INVISIBLE LIGHT: SPECTRO-POLARIMETRIC CONTROL AND DETECTION OF THERMAL RADIATION

Thermal radiation, an omnipresent phenomenon characterized by electromagnetic wave emission from objects above absolute zero, has consistently intrigued scientific exploration throughout history and profoundly influences various technological applications. Traditionally, the primary utilization of thermal radiation has been limited to fields such as lighting, cooling, and energy harvesting. However, the true potential of thermal radiation extends far beyond these energy-oriented applications. Every object imprints a unique signature within its emitted thermal radiation. These signatures, distinguished by their wide-ranging spectral and polarimetric characteristics, represent a rich information source about the emitting objects. The goal of this dissertation is to offer novel prospective and platforms to expand our perception and utilization of the spectral and polarimetric attributes of thermal radiation. It seeks to augment the conventional understanding of thermal radiation as merely an energy source, underlining its immense potential as an information carrier.

This dissertation explores the spectral and polarimetric features left within the thermal radiation and how these features can be manipulated. The research uncovers that the macroscopic spectral, spatial, and particularly spin properties of thermal radiation are intimately connected to the underlying symmetry of the microscopic emitters within a nanophotonic system. This close relationship between symmetry and thermal radiation introduces a universal strategy to gain thorough control over the spectral-polarimetric properties of thermal radiation. The control of these properties may spur pioneering developments in encoding information within thermal radiation.

Furthermore, platforms to decode these spectral and polarimetric properties in thermal radiation are as pivotal as the encoding platforms. These decoding platforms allow us to uncover hidden messages within this invisible light and enable us to push the boundaries of fully passive and physics-aware machine perception. Nevertheless, contemporary methods for spectrum and polarization resolved detection of thermal radiation, especially in imaging form, are cumbersome, lacking robustness, and prohibitively expensive. Hence, this dissertation explores two fundamentally innovative spectral separation schemes: the nonlocal super-dispersion enabled by optically active crystals and the dispersive dichroism in 2D infrared metasurfaces. These methods present compact, cost-effective, and high-performance solutions for spectral-polarimetric thermal imaging, thereby enhancing its efficacy in diverse applications.