Low Reflectance All-Glass Metasurface Lenses Based on Laser Self-generated Nanoparticles
Listen
Radio episode about this paper
Transcript
Introduction to the show: ident: Physics. The day in general physics: instruments, methods and results across the field.
Kai: We're looking at the paper: "Low Reflectance All-Glass Metasurface Lenses Based on Laser Self-generated Nanoparticles".
Title and authors: Kai: The title itself tells us a lot about what they're tackling here: "Low Reflectance All-Glass Metasurface Lenses Based on Laser Self-generated Nanoparticles." It basically sets up the whole story for this work.
Mira: And the authors are Jae Hyuck Yoo, Nathan J. Ray, Mike A. Johnson, Hoang T. Nguyen and Eyal Feigenbaum from Lawrence Livermore National Laboratory. They're bringing together different strengths here with their expertise in materials and optics >
Lev: From my side, I'm thinking about the practical reality of this—they are using a laser to create nanoparticles that act as a mask for etching into fused silica glass, which is solid material. That involves dealing with things like deposition and then reactive ion etching, or RIE >
Kai: And what they’re claiming is that because of this self-organizing process, you can produce large amounts of nano-elements all at once, which suggests a path for aperture scaleup in optics >
The paper's summary: Mira: So the main takeaway from the abstract is that instead of making every single meta-element individually, they use this self-organizing working principle to create many nano-elements simultaneously >
Kai: That means they can generate sub one hundred nanometer elements, which opens up a route to operating at shorter wavelengths, not just longer ones >
Lev: But the abstract also mentions two key advancements they've made toward making this viable for real hardware. One is using a laser scan with in-situ transmission feedback to pattern the etching mask to get the right nanoparticle distribution >
Mira: And another thing is they crafted a beyond-mask-erosion-point etching process, which helps increase the phase difference across the metasurface to at least pi while keeping that reflection extremely low >
The paper's improvements: Kai: The paper highlights these two key advancements we just mentioned. That in-situ feedback for patterning and the enhanced etching process for controlling the phase delay are what they say are going to make this actually work >
Lev: From an engineering standpoint, when you talk about that laser scan with feedback, you're trying to overcome fabrication nonlinearities where the process doesn't behave predictably >
Mira: And they also address the issue of etching time; they found that longer etching times, like seven hundred fifty seconds or one thousand seconds, actually increase the phase delay across the structure beyond one quarter wave for those longer times >
Conclusion: Kai: So to wrap up this paper on "Low Reflectance All-Glass Metasurface Lenses Based on Laser Self-generated Nanoparticles," we see a method that makes large aperture optics possible because it scales the element production and allows for very low reflection at five hundred thirty-two nanometer wavelength > <ref:2603.08673#pg1,Low Reflectance All-Glass Metasurface Lenses Based on Laser>
Mira: It really points toward making high power laser optics more practical by addressing the size, throughput, and durability issues that limited previous metasurface designs >
Lev: From a hardware perspective, the main thing is that they are showing a path to creating durable structures that can handle the necessary phase control for high-power applications >
Kai: And for us working on superconducting qubits, this reminds me of how we need to get highly uniform and scalable fabrication methods if we're going to move beyond small test pieces >
Mira: Exactly. The paper shows a way to handle the scaling issue by using the laser irradiation itself to control the mask fill-factor, which is a critical link between temperature and shape >
Lev: It’s promising, but as they state in the limitations, their previous demonstration only showed rudimenta, meaning they haven't fully optimized it for complex patterns yet >
Lawrence Livermore National Laboratory
physics.optics
Submitted: 2026-03-09
Updated: 2026-03-09
Journal ref: Adv. Optical Mater., 14: e71897 (2026)
DOI: 10.1002/adom.71779
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Importance score: 81/100
The gist: Optical metasurfaces comprised of subwavelength nanostructures hold a great promise to high-power laser optics but also a limited pertinence due to their currently limited aperture size…
Key concepts
- LSG-NPM process
- This is a four-step method for creating the metasurface: depositing a thin metal layer on glass, using laser heat to form self-organized nanoparticles, etching these particles into the glass using reactive ion etching (RIE), and finally washing away the sacrificial metal mask.
- Local Induced Phase Delay (PD)
- This is how much a metasurface changes the phase of incoming light. It depends on the depth of the metasurface layer and its effective index. The mask fill-factor, which is how much area the nanoparticles cover, significantly controls this phase delay.
- In-situ Feedback Loop
- This involves using optical transmission measurements during a laser scan to monitor the evolving nanoparticle distribution in real-time. This feedback allows for iterative adjustments to the laser scanning power pattern, helping the system converge on a desired nanoparticle shape.
