Can InGaP chips convert light colours far more efficiently?
Thin InGaP waveguides double the frequency of telecom light about a hundred times more efficiently than lithium niobate chips and produce very bright entangled photon pairs.
Source
InGaP χ<sup>(2)</sup> integrated photonics platform for broadband, ultra-efficient nonlinear conversion and entangled photon generation
Study at a glance
- Design
- Other — Fabricated thin-film InGaP microrings and a 1.6 mm meander waveguide; measured loss, second-harmonic generation, SPDC pair rates, cross-correlation and two-photon interference
- N
- Device measurements on microring resonators and one main 1.6 mm waveguide; no sample count
- Population
- 110 nm thick InGaP photonic integrated circuits with oxide cladding
- Outcome
- Normalized SHG efficiency, optical loss, photon-pair generation rate and bandwidth, entanglement visibility
Structured fields used in claim comparison tables when every cited study has a complete layer.
What they did
The authors grew 110 nm thick indium gallium phosphide films and patterned them into microring resonators and a folded (meander) waveguide 1.6 mm long, using an aluminium oxide coating to reduce loss. Because InGaP has no birefringence, they phase-matched 1550 nm and 775 nm light by choosing the waveguide width. They measured second-harmonic generation, then pumped the waveguide near 780 nm to make photon pairs by spontaneous parametric down-conversion and tested their entanglement with an unbalanced interferometer.
What they found
The waveguide reached a normalized second-harmonic efficiency of 128,000%/W/cm², close to the simulated 130,000%/W/cm² and nearly two orders of magnitude above thin-film lithium niobate, with a loss of about 0.8 dB/cm. The photon-pair source produced 97 GHz/mW over a 115 nm bandwidth. Two-photon interference visibility was 90.8% raw (about 98.6% after correcting for the interferometer), above the 70.7% classical limit, confirming entanglement.
The limits
What it doesn't show
Longer waveguides did not follow the expected length-squared scaling because the film thickness varies, so the advantage has only been shown over short lengths. Coupling to fibres was lossy, especially for 775 nm light, and detector jitter meant the true coincidence-to-accidental ratio could not be resolved. The results come from a few devices, and applications such as squeezed light or integrated pump lasers are proposed rather than demonstrated.
Key terms
- Second-harmonic generation
- A chi(2) process in which two photons at one frequency combine into one photon at twice the frequency.
- Phase matching
- The condition that interacting waves stay in step along the device so their nonlinear contributions add up constructively.
- Spontaneous parametric down-conversion
- The reverse of SHG: one pump photon splits into a correlated pair of lower-frequency photons.
- Quality factor
- A measure of how long light stays in a resonator; higher intrinsic Q means lower loss.
- Time-energy entanglement
- A quantum correlation in which a photon pair's emission time and energies are linked so that pair interference appears even when single photons show none.
Flashcards
0 of 10 answers reviewed
Research intelligence for this paper
See its role on concept claims, tensions it is part of, placement history, and related discoveries.
Quiz yourself
Why is InGaP suited to SHG of telecom light?
Common questions
Why use InGaP instead of GaAs?
InGaP has a wider bandgap, so it does not absorb the 775 nm second harmonic of telecom light, while keeping a large chi(2) nonlinearity.
How did they phase-match without birefringence?
They tuned the waveguide width so the effective index of the 1550 nm TE mode equalled that of the 775 nm TM mode.
How do they know the photons are entangled?
The two-photon interference visibility exceeded the 70.7% Clauser-Horne limit that classical light cannot beat.
More on Nonlinear optics