Nonlinear optics
Can a metal film make nano-sized frequency doublers efficient?
Open access · cc by · source: Europe PMC
Putting a thin semiconductor waveguide just above a silver film squeezes infrared light into a tiny region and makes frequency doubling far stronger than in the same structure on glass.
Study at a glance
- Design
- Other — AlGaInP waveguides and disks released onto Al2O3/Ag or glass, excited with tunable or broadband infrared pulses; far-field SHG/SFG imaging and spectra compared with COMSOL simulations
- N
- No sample size; several waveguide widths and 1 μm disks were measured
- Population
- Lithographically defined AlGaInP nanostructures on silver versus glass substrates
- Outcome
- SHG/SFG conversion efficiency, far-field visualization of phase matching, enhancement over photonic counterparts
Structured fields used in claim comparison tables when every cited study has a complete layer.
Key findings
Light leaking from higher-order modes let them watch the doubled signal build up or fade along the waveguide, directly showing where phase matching held; at one wavelength they estimated a coherence length of about 12 μm. Structures on silver produced over 1500 times more signal than the same waveguides on glass, which gave essentially nothing. In 1 μm disks, broadband infrared light gave a sum-frequency efficiency of 14.8% per megawatt, about five times that of single-wavelength doubling, and the doubled power scaled with input power to a measured exponent of 2.1.
Methodology
The authors made AlGaInP semiconductor waveguides and disks about 110 nm thick and placed them on a silver film with a thin alumina layer, or on glass for comparison. They coupled telecom-band infrared light in and imaged the red light produced by second-harmonic generation (doubling) and sum-frequency generation (adding two different frequencies) from above. They compared measured patterns and efficiencies with mode simulations.
Limitations
The efficiencies are high per unit volume but absolute output powers were only nanowatts or picowatts, and propagation losses in the plasmonic modes limited amplification to a few micrometres. Some phase-matched modes could not be seen in the far-field setup, and a quoted higher efficiency for a shorter waveguide is an extrapolation rather than a measurement. Only one semiconductor-metal combination was tested.
How this study connects
Role on claims
Each row is a claim on a concept or method page where this paper supports, challenges, or qualifies the statement. Roles are hand-checked — not a model guess.
Squeezing light near metal boosts conversion; the quadratic power law confirms SHG.
Placing a semiconductor waveguide above silver gave over 1500 times more SHG than the same structure on glass, and leaked light let the authors watch phase-matched signal build along the guide (coherence length about 12 um); output power scaled with a measured exponent of 2.1.
Evidence for the claim as stated.
Efficiencies are reported in incompatible units (normalized %/W/cm^2 for waveguides, %/MW for plasmonic disks, absolute fraction for bulk crystals), and the high normalized values correspond to tiny absolute outputs (nanowatts to picowatts in the plasmonic case), so they cannot be ranked directly.
Evidence for the claim as stated.
Open questions
Tensions this paper is part of
From concept pages' “where studies disagree.” Disagreement means the same question; scope means different assays, populations, or outcomes.
Efficiencies are reported in incompatible units (normalized %/W/cm^2 for waveguides, %/MW for plasmonic disks, absolute fraction for bulk crystals), and the high normalized values correspond to tiny absolute outputs (nanowatts to picowatts in the plasmonic case), so they cannot be ranked directly.
Related papers in this topic
Same topic cluster — not a recommendation engine.
- Can we take 3D pictures with single mid-infrared photons?
- Can a crystal grow its own 3D frequency-doubling structure?
- Can a lossy spot perfectly swallow interacting matter waves?
- Can a voltage make a non-chiral crystal respond to light's handedness?
- Can InGaP chips convert light colours far more efficiently?