Phonon-Mediated Coupling of $\mathrm{InGaAs}/\mathrm{GaAs}$ Quantum-Dot Excitons to Photonic Crystal Cavities

We demonstrate that the emission characteristics of site-controlled $\mathrm{InGaAs}/\mathrm{GaAs}$ single quantum dots embedded in photonic crystal slab cavities correspond to single confined excitons coupled to cavity modes, unlike previous reports of similar systems based on self-assembled quantum dots. By using polarization-resolved photoluminescence spectroscopy at different temperatures and a theoretical model, we show that the exciton-cavity interaction range is limited to the phonon sidebands. Photon-correlation and pump-power dependence experiments under nonresonant excitation conditions further establish that the cavity is fed only by a single exciton.

Figure 1

(a) Schematic illustration of a hexagonal array of pyramidal recesses, each with a side length of $\sim 300\mathrm{nm}$ and containing a single PQD (green dots). (b) A PhC structure coaligned with the underlying PQD array with a lattice constant equal to half the pyramid pitch defines the single QD nanocavity system. (c) Secondary electron microscope image of an actual structure (slightly tilted perspective), with the location of the QD marked by the crossing of the dashed lines. (d) Spatial distribution of the electric field intensity of the fundamental cavity mode, computed by 3D finite difference time domain simulation. This mode is mainly polarized along the $V$ axis indicated in (c). (e) PL spectrum of a typical PQD-cavity structure showing the exciton emission ($X$) and the fundamental CM.

Figure 2

Polarization-resolved photoluminescence measurements. (a),(b) Temperature scanning of the dot-cavity detuning for two different PQD-cavity structures. (c),(d) Degree of polarization $\sigma$ versus detuning of the $X$ and CM lines for the two samples. The dashed lines are guides to the eye. The gray-shaded area in (c) and (d) depicts the typical range for $\sigma$ in the case of bare PQDs (i.e., dots that were not incorporated in a cavity structure). In both (a) and (b), the structures were nonresonantly excited with ${\lambda }_{\mathrm{exc}}=700\mathrm{nm}$.

Figure 3

Photon-correlation spectroscopy and power dependence measured at $T=10\mathrm{K}$ ($\Delta E\approx +1.5\mathrm{meV}$) under nonresonant excitation conditions (${\lambda }_{\mathrm{exc}}=700\mathrm{nm}$). (a) Second-order photon-correlation histograms for PQD-cavity structure $A$. Autocorrelation for the QD exciton (top), cross correlation between the cavity mode and the QD exciton (middle), and autocorrelation of the cavity mode photons (bottom). The black and red lines are exponential fits to the data. (b) Power dependence of the integrated PL for the PQD-cavity structure $A$.

Figure 4

Theoretical modeling of phonon-mediated dot-cavity coupling. (a),(b) Computed (dashed lines) and measured (continuous lines) line shapes for the cavity-PQD structure $A$ (a) and $B$ (b). The parameters used in the model are ($A$) $g=150\mu \mathrm{eV}$, $\gamma =200\mu \mathrm{eV}$, and $\kappa =250\mu \mathrm{eV}$ and ($B$) $g=200\mu \mathrm{eV}$, $\gamma =100\mu \mathrm{eV}$, and $\kappa =250\mu \mathrm{eV}$.