Heat is the main source of energy loss in most industrial processes and in electronic devices. To manage and reuse heat more efficiently, a study led by the University of Barcelona has developed a thermal memory prototype capable of controlling heat flow with small electrical voltages. This innovation, published in the journal
Advanced Materials, has potential implications for intelligent heat management, energy conversion and thermal energy harvesting.
The study is led by Professor Eric Langenberg, from the Department of Condensed Matter Physics at the Faculty of Physics and the Institute of Nanoscience and Nanotechnology (IN2UB) at the UB, and the first author is expert Dídac Barneo (UB-IN2UB).
Also taking part are experts from the Center for Research in Biological Chemistry and Molecular Materials (CiQUS), based at the University of Santiago de Compostela; the Materials Science Institute of Madrid (ICMM-CSIC); the Institute of Materials Science of Barcelona (ICMAB-CSIC); and the University of Zaragoza (UNIZAR).
Controlling heat in electronic devices
The study addresses one of the major challenges of modern electronics: controlling heat in electronic devices. “This research could pave the way for the development of a new information storage technology based on heat rather than electricity,” says Professor Langenberg.
The prototype consists of a solid structure formed by a gold layer on top of an ultra-thin layer (seven nanometres thick) of a ferroelectric hafnium-zirconium oxide (Hf0.5Zr0.5O2), which is deposited on an oxygen-ion conductive substrate. (yttria-stabilised zirconia, YSZ).
When small voltages are applied between the gold and the substrate, the thermal conductivity of the ferroelectric oxide reversibly switches between two states: high thermal conductivity (with positive voltages) or low thermal conductivity (with negative voltages).
“The applied voltage makes it possible to define two different thermal states and to modulate at will the ease with which heat passes through the material. Since these two thermal-conductivity states remain stable even when the voltage is removed, the device becomes a non-volatile thermal memory," explains Langenberg.
This behaviour is based on a physical mechanism never before employed to control heat transport with an electric field: the coupling of ferroelectric polarization with the migration of oxygen vacancies. These vacancies are small defects that scatter photons and hinder their passage. Photons are the basic unit of vibration in solid structures, much as electrons do in the case of electricity. “Thus, a high concentration of oxygen vacancies in the ferroelectric oxide hafnium zirconium oxide will impede the passage of phonons (low thermal conductivity), in contrast to a low concentration of vacancies (high thermal conductivity),” explains Dídac Barneo, also a member of the Department of Condensed Matter Physics.
These oxygen vacancies can move through the solid structure under an electric field. Negative voltages attract these vacancies, whereas positive voltages repel them. “If a negative voltage is applied though the gold layer, the ferroelectric oxide exhibits low thermal conductivity. Conversely, when positive voltages are applied, the result is high thermal conductivity,” notes Barneo.
But the applied voltage alone cannot explain the memory phenomenon. “There is a crucial component for memory functionality and the stability of thermal states: the polarization of the ferroelectric oxide, which acts as a valve that prevents the oxygen vacancies from redistributing themselves when the voltage is removed. This effect allows the two thermal states to be preserved and turns the device into a thermal memory,” the experts say.
From electrons to phonons: challenges in solid-state physics
This device is the first functional prototype developed within the e-PHONONBIT project (Design of phononic bits with electrical control in films and epitaxial superstructures based on ferroelectric oxides), which is led by Professor Langenberg. As part of the study, the UB team devised the physical mechanism that explains the operation of thermal memory, coordinated the various experimental and theoretical collaborations and integrated all the data into the study’s overall interpretation.
In the future, the performance of the new device will need to be improved in order to increase the difference in thermal conductivity between the two states, enable it to operate at room temperature (it currently operates at 200°C) and to reduce the switching time between the two thermal states (from minutes to milliseconds).
“The idea is to explore different combinations of ferroelectric materials, ionic conductors (in particular, oxygen vacancies) and substrates to address these three challenges,” explains Eric Langenberg.
“In the longer term, we aim to develop new computing architectures based on phonons. Controlling phonon transport with an electric field is quite a challenge in solid-state physics. We are at the dawn of phononics; only time will tell how far this new discipline will take us,” the researcher concludes.