Recycling Reflections for Perfect Photon Capture

Liang Jiang; Yat Wong

REVIEW 2 major objections 8 minor 10 references

Recycling Reflections for Perfect Photon Capture

T0 review · 2 major / 8 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read This paper claims that recycling a cavity's first reflection through a second, delayed pass can capture single photons with unit efficiency in the noiseless regime, and it gives explicit coupling thresholds for doing so.

desk verdict Two-pass reflection recycling is a genuinely new idea with solid core calculations; the universal-pulse threshold needs a real proof before the headlines are trusted. read the letter →

arxiv 2506.01127 v2 pith:34UY4VPZ submitted 2025-06-01 quant-ph

Yat Wong , Liang Jiang This is my paper

classification quant-ph PACS 42.50.Pq03.67.Hk

keywords photoncapturequantumstatetransfercavityQEDtwo-passprotocolreflectionrecyclingmemorypulseshapingLangevinequation

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

Single-pass photon capture in a cavity always leaves part of the pulse uncaptured when the coupling strength and capture window are finite, so perfect capture of an arbitrary pulse is impossible that way. This paper claims a two-pass 'pitch-and-catch' protocol can recycle the initial reflection by redirecting it back into the cavity after a controllable delay, and by tuning both couplings in real time can reach capture efficiency 1 in the noiseless regime. For any bounded positive pulse the stated requirement is a peak coupling $\kappa_{\max}\ge 2\ln 2\cdot\max_t b_{\mathrm{in}}(t)^2$; for exponential-decay pulses the infimum is $\kappa_{\max}/\gamma > k\approx 1.2834$, where $k$ solves $(k+1)^{k+1}=4k^2$. With intrinsic loss $\kappa_i$, the efficiency becomes $\kappa_i^{\kappa_i/(1-\kappa_i)}$, eliminating the single-pass reflection-loss factor $O(\gamma/\kappa_{\max})$. This matters because it removes a known obstruction to high-fidelity quantum state transfer, memory, and transduction using finite hardware resources.

What carries the argument

The load-bearing object is the two-pass Langevin equation $\frac{d}{dt}a=-\frac{\kappa_1+\kappa_2}{2}a+\sqrt{\kappa_1}b_{\mathrm{in}}+\sqrt{\kappa_2}b'_{\mathrm{in}}$ with $b'_{\mathrm{in}}(t)=b_{\mathrm{out}}(t-\Delta t)$, i.e. the first-pass reflection redirected losslessly into an orthogonal second port after a controllable delay. The control law that carries the argument is $\kappa(t)=\min(\kappa_{\max},(b_{\mathrm{in}}(t)/a(t))^2)$: when the instantaneous reflected amplitude can be cancelled it is cancelled exactly, and otherwise the coupling is held at its maximum. The analysis is organized by the assertion that the square packet maximizes initial reflection and is therefore the hardest bounded packet to capture, which reduces the universal threshold $\kappa_{\max}\ge 2\ln 2$ to a calculable square-pulse case; for exponential decay the analogous calculation fixes the constant $k\approx1.2834$ from $(k+1)^{k+1}=4k^2$ and fixes the required delay through Eqs. (26) and (38).

What would settle it

Run the two-pass control law on smooth bounded pulses such as Gaussian or hyperbolic-secant shapes with $\kappa_{\max}=2\ln 2\cdot\max_t b_{\mathrm{in}}^2$ and the paper's prescribed delays; if any pulse leaves a nonzero residual reflection at the second port, the square-packet extremal lemma is false. Experimentally, a weak coherent pulse through a two-coupler cavity with a low-loss delay line should show reflected power vanishing at the predicted threshold; a reflection floor above shot noise would indicate a missing loss channel.

