Figure 5
From: Deep learning based analysis of microstructured materials for thermal radiation control
Material search algorithm identifying the most optimal microstructures for cooling at surface temperatures of 300, 500, and 1000 K based on the figure of merit defined by Eqs. (5–6). Due to the role of transmission in Eq. (5), the most optimum microstructure for cooling at room temperature is Au (FOM = 0.772) as typical cooling materials such as PDMS and SiO2 transmit thermal radiation in the visible wavelengths, negating cooling for a surface below. At 500 K, we identify SiO2 micropyramids as being most optimal (FOM = 0.852). At 1000 K, the algorithm identifies VO2 as best performing micropyramid structure (FOM = 0.982) among all 41 materials that were predicted by the network. All materials are predicted assuming a minimum wavelength of 0.3 and maximum wavelength of 16 um to capture both thermal emission and solar absorption optical properties.
