Pith. sign in

REVIEW 3 major objections 5 minor 10 references

Bright and Purcell-enhanced single photon emission from a silicon G center

T0 review · 3 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read This paper demonstrates that a single silicon G center coupled to a nanobeam photonic-crystal cavity emits zero-phonon photons with a 0.97 ns lifetime, a six-fold speed-up over uncoupled G centers, with a lower-bound Purcell factor of 31…

desk verdict First Purcell-enhanced lifetime reduction for a single silicon G center, but the brightness numbers contain an arithmetic error and the single-emitter decomposition needs stronger support. read the letter →

arxiv 2412.10603 v2 pith:BXI2OZAM submitted 2024-12-13 physics.optics quant-ph

Kyu-Young Kim , Chang-Min Lee , Amirehsan Boreiri , Purbita Purkayastha , Fariba Islam , Samuel Harper , Je-Hyung Kim , Edo Waks This is my paper · ORCID
classification physics.opticsquant-ph
keywords siliconGcenternanobeamcavityPurcelleffectsinglephotonsourcezero-phononlinetelecomO-bandtime-resolvedphotoluminescenceDebye-Wallerfactor
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

Silicon G centers emit in the telecom O-band but waste most of their light in a phonon sideband and suffer from non-radiative decay, so they have been dim single-photon sources. This paper claims that coupling one G center to a nanobeam photonic-crystal cavity suppresses that waste by accelerating the zero-phonon-line emission through the Purcell effect. The measured excited-state lifetime drops from 5.97 ns outside the cavity to 0.97 ns on resonance, a six-fold speed-up, which the authors report as the fastest single-photon emission rate seen in silicon. From the lifetime change and the known 15% Debye-Waller factor they derive a lower-bound Purcell factor of 31 and a source brightness of 77%, and they measure 14% fiber-coupled brightness. If correct, the result turns a historically inefficient defect into a practical emitter for silicon photonic quantum circuits.

What carries the argument

The load-bearing mechanism is the Purcell effect acting on the zero-phonon line of a single G center. The G center is the silicon defect made of two substitutional carbon atoms plus an interstitial silicon atom, emitting at 1277 nm; its emission is mostly lost into a phonon sideband (Debye-Waller factor 0.15). The central object is a one-dimensional silicon nanobeam photonic-crystal cavity with a linear taper defect, quality factor 4600 and mode volume $0.26(\lambda/n)^3$, tuned to the G center zero-phonon line, with asymmetric mirror hole counts so that the cavity field is emitted into an adiabatic taper mode-matched to a lensed fiber. The argument's quantitative engine is the pair of decay-rate equations $\gamma_{\rm on}=F_P\gamma_{\rm ZPL}+\gamma_{\rm off}$ and $\gamma_{\rm ZPL}=F_{\rm DW}\varepsilon_{\rm QE}/\tau_0$, which convert the three measured lifetimes into $F_P$ and into the source-brightness estimate $1-\tau_{\rm on}/\tau_{\rm off}$.

What would settle it

One decisive check would be to measure the decay curve while spectrally selecting only the narrow zero-phonon line and sweeping the cavity detuning; if the 0.97 ns component does not follow the cavity resonance curve and vanish when the cavity is detuned, then the claimed six-fold enhancement is not from one cavity-coupled G center.

Watch

Extended reading notes

Core claim

On the paper's own terms, the central discovery is that the G center's zero-phonon line can be made the dominant decay channel by a nanobeam cavity. Time-resolved photoluminescence on resonance shows a bi-exponential decay with a fast component of $\tau_{\rm on}=0.97\pm0.01$ ns, compared with $\tau_0=5.97\pm0.01$ ns for G centers in the nanobeam outside the cavity and $\tau_{\rm off}=4.27$ ns from a Lorentzian fit of the lifetime versus detuning. Using $F_P = \tau_0(1/\tau_{\rm on}-1/\tau_{\rm off})/(F_{\rm DW}\varepsilon_{\rm QE})$ with Debye-Waller factor $F_{\rm DW}=0.15$ and quantum efficiency $\varepsilon_{\rm QE}\le1$, the authors obtain a lower bound $F_P\ge31$, and they estimate source brightness $1-\tau_{\rm on}/\tau_{\rm off}=0.77$. The paper also verifies single-photon emission with $g^{(2)}(0)=0.408$ and reports a corrected single-photon count rate of 476 kcps, which after a photon-budget analysis corresponds to 14% fiber-coupled brightness.

