Hybrid quantum networks will be based on nodes that operate at different wavelengths, requiring quantum channel standardization via quantum frequency conversion (QFC). QFC is typically based on highly efficient sum- or difference-frequency generation in second-order nonlinear materials, such as periodically poled lithium niobate waveguides. The presence of the strong pump beam in such a nonlinear medium leads to unwanted nonlinear processes that produce noise. One of these noise processes is spontaneous Raman scattering (SRS). Typically, the pump is chosen to be the longest wavelength in the second-order nonlinear mixing process so that noise photons at the signal wavelength are produced by the less efficient anti-Stokes Raman scattering process rather than the Stokes scattering process. Since SRS is a temperature-dependent process, lowering the temperature reduces the Raman-scattered photons. We discuss the theory of temperature-dependent Raman scattering and present experimental results of the temperature dependence of dark count rates in a guided-wave QFC device.
Hybrid quantum networks will be based on nodes that operate at different wavelengths, requiring quantum channel standardization via quantum frequency conversion (QFC). QFC is typically based on highly efficient sum- or difference-frequency generation in second-order nonlinear materials, such as periodically poled lithium niobate waveguides. The presence of the strong pump beam in such a nonlinear medium leads to unwanted nonlinear processes that produce noise. One of these noise processes is spontaneous Raman scattering (SRS). Typically, the pump is chosen to be the longest wavelength in the second-order nonlinear mixing process so that noise photons at the signal wavelength are produced by the less efficient anti-Stokes Raman scattering process rather than the Stokes scattering process. Since SRS is a temperature-dependent process, lowering the temperature reduces the Raman-scattered photons. We discuss the theory of temperature-dependent Raman scattering and present experimental results of the temperature dependence of dark count rates in a guided-wave QFC device.
Quantum frequency conversion is important in quantum networks to interface nodes operating at different wavelengths and to enable long-distance quantum communication using telecommunications wavelengths. Unfortunately, frequency conversion in actual devices is not a noise-free process. One main source of noise is spontaneous Raman scattering, which can be reduced by lowering the device operating temperature. We explore frequency conversion of 1554 nm photons to 837 nm using a 1813 nm pump in a periodically poled lithium niobate waveguide device. By reducing the temperature from 85°C to 40°C, we show a three-fold reduction in dark count rates, which is in good agreement with theory.
High-efficiency quantum frequency conversion (QFC) for group-IV color centers, such as silicon vacancies (SiV), is required for long-distance quantum networking protocols. Performance of QFC depends on the choices of network and pump wavelengths and the choice between bulk crystal and waveguide, since these directly determine the dominant noise processes. These furthermore lead to fiber-based or free-space, single- or multi-pass setups with different complexities, power requirements, losses, and hence, external efficiencies. Here, we present a QFC device design for SiV QFC with simultaneous state-of-the-art performance and simple setup. Namely, by choosing the C-band pump wavelength of λ p = 1561 nm, which leads to an E-band network wavelength of λ n = 1398 nm, we remain in the anti-Stokes noise regime in a one-stage, single-pass conversion scheme. The high-quality erbium-doped fiber amplifiers in this band aid in maximizing the internal nonlinear process. We measure an external (internal) photon conversion efficiency of 44.5 ± 1.8 % (8.5 ± 1.8 %) at an external (internal) pump power of 1.51 W (0.97 W), limited by fiber coupling efficiencies. This is an increase in efficiency by 9 % (25 %) compared to prior demonstrations for SiV QFC, which were limited by factors such as internal device efficiencies, limited available pump power, and losses. We measure a pump-generated noise spectral density of 4913 counts/s/nm, which can be attributed to the anti-Stokes Raman scattering, in line with other one-stage demonstrations, though higher by an order of magnitude than that obtained in two-stage conversion. Without extra filtering, we expect this to lead to a networking photon signal-to-noise ratio of above 7000 for cavity-integrated SiVs in emission-based schemes.
Quantum frequency conversion (QFC) will be an indispensable ingredient in future quantum technologies. For example, large-scale fibre-based quantum networks will require QFC to interconnect heterogeneous building blocks like emitters, channels, memories, and detectors. The performance of existing QFC devices – typically realized in periodically poled nonlinear crystals – is often severely limited by parasitic noise that arises when the pump wavelength lies between the inter-converted wavelengths. Here we comprehensively investigate the noise spectrum of a QFC device pumped by a CW 1064 nm laser. The converter was realized as a bulk periodically poled potassium titanyl phosphate (ppKTP) crystal quasi-phase-matched for conversion between 637 nm and 1587 nm, which was also polished and coated to resonantly enhance the pump field by a factor of 50. While Raman scattering dominates the noise background from 1140 nm to 1330 nm, at larger energy shifts (beyond 60 THz), parasitic spontaneous parametric down-conversion (SPDC) is the strongest noise source. Further, the noise spectrum was contaminated by a regular succession of narrow-band peaks, which we attribute to a heretofore unidentified higher-order counter-propagating SPDC processes – with quasi-phase-matching orders up to 44 evident in our measurements. This work provides a comprehensive overview of the limiting noise sources in QFC devices that use quasi-phase-matched crystals and will prove an invaluable resource in guiding their future development.
