Broadband Fiber-based Entangled Photon Pair Source at Telecom O-band
Changjia Chen, Calvin Xu, Arash Riazi, Eric Y. Zhu, Alexey V. Gladyshev, Peter G. Kazansky, Li Qian
LLetter Optics Letters 1
Broadband Fiber-based Entangled Photon-Pair Sourceat Telecom O-band C HANGJIA C HEN , C
ALVIN X U , A RASH R IAZI , E RIC
Y. Z HU , A LEXEY
V.G
LADYSHEV , P ETER
G.K
AZANSKY , AND L I Q IAN Dept. of Electrical and Computer Engineering, University of Toronto, 10 King’s College Rd., Toronto, M5S 3G4, Canada Prokhorov General Physics Institute of the Russian Academy of Sciences, Dianov Fiber Optics Research Center, 38 Vavilov Street, 119333 Moscow, Russia Optoelectronics Research Centre, University of Southampton, Southampton SO17 1BJ, United Kingdom * Corresponding author: [email protected] February 26, 2021
In this letter, we report a polarization-entangledphoton-pair source based on type-II spontaneous para-metric down conversion at telecom O-band in periodi-cally poled silica fiber (PPSF). The photon-pair sourceexhibits more than 130 nm ( ∼
24 THz) emission band-width centered at 1306.6 nm. The broad emission spec-trum results in a short biphoton correlation time andwe experimentally demonstrate a Hong-Ou-Mandel in-terference dip with a full width of 26.6 fs at halfmaximum. Owing to the low birefringence of thePPSF, the biphotons generated from type-II SPDC arepolarization-entangled over the entire emission band-width, with a measured fidelity to a maximally entan-gled state greater than 95.4 % . The biphoton source pro-vides the broadest bandwidth entangled biphotons atO-band to our knowledge. © 2021 Optical Society of America http://dx.doi.org/10.1364/ao.XX.XXXXXX In the last few decades, the development of photon-pairsources has laid the foundation for quantum photonic tech-nologies. One of the most convenient methods of producingentangled photon pairs is spontaneous parametric downconver-sion (SPDC) in a second-order nonlinear medium. SPDC hasbeen extensively studied for its applications in quantum infor-mation processing and photonic quantum computation[1]. Byengineering the nonlinear medium, the photon pairs generatedby SPDC can be naturally endowed with nonclassical corre-lations, such as entanglement in polarization[2] or in orbitalangular momentum[3] degrees of freedom. The frequency-timecorrelated biphotons in SPDC sources are also of broad interest,as they enable quantum interferometry that surpasses the limitsof classical sensitivity[4].In particular, broad bandwidth and time-correlated photonpairs are desirable as they enable the investigation of the opti-cal properties of a sample over a wide spectral range, enhanc-ing the spatial and temporal resolution of quantum sensing[5].Broadband photon-pair sources are also significant for quantumcommunication techniques, such as wavelength-multiplexedentanglement distribution[6] and clock synchronization[7]. For example, it is crucial to use broadband time-correlated photonpairs for distant clock synchronization, as its precision is limitedby the biphoton correlation time[7]. However, in fiber-basedquantum communication where the biphotons are transmittedthrough long-distance fiber links, precise dispersion compensa-tion is needed because the fiber’s chromatic dispersion will ob-scure the timing correlation. A biphoton source centered near thezero-dispersion wavelength (around 1310 nm) of the telecom sil-ica fiber may resolve this problem and improve the precision oftiming[8]. Various entangled photon sources at telecom O-band(1260 - 1360 nm) with a biphoton emission bandwidth up to 70nm have been developed[9–16]. However, this bandwidth is stilllacking. Generating broadband and highly time-correlated pho-ton pairs via the process of SPDC remains a challenge. In orderto use SPDC to generate broadband, time-correlated biphotons,various methods have been employed, such as using a shortnonlinear medium[6] and chirped quasi-phase-matching[17],though at the expense of reduced biphoton generation rates andcomplex fabrication requirements. A time-correlated photon-pair source at O-band which has broader bandwidth and iscompatible with telecom fibers is still highly desired.In addition to the time-frequency degree of freedom, the