On-chip generation and demultiplexing of quantum correlated photons using a silicon-silica monolithic photonic integration platform
Nobuyuki Matsuda, Peter Karkus, Hidetaka Nishi, Tai Tsuchizawa, William J. Munro, Hiroki Takesue, Koji Yamada
OOn-chip generation and demultiplexing of quantum correlated photons using asilicon-silica monolithic photonic integration platform
Nobuyuki Matsuda,
1, 2, ∗ Peter Karkus, Hidetaka Nishi,
1, 3
Tai Tsuchizawa,
1, 3
William J. Munro, Hiroki Takesue, and Koji Yamada
1, 3 NTT Nanophotonics Center, NTT Corporation, Atsugi, Kanagawa 243-0198, Japan NTT Basic Research Laboratories, NTT Corporation, Atsugi, Kanagawa 243-0198, Japan NTT Device Technology Laboratories, NTT Corporation, Atsugi, Kanagawa 243-0198, Japan
We demonstrate the generation and demultiplexing of quantum correlated photons on a monolithicphotonic chip composed of silicon and silica-based waveguides. Photon pairs generated in a nonlin-ear silicon waveguide are successfully separated into two optical channels of an arrayed-waveguidegrating fabricated on a silica-based waveguide platform.
I. INTRODUCTION
Integrated waveguide technology has proven useful forthe large-scale integration of quantum information sys-tems on photonic chips [1], thanks to its compactnessand circuit stability. In this context, intense study isunder way on the development of on-chip quantum com-ponents, such as quantum processing circuits [2–8], quan-tum light sources [9–21], and single photon detectors [22–24]. Moreover, to fully exploit the advantages of inte-grated photonics, it is ideal to integrate these differentcomponents on a single substrate. Motivated by thisgoal, several researchers have recently demonstrated thehybrid [25] or monolithic [26–28] integration of differentquantum-optical components.The quantum light sources include photon pair sources,which can serve as entangled photon pairs or heraldedsingle photons. A photon pair source can be realized byemploying nonlinear wave mixing in integrated waveg-uides such as silicon wire waveguides [9–11, 14, 15, 20].Quantum circuits can also be realized by using integratedwaveguides with cores made of Si [5], GaAs [7] or silica-based materials [2–4, 6, 8]. Of these approaches, silica-based waveguide technology has realized planar lightwavecircuits with a significantly large scale for classical opti-cal communication [29, 30]; this capability will facilitatethe construction of large-scale quantum circuits in thenear future. In addition, the low nonlinearity of silica[31] helps us to avoid the generation of unwanted pho-tons by the intense pump fields used for photon pair gen-eration in the quantum light sources. To exploit theseadvantages, the integration of quantum light sources andsilica waveguides is an attractive technology with whichto construct on-chip quantum information systems. Asignificant step in this direction is the integration of aphoton pair source with its interface, namely a photon-pair demultiplexer, for direct connection to a quantumcircuit.In this paper, we demonstrate the monolithic inte-gration of a Si waveguide photon pair source and a ∗ [email protected] photon-pair demutiplexer employing a silica-based ar-rayed waveguide grating (AWG). In Sec. 2, we inves-tigate the photon pair generation property of the mono-lithic waveguide platform. By performing experimentsusing waveguides of various lengths, we confirm that thecontribution of our silica-based waveguide to unnecessaryphoton generation is negligible. In Sec. 3, we demon-strate the on-chip generation and demultiplexing of acorrelated photon pairs using a monolithic circuit con-sisting of a Si-wire photon pair source and a silica-basedAWG. We show that the chip is capable of generatingquantum correlated photons and guiding them into dif-ferent output ports II. PHOTON PAIR GENERATIONPROPERTIES ON A SILICON-SILICAMONOLITHIC WAVEGUIDE PLATFORM