Watch

Extended reading notes

Core claim

The paper's central claim is that a two-pass receiver with independently tunable couplings $\kappa_1(t)$ and $\kappa_2(t)$ and a delayed, losslessly redirected reflection can capture a single photon with unit efficiency in the ideal noiseless regime, both for finite-time bounded pulses and for exponential-decay pulses in the asymptotic limit. The explicit thresholds are $\kappa_{\max}\ge 2\ln 2 \cdot \max_t b_{\mathrm{in}}^2$ for bounded pulses when both ports have the same peak coupling, and $\kappa_{\max}/\gamma>k\approx1.2834$ for exponential-decay pulses, with $k$ the root of $(k+1)^{k+1}=4k^2$. The second-pass reflection is cancelled exactly by choosing the delay $\Delta t$ so that $\sqrt{\kappa_{2,\max}}\,a(\Delta t)=b'_{\mathrm{in}}(\Delta t)$, and the same control law reversed lets the cavity emit an arbitrarily shaped pulse. Under intrinsic loss $\kappa_i$, the paper derives a capture efficiency $\kappa_i^{\kappa_i/(1-\kappa_i)}$ (Eq. 42), which removes the leading $O(\gamma/\kappa_{\max})$ degradation of single-pass capture, at the cost of a slightly shorter optimal capture time.

Load-bearing premise

The argument assumes both that the square packet is genuinely the hardest bounded pulse to capture (a lemma stated without proof) and that the first-pass reflection can be redirected losslessly into a second port with a controllable delay and independently tunable couplings; if either fails, the perfect-capture claims no longer follow.

Editorial extensions

If this is right

Any bounded positive pulse can be stored with unit efficiency in the noiseless regime once the peak coupling satisfies $\kappa_{\max}\ge 2\ln 2\cdot\max_t b_{\mathrm{in}}(t)^2$; a single-pass receiver under the same constraints leaves a residual loss $e^{-\kappa_{\max}t_{\max}}$.
Spontaneous-emission (exponential-decay) pulses, which cannot be perfectly captured in a single pass even with infinite time, are captured perfectly when $\kappa_{\max}/\gamma>k\approx1.2834$, with the delay set by Eq. (26).
With intrinsic loss $\kappa_i$, two-pass capture reaches efficiency $\kappa_i^{\kappa_i/(1-\kappa_i)}$, so the leading single-pass reflection factor $O(\gamma/\kappa_{\max})$ is gone; the paper's example shows the improvement factor scales as $\kappa_{1,\max}^{-1}$.
Reversing the schedule releases an arbitrarily shaped pulse, giving on-demand pulse shaping from the same cavity.
For quantum state transfer, the scheme reduces both residual and reflection loss, with direct relevance to quantum repeaters, memories, and transduction.

Reading between the lines

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

Inference: the universality of the $2\ln 2$ threshold is only as strong as the unproved square-packet lemma; a numerical sweep over Gaussian, sech, and asymmetric bounded pulses at exactly $\kappa_{\max}=2\ln 2\cdot\max b_{\mathrm{in}}^2$ would test whether any shape needs a larger peak coupling.
Inference: because the dynamics are linear in the single-excitation sector, the same equations describe weak coherent states, so the predicted unit capture should be measurable as vanishing reflected power in a classical microwave or optical experiment without single-photon counting.
Inference: in the high-coupling limit the required delay shrinks as $\Delta t\approx2\ln 2/\kappa_{\max}$, so the hardware overhead is modest; a real delay line with small loss should behave like an effective intrinsic $\kappa_i$, degrading efficiency only through the ratio $\kappa_i/\gamma$.
Inference: the two-pass recycling idea is not obviously limited to cavities; any linear scatterer with a second controllable interaction port and a delay line could recycle its reflection to approach unit capture, potentially applying to atomic ensembles or waveguide-QED devices in the linear regime.

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.

Desk Editor's Note

read the letter

Two-pass 'pitch-and-catch' is a genuinely new protocol: instead of absorbing the single-pass reflection loss, you redirect the reflected wavepacket through an orthogonal port after a controlled delay and tune a second coupling to impedance-match it into the cavity. The core calculations for the concrete cases—square pulse and exponential decay—check out. The Sec. III variational derivation of the single-pass limit is clean, and I verified the square-pulse first-pass expressions and the 2 ln 2 threshold for the symmetric-coupling case. Removing the O(gamma/kappa_max) single-pass penalty and replacing it with an intrinsic-loss-limited efficiency is a real advance, and the k ~1.2834 threshold for exponential pulses is new to me.