po-larization correlation within the biphotons generated via SPDCalso attracts much interest. Broadband biphotons with polariza-tion entanglement can be used for polarization sensitive quan-tum sensing[18] and multi-user quantum communication[19].However generating polarization entangled photons over abroad bandwidth often requires stable compensation for thewavelength-dependent birefringence of the nonlinear medium[6,10]. Interferometric methods or compensating components aretherefore needed to achieve a high polarization entanglementquality[20], increasing the experimental difficulties.A promising solution is to use a nonlinear medium of suffi-cient interaction length but low birefringence and low chromaticdispersion to generate broadband entangled photon pairs . Inthis letter, we demonstrate a cw-pumped periodically poled sil-ica fiber (PPSF) based entangled photon-pair source quasi-phase-matched (QPM) at 1306.6 nm, which is near the zero-dispersionwavelength of telecom single-mode silica fiber (e.g. SMF-28).The low chromatic dispersion and the low birefringence of thePPSF at telecom O-band result in a biphoton emission band- a r X i v : . [ qu a n t - ph ] F e b etter Optics Letters 2 width greater than 130 nm ( >
24 THz in frequency), which to ourknowledge is the broadest bandwidth of polarization-entangledbiphotons at O-band. We demonstrate Hong-Ou-Mandel inter-ference with the biphoton source and measure an interferencedip width of 26.6 ± I ( ω s , ω i ) ∝ sinc (cid:20) L (cid:18) k A ( ω p ) − k B ( ω s ) − k C ( ω i ) − π Λ (cid:19)(cid:21) , (1) where k is the wavenumber of the PPSF, L is the length of thePPSF, subscripts p , s , i denote pump, signal and idler respectivelyand Λ is the QPM period. The angular frequencies of photonsobey energy conservation: ω s + ω i = ω p and we assume ω i > ω s . Subscripts A , B and C represent the polarization modes ofthe photons. In the context of this letter, we discuss the type-IISPDC in PPSF, in which a pair of orthogonally polarized photonsare generated by the down-conversion of a vertically polarizedpump photon, i.e. A = V [23]. Therefore, in type-II SPDC,subscripts B , C can be H , V or V , H , where H is for horizontallypolarized photons and V is for vertically polarized photons. Weassume the PPSF is cw-pumped and k A ( ω p ) = k V ( ω p ) ≡ k p ,thus the QPM condition becomes: k p − k H ( ω p /2 ) − k V ( ω p /2 ) − π Λ = (2) Taylor-expanding the wavenumbers k ( ω ) in Eq. (1) about ω p /2up to the second-order, and substituting Eq. (2) into Eq. (1) toeliminate the zeroth order terms, we obtain[24]: I ( ∆ ) ∝ sinc (cid:20) ML ∆ + k L ∆ + O ( ∆ ) (cid:21) , (3) where ∆ = ω p /2 − ω s = ω i − ω p /2, k = ∂ k ( ω ) / ∂ω | ω = ω p /2 is the chromatic dispersion at frequency ω p /2, M =( ∂ k H ( ω ) / ∂ω − ∂ k V ( ω ) / ∂ω ) | ω = ω p /2 is the group velocity mis-match between H and V polarizations, and O ( ∆ ) representsthe terms that have orders 3 or more. The width of I ( ∆ ) hasan inverse relation to M and k . If M and k are small enough,the spectral bandwidth of SPDC can be very broad. In general,the SPDC bandwidth in second-order nonlinear crystals suchas lithium niobate and potassium titanyl phosphate is signif-icantly limited by their high group birefringence M and non-trivial chromatic dispersion k . On the other hand, the groupbirefringence M in PPSF has been shown to be negligible[24]and the polarization entanglement of the biphotons in PPSF canbe naturally generated without the need of any birefringencecompensation[25]. As a result, the bandwidth of the biphotons inPPSF sources is mainly determined by its chromatic dispersion k , and ∆ ∝ √ k L . Taking advantage of its low dispersion atO-band, the PPSF is able to generate entangled photons withvery broad bandwidth. Fig. 1. (a) Experimentally measured and theoretically sim-ulated SHG spectra of the PPSF. The solid-line curves havetheir width based on the dispersion of the fiber, while theirheights are the best fit to the experimental data. The weak bire-fringence in PPSF results in the spectral separation of threepolarization-dependent phase-matchings. (b) Theoretical tun-ing curve of the type-II SPDC emission spectrum in PPSF withrespect to the pump wavelength. (c) Experimentally measuredbiphoton spectrum shows a biphoton emission bandwidthgreater than 130 nm ( >