Figure 1(a) is a schematic diagram of a monolithicwaveguide platform made of Si and silicon-rich silica(SiO x ) [32]. We first fabricate a silicon wire rib waveguideby electron-beam lithography and electron-cyclotron res-onance (ECR) plasma etching on a silicon-on-insulator(SOI) substrate. The SiO x waveguides are fabricatedin a region from which the top Si layer of SOI sub-strate has been removed by reactive ion etching (RIE). Inthat region, the SiO x layer is deposited by ECR plasma-enhanced chemical vapor deposition (PE-CVD) at lowtemperature. Then the cores of the SiO x waveguidesare fabricated by photolithography and RIE. Finally, theSiO layer is deposited by ECR PE-CVD. The SiO x waveguides have a core-cladding index contrast of ∼ x waveguides. The cross-sectional dimensions ofeach waveguide are shown in Fig. 1(b). We use sev-eral devices with different combinations of Si and SiO x waveguide lengths, L Si and L SiO x , as shown in the tablein Fig. 1(a). The propagation losses of the Si and SiO x waveguides, α Si and α SiO x , are estimated by the cut-backmethod to be approximately – 2.1 and – 1.8 dB/cm forthe fundamental transverse-electric (TE) mode, respec-tively. a r X i v : . [ qu a n t - ph ] S e p Si rib wire waveguide, (Nonlinear photon pair source) SiO x waveguide, (Passive circuit) Spot-size converters SiO undercladding (a)
600 nm 100 nm 100 nm Si core
SiO x SiO (BOX) m m SiO x core SiO (BOX) SiO m m (b) SiO x L Si L Waveguides (i) (ii) (iii) (iv) (v) (vi) 0.86 2.37
SiO x L Si L (units in cm) FIG. 1. (a) Si-silica monolithic waveguide platform for anintegrated photonic quantum information system. L Si and L SiO x are the lengths of each waveguide section. The tableshows the waveguide length combinations used in the exper-iment. (b) Cross-sectional views of each waveguide section.The silica overcladding and Si substrate are omitted for clar-ity. IM EDFA Waveguide chip
BPF l p =1551.1 nm CW laser FBG l s =1546.4 nm l i =1556.0 nm s i WDM filter
SPCMs
Time- interval analyzer
FIG. 2. Experimental setup for photon-pair generation exper-iment of Si-silica monolithic waveguide chip. IM: LiNbO in-tensity modulator, EDFA: erbium-doped fiber amplifier, BPF:band-pass filter, WDM filter: wavelength-division multiplex-ing filter, SPCM: single-photon counting module. To investigate the nonlinearity in the monolithicwaveguide, we undertake a photon pair generation exper-iment using the device with a cascaded structure consist-ing of a Si and a SiO x waveguide shown in Fig. 1(a). Theexperimental setup is depicted in Fig. 2. The LiNbO intensity modulator (IM) modulates a continuous beamfrom the light source operating at a wavelength λ p of1551.1 nm into a train of pump pulses with a temporalfull-width at half maximum (FWHM) ∆ t of 200 ps anda repetition rate R of 100 MHz. The pulses are ampli-fied by an erbium-doped fiber amplifier (EDFA), filteredwith a band-pass filter BPF (3-dB bandwidth: 0.2 nm)to eliminate amplified spontaneous emission noise, andthen launched into the waveguides with a lensed fiberfrom the Si waveguide side. The input polarization is setat the fundamental TE mode. The in- and out-couplingefficiency with the chip, η couple , is approximately – 1dB/facet.In nonlinear waveguides such as a Si waveguide, a cor-related pair of signal and idler photons are created via a χ (3) spontaneous four-wave mixing (SFWM) process fol-lowing the annihilation of two photons inside the pumppulse [9, 10]. The χ (3) nonlinearity of Si is 200 timeshigher than that of silica. In addition, the core areaof a Si waveguide is approximately two orders of mag- nitude smaller than that of our SiO x waveguide. Thuswe can expect the Si part to play a major role in cor-related photon pair generation. However, this should beconfirmed experimentally since SiO x has a material com-position different from that of standard silica. We shouldalso investigate the noise photon generation characteris-tics in the SiO x waveguide, since the Raman scatteredphotons in fused silica waveguides such as optical fibers[33, 34] are a potential source of noise.The optical fields output from the side of the SiO x waveguide including the correlated photons were col-lected by another lensed fiber. Then, the light was intro-duced into the fiber-Bragg grating (FBG) filter and thewavelength-division multiplexing (WDM) filter, whichseparated the signal and idler photons into different fiberchannels. Here the total pump-wavelength suppression ofthe FBG and WDM filters exceeds 130 dB. Each outputport of the WDM filter has center wavelengths of 1546.4nm ( λ s ) and 1556.0 nm ( λ i ) with a passband width ∆ ν of 0.12 THz (0.96 nm). Finally, the photons are receivedby avalanche-photodiode-based single photon countingmodules (SPCMs) (id210, id Quantique) that operatedat a gate frequency of 100 MHz synchronized with thepump repetition rate R . The quantum efficiency η QE ,gate width, dark count rate d , and dead time of the de-tectors were 21 %, 1.0 ns, 2.1 kHz, and 10 µ s, respec-tively. The overall transmittance of the filtration system η f is approximately – 3.8 dB. The raw coincidence rate D c (including the accidental coincidence count) and theraw accidental coincidence rate D c , a were determined bymeasuring the time correlation of the signals output fromthe two SPCMs using a time-interval analyzer.Figures 3(a) and 3(b) are the net photon pair gen-eration rate (at the waveguide output end of the SiO x waveguide side) as a function of the waveguide length.Fig. 3(a) shows the L SiO x dependence obtained usingwaveguides (i), (v), (vi), whereas Fig. 3(b) is the L Si dependence with waveguides (i) to (iv). We used threewaveguides for each length condition. Here we estimatedthe net photon pair generation rate at the end of theSiO x waveguide, µ c , using µ c = D c − D c , a Rη , (1)where η total = η couple η f η QE η gate with η gate being the ra-tio of the active gates to the 100 MHz clock rate [20].The number of active detector gates decreases due tothe finite detector dead time set in our experiment. Themeasurement time was 60 s for each data point to obtaingood statistics.In more detail, Fig. 3(a) shows that the pair genera-tion rate decreases monotonically with increases in L SiO x ,in contrast to the L Si dependence shown in Fig. 3(b).This strongly suggests that the contribution of the SiO x waveguide as a photon pair source is negligible. Assum-ing that the SiO x waveguide is a passive transmission lineto the photon pairs generated in the Si waveguide section, µ c should follow µ c ∝ η x , where η SiO x = e − α SiO x L SiO x (i) (v) (vi) (a) (b) (d) (c) SiO x waveguide length (cm) SiO x L N e t pho t on pa i r gene r a t i on r a t e m c ( / pu l s e ) Si waveguide length (cm) Si L N e t pho t on pa i r gene r a t i on r a t e m c ( / pu l s e ) -4 -3 -2 -1 (i) (ii) (iii) (iv) (v) (vi) waveguide N e t s i ng l e c oun t s i n t he s i gna l c hanne l m s ’ ( / pu l s e ) Coupled pump peak power P p (mW) waveguide Coupled pump peak power P