Single-photon sources based on spontaneous photon-pair generation have enabled pioneering experiments in quantum physics. However, next-generation photonic quantum technologies require higher generation probabilities of photons in well-controlled pure states capable of high-visibility interference. We have harnessed bespoke fiber technology to develop virtually alignment-free sources that deliver high-purity heralded single photons in telecoms single-mode fiber. The resulting access to low-loss optical delay enabled us to actively route the heralded output from two almost identical sources to enhance the delivery probability of single photons relative to one individual source. Our results indicate how the scale of photonic quantum technologies might be increased via guided-wave multiplexing of high-purity photons.
© 2016 Optical Society of America
Single photons are a vital resource in photonic quantum-enhanced technologies for information processing, communications, and measurements . Spontaneous photon-pair sources, in which signal and idler photon pairs are generated as pump laser pulses propagate through a nonlinear medium, have established themselves as the workhorses of nonclassical light generation due to their high performance and relative simplicity [2,3]. Although the generation mechanisms of parametric downconversion (PDC) and four-wave mixing (FWM) are probabilistic, the delivery of a single photon can be heralded by detecting its twin. Whether or not the ultimate performance of these sources exceeds that of single-emitter photon sources such as quantum dots [4,5], heralded single-photon sources are essential for the rapid deployment of quantum technologies.
Nevertheless, scaling up devices such as photonic quantum simulators requires an increase in the number of heralded single photons that can be delivered simultaneously by independent sources. PDC and FWM produce a state with a statistical distribution of photons per mode that is thermal. Although a successful heralding detection removes the zero-photon component, the probability of generating a pair must be kept low to limit detrimental contributions from pairs in the same mode, which occurs with probability of order . Typical operation at is tolerable for few-photon demonstrations of quantum effects; however, the waiting time for larger -photon devices increases exponentially with , precluding operation with more than a few single photons. Even in the ideal case, source statistics impose a limit of for delivering one photon into a single mode .
Multiplexed single-photon sources incorporating active routing have the potential to solve this problem. As depicted in Fig. 1, delaying the output of several generation modes allows feed forward to a switch network that routes one successful mode to a master output conditioned on heralding detections [8–10]. This enables the probability of delivering a single photon to be enhanced without a commensurate increase in multi-pair generation, and deterministic operation can be approached in the limit of low switch loss . Any photonic degree of freedom can be exploited, including spatial [11,12] or temporal mode [13,14], polarization , or frequency [16,17], either individually or in combination [18–20]. Active multiplexing has been demonstrated using PDC in bulk crystals [15,21] and waveguides  as well as FWM in integrated photonic crystal  and nanowire waveguides , and recently a delivery probability above the single-source limit was achieved using bulk optics .
Despite these advantages, the single photons that are delivered by multiplexed sources are useful only if they are capable of high-visibility Hong–Ou–Mandel interference; hence they must be identical and in pure quantum states . Typically, energy conservation produces frequency anti-correlation between signal and idler, resulting in heralded photons that vary randomly from shot to shot. The resulting mixed state can be cleaned up with narrowband spectral filtering, but at the cost of significant loss and with a purity that approaches unity only as the filter bandwidth tends to zero . Alternatively, by tailoring the dispersion of the nonlinear medium so that signal, pump, and idler group velocities satisfy , respectively, frequency correlation can be minimized at the point of generation yielding a high-purity state directly upon heralding [27,28]. This has been demonstrated in a number of systems including bulk nonlinear crystals  and waveguides , as well as various types of fiber [31–33]. However, as the technique cannot use narrowband filters, it is yet to be implemented in a fully guided-wave architecture due to the difficulty of removing noise photons. Furthermore, no existing multiplexed source produces heralded photons directly in high-purity states without filtering.
We have developed a single-photon source that implements both high-purity heralded photon generation and active multiplexing simultaneously in a spliced fiber architecture. All essential elements—nonlinear material with dispersion control, wavelength isolation, optical delay, and fast switching—are incorporated in an almost alignment-free package that heralds telecoms-band photons directly in single-mode fiber (SMF).