24 THz).The PPSF we use in the source is a 20-cm-long, weakly bire-fringent step-index silica fiber with an NA ≈ µ m), which is compatible with telecom fibers suchas SMF-28[26]. It exhibits second-order nonlinearity which isinduced through thermal poling. The fabrication details of thePPSF are discussed in Ref.[27]. The QPM condition is achievedthrough periodic UV erasure with Λ = µ m . As shown by theexperimentally measured second harmonic generation (SHG)spectrum in Fig.1(a), the PPSF supports three types of SHG, andcorrespondingly three types of SPDC[23]. The conversion effi-ciency is estimated to be 0.70%/ W for type-0 SHG and 0.16%/ W for type-II SHG. The theoretical ratio of the three peaks (Type 0,I, II) should be 9:1:4[28]. The experimental type-II peak is a bitlower than 4/9 of the type 0 peak, which can be attributed to theimprecise polarization alignment of the fundamental light in theSHG measurement. Among the three types, the type-II SPDCin PPSF can be used for compensation-free broadband polariza-tion entanglement generation[24]. A finite element eigenmodesolver ( Lumerical Inc. ) is used for calculating the birefringenceand dispersion of the fiber modes based on the PPSF’s geometry.Assuming a cw-pump is used, we simulate the type-II SPDCspectrum and obtain the tuning curve as a function of the pumpwavelength, as shown in Fig.1(b). When the pump wavelengthfor SPDC is set at around the type-II SHG peak (653.3 nm), thePPSF can generate broadband biphotons with more than 130 nmbandwidth.The biphoton spectrum of type-II SPDC in PPSF is exper-imentally measured with a pump laser at 653.3 nm at room etter Optics Letters 3
Fig. 2. (a) Experimental setup for the Hong-Ou-Mandel interference Measurement. LD: cw laser diode with emission wavelengthat 653 nm. SPD, single photon detector (IDQ220, ID Quantique, quantum efficiency ∼
20% in O-band); TIA, time interval analyzer(Hydraharp 400). (b) Hong-Ou-Mandel interference shows a dip of 83.2 ± ± ∼ o C ), as shown in Fig.1(c). To measure thebiphoton spectrum, we first use a pair of fine tunable filters atO-band ( O/E land Inc. , tuning range 1260 nm -1360 nm). Outsidethe tuning range of the fine tunable filters, we use two coarsewavelength division multiplexers (CWDMs, each has a 3dBtransmission bandwidth of 17nm) centered at 1370 nm and 1390nm respectively. After calibrating the wavelength-dependentloss and the detector efficiencies, a spectral bandwidth of morethan 130 nm (>24 THz in frequency) of type-II SPDC is obtainedfrom the measured coincidence counts.In addition to the direct spectral brightness measurement, aHong-Ou-Mandel interference (HOMI) experiment is also per-formed, as one may use the temporal correlation[29] of the bipho-tons to infer their bandwidth. The HOMI experimental setup isshown in Fig.2(a). A laser diode that has its wavelength stabi-lized at 653.3 nm by a fiber Bragg grating is used as the pump,with its polarization adjusted to align to the slow axis (definedas the V polarization in PPSF[23]) of the PPSF at the fiber input.At the output end of the PPSF, fiber-pigtailed pump suppressionfilters reduce the pump power by >90 dB, at an insertion lossof 6 dB for the down-converted light. The biphotons then passthrough an interferometer which is composed of a polarizingbeam splitter, an inline controllable delay line (
General PhotonicsInc. MDL-001 ), a polarization controller, and a 50:50 beam split-ter. The coincidence rates are depicted in Fig.2(b). An HOMI dipof 26.6 ± ± λ s and λ i ,three different CWDM sets are used as wavelength splitters. TheCWDMs of 17 nm bandwidth in each set were cascaded such thatconjugated wavelengths near the center wavelengths (shown in Table 1.
The experimental results of the quantum state tomog-raphy: the concurrence[21] and the fidelity to a maximallypolarization-entangled biphoton state that are measured us-ing three CWDM sets at different wavelengths. Each of theCWDM filters has a top-hat transmission profile and a 3 dBbandwidth of 17nm.Wavelengths λ s , λ i Concurrence Fidelity1330 nm, 1290 nm 0.9570 ± ± ± ± * ± ± * See caption of Fig.3(d).