p (mW) C A R FIG. 3. Photon pair generation rate as a function of (a) SiO x and (b) Si length, respectively. The pump peak power P p isfixed at 37 mW. The dashed curves show the fitting resultsas described in the text. (c) The net photon generation rateinside the wavelength band for the signal photons at the endof the Si waveguide versus P p . (d) Coincidence to acciden-tal coincidence ratio (CAR) values as a function of P p . Thecurves show the results of numerical calculations as describedin the text. is the transmittance in the SiO x waveguide. A fittedfunction with this relation is shown as a dashed curve inFig. 3(b). The experimental results are well explained bythe fitting. From the fitting we obtained an α SiO x valueof – 2.4 dB/cm, which is similar to the value obtainedwith the cut-back method. These results suggest thatthe SiO x waveguide works as a passive circuit withoutthe creation of a significant number of unwanted photonpairs inside it.On the other hand, in Fig. 3(b) we find that µ c in-creases with increases in L Si , and then starts to decreasein the L Si > µ (cid:48) c can be written as µ (cid:48) c = ∆ ν ∆ t ( γP p L eff ) η , (2)where γ is the nonlinear constant of the Si waveguide, P p = P/ ( R ∆ t ) is the coupled pump peak power, and L eff is the effective Si waveguide length associated with L eff = − e − α Si L Si α . η Si = e − α Si L Si is the linear transmittanceof the Si waveguide that causes the intrinsic loss of thephoton pairs [35, 36]. Here we assumed that α Si has nowavelength dependence. The dashed curve in Fig. 3(b)shows a fitted function using µ c = η x µ (cid:48) c and Eq. (2)with a fitting parameters γ = 161 /W/m and α Si = 2.0dB/cm. The experimental data are well explained by thefitting. The γ value obtained from the fitting is slightlylower than that of channel-type Si wire waveguides [10,11]. This is because the present rib-type waveguide has a larger effective mode area than a channel waveguide.Noise photons, such as Raman scattered photons, cancontaminate wavelength channels for the correlated pho-tons. To investigate the generation of noise photons inour SiO x waveguide, we plot single count rate µ (cid:48) s as afunction of P p in Fig. 3(c). Here µ (cid:48) s is the photon gen-eration rate in the signal wavelength channel at the endof the Si waveguide part (before the SiO x waveguide) es-timated by µ (cid:48) s = N s Rη total η SiO x , where N s is the raw singlecount rate measured by the SPCM set in the signal chan-nel excluding the dark count rate of the detector. FromFig. 3(c), we find that the µ (cid:48) s values remain the same re-gardless of L SiO x (by comparing the results obtained withwaveguides (i), (v) and (vi)). This means that a negli-gible number of noise photons are generated in the SiO x waveguide. The dashed curve shows a second-order poly-nomial fitting to the data obtained for the waveguide (i).In addition to the P component that originated fromthe SFWM, we can see the P component in the low ex-citation regime. This indicates that processes other thanSFWM, for example inelastic scattering in the Si waveg-uide part, contribute little [37].Finally, Fig. 3(d) shows the measured coincidence toaccidental coincidence ratio (CAR = D c /D c,a ) from threewaveguides with the same L Si with respect to P p . Themaximum CAR of around 100 shows the strong quantumcorrelation of the photon pairs. We see that the overallCAR values decrease with increases in L SiO x . If the SiO x waveguides generated a negligible number of noise pho-tons, this reduction should be explained by the linearpropagation loss of the photons in the SiO x