High-quality single photons from cavity-enhanced biexciton-to-exciton transition

The gist: The study demonstrates that resonant two-photon excitation followed by single photon emission from a cavity-enhanced biexciton-to-exciton transition can generate high-quality single photons with competitive quality quantifiers without systematic optimization, overcoming limitations like emitter re-excitation and spectral overlap between the excitation laser and emitted photons.

System Model and Scheme

The research models semiconductor quantum dots as a four-level system including the ground state G⟩, two orthogonal exciton states XH⟩ and XV⟩, and the biexciton state XX⟩ <ref:2602.18153pg3. The core scheme involves resonant two-photon excitation (TPE) of the biexciton state followed by collection of the emitted single photon with the cavity <ref:2602.18153pg4. This approach circumvents emitter re-excitation and naturally introduces spectral separation between the excitation laser and emitted single photon <ref:2602.18153pg4. The system time evolution is calculated solving the polaron master equation <ref:2602.18153pg3.

Key Advantages of the Approach

The proposed method offers several advantages over established solutions for high-quality single-photon sources <ref:2602.18153pg4. Specifically, it avoids emitter re-excitation by initiating photon generation from a biexciton (XX) state rather than an exciton (X) state <ref:2602.18153pg4. This configuration enables a clear separation – both spectrally and in polarization – between the generated photons and the laser while avoiding re-excitation <ref:2602.18153pg4. Furthermore, it naturally introduces spectral separation of excitation laser and emitted single photon <ref:2602.18153pg4.

Quality Quantifiers and Performance

The quality of single photons is characterized by emission PH, purity PH, and indistinguishability IH <ref:2602.18153pg4. The theoretical analysis shows that with selective Purcell enhancement, the observed quality quantifiers of single-photon emission (purity, equivalently g(2)(0), and HOM visibility V) are competitive with respect to high-quality deterministic quantum-dot single-photon sources <ref:2602.18153pg4. For a specific parameter set, the results show g(2)(0) = 0.0041 and V = 96.07 % at T = 0 K with moderate spectral filtering <ref:2602.18153pg7.

Influence of System Parameters

The performance is highly dependent on the biexciton binding energy Ebind <ref:2602.18153|pg4. The analysis shows that at sufficiently large absolute XX binding energies, as expected for the scenario where only the XX-X transition is resonant with the cavity mode, we observe emission below unity <ref:2602.18153|pg5. A maximum in emission occurs close to zero absolute XX binding energy, where also the X-G transition is near-resonant with the cavity mode <ref:2602.18153|pg4. The two dips in indistinguishability arise for the following reasons: at exact resonance, all photons share the same frequency, resulting in slightly increased indistinguishability <ref:2602.18153|pg5.

Optimization and Future Directions

The paper suggests that to improve performance, emphasis should be placed on reducing cavity loss κ and further suppressing exciton decay <ref:2602.18153pg7. Spectral filtering can be used to enhance the photon quality further, with results showing a slight reduction in emission and an increase in the single-photon character <ref:2602.18153pg7. Temporal filtering may also be considered to separate XX photons from X photons <ref:2602.18153pg7. Switching from InGaAs to GaAs quantum dot systems is expected to further enhance photon quality by reducing phonon interactions and offering larger biexciton binding energies <ref:2602.18153pg7. The scheme's simplicity and absence of aggressive emission post-processing are highlighted as clear advantages <ref:2602.18153pg7.

Entangled Photon Pairs

The setup is also capable of generating high-quality indistinguishable polarization-entangled photon pairs <ref:2602.18153pg6. While initial calculations at 0 K without taking phonon interactions into account reveal that the current setup does not support this (C = 0.13) due to which-path information introduced by the setup <ref:2602.18153pg6, adjustments that eliminate the polarization splittings can lead to perfect entanglement with C = 1 <ref:2602.18153pg6.

Experimental Validation

The theoretical findings are supported by first experimental results demonstrating simultaneously high values of photon indistinguishability and single-photon purity, without the need of any signal post processing <ref:2602.18153pg4. The measured visibility V = 0.94 ± 0.02 and g(2)(0) = 0.023 ± 0.002 of the XX photon are achieved in the present experiments with a device that has not been specifically engineered as a biexciton-based single-photon source <ref:2602.18153pg7. The excellent agreement between experiment and theory is found <ref:2602.18153pg7.