Load-bearing premise

The results rest on the assumption that the measured fast 0.97 ns decay comes from the same single G center whose off-resonance lifetime is 4.27 ns, and that no other emitters or spectral jumps mix into those two numbers.

Editorial extensions

If this is right

  • A 0.97 ns radiative lifetime means the source can in principle be excited at repetition rates above 1 GHz without strong multiphoton contamination, a prerequisite for high-rate quantum key distribution.
  • The order-of-magnitude brightness gain over earlier G-center sources (476 kcps corrected count rate versus about 10 kcps) moves silicon G centers from proof-of-concept to practical on-chip sources.
  • Because the cavity design is adapted from a T-center nanobeam, the same fabrication and tuning recipe should transfer to other telecom silicon color centers.
  • Detuning the cavity by about 1 nm changes the decay time from 0.97 ns to 3.92 ns, confirming that the decay-rate change is cavity-mediated and that the coupling can be switched by tuning.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the emitter's position and dipole orientation were optimized toward the simulated maximum Purcell factor of 1340, the lifetime could approach the cavity-limited regime; the authors identify position and dipole mismatch as the likely reason the observed factor is only 31.
  • A direct experimental cross-check would be to extract the Purcell factor from the power-saturation enhancement and compare it with the lifetime-derived value; agreement would confirm that the fast decay component belongs to the same single G center.
  • The same cavity technique should apply to other silicon color centers with low Debye-Waller factors: engineering the cavity to enhance only the zero-phonon line effectively raises the system's Debye-Waller factor, a strategy that could extend to W and T centers.
  • Because the photon-budget brightness estimate depends on a 50% nanobeam-to-fiber coupling that is assumed rather than directly measured, a direct outcoupling-efficiency measurement would be the cleanest way to confirm the 14% fiber-coupled brightness.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 5 minor

Summary. The manuscript reports a cavity-coupled single G center in a silicon nanobeam photonic crystal. Time-resolved photoluminescence shows a fast decay component of 0.97 ns on resonance, which the authors interpret as Purcell-enhanced emission from a single G center, yielding a six-fold lifetime reduction relative to 5.97 ns outside the cavity and a lower-bound Purcell factor Fp ≥ 31 after accounting for phonon-sideband decay. Second-order autocorrelation gives g(2)(0) = 0.408 at 1 μW, confirming single-photon emission. From saturation measurements and a photon budget, the authors report 4.76% end-to-end, 14% fiber-coupled, and 38% source brightness, which they claim is an order-of-magnitude improvement over previous silicon G-center sources. The central claims rest on a bi-exponential decomposition of the on-resonance decay and on a Lorentzian fit of the detuning-dependent lifetime.

Significance. If the interpretation is correct, this would be an important step for silicon-based quantum photonics: a telecom O-band single-photon emitter with sub-nanosecond lifetime and a Purcell factor exceeding 30 would be competitive with other cavity-coupled color centers. The paper provides a clear derivation of the Purcell-factor estimator, a careful detuning-dependent lifetime measurement, and an explicit photon-budget analysis, which are good practices. However, the quantitative claims in the abstract are not yet supported: the corrected single-photon count rate is arithmetically inconsistent, and the assignment of the 0.97 ns component to a single Purcell-enhanced emitter is not validated against background and uncoupled contributions. The large discrepancy between the lifetime-derived source brightness (0.77) and the photon-budget-derived source brightness (0.38) indicates an internal inconsistency that must be resolved.