waveguides.To confirm this, we estimated CAR usingCAR = η η x µ c ( η total µ (cid:48) s + d )( η total µ (cid:48) i + d ) + 1 , (3)here we calculated µ c using Eq. (2) with γ = 161 /W/m,which we obtained from the fitting above. For µ (cid:48) s we usedthe fitted functions for waveguide (i) represented by thedashed curve in Fig. 3(c); we obtained the µ (cid:48) i function inthe same way. The estimated CAR is shown by the solidcurve in Fig. 2(d), which agrees well with the experi-mental data for waveguide (i). Next, we replace the L SiO x value with 2.93 and 4.49 cm in the calculation above, andplot the results as dashed and dot-dashed curves, respec-tively. The two curves well describe the experimentaldata obtained with waveguides (v) and (vi). Thus, weconfirmed that the decrease in CAR with increases inSiO x waveguide length can be explained by the photonloss in the SiO x part. This also indicates that no signifi-cant noise photons are created in the SiO x waveguide. III. ON-CHIP GENERATION ANDWAVELENGTH-DIVISION DEMULTIPLEXINGOF PHOTON PAIRS
We have shown that our SiO x waveguide works as alow-nonlinear circuit without creating a significant num-ber of noise photons. Next we attempt to monolithicallyintegrate the Si-wire photon pair source and its passivewavelength demultiplexing filter on the same chip.Figure 4(a) is a schematic diagram of the monolithicdevice fabricated in the manner described in Sec. 2,together with the experimental setup. In the device,correlated photon pairs are created via the SFWM inthe first Si rib waveguide ( L Si = 1.37 cm), and subse-quently spectrally separated by the on-chip SiO x AWGinto different output channels. Our AWG has 16 outputchannels designed to have a 200 GHz channel spacing.AWGs are commonly used for separating photon pairsgenerated via SFWM in integrated waveguides [10, 12–14, 19, 20]. The monolithic integration of the AWG andSFWM-based photon pair source thus provides a com-pact and stable photon pair source and a heralded singlephoton source. It is also a key technology for the im-plementation of a multiplexed single photon source [25].Moreover, our monolithic AWG can be used as an in-terface between a photon pair source and a linear-opticquantum circuit.The pump pulses for the experiment are obtained us-ing the setup shown in Fig. 2. We choose to collect pho-tons output from a pair of waveguides that are 3 channelsaway from the center output port. With this channel sep-aration we obtain a pump-to-signal (or idler) wavelengthseparation similar to that of the experiments describedin Sec. 2. We show the transmission spectra of the twoAWG outputs in Fig. 4(b). Here the 3-dB passbandwidths of the transmission windows are approximately80 GHz. The SSCs with Si tapers are fabricated betweenthe Si waveguide and the AWG for their low-loss connec-tion. The output optical fields are collected by opticalfibers with a high numerical aperture. Then the pho-tons are introduced into spectral filters, each of whichconsists of an FBG notch filter and a BPF for the sup-pression of residual pump fields. The 3-dB bandwidthof the BPFs ∆ ν are 100 GHz, which covers the AWGpassbands. Finally, the photons are received by SPCMsand a coincidence measurement is performed with a time-interval analyzer. The overall transmittance of the filters η f is –2.8 dB and the AWG insertion loss η AWG is –7.7dB. The quantum efficiency η QE and dark count rate d of the SPCM are 24% and 5.1 kHz, respectively, in thepresent experiment.To complete the integration and characterization of thedevice, Fig. 5(a) shows the net photon pair generationrate estimated using Eq. (1) as a function of the pumppeak power P p . The data exhibit good P dependence,indicating photon pair generation via the SFWM pro-cess. The solid line shows the estimation obtained withEq. (2) and µ c = η µ (cid:48) c using the same γ value of 161/W/m obtained from the fitting in Sec. 2. The experi-mental result agrees well with the calculation. Hence theobserved photon pairs are considered to be generated viaSFWM in the Si waveguide.Figure 5(b) shows the measured CAR as a function of P p . We obtained a maximum CAR of around 30. The Si waveguide (a)