Table.1) would be transmitted. The biphoton polarization statesare analyzed by two polarization analyzers (PAs,
Hewlett Packard8169A ). Each PA includes one polarizer and two waveplates.The commercial PAs were designed for 1550 nm and thereforethe retardances of the waveplates were carefully calibrated forthe transmission wavelengths of the CWDMs at telecom O-band.The results of the QST are summarized in Table.1, which showsthe concurrences (without subtracting accidentals) of the bipho-ton states of each wavelength pair. The density matrices of thebiphotons states are shown in Fig.3(b)-(d). Neither the measure-ment apparatus nor the optical components of the source areoptimal (they are all designed for 1550nm except for the PPSF),yet we are able to obtain concurrences of more than 0.91 in thebiphoton emission band, even when the biphotons are morethan 120 nm apart.In summary, we have demonstrated a broadbandpolarization-entangled photon-pair source in PPSF by ex-ploiting type-II SPDC at telecom O-band. The cw-pumpedPPSF source emits highly time-correlated and high qual-ity polarization-entangled biphotons of more than 130 nmbandwidth. The concurrence of polarization entanglement ismeasured to be greater than 0.91 even when the signal-idleris greater than 120 nm apart. The O-band PPSF source can befurther optimized. For instance, the PPSF in the context of thisletter has a non-zero chromatic dispersion ( | k | ∼ ps / km ) atthe current QPM wavelength. The biphoton emission spectrumcan be potentially broadened to be more than a few hundredsof nanometers if the QPM wavelength is tuned to be closer tothe zero-dispersion wavelength of the PPSF. Moreover, using etter Optics Letters 4 Fig. 3. (a) Experimental setup for polarization state tomogra-phy. PC, polarization controller; PSF, pump suppression filters;CWDMs, coarse wavelength division multiplexers; WP, wave-plate; Pol., polarizer. (b)-(d) The real and imaginary part of thedensity matrices of the biphoton states that are measured us-ing CWDMs centered at (b) 1330 nm and 1290 nm (c) 1350 nmand 1270 nm; (d) 1370 nm, CWDM centered on 1250nm is notavailable to us. Instead of using a second CWDM at 1250nm,we used the refection port of the 1370nm CWDM, which has abroad reflection spectrum, for the idler’s measurement.better components such as pump suppression filters and beamsplitters with lower loss and better performance will surelyresult in higher entanglement quality as well.This simple fiber-based source is compatible with existingfiber infrastructure. Operating near the zero-dispersion wave-length of silica fiber, the time-correlated and polarization entan-gled biphotons generated in this source are expected to be robustagainst chromatic dispersion and polarization mode dispersionin long-distance quantum fiber communication[32–34]. Throughfurther development, we anticipate that the source can be usedin quantum applications such as wavelength-multiplexed quan-tum communication[19, 35, 36] and high precision quantumsensing[4, 5, 7, 37–39].
Funding.
Natural Sciences and Engineering Research Councilof Canada (RGPIN-2019-07019, RGPAS-2019-00113, CREATE-484907-16), Mitacs Globalink Research Award-Abroad, and USArmy Award W911NF20-2-0242. The U.S. Government has acopyright license in this work pursuant to a Cooperative Re-search and Development Agreement with University of Toronto
Disclosures.
The authors declare no conflicts of interest.
REFERENCES