major comments (3)
  1. [Figure 4(c) and Supporting Information Section 4] The corrected single-photon count rate is arithmetically inconsistent. The authors state I_single = I_sat × (1 − g(2)(0)) and report I_sat = 669 kcps and g(2)(0) = 0.408, which yields I_single = 669 × 0.592 ≈ 396 kcps, not 476 kcps. This error propagates: the end-to-end brightness becomes 3.96% (not 4.76%), the fiber-coupled brightness becomes approximately 11.6% (not 14%), and the source brightness becomes approximately 31% (not 38%). Because the abstract and conclusion advertise the 14% fiber-coupled brightness and the order-of-magnitude improvement, these numbers must be corrected and the analysis repeated.
  2. [Figure 3 and Supporting Information Section 3] The Purcell factor and source brightness estimates rely on the assignment of the 0.97 ns component of the bi-exponential on-resonance decay to the same single G center that has off-resonance lifetime τ_off = 4.27 ns. This assignment is not adequately justified: g(2)(0) = 0.408 at 1 μW indicates a substantial background, and the slow 3.75 ns component is attributed to uncoupled G centers, so the fast component could in principle include emission from background or partially coupled emitters. The authors should demonstrate, for example by spectrally filtering the emission, by measuring the power dependence of the fast component, or by showing that the fast component disappears at large detuning, that the 0.97 ns decay belongs to a single emitter and not to a mixture.
  3. [Source brightness comparison, main text after Fig. 4(c) and SI Section 4] The lifetime-derived source brightness of 0.77 is inconsistent with the photon-budget source brightness of 0.38 by a factor of two. The authors attribute this to errors in coupling efficiencies and detector polarization, but the discrepancy is too large to be dismissed without a quantitative error analysis. Since the source brightness is a central advertised result, the authors should either reconcile the two estimates with a full uncertainty budget or soften the brightness claims.
minor comments (5)
  1. [Main text, Fig. 3(a)] The phrase 'we obtain a spontaneous emission rate of 0.97 ns' should read 'lifetime of 0.97 ns' (the rate is the inverse).
  2. [Supporting Information Eq. (4)] The Lorentzian lineshape factor in the expression for γ_cav(ω) is not fully defined; please specify the prefactor and clarify whether Δω is the half-width or the full width at half maximum.
  3. [Figure 2(b)] Please report the fitting uncertainty of the 0.091 nm ZPL linewidth and state whether the measurement is limited by the spectrometer resolution or by the emitter linewidth.
  4. [Abstract and conclusion] The claim of the 'fastest single photon emission rate reported in silicon' should be qualified by the comparison set (e.g., among G centers or among silicon color centers), since the cited W-center and T-center experiments may involve different collection and excitation conditions.
  5. [Supporting Information Section 2] The statement 'we assumed that the coupling-in and coupling-out efficiencies are the same' should be justified, as this assumption directly enters the 75% lensed-fiber coupling efficiency used in the photon budget.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the Purcell factor and source brightness are computed from directly measured lifetimes and literature parameters, not from the claims themselves.

full rationale

The central claim—a six-fold lifetime enhancement and Purcell factor lower bound Fp≥31—is derived from measured time-resolved decay components (τ_on=0.97 ns, τ_off=4.27 ns, τ_0=5.97 ns) and the literature Debye-Waller factor of 15%, via Supporting Information Eq. (7), which is a standard rate-equation decomposition. The source brightness 1−τ_on/τ_off is likewise a direct algebraic consequence of the same measured lifetimes (SI Eq. 9), not a quantity fitted to itself. The g(2)(0) and pump-power measurements provide independent, external confirmatory evidence for single-emitter behavior and brightness. The only self-citations are to the authors' earlier cavity design and fabrication recipe (Refs. 8 and 9), used for device construction rather than as evidence for the Purcell enhancement; those are published prior experimental results, and the present Purcell-factor derivation does not depend on them. Potential uncertainties about the biexponential decomposition, background contamination, or Debye-Waller factor value are experimental correctness concerns, not circularity. Therefore no load-bearing step reduces to its own inputs, and the paper is self-contained with respect to its main quantitative claims.

Assumptions & free parameters 6 free parameters · 4 assumptions · 0 invented entities

The Purcell factor and brightness claims rest on measured lifetimes, a literature Debye-Waller factor, a single-emitter assignment, and estimated coupling efficiencies. No new physical entities are postulated, and the fitted quantities are all experimental parameters rather than ad hoc model constants.

free parameters (6)
  • tau_on = 0.97 ns
    Fast component of the bi-exponential fit of on-resonance time-resolved photoluminescence; directly enters the Purcell factor and brightness calculations.
  • tau_off = 4.27 ns
    Off-resonance lifetime obtained from a Lorentzian fit over detunings; used to subtract non-ZPL decay in the Purcell factor and source brightness formulas.
  • tau_0 = 5.97 ns
    Reference lifetime of G centers outside the cavity, from a single exponential fit; serves as the unperturbed decay time.
  • I_sat = 669 kcps
    Saturation count rate from a two-level fit to pump-power dependence; combined with g2(0) to obtain the corrected single-photon count rate.
  • eta_wg = 50%
    Assumed cavity-waveguide coupling efficiency used in the photon budget to derive the source brightness.
  • eta_fib_coup = 75%
    Measured lensed-fiber coupling efficiency from off-resonance reflectivity; used in the photon budget.
assumptions (4)
  • domain assumption G center excited-state decay separates into zero-phonon line, phonon sideband, and non-radiative channels, and the cavity enhances only the zero-phonon rate.
    Supporting Information Section 3, Eqs. (1)-(4). This is the central model behind the Purcell factor formula.
  • domain assumption The Debye-Waller factor of a G center without a cavity is 15%.
    Taken from Refs 3, 16, and 20; used as an external constant in the Purcell factor denominator, with no propagated uncertainty.
  • domain assumption The fast 0.97 ns decay component is emitted by a single G center, while the slow component comes from uncoupled G centers.
    Main text Figure 3(a) and Supporting Information Section 3. This assignment is required to interpret the bi-exponential fit and compute brightness.
  • domain assumption The relation I_single = I_sat * (1 - g2(0)) correctly separates single-photon counts from background.
    Main text after Figure 4(c), citing Refs 25 and 27. This formula underlies the reported corrected count rate of 476 kcps.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Bright and Purcell-enhanced single photon emission from a silicon G center." pith.science (2026). https://pith.science/paper/BXI2OZAM

