Single-photon detection and cryogenic reconfigurability in lithium niobate nanophotonic circuits

Abstract

Lithium-Niobate-On-Insulator (LNOI) is emerging as a promising platform for integrated quantum photonic technologies because of its high second-order nonlinearity and compact waveguide footprint. Importantly, LNOI allows for creating electro-optically reconfigurable circuits, which can be efficiently operated at cryogenic temperature. Their integration with superconducting nanowire single-photon detectors (SNSPDs) paves the way for realizing scalable photonic devices for active manipulation and detection of quantum states of light. Here we demonstrate integration of these two key components in a low loss (0.2 dB/cm) LNOI waveguide network. As an experimental showcase of our technology, we demonstrate the combined operation of an electrically tunable Mach-Zehnder interferometer and two waveguide-integrated SNSPDs at its outputs. We show static reconfigurability of our system with a bias-drift-free operation over a time of 12 hours, as well as high-speed modulation at a frequency up to 1 GHz. Our results provide blueprints for implementing complex quantum photonic devices on the LNOI platform.

Fig. 1. Configuration of the chip and the experimental setup.

a Optical microscope image of the integrated device. The reference directional coupler (DC) at the upper left of the figure allows for measuring the splitting ratio of the DCs upstream from the detectors, and for estimating the insertion loss of the MZI. b Schematic of the experimental setup with optical (red) and electrical (blue) access to the LN-chip. Inset in the figure shows a photograph of the photonic chip, the fiber array, and the contact probe employed in our experiment. c Electrode configuration used for the realization of the electro-optic phase shifter. The colormap is the field intensity of the fundamental TE mode supported by the waveguide calculated with a finite-difference mode solver. White arrows are the lines of the electric field applied to the waveguide calculated with a finite-difference mode solver. d Schematic of the nanowire configuration used for the realization of waveguide-integrated SNSPDs, and simulation of the nanowire absorption. The colormap is the field amplitude of the fundamental TE mode supported by the waveguide as calculated with a finite-element model solver. e False-color scanning electron micrograph of a waveguide-integrated SNSPD taken in the proximity of the gold electrodes pads. f Surface topography of a waveguide-integrated SNSPD taken in the proximity of the U-turn by atomic force microscopy. The same color scheme of Fig. 1e is here used for clarity. Both images were taken after waveguide etching, before cladding the device with HSQ.

Fig. 2. Characterization of the device loss.

a Resonance of a racetrack resonator (fabricated on the same chip of the device in Fig. 1) in the critical coupling regime measured at room temperature. The extracted loaded Q factor is $Q_{\mathrm{L}}=5.2\times {10}^5$ (intrinsic $Q\simeq {10}^6$). The inset in the graph shows a sketch of the racetrack cavity used to determine the loss. A bus waveguide connected to two grating couplers is used to couple light into the resonator. b Resonance of the same racetrack resonator measured at cryogenic temperature. The extracted loaded Q factor is $Q_{\mathrm{L}}=6.4\times {10}^5$ (intrinsic $Q\simeq 1.2\times {10}^6$). c Transmission spectrum of the reference DC device at the upper left of Fig. 1a, and of the two outputs of the MZI measured when laser light is injected into In2. The insertion loss of the MZI is estimated by fitting the transmission spectrum of the reference DC with a Gaussian function, the spectrum of Out1/Out2 with a Gaussian function multiplied by a sinusoidal function, and by comparing the maxima of the two Gaussians. The insertion loss averaged over six identical devices fabricated on the same chip is found equal to $\left(0.82\pm 0.24\right)$ dB.

Fig. 3. Performance of the waveguide-integrated detectors.

a, b Measured on-chip detection efficiency (OCDE) of the two detectors as a function of the bias current. The dashed-dotted line indicates the value of the current I = 0.85 × I_c, at which the two detectors were biased to measure the timing jitter and to show the joint operation of the EOM and the SNSPDs. c, d Output electrical signals of the two detectors registered with a digital oscilloscope upon the absorption of a photon. e, f Time histograms of the photon counts used to measure the timing jitter of the two detectors (see main text for an explanation). Det2 is connected to a cryogenic temperature amplifier, which allows to reduce the electrical noise of the amplified signal and measure a lower timing jitter.

Fig. 4. Combined operation of EOM and SNSPDs: slow-speed and DC operation.

a Normalized count rates collected from the two detectors with a time tagging module (lower image) when the EOM is driven with a ramp function with an amplitude of 20 V_pp and a frequency of 1 kHz (upper image). b Zoom in on the detector counts showing a half-modulation period of the time histogram of Fig. 3a. The plotted data is used to estimate the half-wave voltage of the EOM. c Collected count rates from Det1 (upper image) and from Det2 (lower image) as a function of the laser wavelength for an applied DC bias of 0 V (solid line) and of 16.5 V (dashed line). Data are plotted on a dB scale. d Count rate collected from Det2 over a period of 12 h at an applied DC bias of 16.5 V. At each data point the photon counts are integrated over a time of 10 s. The inset in the figure shows the normalized count rate plotted in a dB scale. During the measurement, the laser wavelength was set to a value for which all the light entering the MZI is directed to Det1 for a zero applied voltage to the EOM. Upon the application of a DC bias equal to the $V_{\mathrm{\pi }}$ of the EOM, the optical outputs of the MZI are inverted and all the input light directed to Det2.

Fig. 5. Combined operation of EOM and SNSPDs: high-speed operation.

a Modulation bandwidth of the EOM measured with a vector network analyzer at cryogenic temperature (blue trace) and room temperature (red trace) inside the same cryostat. The contribution of the RF lines of the cryostat is not calibrated out of the measurements. b Normalized count rates collected from Det2 with a time tagging module when the EOM is driven with a sinusoidal function with an amplitude of 5 V_pp and a frequency of 100 MHz. Time bin width is set equal to 100 ps. c Normalized count rates collected from Det2 with a time tagging module when the EOM is driven with a sinusoidal function with an amplitude of 5 V_pp and a frequency of 1 GHz. Time bin width is set equal to 50 ps.