pith. sign in

arxiv: 2605.15481 · v1 · pith:CSQN27DCnew · submitted 2026-05-14 · ⚛️ physics.app-ph

High-Efficiency InGaP-on-Insulator Microresonator Nonlinear Conversion and Entanglement Generation

Xuefeng Li , Lillian Thiel , Yiming Pang , Amalu Shimamura , Lucas Wang , Joshua Castro , Max Meunier , Nicholas Lewis
show 4 more authors
John Bowers Kevin Silverman Richard Mirin Galan Moody
This is my paper

Pith reviewed 2026-05-19 14:21 UTC · model grok-4.3

classification ⚛️ physics.app-ph
keywords InGaP-on-insulatormicroresonatorsecond harmonic generationphoton pair generationquantum entanglementnonlinear opticsintegrated photonicssurface treatment
0
0 comments X

The pith

Surface treatment reduces loss in InGaP-on-insulator microresonators to deliver 3.01×10^5 %/W second-harmonic generation and 11.7 MHz/μW photon-pair rates.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

High propagation loss at visible wavelengths has restricted the usefulness of InGaP-on-insulator for integrated frequency conversion and on-chip entanglement sources. Mode-profile analysis identifies surface effects as the dominant loss source. A targeted surface treatment reduces that loss by a factor of 3.5 to 4, producing propagation losses of 0.49 dB/cm at 1560 nm and 4.31 dB/cm at 780 nm. Statistical mapping of quality factor versus radius selects the ring geometry that keeps nonlinear interaction strong while holding bending loss in check. The resulting devices reach the stated efficiencies and pair rates while confirming quasi-phase matching across the full parameter space.

Core claim

Mode-profile analysis identifies the dominant loss mechanism in InGaP-on-insulator waveguides at visible wavelengths. A surface treatment mitigates this loss without degrading the intrinsic nonlinearity, achieving propagation losses as low as 0.49 dB/cm at 1560 nm and 4.31 dB/cm at 780 nm. Optimal ring radius is selected via quality factor statistics to maintain strong nonlinear interaction while suppressing bending loss. This enables second-harmonic generation with efficiency 3.01×10^5 %/W and photon-pair generation at 11.7 MHz/μW with CAR up to 10,000, with experimental verification of quasi-phase matching across parameters.

What carries the argument

The surface treatment process that targets and reduces the dominant surface-induced propagation loss identified through mode-profile analysis in the InGaP-on-insulator microresonators

If this is right

  • Establishes InGaP-on-insulator as a scalable platform for both classical frequency conversion and on-chip quantum entanglement generation.
  • Provides a concrete method to achieve low-loss waveguides in the visible range while preserving high chi^(2) nonlinearity.
  • Allows systematic characterization of nonlinear processes over the full parameter space in optimized resonators.
  • Supports wafer-scale integration for photonic circuits requiring high-efficiency nonlinear interactions.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Similar surface treatments could be adapted to other high-nonlinearity materials facing surface loss issues in integrated photonics.
  • Lower loss may enable longer coherence lengths or more complex circuits involving multiple resonators for advanced quantum protocols.
  • Future work could test whether the same treatment improves performance in other nonlinear processes such as four-wave mixing or electro-optic modulation.

Load-bearing premise

The surface treatment specifically mitigates the identified dominant loss mechanism without introducing new losses or reducing the material's intrinsic optical nonlinearity.

What would settle it

Fabricate and test additional InGaP-on-insulator resonators with and without the surface treatment to check if the reported loss reduction, quality factors, SHG efficiency, and photon pair rates are consistently reproduced.

Figures

Figures reproduced from arXiv: 2605.15481 by Amalu Shimamura, Galan Moody, John Bowers, Joshua Castro, Kevin Silverman, Lillian Thiel, Lucas Wang, Max Meunier, Nicholas Lewis, Richard Mirin, Xuefeng Li, Yiming Pang.

Figure 1
Figure 1. Figure 1: (a) Depiction of the e"ect of S treatment on surface roughness and surface states. For the same resonator configuration, the device with S treatment exhibits critical coupling, whereas the untreated device is significantly undercoupled and shows much higher propagation loss. (b) FDE simulated e"ective index of the 1560 nm TE00 mode and the 780 nm TM00 and TM01 modes. (c) Simulated mode profiles of the 1560… view at source ↗
Figure 2
Figure 2. Figure 2: Normalized XPS spectra of (a) Ga 2𝑈, (b) In 3𝑄, and (c) P 2𝑈 under di"erent surface treatments. AFM images of samples (d) as-grown, (e) OH-treated, and (f) S-treated. photon energies approach defect levels [27]. The oxidation of InGaP surfaces in air occurs almost instantaneously, and e"ective surface passivation requires not only dielectric deposition but also prevention of native oxide formation. 3. Low-… view at source ↗
Figure 3
Figure 3. Figure 3: InGaP-on-insulator nanofabrication process flow. (a) As-grown InGaP epitaxial. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Normalized transmission spectrum of a 20 𝑀m-radius all-pass ring resonators in (a) IR and (b) VIS range. Resonance fitting for (c) IR TE00, (d) VIS TM00, and (e) VIS TM01 modes, corresponding to the dashed box in [PITH_FULL_IMAGE:figures/full_fig_p006_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: (a) Schematic of the entangled-photon pair generation measurement setup. [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: (a) Normalized SHG intensity under di"erent thermal current biases. The SHG spectrum redshifts with increasing thermal current. (b) On-chip SHG power as a function of pump power, confirming a quadratic dependence. (c) Conversion e!ciency as a function of on-chip pump power. ↓ 3.5–4→ reduction in propagation loss and intrinsic quality factors approaching ↓ 200k in the visible wavelength range. Statistical a… view at source ↗
read the original abstract