The main soft spot: the universality claim that any bounded positive packet is perfectly captured for kappa_max >= 2 ln 2 rests entirely on the asserted lemma that the square packet maximizes initial reflection. That lemma is not proved. I share the stress-test note's worry that for a shaped pulse the binding quantity is sup_t (b'_in/a), and that Eq. (19)'s comparison to a^2(t=1) is computed only for the square pulse. I don't have a counterexample—a late subpulse likely arrives after a has grown, so the early square reflection may indeed be the worst case—but the paper needs to prove it or back it with numerics, not assert it. If the universal claim is weaker, the specific square and exponential results still stand.

Secondary issues: Sec. VI (intrinsic loss) omits enough algebra that I couldn't independently verify Eqs. (35)-(42); the asymptotic statements are plausible but not checkable from the text. The abstract says 'arbitrary photonic states' while the model is explicitly single-excitation; that overstates the scope. The figures have no code or data, but for a theory paper that's minor.

Bottom line: the central idea is solid and worth a serious referee. Send it out, and ask for (1) a proof or numerical evidence for the square-packet worst-case lemma and the universal threshold, (2) fuller derivation in Sec. VI. If those land, this is a useful contribution to the quantum memory and repeater toolkit.

Referee Report

2 major / 8 minor

Summary. This paper proposes a two-pass 'pitch-and-catch' protocol for capturing a single-photon pulse in a cavity with tunable couplings. In the first pass, the pulse partially excites the cavity; the reflected field is redirected through an orthogonal port with a controllable delay and interacts with the cavity a second time. The authors derive the single-pass capture limit (Section III), then show that for square and exponential-decay pulses the two-pass protocol achieves unit efficiency with finite peak couplings: the stated thresholds are κmax ≥ 2 ln 2 for bounded positive packets (Section IV) and κmax > k ≈ 1.2834 for exponential decay (Section V). With intrinsic loss, the capture efficiency becomes κ_i^{κ_i/(1-κ_i)} (Eq. 42). The method is also proposed for time-reversed emission and pulse shaping.

Significance. The core idea is novel and potentially important. If the universal claim holds, the scheme removes the O(γ/κmax) reflection loss inherent in single-pass capture and enables near-perfect storage with finite couplings and capture windows. The variational derivation of the single-pass limit (Eqs. (8)-(12)) is clean and self-contained, and the explicit square- and exponential-pulse formulas are concrete and testable. The paper also correctly identifies that the two-pass scheme changes the loss scaling in the noisy case. However, the headline universality result rests on an unproven worst-case lemma, which currently tempers the significance.

major comments (2)

[Section IV] The universal claim that 'for any bounded packet, κmax ≥ 2 ln 2 would guarantee perfect two-pass capture' rests entirely on the assertion that 'The square packet ... maximizes initial reflection, making it the most challenging case for complete capture.' This worst-case lemma is stated without proof. The binding quantity for the second pass is sup_t |b'_in(t)/a(t)|, because the control law (Eq. 18) sets κ2(t) = (b'_in(t)/a(t))^2 whenever this is below κ2,max. For a square pulse, this supremum can be related to the first-pass final amplitude a^2(1), giving Eq. (19). For a non-square bounded packet, however, the reflected wavepacket b'_in can have a later maximum while a(t) is still growing, so the bound in Eq. (19) does not control the supremum. For example, a packet with a low-amplitude early section followed by a late subpulse produces a reflected field whose peak arrives when a(t) has already increased, potentially requiring a larger κ2,max than the square-pulse threshold. Thus the central universality result is not established. I recommend either proving the lemma (for example, by a rearrangement or majorization argument) or restricting the stated guarantee to the pulse families analyzed in Sections IV and V.
[Section V] The derivation of the exponential-decay threshold assumes that the second-pass capture condition is satisfied once the leading edge of the reflected wavepacket can be absorbed, i.e., 1 − sqrt(κ2,max) a(Δt) = b'_out(Δt) = 0 (Eq. 23). However, the reflected wavepacket extends over a finite interval [Δt, Δt + t1], and the control law (Eq. 28) must keep the ratio b'_in(t)/a(t) below sqrt(κ2,max) for all t in that interval. The paper does not prove that the ratio is maximized at the leading edge; it only verifies the condition at t = Δt. Without such a bound, the infimal threshold k ≈ 1.2834 is not rigorously justified. Please add an explicit proof that sup_{t≥Δt} [b'_in(t)/a(t)] = sqrt(κ2,max) for the chosen Δt, or derive the threshold from a direct optimization over the full second-pass interval.

