This project investigates a new means to store computed information that is not sensitive to electromagnetic pulses. In this method heat is conducted through a material that can be in two different states. In one state, the material will have like atoms arranged in a specific order, while in the other state the atoms will inhabit random positions. Laser pulses are used to locally heat the material and transform it from one state to the other, which then serve as either on-state or off-state memories. It has been computationally predicted that solid solutions of various atoms will conduct heat differently depending on the chemical order, but has not been experimentally proven. This hardened computing method enhances information security and computing resiliency against unavoidable electromagnetic pulses, such as those resulting from solar flares. This project also seeks to educate the public on the magnitude of energy wasted as heat and how it can be harnessed and converted for useful energy.
This research project studies the role of chemical order on thermal conductivity in a material system where heat is carried by phonons, and will advanced the fundamental understanding of how local chemical order can be manipulated to achieve a desired phononic or thermal response. The material studied is lead scandium tantalate, a perovskite oxide that can be chemically ordered or disordered on the octahedrally coordinated cation site depending on the thermal history. The project employs epitaxial growth via pulsed laser ablation on single crystal strontium titanate substrates with an epitaxial strontium ruthanate conductive oxide electrode. Thermal conductivity evaluated using time domain thermoreflectance establishes the thermal conductivity of the lead scandium tantalate with varying degrees of chemical order. High energy optical laser pulses heat the lead scandium tantalate material locally resulting in spatial control of chemical order and thermal conductivity. The perovskite lead scandium tantalate is isostructural with lead zirconium titanate, which has been previously shown to serve as a phonon regulating material. This project develops the necessary perovskite thermal memory to accompany the thermal regulator, which is an important step toward developing a phonon computer. Furthermore, this research tests computational predictions that posit a chemically ordered material possesses a higher thermal conductivity at low temperatures than a disordered counterpart, but that will exhibit a cross-over in thermal conductivity at elevated temperatures, which is not yet experimentally validated. The ability to compute and store information via phonons is a potential path toward electro-magnetic pulse resilient computing, which is vital for information security.
This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
