Using ferroelectric domain walls for active control of heat flow at the nanoscale

In order to satisfy societal demand for continual improvements in microelectronic device performance, there is an ongoing drive for transistor miniaturisation so that spatial packing densities can be maximised. However, the associated increases in operational power density leads to increased heat generation and rises in on-chip temperature that can prevent reliable device performance. This represents a tremendous technological challenge and there is a clear need to identify and characterise materials with novel thermal properties that will enable superior thermal energy management at the nanoscale. In particular, the ability to actively control heat flow with an external stimulus (e.g. voltage) could have dramatic implications for the thermal management demands and lifetimes of next generation microelectronics. In this regard, oxide ferroelectric materials present an exciting opportunity.

In ferroelectric materials, there exist atomically sharp structural interfaces called 'domain walls' (DWs) that are known to impede heat-flow by disrupting thermal vibrations. What is unique about DWs is their remarkable ability to be created, erased or repositioned inside the material in a fully reversible way by using applied voltages or pressure. This property provides an unprecedented means to actively control heat flow by being able to alter the number of DWs present in the material at a given time and the way in which they are arranged. However, to realise heat flow control using DWs, definitive estimates for the thermal interfacial resistance presented by DWs in different materials must first be determined. Therefore, one of the main goals of this project is to quantify DW thermal resistances through direct thermal conductivity measurements. Ferroelectric material systems having DWs that effectively inhibit heat flow will then be identified. Following this, prototype thermal devices will be fabricated where the relative ease of heat flow through the material will be changed by using applied voltages to reversibly alter the DW pattern. This will also provide the foundation for a longer-term research vision to create a more exotic nanostructured 'thermal mirror' device. In this case, it is envisaged that DWs can be engineered to behave as periodic reflectors of thermal waves in order to maximise the rejection of thermal energy, much like how light is reflected with high efficiency by the multiple layers in a dielectric mirror.

Over the last decade, it has become clear that DWs can be considered as a new type of sheet-like functional material with properties that can be remarkably different than bulk. For example, electrical conduction within DWs can be metallic, or even superconducting, when the bulk is comparatively insulating. Prototype active devices have been fabricated where functionality is derived entirely from deployment of electrically conducting DWs. However, the complementary idea that the narrow DW region may have thermal properties entirely of its own is completely new and unexplored. Within conducting DWs, it is likely that heat flow will be enhanced, due to the availability of extra heat carriers (e.g. mobile electrons), and thermal conductivity measurements will be carried out to confirm this. Conducting DWs will also be explored for conversion of waste heat into electricity since recent predictions indicate that the thermoelectric power can be enhanced by up to 100% within DWs, compared to bulk.

Overall, ferroelectric DWs are exciting candidates for use as the active elements in thermal devices since the DWs may behave functionally to either enhance or restrict heat-flow. However, neither case is currently well characterised nor understood. The innate reconfigurability of these DWs means there is real potential to design and build new types of active thermal devices based on ferroelectric materials that has yet to be capitalised upon.