minor comments (8)

[Eq. (16)] Equation (16) appears to be missing a division sign: it should read a^2(t=1) = 4/κ1,max (1 − e^{−κ1,max/2})^2, not '4 κ1,max'.
[Eqs. (17) and (22)] Equations (17) and (22) use ambiguous notation '1 − 2 ln 2− 1 / κ1,max'; please write 1 − (2 ln 2 − 1)/κ1,max (and similarly for Eq. (22)).
[Abstract and Section IV] The abstract and introduction promise 'perfect capture of arbitrary pulses,' but Section IV explicitly restricts to positive bounded pulses (0 < bin(t) < bmax), with only a brief remark about tuning the phase for complex pulses. Please qualify the abstract accordingly.
[Section VII] In Section VII, 'κmax/maxt bin(t)' should be 'κmax / max_t bin(t)'.
[Section VI] The statement in Section VI that 'the capture efficiency is now 1 − e^{O(κi/γ)}' is imprecise; Eq. (42) implies the loss is κ_i^{κ_i/(1-κ_i)}, which for small κ_i behaves as κ_i ln(1/κ_i). Please rephrase.
[Reference [9]] Reference [9] is incomplete; it should include the article number or page range.
[Figure captions] Figure captions 2 and 4 should explicitly state the units of κ1,max and κ2,max (b_max^2 and γ, respectively), as the text does.
[Eq. (5)] Equation (5) assumes lossless, dispersionless redirection of the reflected field with a controllable delay; this should be stated as an explicit assumption at the point of introduction.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the two-pass control laws are explicit constructions, and the finite thresholds follow from the dynamics rather than from fitting or self-citation.

full rationale

I find no step in which a claimed prediction reduces to an input by construction. The reflection-nulling controls (Eqs. 15 and 18) are explicit feedback laws: κ1(t)=min(κ1,max,(bin/a)^2) is chosen to maximize the instantaneous capture rate in Eq. (14), and κ2(t) is the same construction for the recycled mode. The quantitative claims—κmax≥2ln2 for bounded packets (Sec. IV), the exponential-decay threshold k≈1.2834 (Sec. V), and the loss scaling κi^{κi/(1-κi)} (Eq. 42)—are obtained by integrating the resulting nonlinear Langevin dynamics for specific packets (square and exponential), not by fitting parameters to the target efficiency. No parameter is fitted to data, and no result is imported from a self-citation: reference [7] (with author overlap) is used only as an experimental demonstration, not as a load-bearing theorem. The main weakness is Section IV's unproven assertion that the square packet 'maximizes initial reflection, making it the most challenging case for complete capture'; this is a correctness gap in the universality claim, but it is not circular, because the worst-case lemma is not defined in terms of the threshold it is used to prove, and the square-packet threshold itself is computed independently from the dynamics. Accordingly the circularity score is 0.

Assumptions & free parameters 3 free parameters · 7 assumptions · 0 invented entities

No constants are fitted to data anywhere in this paper; the analysis is fully analytic. The entries above are hand-chosen protocol or environment parameters. The paper-specific assumptions beyond standard input-output theory are the controllable-lossless delay redirect and the unproven square-packet worst-case lemma. No invented entities (particles, mediators, forces, dimensions, or conserved quantities) are introduced; 'recycling' redirects the existing reflected field.

free parameters (3)

Delta t (feedback delay)
Set by Eqs. (26)/(35)/(38) to satisfy the zero-reflection condition at the second port; a hand-chosen design parameter, not fitted to data.
kappa_i (intrinsic loss rate)
Environmental parameter hand-set equal for cavity and delay line (Section VI); enters all lossy-regime efficiencies.
kappa_1,max, kappa_2,max (coupling bounds)
Resources of the protocol, not fitted; the paper derives required thresholds (2 ln 2; k about 1.2834) as functions of them.

