Liquid-Crystal Metalens Switching Enables Dual Bright-Field and Synthesized Edge-Enhancement Modes

Researchers have proposed a liquid-crystal-integrated polarization-multiplexed metalens that electrically selects two bright-field states and an interference-synthesized edge-enhanced state. The two polarization channels share a focal plane but have unequal effective apertures and different spatial-frequency bandwidths.

Under 45°-polarized illumination a followed by a 45° analyzer, a designed π-phase offset coherently cancels shared low-frequency components while retaining higher spatial frequencies. Experiments at 635 nm validate focusing, resolution-target imaging, biological imaging, and voltage-controlled switching without digital post-processing.

In an optical imaging system, the transfer function determines which spatial-frequency components of the object field reach the image plane and with what amplitude and phase. Engineering this function is essential for visualizing transparent or weakly absorbing specimens, whose internal boundaries may produce little intensity contrast under bright-field illumination. Dark-field, Zernike phase-contrast and differential-interference-contrast microscopy create contrast by modifying the pupil function or interfering selected field components. These methods rely on stops, phase plates, birefringent prisms and accurately aligned conjugate planes, which complicate miniaturization and integration.

Metasurfaces provide a planar platform for engineering amplitude, phase and polarization and have been used for spatial differentiation, edge enhancement and related analog image operations. Most metasurface image processors implement each prescribed transfer function in a separate optical channel. Multiplexed designs may assign different responses to polarization, wavelength or angle, but the constituent channels are generally designed and read out independently. Their coherent interaction has rarely been used to generate a transfer function beyond those directly encoded in the individual channels. The present work makes that coherent interaction part of the transfer-function design.

A voltage-controlled liquid-crystal retarder selects X-, Y- or 45°-linear polarization at a polarization-multiplexed metalens. The X- and Y-polarized channels share a nominal focal plane but have different effective apertures and therefore different cutoff spatial frequencies. X- or Y-polarized input accesses the corresponding bright-field response. With 45°-polarized input and a 45° analyzer, the two complex fields are projected onto the same polarization basis. Within the spatial-frequency range transmitted by both channels, a designed relative phase difference of π makes their field contributions interfere destructively. Frequencies transmitted only by the larger-aperture X-polarized channel are retained, so the resulting transfer function enhances object boundaries. Thus, two encoded polarization channels provide three voltage-selectable imaging responses without moving the meta-device or applying digital post-processing.

The research group of Prof. Junsuk Rho from Pohang University of Science and Technology (POSTECH), in collaboration with the groups of Prof. Yuchuan Shao from the Shanghai Institute of Optics and Fine Mechanics (SIOM), Chinese Academy of Sciences, and Prof. Trevon Badloe from Korea University, introduces a liquid-crystal-integrated polarization-multiplexed metalens that enables two bright-field imaging states and one interference-synthesized edge-enhanced state. The metalens consists of rectangular hydrogenated amorphous silicon (a-Si:H) nanopillars on a silica (SiO2) substrate and has a nominal focal length of 800 μm. Its two orthogonal polarization channels are designed to be cofocal but have effective aperture diameters of Dx = 570 μm and Dy = 120 μm, corresponding to numerical apertures of approximately 0.336 and 0.075. The unequal apertures give the two channels different spatial-frequency cutoffs. Within their overlapping aperture, the meta-atom design imposes a relative phase difference of π between the two channels to establish the required interference condition.

With X- or Y-polarized input, the corresponding channel forms a conventional bright-field image. Under 45°-polarized illumination followed by a 45° analyzer, the two transmitted fields are projected onto a common polarization basis and interfere coherently. Low-frequency components transmitted by both channels acquire approximately opposite phases and are suppressed, while higher spatial frequencies supported by the larger X-polarized aperture remain. This coherent field difference produces the edge-enhancing transfer-function response. The third response thus arises from interference between the two encoded channels rather than from a separately encoded optical channel. The voltage-controlled liquid-crystal retarder selects the input polarization at the metalens, allowing three calibrated voltages to address the three imaging states without moving the optical element.

Experimental characterization at 635 nm produced focal-spot full widths at half maximum of 1.2 μm, 4.8 μm and 1.1 μm for the X-polarized, Y-polarized and 45°-polarized states, respectively. The measured focusing efficiencies were 43%, 36% and 24%. The lower efficiency in the 45° state arises partly from the intended cancellation of shared low-frequency content. Imaging of a USAF 1951 target confirmed the distinct resolution ranges of the two bright-field channels and the paired-edge contours generated by the synthesized state. A benchmark meta-device with the same focal length and aperture geometry but without the designed phase relationship retained the two bright-field responses yet failed to reproduce the 45°-polarized edge-enhanced response, confirming that the designed relative phase relationship is necessary for the interference-synthesized response.

The team further imaged an earthworm cross-section, onion epidermal cells and a cross-section of a young Ligustrum stem. The X-polarized state recorded conventional bright-field views, whereas the 45° state accentuated cellular and tissue boundaries directly in the optical domain, without digital post-processing. The concept therefore offers a route to compact, electrically selectable, label-free contrast enhancement. The proof-of-concept device is optimized at 635 nm. In particular, the edge-enhanced state requires sufficiently high spatial coherence, a sufficiently narrow spectral bandwidth and adequate matching of the two channel fields. Wavelength shifts, fabrication errors, defocus and off-axis incidence can reduce cancellation quality. Future improvements may combine higher-transmission meta-atoms, lower-loss liquid-crystal and polarization components, and dispersion engineering to broaden the usable wavelength range and improve efficiency.

This work was financially supported by the POSCO-POSTECH-RIST Convergence Research Center program funded by POSCO, and the National Research Foundation (NRF) grants (RS-2026-25507992, RS-2025-17492968, RS-202502317602, RS-2025-02217649, RS-2024-00462912, RS-2024-00356928) funded by the Ministry of Science and ICT (MSIT) of the Korean government. This research was also supported by a grant of Korean ARPA-H Project through the Korea Health Industry Development Institute (KHIDI), funded by the Ministry of Health & Welfare, Republic of Korea (Grant number: RS-2025-25454431). T.B. acknowledges the NRF grant (RS-2026-25598896) funded by the MSIT of the Korean government, and the ANCHOR program through the Sejong ANCHOR Center, funded by the Ministry of Education (MOE) and the Sejong City of the Korean government. (2026-ANCHOR-08-001).

Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type
Submit

While we only use edited and approved content for Azthena answers, it may on occasions provide incorrect responses. Please confirm any data provided with the related suppliers or authors. We do not provide medical advice, if you search for medical information you must always consult a medical professional before acting on any information provided.

Your questions, but not your email details will be shared with OpenAI and retained for 30 days in accordance with their privacy principles.

Please do not ask questions that use sensitive or confidential information.

Read the full Terms & Conditions.