InGaP-on-insulator, with its intrinsically high $\chi^{(2)}$ optical nonlinearity, has emerged as an efficient and bright integrated photonic platform for frequency conversion and on-chip entanglement generation, but high waveguide propagation loss in the visible wavelength range has limited its overall performance. Here, we identify the dominant loss mechanism through mode-profile analysis and effectively mitigate the loss using a surface treatment method. Statistical analysis of the resonator quality factor and propagation loss reveals the optimal ring radius that maintains a strong nonlinear interaction while suppressing significant bending related loss, resulting in loss as low as 0.49 dB/cm (4.31 dB/cm) at 1560 nm (780 nm). The method provides a 3.5--4$\times$ linear performance enhancement, enabling a second-harmonic generation efficiency of $3.01\times10^{5}$ %/W and a photon-pair generation rate of $11.7,\mathrm{MHz}/\mu\mathrm{W}$ and coincidence-to-accidental ratio as high as 10,000. The quasi-phase matching condition is experimentally verified, and nonlinear conversion is systematically characterized across the entire parameter space. This work establishes a scalable pathway for classical and quantum photonics in a low-loss, highly nonlinear, and wafer-scale integration platform.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The manuscript describes an InGaP-on-insulator microresonator platform in which a surface treatment is used to mitigate the dominant propagation loss identified via mode-profile analysis. Statistical analysis of quality factor versus ring radius identifies an optimal geometry that balances nonlinear interaction strength against bending loss, yielding propagation losses of 0.49 dB/cm at 1560 nm and 4.31 dB/cm at 780 nm. This enables a second-harmonic generation efficiency of 3.01×10^5 %/W, a photon-pair generation rate of 11.7 MHz/μW, and coincidence-to-accidental ratios up to 10,000, with experimental verification of quasi-phase matching across parameter space.

Significance. If the reported performance metrics and the causal attribution of the 3.5–4× enhancement to the surface treatment are substantiated, the work would represent a meaningful advance for integrated nonlinear and quantum photonics. The platform combines high intrinsic χ(2), wafer-scale compatibility, and low visible-wavelength loss, directly addressing a key limitation for bright on-chip entanglement sources. The inclusion of statistical resonator characterization and systematic nonlinear mapping strengthens the practical utility of the results.

major comments (2)
  1. [Results on surface treatment and loss characterization] The central claim that the surface treatment provides a 3.5–4× linear performance enhancement and directly enables the quoted SHG efficiency (3.01×10^5 %/W) and pair rate (11.7 MHz/μW) rests on the assumption that the treatment specifically mitigates the identified dominant loss without introducing compensating losses or altering intrinsic nonlinearity. The manuscript does not appear to describe a controlled split-wafer comparison (treated versus reference devices fabricated and measured under matched conditions) that would isolate the treatment effect from run-to-run variations in etch depth, sidewall roughness, or material quality. This weakens the causal link between the reported loss values and the surface-treatment method.
  2. [Mode-profile analysis and loss mitigation] The mode-profile analysis that identifies the dominant loss mechanism should be accompanied by quantitative before-and-after loss spectra or Q-factor distributions on the same devices (or matched pairs) to confirm that the treatment reduces the targeted loss channel without degrading the χ(2) response or adding scattering. Without such data, it remains possible that the observed improvements arise from other process variations.
minor comments (2)
  1. [Figure 3 or equivalent statistical analysis figure] Figure captions for the resonator Q versus radius plots should explicitly state the number of devices measured per radius and the fitting procedure used to extract propagation loss from the statistical data.
  2. [Abstract and Section 4] The abstract and main text use slightly inconsistent notation for the pair-generation rate units (MHz/μW versus MHz/μW); a single consistent format should be adopted throughout.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive and detailed review. The comments correctly identify the need for stronger substantiation of the surface treatment's specific contribution. We address each major comment below and indicate where revisions will be made to the manuscript.

read point-by-point responses

  1. Referee: [Results on surface treatment and loss characterization] The central claim that the surface treatment provides a 3.5–4× linear performance enhancement and directly enables the quoted SHG efficiency (3.01×10^5 %/W) and pair rate (11.7 MHz/μW) rests on the assumption that the treatment specifically mitigates the identified dominant loss without introducing compensating losses or altering intrinsic nonlinearity. The manuscript does not appear to describe a controlled split-wafer comparison (treated versus reference devices fabricated and measured under matched conditions) that would isolate the treatment effect from run-to-run variations in etch depth, sidewall roughness, or material quality. This weakens the causal link between the reported loss values and the surface-treatment method.

