow. This method is contactless and uses a linear heater for sample heating. The student will be supervised by an established team of PIs (R. Rurali, theory; J. S. Reparaz, experiments) with a long collaborative history.