Theoretical Foundations

The theoretical framework involves solving the polaron master equation, which includes coupling of the QD to its environment (full details on the modelling are given in Appendix B) <ref:2602.18153pg3. The full Hamiltonian in the lab frame is given by H = HQD + Hcav + HQD-cav + HQD-light <ref:2602.18153pg1. The analysis in Appendix C shows that cavity enhancing the XX decay in the Purcell regime leads to replacement of the natural XX lifetime with the inverse of the Purcell decay rate 4g 2/κ [4] <ref:2602.18153pg7. The generalized theory allows for a detailed description of experimental situations involving a tunable cavity coupled to radiatively decaying QD transitions <ref:2602.18153pg9.

Conclusion

Overall, the scheme possesses clear advantages that lie in its simplicity and the absence of any need for aggressive emission post-processing <ref:2602.18153pg7. With targeted device optimization, this approach may ultimately outperform existing sources <ref:2602.18153pg7. The intrinsic problems that plague resonant single-photon generation from two-level systems – namely quantum-emitter re-excitation and difficult (spectral) separation of excitation laser from generated single photons – play no role in our approach <ref:2602.18153pg7.

Acknowledgments

N.H. thanks Nikolas K¨ocher for various fruitful discussions <ref:2602.18153pg8. The work in Paderborn was supported by the Deutsche Forschungsgemeinschaft (German Research Foundation) through the transregional collaborative research center TRR 142/3-2022 (231447078, project C09) and by Paderborn Center for Parallel Computing, PC2 <ref:2602.18153pg8. This work is further supported by the German Federal Ministry of Research, Technology and Space (BMFTR) through the project QR.N (16KIS2206) <ref:2602.18153pg8.

Appendix B Details

The polaron master equation model includes phonon-induced pure dephasing Ldeph ρ = -γdeph squared Xσˆ̸=ˆσ'σρˆ σˆ' <ref:2602.18153pg12. The phonon-mediated QD-cavity interactions are modeled by a Lindblad operator LPh-I ρ = -1/ħ 2 Z ∞ 0 dτ X i=g,u n Gi(τ) × · · · · × h Xi(t, t − τ)ρ(t − τ) i + h.c.o <ref:2602.18153pg15. The cavity emission spectrum is calculated using Si(ω) = Z dtSi(t, ω), with the time-dependent spectrum at time t Si(t, ω) = κ Z dτ e-iωτG(1) i (t, τ) <ref:2602.18153pg13. The algebraic solution for the cavity annihilation operator reads a σˆ(ω) = g σˆω - ωcav + i κωcav 2/σˆ(ω) <ref:

1. textbfConcentration on Spectral Filtering for Purity Enhancement + Specific Action: The improved AI system can optimize spectral filtering parameters (filter bandwidth µS) in real-time to minimize X photons emitted offresonantly into the cavity by using the sensor method [76, 77] and reducing the negative influence of the X photons on the photon output.

2. textbfModel Integration of Phonon-Mediated Cavity Feeding + Specific Action: The system can incorporate the analytical rate ΓCav,T QD−trans(∆) to model phonon-mediated cavity feeding and determine if temporal filtering is viable for its elimination, as suggested by the finding that XX photons are emitted after tens of picoseconds while X photons decay on longer timescales.

3. textbfParameter Regime Identification for High Quality Photons + Specific Action: The AI can systematically identify QD-cavity parameter regimes that lead to the generation of high-quality single photons, specifically by searching for conditions where the XX binding energy is sufficiently large and applying very moderate spectral filtering applied to achieve results like g(2)(0) = 0.0041 and V = 96.07 %.

4. textbfRobustness Assessment Under Excitation Perturbation + Specific Action: The system can assess the resilience of the scheme against excitation variations by checking the single-photon character of the H polarized cavity photons is largely unaffected by TPE in the V mode, confirming that perpendicular polarization channels for QD excitation and photon generation ensure robustness.

5. textbf Entanglement Feasibility Mapping + Specific Action: The AI can map out conditions for entanglement generation, determining that while initial calculations at 0 K without taking phonon interactions into account reveal that our current setup does not support this (C = 0.13), adjustments like zero cavity mode splitting and zero fine structure splitting are required to achieve perfect entanglement with C = 1.

6. textbf Time-Resolved Spectral Analysis for Source Discrimination + Specific Action: The system can use the time-resolved cavity emission spectra, specifically looking at the difference between the XX (and the associated intense peak in the spectrum) is emitted on a short timescale of a few tens of picoseconds and the X and associated peaks (the tail of the intense peak at cavity resonance and the side peak) decay on much longer timescales.

7. textbf Transition to InGaAs/GaAs Systems + Specific Action: The AI can use the findings that switching from InGaAs to GaAs quantum dot systems is expected to further enhance photon quality by reducing phonon interactions and offering larger biexciton binding energies as a guidance for future hardware selection.