1. S. Slussarenko and G. J. Pryde, Appl. Phys. Rev. , 041303 (2019).2. P. G. Kwiat, K. Mattle, H. Weinfurter, A. Zeilinger, A. V. Sergienko, andY. Shih, Phys. Rev. Lett. , 4337 (1995).3. Y. Zhang, F. S. Roux, T. Konrad, M. Agnew, J. Leach, and A. Forbes,Sci. advances , e1501165 (2016).4. F. Kaiser, P. Vergyris, D. Aktas, C. Babin, L. Labonté, and S. Tanzilli,Light. Sci. & Appl. , 17163 (2018).5. M. B. Nasr, B. E. A. Saleh, A. V. Sergienko, and M. C. Teich, Phys. Rev.Lett. , 083601 (2003). 6. H. C. Lim, A. Yoshizawa, H. Tsuchida, and K. Kikuchi, Opt. express ,16052 (2008).7. A. Valencia, G. Scarcelli, and Y. Shih, Appl. Phys. Lett. , 2655 (2004).8. J. A. Grieve, Y. Shi, H. S. Poh, C. Kurtsiefer, and A. Ling, Appl. Phys.Lett. , 131106 (2019).9. T. Zhong, F. N. C. Wong, T. D. Roberts, and P. Battle, Opt. Express ,12019 (2009).10. T. Zhong, X. Hu, F. N. C. Wong, K. K. Berggren, T. D. Roberts, andP. Battle, Opt. Lett. , 1392 (2010).11. A. Martin, V. Cristofori, P. Aboussouan, H. Herrmann, W. Sohler, D. Os-trowsky, O. Alibart, and S. Tanzilli, Opt. Express , 1033 (2009).12. A. Martin, A. Issautier, H. Herrmann, W. Sohler, D. B. Ostrowsky,O. Alibart, and S. Tanzilli, New J. Phys. , 103005 (2010).13. M. T. Liu and H. C. Lim, Opt. Express , 23261 (2014).14. Y. Shi, S. M. Thar, H. S. Poh, J. A. Grieve, C. Kurtsiefer, and A. Ling,Appl. Phys. Lett. , 124002 (2020).15. Z.-H. Xiang, J. Huwer, J. Skiba-Szymanska, R. M. Stevenson, D. J. P.Ellis, I. Farrer, M. B. Ward, D. A. Ritchie, and A. J. Shields, Commun.Phys. (2020).16. M. A. Hall, J. B. Altepeter, and P. Kumar, Opt. Express , 14558(2009).17. M. B. Nasr, S. Carrasco, B. E. A. Saleh, A. V. Sergienko, M. C. Teich,J. P. Torres, L. Torner, D. S. Hum, and M. M. Fejer, Phys. Rev. Lett. , 183601 (2008).18. M. C. Booth, B. E. Saleh, and M. C. Teich, Opt. Commun. , 2542(2011).19. S. K. Joshi, D. Aktas, S. Wengerowsky, M. Lonˇcari´c, S. P. Neumann,B. Liu, T. Scheidl, G. C. Lorenzo, Ž. Samec, L. Kling, A. Qiu, M. Razavi,M. Stipˇcevi´c, J. G. Rarity, and R. Ursin, Sci. Adv. , eaba0959 (2020).20. T. Kim, M. Fiorentino, and F. N. C. Wong, Phys. Rev. A , 012316(2006).21. W. K. Wootters, Phys. Rev. Lett. , 2245 (1998).22. A. Christ, A. Fedrizzi, H. Hübel, T. Jennewein, and C. Silberhorn, Single-Photon Generation and Detection: Chapter 11. ParametricDown-Conversion , vol. 45 (Elsevier Inc. Chapters, 2013).23. E. Y. Zhu, Z. Tang, L. Qian, L. G. Helt, M. Liscidini, J. E. Sipe, C. Corbari,A. Canagasabey, M. Ibsen, and P. G. Kazansky, Phys. Rev. Lett. ,213902 (2012).24. C. Chen, E. Y. Zhu, A. Riazi, A. V. Gladyshev, C. Corbari, M. Ibsen,P. G. Kazansky, and L. Qian, Opt. Express , 22667 (2017).25. E. Y. Zhu, Z. Tang, L. Qian, L. G. Helt, M. Liscidini, J. E. Sipe, C. Corbari,A. Canagasabey, M. Ibsen, and P. G. Kazansky, Opt. Lett. , 4397(2013).26. Corning, Corning SMF-28 Optical Fibers .27. A. Canagasabey, C. Corbari, A. V. Gladyshev, F. Liegeois, S. Guillemet,Y. Hernandez, M. V. Yashkov, A. Kosolapov, E. M. Dianov, M. Ibsen,and P. G. Kazansky, Opt. Lett. , 2483 (2009).28. E. Y. Zhu, L. Qian, L. G. Helt, M. Liscidini, J. E. Sipe, C. Corbari,A. Canagasabey, M. Ibsen, and P. G. Kazansky, Opt. Lett. , 1530(2010).29. C. K. Hong, Z. Y. Ou, and L. Mandel, Phys. Rev. Lett. , 2044 (1987).30. A. M. Bra´nczyk, “Hong-ou-mandel interference,” (2017).31. D. F. V. James, P. G. Kwiat, W. J. Munro, and A. G. White, Phys. Rev. A , 052312 (2001).32. D. E. Jones, B. T. Kirby, and M. Brodsky, npj Quantum Inf. (2018).33. M. Brodsky, E. C. George, C. Antonelli, and M. Shtaif, Opt. Lett. , 43(2010).34. M. Shtaif, C. Antonelli, and M. Brodsky, Opt. Express , 1728 (2011).35. S. Wengerowsky, S. K. Joshi, F. Steinlechner, H. Hübel, and R. Ursin,Nature , 225 (2018).36. E. Y. Zhu, C. Corbari, A. Gladyshev, P. G. Kazansky, H.-K. Lo, andL. Qian, J. Opt. Soc. Am. B , B1 (2019).37. A. Riazi, C. Chen, E. Y. Zhu, A. V. Gladyshev, P. G. Kazansky, J. E.Sipe, and L. Qian, npj Quantum Inf. , 1 (2019).38. D. A. Kalashnikov, A. V. Paterova, S. P. Kulik, and L. A. Krivitsky, Nat.Photonics , 98 (2016).39. A. Riazi, E. Y. Zhu, C. Chen, A. V. Gladyshev, P. G. Kazansky, andL. Qian, Opt. letters44