Experimental Realization of Highly Efficient Broadband Coupling of Single Quantum Dots to a Photonic Crystal Waveguide

We present time-resolved spontaneous emission measurements of single quantum dots embedded in photonic crystal waveguides. Quantum dots that couple to a photonic crystal waveguide are found to decay up to 27 times faster than uncoupled quantum dots. From these measurements $\beta$-factors of up to 0.89 are derived, and an unprecedented large bandwidth of 20 nm is demonstrated. This shows the promising potential of photonic crystal waveguides for efficient single-photon sources. The scaled frequency range over which the enhancement is observed is in excellent agreement with recent theoretical proposals taking into account that the light-matter coupling is strongly enhanced due to the significant slow-down of light in the photonic crystal waveguides.

Figure 1

Sketch of the experimental setup, which is described in detail in the text. The inset to the bottom right is a scanning electron micrograph of a fabricated PCW. The red sketched region illustrates the size of the area from which spontaneous emission is collected.

Figure 2

(a) Spontaneous emission spectra (solid blue curve) displaying single QD emission lines. The spectra were recorded at two different positions on a PCW with $a=256\mathrm{nm}$. The excitation density was $\sim 3\mathrm{W}/{\mathrm{cm}}^2$. The crosses indicate the measured decay rates (right axis) of the different QD emission lines. (b) Decay curves from two single QD lines. The blue (fast) decay curve is measured at 981.0 nm where the QD is coupled to the PCW. The red (slow) decay curve is measured at 969.7 nm and is an example of an uncoupled QD. The black lines are the fitted single (red curve) or double (blue curve) exponential decay models.

Figure 3

(a) Measured QD decay rates on the PCW sample with $a=256\mathrm{nm}$ as a function of scaled frequency. The excitation power for these measurements was $3\mathrm{W}/{\mathrm{cm}}^2$. The data points marked by open (filled) circles are modeled using a double (single) exponential decay model. The solid red line (right axis) displays the decay rates calculated from numerical simulations with uncertainties in $a$ of $\pm 2\mathrm{nm}$ given by the hatched area. The black line at $a/\lambda =0.253$ is the calculated band edge of the 2D photonic bandgap. The dotted green line is the mean decay rate of uncoupled QDs used for the estimation of the $\beta$-factor. (b) Measured decay rates on a photonic crystal sample with no PCW ($a=256\mathrm{nm}$).

Figure 4

Measured decay rates for QDs in a PCW sample with $a=248\mathrm{nm}$. The pump excitation power was $1.5\mathrm{kW}/{\mathrm{cm}}^2$. The data points marked by open (solid) squares are modeled using a double (single) exponential decay model. The red line (right axis) shows the decay rates calculated from numerical simulations with uncertainties in $a$ of $\pm 2\mathrm{nm}$ given by the hatched area. The dotted green line is the mean decay rate of uncoupled QDs used for the estimation of $\beta$. The inset displays the $\beta$-factors above 0.5 calculated from the data in Fig. 3 (green circles) and Fig. 4 (blue squares) as described in the text. The green (blue) bar shows the coupling bandwidth of the PCW sample with $a=256\mathrm{nm}$ ($a=248\mathrm{nm}$).