arXiv:2611.01205
Pump-scheme dependence of few-emitter lasers: exact multilevel correlations and photon statistics
Stephan Hartmann
Abstract
Few-emitter laser models usually replace the excitation cycle of the gain medium by a single Markovian pump jump of a two-level emitter. We test this replacement with the cavity and laser-transition parameters held fixed and the two-level pump rate fitted to the same mean photon number, using an exact damping-basis continued-fraction solution of the cavity--emitter master equation for $N$ identical $d$-level emitters with arbitrary independent incoherent transitions and one cavity-coupled transition, in the known reduced space of level populations and lasing coherences. First, a pump-quantum filtration shows that the inter-emitter coherence of $Λ$ and ladder media starts at second rather than first order in the pump; we evaluate its leading coefficient from the exact finite-system series and map the recycling rate at which its sign changes. Second, at matched photon number, rate-limited steps in the pump cycle push the long-time output-counting Fano factor of two and three emitters well below one: for two emitters it reaches 0.66 at a mean cavity occupation of 0.3 and is still 0.74 at an occupation of one, where $g^{(2)}(0)$ is within 3% of unity, because the anticorrelation sits at delays of about half a cavity lifetime rather than at zero delay. For two and three emitters the fitted two-level model gets both statistics wrong and places the output above the shot-noise level. Against independent single-emitter lasers at the same output, the common cavity erodes rather than enhances this single-emitter effect.
AI Summary
The paper asks when a single Markovian pump jump for a two-level emitter reproduces a few-emitter laser with a multilevel excitation cycle. Comparisons hold the cavity, coupling, laser-transition decay and emitter number fixed, fitting only the two-level pump rate to the mean photon number. It extends a damping-basis matrix-continued-fraction solution to identical $d$-level emitters with arbitrary independent incoherent transitions and one cavity-coupled transition. The known population-and-lasing-coherence reduction gives atomic blocks of dimension $\binom{N+d+1}{d+1}$. A pump-quantum filtration establishes that pair coherence begins at $O(P^2)$ for $\Lambda$ and ladder cycles, versus $O(P)$ for two-level and V cycles; finite-system series calculations determine its recycling-dependent sign. A flux bound proves strong-pump quenching when pumping empties either laser level, whereas the studied $\Lambda$ and ladder cycles saturate to explicit limiting models. At matched mean occupation $0.3$, two ladder emitters reach a long-time output-counting Fano factor of $0.663$. At occupation one, it is $0.740$ despite $g^{(2)}(0)=0.974$: the correlation minimum occurs near $0.43/A$. The fitted two-level model instead gives counting noise above shot noise for two and three emitters. Independent Fock-space calculations reproduce selected counting results to approximately $10^{-12}$. Independent single-emitter sources are quieter at equal total output, so the common cavity erodes the suppression. The construction excludes emitter inhomogeneity, coherent pumping and multiple cavity-coupled transitions. Infinite-dimensional stationary convergence requires explicit uniqueness and pivot assumptions at zero cavity temperature. Matched-output comparisons cover $N\leq3$; larger-$N$ calculations concern one benchmark. Fits with additional two-level parameters are not tested.
Written by gpt-6.1-sol from the paper's full text. Automated: it can be wrong.
- Comments
- 22 pages, 10 figures, 8 tables