Quantum frequency conversion (QFC) enables optical interfaces between dissimilar quantum systems, but suppressing noise while maintaining high conversion efficiency remains challenging. Here, we demonstrate that adapted poling in thin-film lithium niobate (TFLN) waveguides reduces the pump-power requirement while shifting the dominant noise mechanism and its power scaling. In 2-cm-long waveguides, adapted poling enables maximum conversion efficiency at an on-chip pump power of 19±0.8~mW, reducing the required pump power by 8~dB and the total conversion noise by 14~dB relative to conventional periodic poling at their respective maximum-conversion operating points. Controlled measurements with different pump conditions and device configurations reveal a transition from predominantly quadratic noise scaling in periodically poled devices to predominantly linear scaling in adapted-poling devices. This transition accompanies a shift from noise dominated by cascaded second-harmonic generation and Stokes spontaneous Raman scattering to direct anti-Stokes spontaneous Raman scattering. These results establish adapted poling as an effective approach to suppressing nonlinear noise, enabling low-noise integrated frequency converters for quantum networks.
Single-photon sources based on quantum dots have been shown to exhibit almost ideal properties such as high brightness and purity in terms of clear anti-bunching as well as high two-photon interference visibilities of the emitted photons, making them promising candidates for different quantum information applications such as quantum computing, quantum communication and quantum teleportation. However, as most single-photon sources also quantum dots typically emit light at wavelengths of electronic transitions within the visible or the near infrared range. In order to establish quantum networks with remote building blocks, low-loss single photons at telecom wavelengths are preferable, though. Despite recent progress on emitters of telecom-photons, the most efficient single-photon sources still work at shorter wavelengths. On that matter, quantum frequency down-conversion, being a nonlinear optical process, has been used in recent years to alter the wavelength of single photons to the telecom wavelength range while conserving their nonclassical properties. Characteristics such as lifetime, first-order coherence, anti-bunching and entanglement have been shown to be conserved or even improved due to background suppression during the conversion process, while the conservation of indistinguishability was yet to be shown. Here we present our experimental results on quantum frequency down-conversion of single photons emitted by an InAs/GaAs quantum dot at 903.6 nm following a pulsed excitation of a p-shell exciton at 884 nm. The emitted fluorescence photons are mixed with a strong pump-field at 2155 nm inside a periodically poled lithium niobate ridge waveguide and converted to 1557 nm. Common issues of a large background due to Raman-scattered pump-light photons spectrally overlapping with the converted single photons could largely be avoided, as the pump-wavelength was chosen to be fairly longer than the target wavelength. Additional narrowband spectral filtering at the telecom regime as a result of the small conversion bandwidth and using a high-performance fiber-Bragg-grating solely left the detector dark counts as the only noise source in our setup. Therefore, we could achieve conversion efficiencies of more than 20 %. In order to test the indistinguishability, sequentially emitted photons were fed into a Mach-Zehnder interferometer and spatially as well as temporally overlapped at the output beam splitter. Cross-correlation measurements between both output-ports of the beam splitter exhibit two-photon interference contrasts of more than 40 % prior to and after the down-conversion step. Accordingly, we demonstrate that the process of quantum frequency conversion preserves photon indistinguishability and can be used to establish a versatile source of indistinguishable single photons at the telecom C-Band. Furthermore our scheme allows for converting photons in a wavelength band from 900 nm to 910 nm to the same telecom target wavelength. This enables us to test indistinguishability of frequency-converted photons, originally stemming from different sources with dinstinguishable wavelengths.
We present a quantum-mechanical model that describes fiber-based frequency conversion by four-wave-mixing Bragg scattering in the presence of Raman interactions. In the case of continuous-wave pumps, we find closed-form expressions for the conversion efficiency and photon statistics characterized by the second-order correlation function. For pulsed pumps, we derive a highly general model based on Green functions and provide a numerical solution method using a split-step scheme. In both cases, we find that noise from spontaneous Raman scattering can pose a serious challenge to this type of frequency conversion if the pumps are less than 30 THz from the quantum fields. However, this impact can be mitigated with cross-polarized pumps and, on the anti-Stokes side, through fiber cooling.