To achieve this we first designed sources based on FWM in photonic crystal fiber (PCF) dispersion-engineered to minimize frequency correlation between signal and idler by group-velocity matching when pumped at 1064 nm. The two-photon component of the generated state can be written
We designed and fabricated PCF to generate the simulated joint spectrum shown in Fig. 2(b), corresponding to . We characterized the FWM joint spectral intensity, , with bright light using a process known as stimulated emission tomography . Seeding photon-pair generation processes with a tunable narrowband laser at produces a bright signal whose spectral intensity [35,36]. We built a setup capable of rapid measurements of joint spectra as shown in Fig. 2(a). A 1450–1610 nm tunable CW seed laser was coupled into lengths of PCF along with the pump pulses and the corresponding signal spectra recorded to build up a classical analogue of the joint spectrum. This rapid measurement method enabled us to mitigate the effect of small variations in the PCF structure on the FWM joint spectra, by both selecting sections of PCF with the greatest overlap and identifying the length with the highest level of factorability. The results for one source are shown in Fig. 2(b). The Schmidt decomposition of placed an upper bound on the purity of the heralded single photons of . The spectral overlap between the two sections of PCF was 95%.
Two sections of PCF were built into fully fiber-integrated sources whose design is shown in Fig. 3(a). The PCF was pumped with a 1064 nm, 10 MHz modelocked fiber laser (Fianium FemtoPower 1060-PP) to produce photon pairs at highly nondegenerate wavelengths: 810 nm to enable efficient heralding with silicon detectors and 1550 nm to access the low-loss telecoms window. Fiber Bragg gratings (FBGs) (3 dB bandwidth 28 nm at 1064 nm) and a custom fiber wavelength division multiplexer (WDM) separated the daughter photons and provided more than 90 dB of pump attenuation in the signal and idler channels.
To isolate the signal and idler from noise outside the FBG rejection band (for example, from Raman scattering and unwanted FWM processes) while also preserving the spectrum of the photon pairs, we developed broadband transmission filters based on a solid photonic band gap fiber (PBGF). We fabricated two PBGFs with periodic claddings of high-index germanium-doped rods with photonic band gaps to create low-loss fiber transmission bands centered at 800 nm (PBGF-800) and 1550 nm (PBGF-1550) and high loss for out-of-band noise photons. Due to the large core diameter of PBGF-1550, custom fiber tapers were fabricated using the flame-brush technique to mode-match efficiently with telecoms SMF and minimize loss. Hence the outputs at 800 nm and 1550 nm were delivered into conventional SMF (SM-800 and SMF-28, respectively). The total attenuation through all source components following the PCF was measured by cut-back to be at 800 nm and at 1550 nm [Fig. 3(c)]. We estimate that these could be reduced to approximately (see Supplement 1).
We measured the marginal second-order coherence of the unheralded 1550 nm output, , which is related to the heralded single-photon purity by . For an ideal single-mode state with thermal photon statistics , giving a heralded single-photon purity of ; as the number of Schmidt modes and frequency correlation increases the statistics become Poissonian and . The results illustrated in Fig. 3(d) demonstrate that matching the pump bandwidth with the phasematching function to minimize spectral correlation yielded a maximum value of corresponding to . The reduction in the heralded purity from the limit of imposed by the joint spectrum is primarily due to background counts from the pump and unwanted nonlinear processes. Imperfect isolation adds a Poissonian background with to the FWM state that would otherwise display statistics closer to a thermal distribution.
To multiplex the high-purity heralded photons, the output of each source was delayed by 200 ns using 42 m of SMF-28 as shown in Fig. 4(a). The delay lines were spliced to a fiber-coupled optical switch (bandwidth 1 MHz, measured insertion loss 0.94 dB at 1550 nm) to route the heralded photons to a single output via an in-line fiber polarizer (insertion loss 0.6 dB). One pump beam contained an optical delay to match arrival times between idler photons from each source. The switch was controlled by a field programmable gate array (FPGA) conditioned on heralding detection events, and the switch output was monitored by an InGaAs single-photon detector.
We quantified the source performance by measuring single and coincidence count rates after the output of the switch. We evaluated the coincidence-to-accidentals ratio (CAR; see Supplement 1) with and without multiplexing, and measured the heralded second-order coherence, , by splicing a fiber coupler to the output. The benefit of multiplexing can be seen clearly from the results in Figs. 4(b) and 4(c). At a CAR of 40, the fitted data show that activating the switch improved the heralded single-photon rate relative to each individual source by a factor of 1.86, close to the ideal value of 2. At this point the multiplexed coincidence count rate was with a heralded , whereas the individual sources returned to achieve the same count rates. Hence our results demonstrate that the multiplexed source achieves an increase in the probability of delivering a single photon without a corresponding increase in the probability of multi-photon noise. Furthermore, multiplexing remains advantageous when taking into account the effect of switch insertion loss on the individual sources; removing the static switch insertion loss of 0.94 dB from the single-source data yields an improvement factor of 1.50 at a CAR of 40. Comparing the multiplexed source to the mean performance of the individual sources across the range over which full coincidence count rate data was available yielded a mean improvement factor of 1.75 (1.41 with switch loss removed).
