26th Australian Institute of Physics Congress and the Australian and New Zealand Conference on Optics and Photonics

Performance Limits for Radiative Cooling Structures

Not scheduled

20m

Belinda Hutchinson Building (The University of Sydney )

Contributed Oral ANZOS | Photonics and Optics (ANZCOP)

Speaker

Ben Nel (University of Syndey)

Description

Passive daytime radiative cooling structures can reach sub-ambient temperatures under direct sunlight by reflecting solar radiation while emitting heat through the atmospheric transparency window. Amidst rising global temperatures and the urgent need to slash greenhouse emissions, radiative cooling offers a promising sustainable complement to conventional cooling. Although designs are advancing rapidly, existing performance bounds remain too loose to indicate how close these designs sit to their fundamental limits. Tighter bounds are required because they reveal where substantial improvements remain and identify when a design has reached its constrained global optimum.

Any solar radiation absorbed by a radiative cooler heats the structure, counteracting the cooling. Minimizing solar absorption is therefore essential. Here, we consider a design region shielding an uncontrollable structure from an incident plane wave. The system consists of a top layer, within which material may be structured freely, shielding a bottom layer that cannot be altered. Each layer is specified by its thickness and relative permittivity. For this configuration, we derive bounds on the minimum achievable absorption. A bound is a rigorous performance limit that holds for every possible structure within the design region, implying that no geometry or material distribution, however intricate, can absorb less.

The bounds are obtained by optimizing directly over the electromagnetic currents induced in the design region rather than over specific geometries. The currents are constrained by conservation of real and reactive power.

The bounds are applied to both finite and infinite periodic structures, the latter greatly reducing the computational cost of evaluating large-scale structures. Our results demonstrate the importance of the constraints, with each added constraint significantly tightening the bounds. Future work will enforce causality to tighten the bounds over broad frequency ranges.