Source-linked AI summary
Metasurface Optics for Full-color Computational Imaging
Shane Colburn, Alan Zhan, Arka Majumdar
TL;DR
Metasurface optics has struggled to provide full-visible-spectrum imaging because of chromatic aberration and limited achromatic bandwidth. This paper combines a single NA ~0.45 metalens with computational reconstruction to produce in-focus white-light images across visible colors.
Problem
Full-visible-spectrum imaging remains unsupported by metasurfaces with narrow or discrete achromatic bandwidth.
Method
The system combines computational imaging with an EDOF metasurface using a single metalens with NA ~0.45.
Results
The system demonstrates in-focus full-visible-spectrum imaging directly under white light, including red, green, blue, and intermediate colors after deconvolution.
Takeaways & Limitations
The demonstrated system combines computational imaging and metasurface optics to reduce chromatic aberration while downsizing imaging with simpler optics.
Takeaways & Limitations
Required post-processing complicates the system and introduces delay during image deconvolution.
Abstract
from arXiv · showhide
Conventional imaging systems comprise large and expensive optical components which successively mitigate aberrations. Metasurface optics offers a route to miniaturize imaging systems by replacing bulky components with flat and compact implementations. The diffractive nature of these devices, however, induces severe chromatic aberrations and current multi-wavelength and narrowband achromatic metasurfaces cannot support full visible spectum imaging (400-700 nm). We combine principles of both computational imaging and metasurface optics to build a system with a single metalens of NA ~ 0.45 which generates in-focus images under white light illumination. Our metalens exhibits a spectrally invariant point spread function which enables computational reconstruction of captured images with a single digital filter. This work connects computational imaging and metasurface optics and demonstrates the capabilities of combining these disciplines by simultaneously reducing aberrations and downsizing imaging systems with simpler optics.
Discussion
The discussion establishes direct white-light, full-visible-spectrum imaging with a single metalens and deconvolution, while identifying post-processing and scale-related limitations. It frames optical–digital co-design as a route to compact imaging systems with simpler hardware.
- Full-spectrum imaging: The work is the first reported demonstration of in-focus full visible spectrum imaging directly under white light, including intermediate colors after deconvolution.Captured images show red, green, and blue light in focus after deconvolution, along with yellow, orange, and violet.
- Design approach: Simple transmissive scatterers can extend across geometries and materials, unlike broadband achromatic metalenses requiring carefully designed dispersion characteristics.This design flexibility distinguishes the reported approach from related broadband achromatic designs.
- Limitations: Digital filtering enables broadband color imaging but complicates the system and introduces deconvolution delay, while acceleration via O(NlogN) FFT hardware adds circuitry and cost.Offline deconvolution suits photography and video, whereas real-time implementations require balancing requirements, hardware, and cost.
- Limitations: A 200 μm focal length and small aperture limit space-bandwidth product and light collection, reducing information capacity and SNR while requiring higher power or longer exposure.These constraints arise from the short focal length and would occur at the same length scale in other implementations.
- System significance: NA ~0.45 enables full-visible-spectrum object imaging with minimal chromatic aberrations, supporting microscopy, hyperspectral imaging, and ultra-thin cameras.The metasurface also achieves more than a 2.4X reduction in focal length relative to the shortest previously reported value.
- Hybrid-system outlook: Geometric aberrations can be addressed through co-optimizing optical elements and post-processing, illustrating hybrid systems that minimize imaging-system size and complexity.The proposed design model treats optical hardware and software as jointly generating high-quality images.
Material and Methods · Design and simulation
The metasurfaces were designed across the visible regime using RCWA-based nanopost phase mapping and simulated as complex amplitude masks through Rayleigh–Sommerfeld diffraction. Fabrication used silicon nitride deposition, electron-beam lithography, etching, and microscopy to produce the devices.
- Design and simulation: 400-700 nm visible-regime coverage was targeted by varying nanopost diameters to provide 0 to 2π phase.Transmission amplitude and phase were calculated at 400 nm, 550 nm, and 700 nm using RCWA.
- Design and simulation: Dispersion was incorporated through the refractive indices of the silicon nitride posts and silicon dioxide substrate.
- Design and simulation: Weakly coupled pillars were validated by sweeping lattice constant and observing minimal phase change over a wide lattice-constant range.This supported use of the unit cell approximation for implementing phase profiles.
- Design and simulation: 550 nm nominal metasurface designs used RCWA data as a lookup table mapping desired local phase to the most accurate nanopost diameter.
- Design and simulation: Full FDTD simulation was infeasible because the designs’ large spatial extent exceeded available computational memory.Instead, RCWA transmission coefficients represented the metasurfaces as complex amplitude masks.
- Design and simulation: Metasurface performance was simulated by evaluating the Rayleigh-Sommerfeld diffraction integral using an angular approach.
- Material and Methods: 633 nm silicon nitride was deposited on fused silica before protective coating, cleaning, resist processing, charge-dissipation coating, and electron-beam lithography.The fabrication sequence included a JEOL JBX6300FS system and 8 nm of Au/Pd as a charge dissipation layer.
- Material and Methods: Aluminum masking, CHF3 and O2 plasma etching, and aluminum removal completed fabrication, with scanning electron micrographs presented in Fig. S3.
Device characterization
The fabricated metasurfaces were experimentally characterized by imaging their focal planes and measuring efficiency and modulation transfer functions. Efficiency was obtained from focal-plane and incident-beam power measurements using calibrated optical components.
- Focal-plane characterization: Focal planes were characterized experimentally using fiber-coupled LED illumination and a custom microscope with a translatable stage, objective, tube lens, and camera.The camera captured snapshots of the devices’ focal planes.
- Efficiency measurement: Efficiency was calculated as the ratio of power at the focal plane to incident-beam power.Measurements used a flip mirror, pinhole, and Newport 818-SL photodetector.
- Efficiency measurement: Incident-beam power was measured through glass with the pinhole aperture set to image a region equal to the metalens width.This established the incident-power reference for efficiency calculations.
- Measurement correction: Camera dark noise was corrected using calibration images acquired with the lens cap on.The correction was applied before analyzing camera measurements.
- Optical characterization: Modulation transfer functions of the lenses were determined by Fourier analysis.The supplied passage identifies Fourier processing as the method for determining the MTFs.
transforming and then taking the magnitude of the measured focal spot.
The experiment used a fiber-coupled LED and a metasurface-based imaging setup to capture images near the focal plane, with camera dark noise subtracted before analysis.
- A fiber-coupled LED illuminated an off-axis pattern printed on standard 8.5 x 11 paper.The setup is shown in Fig. S5.
- The metasurface formed an image near its focal plane by focusing light scattered from the printed pattern.A translatable microscope comprising an objective, tube lens, and camera captured the image.
- Before image capture, camera dark noise was subtracted after recording a sequence of pictures with the lens.