Concept · physics
Metalenses and active metasurfaces
7 studiesEvidence last moved Sep 27, 2026
Metasurfaces reshape a wavefront by giving each sub-wavelength element a designed phase (and sometimes polarization) response, letting a flat layer act as a lens, beam steerer or modulator. This page covers focusing efficiency and chromatic correction in metalenses, inverse (topology) design, and metasurfaces made tunable with graphene or liquid crystals.
Flat optics are often advertised as replacements for glass lenses; the evidence shows where they already perform well and where bandwidth, efficiency and off-axis imaging remain open. It also clarifies that 'tunable' results often rest on extrapolated speeds or tiny pixel counts.
Studies
7
Findings
5
7 supporting · 0 challenging · 1 qualifying citations
Open tensions
1
Latest change
Concept page published
Metalenses and active metasurfaces
Currently
What we know
- Achromatic flat lenses work, but bandwidth trades against numerical aperture.
- Computer inverse design can keep metalenses efficient at steep angles.
- Metasurfaces can shape polarization, not just phase, but real metals lose efficiency.
- Graphene turns a fixed metasurface into a fast modulator.
- Liquid crystal plus anisotropic nanostructures gives tunable colour pixels.
Largest unresolved question
The topology-optimised lenses reach high efficiency but only at one wavelength, while the achromatic dispersion-engineered lenses cover a broad band but with lower efficiency than the best single-wavelength metalenses; no study here achieves both.
Common misconceptions
Metalenses have already matched refractive lenses for imaging.
The achromatic lens was tested only for on-axis focusing of a small singlet in the near infrared; off-axis aberrations and real imaging were not shown.
The graphene metasurface was measured switching at 7.2 GHz.
The measurement setup stopped at 1 GHz; 7.2 GHz is inferred from a circuit model, and the 23 kHz imaging frame rate is extrapolated from a 36-pixel demo.
Related
Claim ledger
What the evidence shows
Drawn from 7 studies in this library. Mix labels say which citation roles are present; they are not a strength score. Supports means evidence for a finding; Challenges means evidence against a stated position; Qualifies marks scope.
Achromatic flat lenses work, but bandwidth trades against numerical aperture.
Designing nanopillars to control both phase and its wavelength dependence produced near-infrared metalenses that stayed focused from 1200 to 1650 nm with focal length shifting only 2-5% and Strehl ratios above 0.8; a higher-NA (about 0.88) lens was achromatic only over 1200-1400 nm.
Computer inverse design can keep metalenses efficient at steep angles.
Topology optimisation by independent sections made millimetre-scale lens design feasible on a PC in under a day; fabricated cylindrical lenses reached above 89% relative and 67% absolute efficiency with little drop at high NA.
- Can we design big, efficient metasurfaces without supercomputers?— Single-wavelength, one-dimensional lenses; not achromatic.
Metasurfaces can shape polarization, not just phase, but real metals lose efficiency.
Giving each gold antenna a tailored polarization response let a reflective metasurface steer light with 85% absolute efficiency at 1550 nm and generate beams with position-dependent polarization and a plasmon vortex (coupling about 34% versus 61.4% simulated).
- Can one flat metasurface shape both a beam's wavefront and polarization?
- Can one flat lens read both a light beam's spin and its twist?
Study Role Design N Population Outcome Can one flat metasurface shape both a beam's wavefront and polarization? Supports OtherTheory plus fabrication and optical measurement of three metal-insulator-metal metasurface devices, compared with Green's-function calculations and numerical simulations. Not applicable: three fabricated meta-devices (plus characterisation samples of individual meta-atoms) are measured; no sample-size statistic. Gold/SiO2/gold metal-insulator-metal metasurfaces working at near-infrared wavelengths (1550 nm and 1064 nm) Reflection angle and efficiency, polarization conversion ratio, orbital angular momentum (spiral interference patterns), local polarization maps and surface plasmon coupling efficiency and propagation length Can one flat lens read both a light beam's spin and its twist? Supports OtherTheory plus lab experiment: azimuthal-quadratic phase metasurfaces fabricated by e-beam lithography and atomic layer deposition were illuminated with vortex beams at visible wavelengths, and focal patterns compared with angular-spectrum simulations. No sample count; two metasurface designs (geometric-phase and spin-decoupled) tested with vortex beams of many topological charges. TiO2 nanopillar metasurfaces illuminated by visible vortex and vector beams (480–633 nm) Azimuthal position and screen half of the focal spot, used to identify orbital and spin angular momentum Graphene turns a fixed metasurface into a fast modulator.
Adding gate-tuned graphene to antenna metasurfaces makes them electrically active: a few volts shifted a mid-infrared resonance from about 7.3 to 8.3 um with about 90% modulation depth and no decay up to the 1 GHz measurement limit.
- Can graphene switch mid-infrared light fast with low voltage?
- Can graphene's conductivity make infrared sensors detect tiny molecules?
Study Role Design N Population Outcome Can graphene switch mid-infrared light fast with low voltage? Supports OtherDevice experiment: gold nanoantenna metamaterial absorber with graphene on a thin Al2O3 gate over amorphous silicon; reflection spectra, near-field imaging, modulation speed and a 6 × 6 pixel spatial light modulator were measured, with a circuit model to infer intrinsic speed. No sample count; results are from fabricated devices, including one 6 × 6 pixel array. Hybrid graphene-gold nanoantenna metasurface modulators operating at mid-infrared wavelengths of roughly 7.3 to 8.5 μm Reflection change and modulation depth versus gate voltage and wavelength, modulation bandwidth, and single-pixel imaging performance Can graphene's conductivity make infrared sensors detect tiny molecules? Supports OtherGraphene-covered gold nanorod metasurfaces on a SiO2/Pt cavity exposed to dopant molecules, aluminium nanoparticles and glucose; mid-IR resonance shifts measured and compared with FDTD, circuit and perturbation models across antenna shapes and gap sizes. Multiple fabricated devices of several geometries; no device count reported. Hybrid graphene-metal metasurface sensors Shift of the mid-IR plasmonic resonance frequency, resonance quality factor, and glucose detection limit Liquid crystal plus anisotropic nanostructures gives tunable colour pixels.
