Kaipeng Liu

The co‐manipulation of orbital angular momentum (OAM) in real and momentum spaces presents challenges in electromagnetic wave engineering, primarily due to the counteraction between topological defects and disorder, particularly in overcoming the intrinsic bandwidth limitations of topological photonic systems. Recent advancements have shown the generation of real‐momentum space vortices through bound states in the continuum in topological photonic crystals (PhCs), but these implementations face critical trade‐offs between spatial disorder tolerance and operating bandwidth. Here, we present a disorder‐empowered metasurface architecture that breaks this paradigm through geometric‐phase engineering combined with topological PhC design. Our platform achieves simultaneous broadband OAM generation in both real and momentum spaces, notably demonstrating a 40% fractional bandwidth in the momentum domain—far surpassing previous records. Experimental results break the prevailing notion that real‐space disorder inevitably degrades momentum‐space performance, demonstrating instead that engineered disorder actively stabilizes dual‐space broadband characteristics over an ultra‐wide frequency range. This phenomenon originates from geometrically locked Pancharatnam‐Berry phase robustness and self‐compensated momentum‐space topology, enabling wavelength‐independent vortex generation with preserved charge orthogonality across domains. This dual‐broadband capability expands the information‐carrying potential of topological photonic devices, paving the way for high‐capacity communication systems that leverage disorder as a beneficial design feature rather than a limiting factor.

Solitons are self-sustained wave packets that arise in wave systems and maintain their shape during propagation by balancing nonlinear and dispersive effects, exhibiting stability, robustness, and particle-like interactions. However, localized traveling temperature pulses are difficult to sustain in thermal media, as diffusion rapidly broadens and attenuates localized profiles and intrinsic driving and nonlinearity are generally absent. Here, we show that electronic driving offers a pathway to circumvent these limitations through programmable thermoelectric interfaces, enabling precise, dynamic modulation and the construction of reconfigurable coupling networks. Using this approach, we experimentally demonstrate wave-like transport behavior in a thermoelectric metamaterial with electronically controlled couplings. Within a non-Hermitian framework, a pseudo-convection effect propels thermal fields; further incorporating nonlinearity leads to soliton-like thermal pulses that display markedly reduced amplitude decay and relative broadening compared with linear diffusive dynamics. Our observations reveal the synergistic effect of circuit-mediated non-Hermicity and nonlinearity, providing a mechanism for localized energy propagation and information transmission.

Tailoring the angular distribution of thermal emission is crucial for applications in space thermal management, infrared stealth, and energy systems. The inverse design of nonlocal metasurfaces remains challenging due to the high-dimensional parameter space and computationally expensive full wave simulations. This work presents and experimentally validates an efficient, deep learning driven inverse design platform for nonlocal thermal photonic metasurfaces with customized directional emissivity. First, by applying particle swarm optimization (PSO) to introduce perturbations to a bound state in the continuum (BIC) metasurface, we demonstrate precise control over the angular emissivity profile within a ±60° range. Fourier-transform infrared spectroscopy and long wave infrared thermal imaging confirm the intended asymmetric emission, showing excellent agreement between experiments and simulations. To overcome the limited design freedom of perturbation-based strategies, we further develop a global optimization framework. It employs Non-Uniform Rational B-Splines (NURBS) for high-degree-of-freedom meta-atom parameterization, accelerated by an end-to-end deep learning surrogate model linked to spatiotemporal coupled-mode theory (STCMT). This differentiable surrogate bypasses repetitive full-wave simulations and enables efficient gradient based optimization without adjoint methods. Metasurfaces fabricated using this workflow exhibit thermal emissivity profiles at 150°C in excellent agreement with predictions, achieving a normalized root mean square error below 0.1. This work demonstrates a robust and scalable platform for the inverse design of high performance nonlocal thermal metasurfaces with complex spectral-angular responses.