In conclusion, by using advanced fiber technology we have successfully demonstrated an improvement in performance by multiplexing two sources that deliver high-purity photons in an all-fiber architecture. Low average pump power is required per fiber enabling additional sources to be multiplexed from the same pump laser, and a switchable delay network would allow time-domain multiplexing to combine the output from consecutive pump pulses. Hence our unique source requiring minimal alignment and no narrow spectral filtering is amenable to multiplexing larger numbers of modes, providing a route to high-performance single-photon sources capable of delivering large numbers of independent photons simultaneously. All data underlying the results presented in this Letter can be found at .
UK Engineering and Physical Sciences Research Council (EPSRC) (EP/K022407/1, EP/M013243/1).
We thank S. Yerolatsitis and T. A. Birks for fiber tapering.
See Supplement 1 for supporting content.
1. T. D. Ladd, F. Jelezko, R. Laflamme, Y. Nakamura, C. Monroe, and J. L. O’Brien, Nature 464, 45 (2010). [CrossRef]
2. D. C. Burnham and D. L. Weinberg, Phys. Rev. Lett. 25, 84 (1970). [CrossRef]
3. X. Li, J. Chen, P. Voss, J. Sharping, and P. Kumar, Opt. Express 12, 3737 (2004). [CrossRef]
4. N. Somaschi, V. Giesz, L. De Santis, J. C. Loredo, M. P. Almeida, G. Hornecker, S. L. Portalupi, T. Grange, C. Anton, J. Demory, C. Gomez, I. Sagnes, N. D. Lanzillotti Kimura, A. Lemaitre, A. Auffeves, A. G. White, L. Lanco, and P. Senellart, Nat. Photonics 10, 340 (2016). [CrossRef]
5. X. Ding, Y. He, Z.-C. Duan, N. Gregersen, M.-C. Chen, S. Unsleber, S. Maier, C. Schneider, M. Kamp, S. Höfling, C.-Y. Lu, and J.-W. Pan, Phys. Rev. Lett. 116, 020401 (2016). [CrossRef]
6. M. A. Broome, M. P. Almeida, A. Fedrizzi, and A. G. White, Opt. Express 19, 22698 (2011). [CrossRef]
7. A. Christ and C. Silberhorn, Phys. Rev. A 85, 023829 (2012). [CrossRef]
8. A. L. Migdall, D. Branning, and S. Castelletto, Phys. Rev. A 66, 053805 (2002). [CrossRef]
9. T. B. Pittman, B. C. Jacobs, and J. D. Franson, Phys. Rev. A 66, 042303 (2002). [CrossRef]
10. E. Jeffrey, N. A. Peters, and P. G. Kwiat, New J. Phys. 6, 100 (2004). [CrossRef]
11. L. Mazzarella, F. Ticozzi, A. V. Sergienko, G. Vallone, and P. Villoresi, Phys. Rev. A 88, 023848 (2013). [CrossRef]
12. D. Bonneau, G. J. Mendoza, J. L. O’Brien, and M. G. Thompson, New J. Phys. 17, 043057 (2015). [CrossRef]
13. C. T. Schmiegelow and M. A. Larotonda, Appl. Phys. B 116, 447 (2014). [CrossRef]
14. P. P. Rohde, L. G. Helt, M. J. Steel, and A. Gilchrist, Phys. Rev. A 92, 053829 (2015). [CrossRef]
15. X.-S. Ma, S. Zotter, J. Kofler, T. Jennewein, and A. Zeilinger, Phys. Rev. A 83, 043814 (2011). [CrossRef]
16. R. Kumar, J. R. Ong, J. Recchio, K. Srinivasan, and S. Mookherjea, Opt. Lett. 38, 2969 (2013). [CrossRef]
17. C. Joshi, A. Farsi, S. Ramelow, S. Clemmen, and A. L. Gaeta, in Conference on Lasers and Electro-Optics (2016), paper FTh1C.2.
