I experimentally realized convolution operations in the synthetic frequency dimension. Using a modulated ring resonator, I synthesize arbitrary convolution kernels using a pre-determined modulation waveform with high accuracy. I demonstrate the convolution computation between input frequency combs and synthesized kernels. I also introduce the idea of an additive offset to broaden the kinds of kernels that can be implemented experimentally when the modulation strength is limited. This work demonstrates the use of synthetic frequency dimension to efficiently encode data and implement computation tasks, leading to a compact and scalable photonic computation architecture.

I introduced a scheme to achieve arbitrary convolution kernels in the synthetic frequency dimension with a simple setup consisting of a ring resonator incorporating a phase and an amplitude modulator. This scheme can be used to perform multidimensional convolutions. I provided an analytic approach that determines the required modulation profile for any convolution kernel. The convolution operation is widely used in signal and image processing and represents the most computationally intensive step in convolutional neural networks. My work points to a direction of using optical computing to remove the computational bottleneck in traditional electronic circuits and will be useful in improving machine-learning hardware in artificial-intelligence applications.

I theoretically investigated the dipolar whispering-gallery modes (WGMs) with different mode orders supported by spherical hyperbolic metamaterial (HMM) cavities consisting of alternating metal and dielectric layers. By integrating a nonlinear nanocrystal into the HMM cavities, I show enhancements of intensity of second harmonic generation up to a factor of 3.9 × 10^10, which is two orders of magnitude higher than the largest enhancement achieved in the single-layer plasmonic core-shell cavities.

Contribution: I proposed the EnergyPlus model to evaluate HVAC energy-saving performance for the low-emissivity film. My developed EnergyPlus model comprehensively considers the heat exchange of the whole building coupled with complicated HVAC systems and outputs the final HVAC energy use of the building. In addition, I also independently proposed a customized thermal model that focuses on the heat transmission reduction of a wall. This model calculates the all-year reduction of heat gain/loss through walls for 16 cities when the coloured low-emissivity film is installed on wall exterior surface, interior surface, and double-sided surfaces, respectively. Impact: This study on energy-efficient buildings play an essential role in sustainability. My work is imperative to develop innovations to achieve improved energy efficiency of building HVAC with a reduced carbon footprint. My work successfully overcomes the limits where traditional building materials can hardly achieve both cooling and heating energy savings throughout the year due to excessive heat transfer arising from high emissivity. Combined with experimental results from other components of the team, I demonstrate a new approach of colored low-emissivity films for building wall thermal envelopes to provide year-round energy-saving solutions. Our work can help reduce heat gain and loss by up to 257.6 MJ per installation wall area annually, which would be beneficial for global carbon neutrality and sustainability.

I show that these parasitic losses can be reduced through thermal engineering. I developed an intuitive model that allows system design with maximal power generation at nighttime, showing the optimum power density can be approached by controlling the relation between the emitter area and the thermal resistance of the thermoelectric generator. I am involved in showing that the stacking of multiple thermoelectric generators is an effective way to approach this optimum. I am involved in experimentally demonstrating a generated electric power density >100 mW/m2, representing > 2-fold improvement over the previous results for nighttime radiative cooling, four times over the previous record.

Contribution: I present a systematic optimization of nighttime thermoelectric power generation system utilizing radiative cooling. I show that an electrical power density two orders of magnitude higher than the previously reported experimental result, is achievable using existing technologies. I study the influence of emissivity spectra, thermal convection, thermoelectric figure of merit and the area ratio between the TEG and the radiative cooler on the power generation performance. I optimize the thermal radiation emitter attached to the cold side and propose practical material implementation. The importance of the optimal emitter is elucidated by the gain of 153% in power density compared to regular blackbody emitters. I showed that this nighttime power generation system can achieve performance close to that of the thermodynamic limit set by the Carnot heat engine. Impact: This system combines radiative cooling and thermoelectric power generation and operates at night when solar energy harvesting is unavailable. The thermoelectric power generator (TEG) itself covers less than 1 percent of the system footprint area when achieving this optimal power generation, showing economic feasibility. This result is significantly higher than the previous reported results and points to the potential applicability of harvesting electrical power at night.

Contribution: I extended the functionality of solar panels into the nighttime for water harvesting, using nighttime radiative cooling. I show that, through emissivity engineering, both the water generation rate and the suitable temperature and humidity range can be significantly improved. As a case study, I proved that it can be significantly enhanced up to 681 mL/m2 with further emissivity engineering. The collected water can also be used for other applications including agrophotovoltaic and evaporative cooling of solar panels during the day, and can be extended to other solar energy harvesting systems. My results point to new avenues to explore the nighttime utilization of a wide range of existing sky-facing solar energy harvesting systems and highlight the opportunities to use both the sun and outer space in existing energy systems.