assumptions (7)

domain assumption Linear quantum Langevin/input-output dynamics for the cavity (Eq. 2) with the standard input-output relation b_out = b_in - sqrt(kappa) * a (used in Eq. 5).
Underpins the entire commutator-amplitude analysis; standard in cavity QED but still a modeling assumption (high-finesse, single-mode, Markovian port).
domain assumption Cavity starts in vacuum and the input state is a single mode with at most one excitation, so a(t) = [a(t), c^dagger] is a c-number (Eq. 3).
Required for the beamsplitter reduction; the 'arbitrary photonic states' language in the abstract outruns this at-most-one-excitation restriction.
domain assumption The reflection is redirected losslessly (noiseless sections) into an orthogonal port after a controllable delay Delta t (Eq. 5).
The delayed-feedback model makes b'_in a deterministic function of b_out; loss is only added in Section VI as symmetric kappa_i.
domain assumption kappa_1(t) and kappa_2(t) are independently tunable in real time and can follow the impedance-matching control law instantaneously (Eqs. 15, 18).
Perfect capture requires bang-bang switching at times t1 and Delta t and continuous tracking of (b_in/a)^2; no bandwidth or actuation limits are modeled.
ad hoc to paper The square packet maximizes initial reflection and is the hardest case for complete capture.
Asserted without proof in Section IV; it carries the universal 'any bounded packet, kappa_max >= 2 ln 2' guarantee.
domain assumption kappa_i,cav = kappa_i,delay = kappa_i much less than kappa_max, gamma, with optimal stop times as derived (Eqs. 32, 41).
Explicitly stated 'For simplicity' in Section VI; unequal cavity/delay losses are deferred to future work.
standard math Numerical evaluation of k about 1.2834 from (k+1)^(k+1) = 4k^2 and of the figure contours.
Routine root-finding and plotting; no code shipped.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Recycling Reflections for Perfect Photon Capture." pith.science (2026). https://pith.science/paper/34UY4VPZ

@misc{pith2026250601127,
  author       = {Pith},
  title        = {Pith review of: Recycling Reflections for Perfect Photon Capture},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/34UY4VPZ}},
  note         = {Machine review of arXiv:2506.01127}
}

read the original abstract

Efficient photon capture in optical cavities is essential for quantum networks and computing, yet single-pass methods suffer from uncaptured reflections due to finite capture windows and coupling strengths, precluding perfect transfer of arbitrary photonic states. We introduce a two-pass `pitch-and-catch' method that recycles initial reflections to achieve perfect capture of arbitrary pulses in the noiseless regime and enhances fidelity under intrinsic losses. The method extends to photon emission, enabling arbitrary pulse shaping. This advance offers significant improvements for quantum repeaters, memories, and transduction, enhancing the toolkit for quantum information processing.

Figures

Figures reproduced from arXiv: 2506.01127 by the authors.

**Figure 2.** FIG. 2: Logarithm of minimum loss achievable with [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗

**Figure 3.** FIG. 3: Example pulse of single-pass with [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

Figures from the paper (4 more)

**Figure 4.** Figure 4: FIG. 4: Logarithm of minimum loss achievable with [PITH_FULL_IMAGE:figures/full_fig_p004_4.png]

**Figure 5.** Figure 5: FIG. 5: Example pulse of single-pass with [PITH_FULL_IMAGE:figures/full_fig_p004_5.png]

**Figure 6.** Figure 6: FIG. 6: Example pulse of single-pass with [PITH_FULL_IMAGE:figures/full_fig_p005_6.png]

**Figure 7.** Figure 7: FIG. 7: Various setups for achieve two-pass capture. In [PITH_FULL_IMAGE:figures/full_fig_p007_7.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

10 extracted references · 8 canonical work pages

[1]

H. J. Kimble, The quantum internet, Nature 453, 1023–1030 (2008)

work page 2008
[2]

N. Lauk, N. Sinclair, S. Barzanjeh, J. P. Covey, M. Saffman, M. Spiropulu, and C. Simon, Perspectives on quantum transduction, Quantum Science and Tech- nology 5, 020501 (2020)

2020
[3]