18. B. L. Glebov, J. Fan, and A. Migdall, Appl. Phys. Lett. 103, 031115 (2013). [CrossRef]
19. P. Adam, M. Mechler, I. Santa, and M. Koniorczyk, Phys. Rev. A 90, 053834 (2014). [CrossRef]
20. X. Zhang, I. Jizan, J. He, A. S. Clark, D.-Y. Choi, C. J. Chae, B. J. Eggleton, and C. Xiong, Opt. Lett. 40, 2489 (2015). [CrossRef]
21. G. J. Mendoza, R. Santagati, J. Munns, E. Hemsley, M. Piekarek, E. Martín-López, G. D. Marshall, D. Bonneau, M. G. Thompson, and J. L. O’Brien, Optica 3, 127 (2016). [CrossRef]
22. T. Meany, L. A. Ngah, M. J. Collins, A. S. Clark, R. J. Williams, B. J. Eggleton, M. J. Steel, M. J. Withford, O. Alibart, and S. Tanzilli, Laser. Photon. Rev. 8, L42 (2014). [CrossRef]
23. M. J. Collins, C. Xiong, I. H. Rey, T. D. Vo, J. He, S. Shahnia, C. Reardon, T. F. Krauss, M. J. Steel, A. S. Clark, and B. J. Eggleton, Nat Commun. 4, 2582 (2013). [CrossRef]
24. C. Xiong, X. Zhang, Z. Liu, M. J. Collins, A. Mahendra, L. G. Helt, M. J. Steel, D. Y. Choi, C. J. Chae, P. H. W. Leong, and B. J. Eggleton, Nat. Commun. 7, 10853 (2016).
25. F. Kaneda, B. G. Christensen, J. J. Wong, H. S. Park, K. T. McCusker, and P. G. Kwiat, Optica 2, 1010 (2015). [CrossRef]
26. J. W. Silverstone, D. Bonneau, K. Ohira, N. Suzuki, H. Yoshida, N. Iizuka, M. Ezaki, C. M. Natarajan, M. G. Tanner, R. H. Hadfield, V. Zwiller, G. D. Marshall, J. G. Rarity, J. L. O’Brien, and M. G. Thompson, Nat. Photonics 8, 104 (2013). [CrossRef]
27. W. P. Grice, A. B. U’Ren, and I. A. Walmsley, Phys. Rev. A 64, 063815 (2001). [CrossRef]
28. K. Garay-Palmett, H. J. McGuinness, O. Cohen, J. S. Lundeen, R. Rangel-Rojo, A. B. U’Ren, M. G. Raymer, C. J. McKinstrie, S. Radic, and I. A. Walmsley, Opt. Express 15, 14870 (2007). [CrossRef]
29. P. J. Mosley, J. S. Lundeen, B. J. Smith, P. Wasylczyk, A. B. U’Ren, C. Silberhorn, and I. A. Walmsley, Phys. Rev. Lett. 100, 133601 (2008). [CrossRef]
30. A. Eckstein, A. Christ, P. J. Mosley, and C. Silberhorn, Phys. Rev. Lett. 106, 013603 (2011). [CrossRef]
31. M. Halder, J. Fulconis, B. Cemlyn, A. Clark, C. Xiong, W. J. Wadsworth, and J. G. Rarity, Opt. Express 17, 4670 (2009). [CrossRef]
32. O. Cohen, J. S. Lundeen, B. J. Smith, G. Puentes, P. J. Mosley, and I. A. Walmsley, Phys. Rev. Lett. 102, 123603 (2009). [CrossRef]
33. C. Söller, O. Cohen, B. J. Smith, I. A. Walmsley, and C. Silberhorn, Phys. Rev. A 83, 031806 (2011). [CrossRef]
34. M. Liscidini and J. E. Sipe, Phys. Rev. Lett. 111, 193602 (2013). [CrossRef]
35. A. Eckstein, G. Boucher, A. Lematre, P. Filloux, I. Favero, G. Leo, J. E. Sipe, M. Liscidini, and S. Ducci, Laser Photon. Rev. 8, L76 (2014). [CrossRef]
36. B. Fang, O. Cohen, M. Liscidini, J. E. Sipe, and V. O. Lorenz, Optica 1, 281 (2014). [CrossRef]