Silica-based AWG
Pump pulses Single-photon counting modules Idler photons (1565.2 nm)
Signal photons (1556.0 nm)
Time-interval analyzer BPF1 BPF2 FBG2 FBG1 (b)
Wavelength (nm) T r an s m i ss i on i n t en s i t y ( a r b . un i t s i n d B ) FIG. 4. (a) Monolithic chip housing a Si wire photon pairsource and a silica-based arrayed waveguide grating, illus-trated with the experimental setup. (b) Transmission spectraof AWG output ports, from which photon pairs are collected. -5 -4 -3 -2 -1 (a) (b) Coupled pump peak power P p (mW) Coupled pump peak power P p (mW) N e t pho t on pa i r gene r a t i on r a t e m c ( / pu l s e ) C A R FIG. 5. (a) Net photon pair generation rate as a function ofpump peak power P p . The dashed curve shows the theoreticalcalculation result. (b) CAR values as a function of P p . Thesolid curve shows a calculation result. The dashed and dot-dashed curves are estimations of the lower AWG insertion lossand lower dark count rate, respectively, described in the text. value is well above CAR = 2, which is the limit accessiblewith classical light [38]. This indicates that the quan-tum correlation between photons was preserved even af-ter they had passed through the integrated AWG. Thus,our chip successfully generated and demultiplexed quan-tum correlated photons on the monolithic device. Nextwe analyze the obtained CAR values. The solid curveshows CAR values estimated using Eq. (3) in accordancewith the procedure employed to obtain the solid curve inFig. 3(c). The curve agrees well with the experimentaldata. The dashed curve shows the CAR values for η AWG = 0, exhibiting a maximum CAR of up to 80. Hence,reducing the AWG insertion loss is effective in improv-ing CAR. We also show the CAR values obtained whenthe detector dark count rate d = 20 (Hz). This suggeststhat we can further improve the maximum CAR by usingSPCMs with low dark count rates such as superconduct-ing single photon detectors with similar dark count rates[39]. IV. CONCLUSION
We have demonstrated the on-chip generation and de-multiplexing of quantum correlated photon pairs usinga monolithic waveguide platform composed of Si andsilica-based waveguides. Furthermore, we have shownthat the silica part of the monolithic platform does notcontribute to noise photon generation. The device can beused as a compact correlated photon pair source, and willbe useful for many quantum information applications in-cluding wavelength-division multiplexing quantum com-munication technologies [40] and heralded single photonsources [15]. Moreover, the silica-based AWG can pro- vide an interface between a Si-based photon pair sourceand silica-based lightwave circuits, which are useful aslinear-optics-based quantum circuits. The wavelength-multiplexing capability is also beneficial for construct-ing a circuit that harnesses high-dimensional quantumstates using path and frequency degrees of freedom [41].Thus, the present platform will prove useful for mono-lithic source-circuit integration with a view to achievingthe full-scale integration of on-chip quantum processors.
ACKNOWLEDGEMENT
