消 色 差 超 构 透 镜:过 去 ,现 在 ,未 来

Significance Lenses are fundamental to human scientific exploration and technological advancement, extending vision from microscopic cells to distant galaxies. Traditional refractive lens systems, however, rely on multiple bulky elements to correct chromatic aberration, conflicting with the urgent demand for miniaturized and integrated optical systems in modern applications like mobile imaging, augmented reality, and biomedical devices. Metasurfaces, composed of subwavelength nanostructures, offer unprecedented control over light’s amplitude, phase, and polarization within an ultra-thin platform. Metalenses, the focusing variant of metasurfaces, promise to revolutionize compact optics. A critical challenge for their practical adoption is inherent chromatic aberration due to the strong dispersion of meta-atoms. Therefore, developing achromatic metalenses capable of broadband, diffraction-limited performance is paramount, serving as the key bridge from laboratory prototypes to widespread industrial and consumer applications. This field represents the convergence of nanophotonics, advanced materials, computational design, and precision manufacturing, aiming to deliver high-performance flat optics that are both miniaturized and color-corrected. Progress The evolution of achromatic metalenses can be segmented into distinct phases. Initially, efforts focused on discrete wavelength correction. A paradigm shift occurred with the establishment of dispersion engineering methods around 2017—2018, which became the mainstream approach for achieving continuous broadband achromatism. These methods, including dispersion compensation, dispersion-phase space, and Taylor expansion, mathematically formalize the requirement for meta-atoms to exhibit a linear or specifically tailored phase dispersion to compensate for the chromatic focal shift. This enabled pioneering demonstrations of broadband achromatic focusing in the near-infrared and visible spectra. However, a fundamental trade-off was identified linking the achievable aperture size, numerical aperture (NA), and bandwidth to the limited phase dispersion range of a single-layer meta-atom, imposing a significant constraint on device scalability. To overcome these limitations, recent research has advanced on multiple fronts. In terms of physical principles, novel concepts like discrete-wavelength dispersion folding and nonlinear dispersion fitting for ultra-broadband correction have been proposed to extend design freedom. Structurally, moving beyond single-layer designs has been highly fruitful. Multilayer metalenses, including dual-layer architectures with complementary dispersion and triple-layer designs enabled by topology optimization, have significantly enhanced dispersion control. A particularly promising strategy involves hybridizing a metasurface with a dielectric phase plate, which simplifies fabrication and alignment while offering excellent achromatic performance. Furthermore, hybrid refractive-metasurface systems combine the advantages of traditional optics (e.g., large aperture, low cost) and metasurfaces (precise aberration correction), showing great potential for practical imaging systems, especially in the infrared regime. Concurrently, the integration of artificial intelligence has brought transformative breakthroughs. For front-end device design, inverse design and deep learning algorithms, such as neural networks and adjoint optimization, have dramatically accelerated the discovery of optimal meta-atom geometries and multilayer configurations, achieving performance metrics difficult to reach with conventional methods. For back-end system integration, a new paradigm of“minimal optics + computational imaging”has emerged. Here, a metalens with partial correction serves as the optical front-end, while deep learning-based image reconstruction algorithms, such as fully differentiable frameworks or specialized neural networks (e.g., MIRUnet), digitally correct residual aberrations, enabling high-quality, full-color, wide-field imaging with single-element metalenses. This synergy between nanophotonics and AI effectively decouples optical and digital complexity. Conclusions and Prospects Achromatic metalens research has progressed from foundational demonstrations to sophisticated designs integrating novel principles, multilayer architectures, and computational intelligence. The field is transitioning from addressing fundamental physical limits to tackling system-level integration and manufacturability challenges. Future development will hinge on the deep convergence of multidisciplinary expertise. In design, AI-driven inverse design and multi-physics optimization will enable more complex, multifunctional meta-optical systems. In fabrication, the transition from high-precision but low-throughput techniques like electron-beam lithography (EBL) to scalable, cost-effective methods such as nanoimprint lithography (NIL) and deep ultraviolet lithography (DUV) is crucial for wafer-scale production and commercialization. The fusion of metasurface optics with computational imaging algorithms presents a powerful pathway toward ultra-compact, high-performance imaging systems. Looking forward, sustained innovation in design algorithms, nanofabrication processes, and heterogeneous integration will be essential to propel achromatic metalenses from advanced laboratory prototypes into mainstream optical products, ultimately fulfilling their promise of miniaturizing and enhancing optical systems across diverse applications.