@misc{pith2026241210603,
  author       = {Pith},
  title        = {Pith review of: Bright and Purcell-enhanced single photon emission from a silicon G center},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/BXI2OZAM}},
  note         = {Machine review of arXiv:2412.10603}
}
read the original abstract

Silicon G centers show significant promise as single photon sources in a scalable silicon platform. But these color centers have large non-radiative decay and a low Debye-Waller factor, limiting their usability in quantum applications. In this work, we demonstrate bright Purcell-enhanced emission from a silicon G center by coupling it to a nanophotonic cavity. The nanobeam cavity enhances the spontaneous emission rate of a single G center by a factor of 6, corresponding to a Purcell factor greater than 31 when accounting for decay into the phonon sideband. We obtain a spontaneous emission rate of 0.97 ns, which is the fastest single photon emission rate reported in silicon. With this radiative enhancement, we achieve an order of magnitude improvement in emitter brightness compared to previously reported values. These results pave the way for scalable quantum light sources on a silicon photonic chip.

Figures

Figures reproduced from arXiv: 2412.10603 by the authors.

Figure 1
Figure 1. (a) Atomic structure of a silicon G center. (b) A schematic image of a nanobeam cavity array and a lensed fiber. (c) Scanning electron microscope image of the cavity part of the fabricated nanobeam cavity. Scale bar is 1 µm [PITH_FULL_IMAGE:figures/full_fig_p004_1.png] view at source ↗
Figure 2
Figure 2. (a) Measured reflectivity spectrum of a nanobeam cavity. The green circles are measured [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. (a) Time-resolved photoluminescence using above-band pulsed excitation with a repetition rate of 10 MHz. Black (blue) circles are the measured data at detuning of 𝛿 = 0.02 nm (𝛿 = 1.03 nm). Green circles are obtained from a G center located in the nanobeam outside the cavity region. Solid curves correspond to exponential decay function fit. (b) Reduced lifetime (black dots) and the estimated Purcell factor FP (red d… view at source ↗
Figures from the paper (1 more)
Figure 4
Figure 4. Figure 4: (a,b) Second order correlation measurement of the zero-phonon line using a pulsed excitation with a repetition rate of 20 MHz, at average power of (a) 1.0 μW and (b) 10 μW. (c) Pump power dependence of detected count rates at 𝛿 = 0.02 nm (black circles) and 𝛿 = 1.03 nm…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

10 extracted references · 5 canonical work pages

  1. [3]

    We performed the same measurement at a detuning of 𝛿=1.03 nm and achieved 𝐼'*33=25.4 kcps

    to potential error in the nanobeam-to-fiber coupling efficiency, polarization of the single photon input of the single photon detectors, and transmission of fiber components. We performed the same measurement at a detuning of 𝛿=1.03 nm and achieved 𝐼'*33=25.4 kcps. This value is 19 times smaller than the on-resonance photon count rate, indicating that the...

