Session: 11-10-04: Radiative Heat Transfer Across Scales

Paper Number: 145996

145996 - Inverse Design of Colored Radiative Cooling Paint 

Radiative cooling materials have found wide-ranging applications in energy-saving technologies. However, current state-of-the-art radiative cooling materials are primarily designed to possess high reflectivity across the entire solar spectrum, resulting in a white or silver appearance. This limitation in color can be aesthetically undesirable, particularly in commercial and social facilities where color plays a significant role in shaping the environment. Moreover, the intense brightness of white surfaces can be harmful to human eyes. Therefore, there is a growing need to develop radiative cooling materials that can exhibit various colors without compromising their cooling efficiency. There are several studies on developing colored radiative cooling paints. However, current approaches to colored radiative cooling paint are primarily forward design processes. In these approaches, the absorber is selected first, and the color is determined afterward. In real-world applications, manufacturers may need to provide paint based on customer color preferences. Achieving the inverse design of colored radiative cooling paint, where the paint formulation is determined to achieve maximum cooling power for a given color, remains a challenge.

In this work, we present an inverse design framework for colored radiative cooling paint. This framework involves two key processes. First, it aims to determine the spectra that maximize cooling power while adhering to a specified color. This inverse problem is turned into an optimization problem, where the cooling power is maximized while constraining the color. Subsequently, the optimized reflective spectra can be realized in different systems. In this work, we focus on a BaSO4-acrylic nanocomposite containing Ag@SiO2 core-shell nanoparticles with a small volume fraction. Similar compositions have been shown to exhibit narrowband absorption characteristics. We employ machine learning to facilitate the inverse design of the optimized spectrum based on various design parameters. In this way, we achieve the inverse design of colored radiative cooling paint. Our framework has the potential to be expanded to other nanophotonic systems, such as 2D photonic crystals and multilayer systems, opening up new possibilities for designing efficient and aesthetically pleasing radiative cooling materials.

Presenting Author: Ziqi Guo Purdue University

Presenting Author Biography: Ziqi Guo is a third-year PhD candidate at Purdue University. His research interest lies at the intersection of physics simulation, energy transport and machine learning.

Authors:

Ziqi Guo Purdue University
Dudong Feng Purdue University
Daniel Carne Purdue University
Guang Lin Purdue University
Xiulin Ruan Purdue University