Contribution: Space heating and cooling consume ~13% of global energy every year. The development of advanced materials that promote energy savings in heating and cooling is gaining increasing attention. To thermally isolate the space of concern and minimize the heat exchange with the outside environment has been recognized as one effective solution. To this end, here, we develop a universal category of colorful low-emissivity paints to form bilayer coatings consisting of an infrared (IR)-reflective bottom layer and an IR-transparent top layer in colors. The colorful visual appearance ensures the aesthetical effect comparable to conventional paints. High mid-infrared reflectance (up to ~80%) is achieved, which is more than 10 times as conventional paints in the same colors, efficiently reducing both heat gain and loss from/to the outside environment. The high near-IR reflectance also benefits reducing solar heat gain in hot days. The advantageous features of these paints strike a balance between energy savings and penalties for heating and cooling throughout the year, providing a comprehensive year-round energy-saving solution adaptable to a wide variety of climatic zones. Taking a typical midrise apartment building as an example, the application of our colorful low-emissivity paints can realize positive heating, ventilation, and air conditioning energy saving, up to 27.24 MJ/m2/y (corresponding to the 7.4% saving ratio). Moreover, the versatility of the paint, along with its applicability to diverse surfaces of various shapes and materials, makes the paints extensively useful in a range of scenarios, including building envelopes, transportation, and storage.

Contribution: Changing how materials emit heat can help save energy in buildings. Such examples include low-emissivity windows and radiatively cooling paints. The past strategies have mostly focused on the spectral properties of radiated heat, whereas changing the direction of radiated heat can bring further energy-saving benefits. We demonstrate a flexible, micro-structured film capable of emitting heat in specific directions. This material allows us to achieve energy savings compared with conventional isotropic emitters, especially in scenarios where environmental temperatures are inhomogeneous, for example, in outdoor radiative cooling for building walls or heating specific areas indoors.

I studied the consequence of breaking reciprocity within the context of near-field radiative heat transfer between two planar bodies. My findings introduce a thermodynamic constraint, which states that the heat transferred from one planar body to another at each frequency and in-plane wave vector is unchanged upon interchanging the two bodies, regardless of whether the materials are reciprocal or not. I further identified a unique signature of nonreciprocity, which is the breaking of the symmetry of the heat flux density between positive and negative in-plane wave vectors. I numerically demonstrated our findings in an example system consisting of magneto-optical materials. My formalism applies to both near- and far-field regimes, opening opportunities for exploiting nonreciprocity in two-body radiative heat transfer systems.

I developed a analytical framework to derive upper bounds to 2D near-field light-matter interactions. My framework connects the classic complex-analytic properties of causal fields with newly developed energy-conservation principles, resulting in a new class of power-bandwidth limits. These limits demonstrate the possibility of orders-of-magnitude enhancement in near-field optical response with the right combination of material and geometry. At specific frequency and bandwidth combinations, the bounds can be closely approached by canonical plasmonic geometries. I also map the bounds as a material “figure of merit,” which determines the maximum response of any material for any frequency and bandwidth.

I developed a theoretical formalism based on the temporal coupled-mode theory that analytically describes the lineshapes of force spectra and their dependencies on resonant scatterers for arbitrary incident wavefronts. I obtained closed-form formulae and discuss the conditions for achieving symmetric as well as asymmetric lineshapes, pertaining, respectively, to a Lorentzian and Fano resonance. I exemplified the relevance of formalism as a design tool with a conceptual scheme of the size-sorting mechanism of small particles, which plays a role in biomedical diagnosis.

I developed the theory of a thermal switch based on resonant coupling of three photonic resonators, in analogy to the field-effect electronic transistor composed of a source, a gate, and a drain. As a material platform, I capitalize on the extreme tunability and low-loss resonances observed in the dielectric function of monolayer hexagonal boron nitride (h-BN) under controlled strain. I controbute to proposing a strain-controlled h-BN–based thermal switch that modulates the thermal conductance by more than an order of magnitude, corresponding to a contrast ratio in the thermal conductance of 98%, in a deep-subwavelength nanostructure. We therefore presented a scheme of robust on/off states in switching the flow of thermal currents, in analogy to the electrical conductance that can be widely modulated within the same material.

I identified force and torque constants, derived within a more general quadratic scattering-channel framework for upper bounds to optical force and torque for any illumination field. I also extended this framework to solve the reverse problem: computing globally optimal “holographic” incident beams, for a fixed collection of scatterers. I analyze structures and incident fields that approach the bounds, which for wavelength-scale bodies show a rich interplay between scattering channels. I show that spherically symmetric structures are forbidden from reaching the plane-wave force/torque bounds. This framework should enable optimal mechanical control of nanoparticles with light.