Liquid crystals over polarization-sensitive silicon Mie resonators tuned reflective colour pixels almost linearly from bright colour to black, with contrast ratios of 42, 37 and 47 for red, green and blue.
Debates
Tensions and limits
Some items are genuine disagreements on the same question. Others mark different assays, populations, or outcomes.
The topology-optimised lenses reach high efficiency but only at one wavelength, while the achromatic dispersion-engineered lenses cover a broad band but with lower efficiency than the best single-wavelength metalenses; no study here achieves both.
The topology-optimised lenses reach high efficiency but only at one wavelength, while the achromatic dispersion-engineered lenses cover a broad band but with lower efficiency than the best single-wavelength metalenses; no study here achieves both.
- Can a flat lens focus all colours to the same spot?
- Can we design big, efficient metasurfaces without supercomputers?
Study Role Design N Population Outcome Can a flat lens focus all colours to the same spot? Supports OtherTheory of achromatic focusing limits plus design, fabrication and optical characterisation of several amorphous-silicon metalenses built from three meta-unit libraries Four fabricated metalenses were characterised; no statistical sample. Amorphous silicon nanopillar metalenses on quartz operating in the near infrared Focal length versus wavelength, focal spot size versus diffraction limit, Strehl ratio, focusing efficiency Can we design big, efficient metasurfaces without supercomputers? Supports Computational / modellingAdjoint topology optimisation of wavelength-scale sections stitched into cylindrical metalenses, benchmarked in simulation and then fabricated in crystalline silicon and measured optically. Design/validation study; simulated lenses across a range of numerical apertures plus three fabricated lenses — no sample N. Crystalline-silicon nanoridge metalenses designed for 640 nm light Computation time scaling, relative and absolute focusing efficiency versus numerical aperture, and focal spot size
PaperFren reads this as a limit on how far one study travels — different assays, populations, or outcomes — not a forced fight between papers.
Timeline
How understanding moved
Study years are when the paper was published. Evidence edits are dated changes to this page's claims. Explanations are when PaperFren added a Discovery — not a claim that the science happened that day.
2026
Concept page published
Metalenses and active metasurfaces
Change log
What changed
Dated edits to this page's evidence: studies added or removed from a claim, claims added or withdrawn, and new explanations tagged here. Rewordings are not listed.
- Concept page published
Papers
7 studies in this library bear on Metalenses and active metasurfaces, ordered by citations.
- Can a flat lens focus all colours to the same spot?
By designing nanopillars that control both the phase and how phase changes with wavelength, the authors built flat lenses that focus a broad band of infrared light to nearly the same point.
- Can one flat lens read both a light beam's spin and its twist?
A single patterned metasurface turns each twisted light beam into a focal spot whose angle reveals its orbital angular momentum and whose half of the screen reveals its spin.
- Can we design big, efficient metasurfaces without supercomputers?
Breaking a lens design into small straight-line sections, optimising each by computer and stitching them together gives large metasurfaces that focus light efficiently even at steep angles.
- Can graphene switch mid-infrared light fast with low voltage?
Adding gate-tunable graphene to a nanoantenna metasurface lets a few volts switch mid-infrared reflection by up to 90% at gigahertz speeds, fast enough for high-frame-rate single-pixel imaging.
- Can nanostructures give switchable colors with true blacks?
Elongated silicon nanostructures reflect a bright color for one light polarization and almost nothing for the perpendicular one, so a liquid-crystal layer that rotates polarization can dial each pixel smoothly from vivid color to deep black.
- Can graphene's conductivity make infrared sensors detect tiny molecules?
Putting graphene over metal nano-antennas lets molecules that dope the graphene shift the infrared resonance strongly, detecting glucose down to picomolar levels.
- Can one flat metasurface shape both a beam's wavefront and polarization?
By giving each tiny antenna a fully tailored polarization response, a flat metasurface can produce light beams with designed wavefronts and position-dependent polarization, including on-surface plasmon vortices.
Compare studies
Select 2–10 studies. Design and N are labels, not a ranking.
Nothing selected yet.
Questions
What is still open
The topology-optimised lenses reach high efficiency but only at one wavelength, while the achromatic dispersion-engineered lenses cover a broad band but with lower efficiency than the best single-wavelength metalenses; no study here achieves both.
Ask PaperFren about Metalenses and active metasurfaces
Study this conceptflashcards and short-answer questions
What design trade-off limits broadband achromatic metalenses, according to the 2018 dielectric metalens study?
To stay achromatic, each pillar must supply not only the right phase but the right change of phase with wavelength, and the range of achievable dispersion is limited. The low-NA lens corrected colour from 1200 to 1650 nm, but a high-NA lens (about 0.88) was achromatic only from 1200 to 1400 nm. Efficiencies were also below the best single-wavelength lenses. So bandwidth, aperture and efficiency trade against one another.
Describe how graphene can make a metasurface tunable and how the resulting speed claim should be read.
Gating graphene changes its carrier density and hence its optical conductivity, which shifts the antenna resonance; in the mid-infrared device this moved the resonance by about 1 um and gave around 90% modulation depth. Modulation showed no decay up to 1 GHz, the instrument limit. The 7.2 GHz cut-off is a model-based inference, not a direct measurement, so the demonstrated speed is 'at least 1 GHz'.