Terminology
Summary
Optical metasurfaces comprised of subwavelength nanostructures hold a great promise to high-power laser optics but also a limited pertinence due to their currently limited aperture size, throughput and durability. p. 0 Here, an alternative approach is presented, reliant on laser-controlled self-organizing mask formation followed by ion etching which results in an allfused-silica-glass metasurface. p. 0 Two 1 mm diameter optical elements (an axicon lens and a shadower) are fabricated and their optical performance is validated at 532 nm wavelength with an extremely low broadband reflection (<0. p. 015%) - a result of the unique metasurface elements shape. p. 0 p. 0 The self-organizing working principle enables producing large amounts of nano-elements at-once, thus a path for aperture scaleup. p. 0 It also enables generation of sub-100 nm nanoelements, thus a path to short wavelengths operation. p. 0 Two key advancements towards viability are presented: a laser scan with in-situ transmission feedback enables patterning the etching mask to a prescribed nanoparticle distribution, and a crafted beyond-mask-erosion-point etching of the mask enables increasing the metasurface phase difference to at least , while keeping extremely low reflection across it. p. 0 This paves a path to high-power lasers optics, requiring large aperture, high throughput and laser light durability. p. 0
Methodology for Mask Generation
The LSG-NPM process consists of a 4-step procedure: depositing of an ultra-thin metal layer on the fused silica glass, applying heat with laser to dewet the thin film and generate self-organized formation of metal nanoparticles, and use thereafter the nanoparticle as directional etching mask to imprint it into the fused silica glass (typically, reactive ion etching (RIE)), where the sacrificial metal mask is finally washed away. The local induced phase delay (PD) by the metasurface on the incoming wave depends on the depth of the MS layer and the effective index of that layer. p. 2 One of the determining factors of the PD is the mask fill-factor, FF, (i. p. 2e. masking factor, mean area covered by the mask nanoparticles), and we have previously shown that this FF is controlled by the local temperature induced by the laser irradiation. p. 2
Iterative Patterning and Feedback
To obtain a nanoparticle mask with a spatial distribution having an FF that tracks a goal prescription requires a method that could overcome the fabrication nonlinearities and nonidealities. An iterative approach was devised that converges to a desired prescribed shape by using optical transmission measurement as in-situ feedback. p. 4 The principal process involves a radial laser scan with the laser beam power being a function of only the radius, adequate for the azimuthally symmetrical optics implemented in this study. p. 4 The beam power is computer-controlled using a rotation of a half-wave plate (HWP) sandwiched between two polarizers, and the beam scan is controlled by two axis galvanometers. p. 4
Etching Control and Phase Delay
The RIE process length affects the end-result metasurface effective index, where in the long etch case, at some of the segments the nanoparticles mask is completely eroded and disappear during the RIE process. For longer etches, 750 s and 1000 s, the long etching conditions are observed where the PD across the structure increases for longer etching time, exceeding ¼ wave for the 1000 s conditions (i. p. 9e., PD > /2). The end-result MS nano-elements are much longer and tapered shaped, where the length of the features is more dominant in determining the local PD than the FF (opposite to the short etch paradigm). p. 9
Experimental Validation of Optical Performance
The optical performance was validated at 532 nm wavelength with an extremely low broadband reflection (<0.15%) - a result of the unique metasurface elements shape. The MS axicon optical function as a lens was demonstrated at 532 nm wavelength using both beam propagation test and simulation. p. 0 The reflectance of the two MS elements has been measured – showing extremely low and spectrally broadband reflectance. p. 0 The spectral reflection of the MS was measured at three locations: the two MS elements and a reference area free of optical elements (using Cary 7000 Universal Spectrophotometer), as presented in Fig. p. 3 9a. The measurements show that the reflectance in all three measured locations is substantially lower compared to native fused silica surface. p. 9
Conclusion and Future Directions
The LSG-NPM approach has several potential advantages for the applications of high energy and power lasers (HEPL), in terms of aperture scalability, thick substrates compatibility, meta-elements small enough to fit even short wavelengths in the ultraviolet, low reflectance, and high laser damage and mechanical durability. The converging nature of the process described here is intended to accommodate the nonidealities of the manufacturing process. p. 9 Further exploration could involve dynamically varying the beam size and raster speed to produce large optics masks with practical manufacturing time. p. 3 The potential utility for other optical elements, e. p. 0g. printing optical grating with prescribed profile and low reflection, is illustrated by an unoptimized demonstration of a diffractive optics-type shaped NP mask. p. 0
Acknowledgements
Funding was provided by Lawrence Livermore National Laboratory (LLNL) Laboratory Directed Research and Development grant. This work was performed under the auspices of the U. p. 3S. Department of Energy by Lawrence Livermore National Laboratory under Contract DE-AC52-07NA27344. p. 0
Green — the paper says it, and the page is printed. Red — no passage carries it, so treat that sentence as unverified. The same check runs on your own text: check a summary against its source, $5.
Improvements for AI systems
-
Improved AI can design nanoparticle mask power profiles for prescribed transmission using an iterative converging mask patterning setup based on in-situ feedback and transmission measurement as a proxy for mask fill-factor. This allows the AI to correct laser scanning power patterns to converge to a desired nanoparticle mask shape, overcoming fabrication nonlinearities.
-
Improved AI can optimize reactive ion etching time based on the resulting phase delay profile to achieve an adequate phase delay range, specifically by studying how
the competition of the contributing conditions results in a non-monotonic behavior
and tailoring themask FF range that for these etching conditions result in the monotonic change range observed.
-
Improved AI can predict and compensate for etching nonlinearities by developing a method to better predict
the connection between the mask transmission and the resulting MS PD due to the etching nonlinearity,
supplementing the ability to implement a prescribed transmission etch mask. -
Improved AI can design fabrication recipes that account for mask erosion effects by utilizing knowledge of
the etch transfer function at the intended etch time into the pattern look-up-table (LUT)
to accommodate nonidealities of manufacturing process and ensure monotonic control over laser patterning.
Related papers
- Raman optical activity and helicity-dependent phonon splitting in chiral CoSi
- Stability and optical quality of "windmill"-formed 8CB liquid crystal films for replenishable plasma mirrors
- Experimental Scaling of Diffraction Efficiency in Laser-Induced Plasma Gratings
- Photonic Exceptional Topological Bound States in the Continuum
- SHAPE: a sparse Hamiltonian approach for prescribing eigenstates in metamaterials
- All-photonic millimeter-wave and terahertz signal core at room and cryogenic temperatures