Two-Photon Interference Using Background-Free Quantum Frequency Conversion of Single Photons Emitted by an InAs Quantum Dot

We show that quantum frequency conversion (QFC) can overcome the spectral distinguishability common to inhomogeneously broadened solid-state quantum emitters. QFC is implemented by combining single photons from an InAs/GaAs quantum dot (QD) at 980 nm with a 1550 nm pump laser in a periodically poled lithium niobate (PPLN) waveguide to generate photons at 600 nm with a signal-to-background ratio exceeding $100:1$. Photon correlation and two-photon interference measurements confirm that both the single photon character and wave packet interference of individual QD states are preserved during frequency conversion. Finally, we convert two spectrally separate QD transitions to the same wavelength in a single PPLN waveguide and show that the resulting field exhibits nonclassical two-photon interference.

Figure 1

(a) Experimental setup used within this work (see the Supplemental Material [23]). $\mathrm{HBT}=$ Hanbury-Brown–Twiss setup; $\mathrm{HOM}=$ Hong-Ou-Mandel interferometer. (b) Converted 600 nm band wavelength vs PPLN waveguide temperature. The inset shows the quasi-phase-matching response of the PPLN waveguide. (c) Signal-to-background ratio (left $y$ axis, blue points) and external conversion efficiency (right $y$ axis, red points) as a function of 1550 nm pump power. The external conversion efficiency includes all losses in the system.

Figure 2

(a) Low-temperature $\mu$-PL spectrum of device M1. Bright QD emission and cavity mode emission are visible around 977 nm. (b) Spectrum of QD emission filtered by a volume Bragg grating. (c),(d) Second-order autocorrelation function measurements performed on the QD emission line before and after frequency conversion.

Figure 3

(a) Autocorrelation of the $X1$ emission line from device $M2$ under cw excitation ($\mu$-PL spectrum inset). (b),(c) Two-photon interference of the $X1$ line under parallel and orthogonal polarization configurations of the interferometer arms, respectively. (d) Autocorrelation of the $X1$ line after frequency conversion (frequency-converted spectrum inset). (e),(f) Two-photon interference of the frequency-converted $X1$ line under parallel and orthogonal polarization configurations of the interferometer arms, respectively. The dashed lines are fits to the experimental data (Supplemental Material [23]), and the solid line marks $g^{(2)}(0)=0.5$ level.

Figure 4

(a) $\mu$-PL spectrum of device $M2$ under above-band excitation. (b) Cross-correlation measurement performed on $X1$ and $X2$ emission lines. (c) PL spectrum after both lines are converted to the same wavelength at 600 nm. (d) Autocorrelation measurement of the combined frequency-converted signal of $X1$ and $X2$.

Figure 5

(a) $\mu$-PL spectrum of device $M3$ under above-band excitation. Two bright excitonic emission lines (named $X1$ and $X2$) are observed with nearly equal intensity. (b) Two-photon interference of the combined $X1$ and $X2$ signal after both lines are frequency-converted to the same wavelength at 600 nm and measured in the parallel polarization configuration. (c) Zoom-in near the central dip of part (b). The solid red line is a fit to the data, while the black dashed line corresponds to the orthogonal polarization configuration. $g_{\parallel }^{(2)}(0)<g_{\perp }^{(2)}(0)$ is due to the two-photon interference effect.