J. I. Cirac, P. Zoller, H. J. Kimble, and H. Mabuchi, Quantum state transfer and entanglement distribution among distant nodes in a quantum network, Physical Re- view Letters 78, 3221–3224 (1997)

work page 1997
[4]

D. N. Matsukevich and A. Kuzmich, Quantum state transfer between matter and light, Science 306, 663–666 (2004)

work page 2004
[5]

A. D. Boozer, A. Boca, R. Miller, T. E. Northup, and H. J. Kimble, Reversible state transfer between light and a single trapped atom, Physical Review Letters 98, 193601 (2007). 8

work page 2007
[6]

Stute, B

A. Stute, B. Casabone, B. Brandst¨ atter, K. Friebe, T. E. Northup, and R. Blatt, Quantum-state transfer from an ion to a photon, Nature Photonics 7, 219–222 (2013)

work page 2013
[7]

C. J. Axline, L. D. Burkhart, W. Pfaff, M. Zhang, K. Chou, P. Campagne-Ibarcq, P. Reinhold, L. Frun- zio, S. M. Girvin, L. Jiang, M. H. Devoret, and R. J. Schoelkopf, On-demand quantum state transfer and en- tanglement between remote microwave cavity memories, Nature Physics 14, 705–710 (2018)

work page 2018
[8]

Stobi´ nska, G

M. Stobi´ nska, G. Alber, and G. Leuchs, Perfect excita- tion of a matter qubit by a single photon in free space, Europhysics Letters 86, 14007 (2009)

work page 2009

Show all 10 references

[9]

H. Yan, Y. Zhong, H.-S. Chang, A. Bienfait, M.-H. Chou, C. R. Conner, E. Dumur, J. Grebel, R. G. Povey, and A. N. Cleland, Entanglement purification and protection in a superconducting quantum network, Physical Review Letters 128, 10.1103/physrevlett.128.080504 (2022)

2022 doi
[10]

Ritter, C

S. Ritter, C. N¨ olleke, C. Hahn, A. Reiserer, A. Neuzner, M. Uphoff, M. M¨ ucke, E. Figueroa, J. Bochmann, and G. Rempe, An elementary quantum network of single atoms in optical cavities, Nature 484, 195–200 (2012)

2012

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

H. J. Kimble, The quantum internet, Nature 453, 1023–1030 (2008)

work page 2008

[2] [2]

N. Lauk, N. Sinclair, S. Barzanjeh, J. P. Covey, M. Saffman, M. Spiropulu, and C. Simon, Perspectives on quantum transduction, Quantum Science and Tech- nology 5, 020501 (2020)

2020

[3] [3]

J. I. Cirac, P. Zoller, H. J. Kimble, and H. Mabuchi, Quantum state transfer and entanglement distribution among distant nodes in a quantum network, Physical Re- view Letters 78, 3221–3224 (1997)

work page 1997

[4] [4]

D. N. Matsukevich and A. Kuzmich, Quantum state transfer between matter and light, Science 306, 663–666 (2004)

work page 2004

[5] [5]

A. D. Boozer, A. Boca, R. Miller, T. E. Northup, and H. J. Kimble, Reversible state transfer between light and a single trapped atom, Physical Review Letters 98, 193601 (2007). 8

work page 2007

[6] [6]

Stute, B

A. Stute, B. Casabone, B. Brandst¨ atter, K. Friebe, T. E. Northup, and R. Blatt, Quantum-state transfer from an ion to a photon, Nature Photonics 7, 219–222 (2013)

work page 2013

[7] [7]

C. J. Axline, L. D. Burkhart, W. Pfaff, M. Zhang, K. Chou, P. Campagne-Ibarcq, P. Reinhold, L. Frun- zio, S. M. Girvin, L. Jiang, M. H. Devoret, and R. J. Schoelkopf, On-demand quantum state transfer and en- tanglement between remote microwave cavity memories, Nature Physics 14, 705–710 (2018)

work page 2018

[8] [8]

Stobi´ nska, G

M. Stobi´ nska, G. Alber, and G. Leuchs, Perfect excita- tion of a matter qubit by a single photon in free space, Europhysics Letters 86, 14007 (2009)

work page 2009