    2024

  2. [5]

    We then measured the photoluminescence spectrum of the nanobeam waveguide from 1260 to 1500 nm wavelength range using the same measurement setup (Figure S4(a))

    G center ensemble photoluminescence To investigate general properties of the defect centers, we fabricated nanobeam waveguide structures using the same implanted silicon wafer described in the main manuscript. We then measured the photoluminescence spectrum of the nanobeam waveguide from 1260 to 1500 nm wavelength range using the same measurement setup (F...

    work page 2022

  3. [10]

    Genuine and Faux Single G Centers in Carbon-Implanted Silicon

    (3) Durand, A.; Baron, Y.; Cache, F.; Herzig, T.; Khoury, M.; Pezzagna, S.; Meijer, J.; Hartmann, J.-M.; Reboh, S.; Abbarchi, M.; Robert-Philip, I.; Gérard, J.-M.; Jacques, V.; Cassabois, G.; Dréau, A. Genuine and Faux Single G Centers in Carbon-Implanted Silicon. Phys. Rev. B. 2024, 110 (2). https://doi.org/10.1103/physrevb.110.l020102

    doi:10.1103/physrevb.110.l020102 2024

  4. [2023]

    (9) Islam, F.; Lee, C.-M.; Harper, S.; Rahaman, M

    https://doi.org/10.1021/acsphotonics.3c01142. (9) Islam, F.; Lee, C.-M.; Harper, S.; Rahaman, M. H.; Zhao, Y.; Vij, N. K.; Waks, E. Cavity-Enhanced Emission from a Silicon T Center. Nano Lett. 2024, 24 (1), 319–325. (10) Johnston, A.; Felix-Rendon, U.; Wong, Y.-E.; Chen, S. Cavity-Coupled Telecom Atomic Source in Silicon. Nat. Commun. 2024, 15 (1),

    doi:10.1021/acsphotonics.3c01142 2024

  5. [2024]

    (12) Redjem, W.; Durand, A.; Herzig, T.; Benali, A.; Pezzagna, S.; Meijer, J.; Kuznetsov, A

    http://arxiv.org/abs/2406.01704. (12) Redjem, W.; Durand, A.; Herzig, T.; Benali, A.; Pezzagna, S.; Meijer, J.; Kuznetsov, A. Y.; Nguyen, H. S.; Cueff, S.; Gérard, J.-M.; Robert-Philip, I.; Gil, B.; Caliste, D.; Pochet, P.; Abbarchi, M.; Jacques, V.; Dréau, A.; Cassabois, G. Single Artificial Atoms in Silicon Emitting at Telecom Wavelengths. Nature Electr...

    arXiv 2020

  6. [2380]

    Cavity-Enhanced Single Artificial Atoms in Silicon

    (16) Saggio, V.; Errando-Herranz, C.; Gyger, S.; Panuski, C.; Prabhu, M.; De Santis, L.; Christen, I.; Ornelas-Huerta, D.; Raniwala, H.; Gerlach, C.; Colangelo, M.; Englund, D. Cavity-Enhanced Single Artificial Atoms in Silicon. Nat. Commun. 2024, 15 (1),

    work page 2024

  7. [3321]

    Individually Addressable and Spectrally Programmable Artificial Atoms in Silicon Photonics

    (15) Prabhu, M.; Errando-Herranz, C.; De Santis, L.; Christen, I.; Chen, C.; Gerlach, C.; Englund, D. Individually Addressable and Spectrally Programmable Artificial Atoms in Silicon Photonics. Nat. Commun. 2023, 14 (1),

    work page 2023

  8. [5296]

    Indistinguishable Photons from an Artificial Atom in Silicon Photonics

    (17) Komza, L.; Samutpraphoot, P.; Odeh, M.; Tang, Y.-L.; Mathew, M.; Chang, J.; Song, H.; Kim, M.-K.; Xiong, Y.; Hautier, G.; Sipahigil, A. Indistinguishable Photons from an Artificial Atom in Silicon Photonics. Nat. Commun. 2024, 15 (1),

    work page 2024

Show all 10 references
  1. [6920]

    Single G Centers in Silicon Fabricated by Co-Implantation with Carbon and Proton

    (18) Baron, Y.; Durand, A.; Herzig, T.; Khoury, M.; Pezzagna, S.; Meijer, J.; Robert-Philip, I.; Abbarchi, M.; Hartmann, J.-M.; Reboh, S.; Gérard, J.-M.; Jacques, V.; Cassabois, G.; Dréau, A. Single G Centers in Silicon Fabricated by Co-Implantation with Carbon and Proton. App...

    2022

  2. [7683]

    Y.; Dhuey, S.; Schwartzberg, A.; Tan, L

    (14) Redjem, W.; Zhiyenbayev, Y.; Qarony, W.; Ivanov, V.; Papapanos, C.; Liu, W.; Jhuria, K.; Al Balushi, Z. Y.; Dhuey, S.; Schwartzberg, A.; Tan, L. Z.; Schenkel, T.; Kanté, B. All-Silicon Quantum Light Source by Embedding an Atomic Emissive Center in a Nanophotonic Cavity. N...

    2023

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.