    Authors: We agree that an explicit split-wafer comparison would provide the strongest possible isolation of the treatment effect. The current manuscript relies on mode-profile analysis to identify the dominant surface-related loss channel, followed by statistical Q-factor and loss measurements across many treated resonators of varying radii, which show consistent improvement and an optimal geometry. These data are compared against prior InGaP reports in the literature. To address the concern, we will revise the manuscript to include a dedicated discussion of fabrication process controls, run-to-run statistics, and any available reference-sample data from the same process lot. We maintain that the combination of targeted loss identification and systematic parameter-space characterization supports the reported enhancement, but we acknowledge the value of the suggested control experiment. revision: yes

  2. Referee: [Mode-profile analysis and loss mitigation] The mode-profile analysis that identifies the dominant loss mechanism should be accompanied by quantitative before-and-after loss spectra or Q-factor distributions on the same devices (or matched pairs) to confirm that the treatment reduces the targeted loss channel without degrading the χ(2) response or adding scattering. Without such data, it remains possible that the observed improvements arise from other process variations.

    Authors: The mode-profile simulations in the manuscript pinpoint surface scattering as the dominant mechanism at both wavelengths, and the surface treatment is applied uniformly after etching. Post-treatment loss values and nonlinear performance are reported with statistical distributions over multiple devices. We will expand the revised manuscript with additional quantitative details on the measurement protocol and any pre- versus post-treatment comparisons that can be extracted from the existing dataset or matched reference runs. We note that repeated measurement on identical devices before and after treatment is experimentally constrained by the need for non-destructive characterization, but the systematic radius-dependent trends help rule out unrelated process variations. revision: partial

Circularity Check

0 steps flagged

No significant circularity; experimental measurements are self-contained

full rationale

This is an experimental fabrication and characterization paper. Reported values (SHG efficiency 3.01×10^5 %/W, pair rate 11.7 MHz/μW, losses 0.49 dB/cm and 4.31 dB/cm, Q-factor statistics) are directly measured from devices rather than obtained via any derivation chain. Mode-profile analysis and statistical radius optimization are data-driven comparisons, not equations that reduce to fitted inputs by construction. No self-definitional steps, fitted predictions, or load-bearing self-citations appear in the abstract or described methods. The central performance claims rest on empirical before/after or comparative measurements, which remain falsifiable outside any internal fit.

Axiom & Free-Parameter Ledger

0 free parameters · 1 axioms · 0 invented entities

Experimental optimization study relying on standard nonlinear-optics assumptions; no new free parameters or invented entities are introduced in the abstract.

axioms (1)
  • domain assumption Quasi-phase matching condition can be satisfied and verified in the fabricated ring resonators
    Stated as experimentally verified in the abstract.

pith-pipeline@v0.9.0 · 5798 in / 1377 out tokens · 53140 ms · 2026-05-19T14:21:27.407956+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. A Wafer-Scale Heterogeneous III-V-on-Silicon Nitride Quantum Photonic Platform

    physics.optics 2026-05 unverdicted novelty 6.0

    Demonstration of wafer-scale III-V-on-SiN integration with <25 mdB coupler loss, >10^6 Q resonators, 15x brighter entanglement sources, and high-efficiency detectors for quantum photonics.

Reference graph

Works this paper leans on

2 extracted references · 2 canonical work pages · cited by 1 Pith paper

  1. [1]

    Enhanced second-harmonic generation in algaas microring resonators,

    Z. Y ang, P . Chak, A. D. Bristow,et al., “Enhanced second-harmonic generation in algaas microring resonators,” Opt. letters 32, 826–828 (2007)

    work page 2007

  2. [2]

    On-and o!-resonance second-harmonic generation in gaas microdisks,

    P . S. Kuo and G. S. Solomon, “On-and o!-resonance second-harmonic generation in gaas microdisks,” Opt. express 19, 16898–16918 (2011). Table 1. ω𝑅 and ω𝑂 for di!erent combinations of IR and VIS resonances. 𝑅VIS[nm] /𝑂VIS 𝑅IR[nm] /𝑂IR 1515.6(-1) 1522.13(-2) 1528.75(-3) 1535.44(-4) 1542.2(-5) 1549.03(-6) 1555.95(-7) 1562.94(-8) 1570.02(-9) 1577.16(-10) 158...

    work page 2011