We are grateful to Dr. Kaoru Shimizu for fruitful dis-cussions. [1] T. D. Ladd, F. Jelezko, R. Laflamme, Y. Nakamura, C.Monroe, and J. L. O’Brien, Nature , 45–53 (2010).[2] A. Politi, M. J. Cryan, J. G. Rarity, S. Yu, and J. L.O’Brien, Science , 646–649 (2008).[3] B. J. Smith, D. Kundys, N. Thomas-Peter, P. G. R.Smith, and I. A. Walmsley, Opt. Express , 13516–13525 (2009).[4] A. Peruzzo, M. Lobino, J. C. F. Matthews, N. Matsuda,A. Politi, K. Poulios, X.-Q. Zhou, Y. Lahini, N. Ismail, K.W¨orhoff, Y. Bromberg, Y. Silberberg, M. G. Thompson,and J. L. O’Brien, Science , 1500–1503 (2010).[5] D. Bonneau, E. Engin, K. Ohira, N. Suzuki, H. Yoshida,N. Iizuka, M. Ezaki, C. M. Natarajan, M. G. Tanner, R.H. Hadfield, S. N. Dorenbos, V. Zwiller, J. L. O’Brien,and M. G. Thompson, New. J. Phys. , 045003 (2012).[6] J. B. Spring, B. J. Metcalf, P. C. Humphreys, W.S. Kolthammer, X.-M. Jin, M. Barbieri, A. Datta, N.Thomas-Peter, N. K. Langford, D. Kundys, J. C. Gates,B. J. Smith, P. G. R. Smith, and I. A. Walmsley, Science , 798–801 (2013).[7] J. Wang, A. Santamato, P. Jiang, D. Bonneau, E. Engin,J. W. Silverstone, M. Lermer, J. Beetz, M. Kamp, S.H¨ofling, M. G. Tanner, C. M. Natarajan, R. H. Hadfield,S. N. Dorenbos, V. Zwiller, J. L. O’Brien, and M. G.Thompson, Opt. Commun. , 49–55 (2014).[8] N. Spagnolo, C. Vitelli, M. Bentivegna, D. J. Brod, A.Crespi, F. Flamini, S. Giacomini, G. Milani, R. Ramponi,P. Mataloni, R. Osellame, E. F. Galv˜ao, and F. Sciarrino,Naure Photon. , 615–620 (2014).[9] J. E. Sharping, K. F. Lee, M. A. Foster, A. C. Turner,B. S. Schmidt, M. Lipson, A. L. Gaeta, and P. Kumar,Opt. Express , 12388–12393 (2006).[10] H. Takesue, Y. Tokura, H. Fukuda, T. Tsuchizawa, T.Watanabe, K. Yamada, and S. Itabashi, Appl. Phys.Lett. , 201108 (2007).[11] S. Clemmen, K. P. Huy, W. Bogaerts, R. G. Baets, P.Emplit, and S. Massar, Opt. Express , 16558–16570(2009).[12] C. Xiong, L. G. Helt, A. C. Judge, G. D. Marshall, M. J.Steel, J. E. Sipe, and B. J. Eggleton, Opt. Express ,16206–16216 (2010).[13] M. Lobino, G. D. Marshall, C. Xiong, A. S. Clark,D. Bonneau, C. M. Natarajan, M. G. Tanner, R. H. Hadfield, S. N. Dorenbos, T. Zijlstra, V. Zwiller, M.Marangoni, R. Ramponi, M. G. Thompson, B. J. Eggle-ton, and J. L. O’Brien, Appl. Phys. Lett. , 081110(2011).[14] C. Xiong, C. Monat, A. S. Clark, C. Grillet, G. D. Mar-shall, M. J. Steel, J. Li, L. O’Faolain, T. F. Krauss, J.G. Rarity, and B. J. Eggleton, Opt. Lett. , 3413–3415(2011).[15] M. Davan¸co, J. R. Ong, A. B. Shehata, A. Tosi, I. Agha,S. Assefa, F. Xia, W. M. J. Green, S. Mookherjea, andK. Srinivasan, Appl. Phys. Lett. , 261104 (2012).[16] R. Horn, P. Abolghasem, B. J. Bijlani, D. Kang, A. S.Helmy, and G. Weihs, Phys. Rev. Lett. , 153605(2012).[17] N. Matsuda, H. Le Jeannic, H. Fukuda, T. Tsuchizawa,W. J. Munro, K. Shimizu, K. Yamada, Y. Tokura, andH. Takesue, Sci. Rep. , 817 (2012).[18] L. Olislager, J. Safioui, S. Clemmen, K. P. Huy, W. Bo-gaerts, R. Baets, P. Emplit, and S. Massar, Opt. Lett. (11), 1960–1962 (2013).[19] A. S. Clark, C. Husko, M. J. Collins, G. Lehoucq, S.Xavier, A. De Rossi, S. Combri´e, C. Xiong, and B. J.Eggleton, Opt. Lett. , 649–651 (2013).[20] N. Matsuda, H. Takesue, K. Shimizu, Y. Tokura, E. Ku-ramochi, and M. Notomi, Opt. Lett. , 8596–8604(2013).[21] F. Boitier, A. Orieux, C. Autebert, A. Lemaˆıtre, E. Ga-lopin, C. Manquest, C. Sirtori, I. Favero, G. Leo, and S.Ducci, Phys. Rev. Lett. , 183901 (2014).[22] W. H. Pernice, C. Schuck, O. Minaeva, M. Li, G. N.Goltsman, A. V. Sergienko, and H. X. Tang, NatureCommun. , 1325 (2012).[23] D. Sahin, A. Gaggero, Z. Zhou, S. Jahanmirinejad, F.Mattioli, R. Leoni, J. Beetz, M. Lermer, M. Kamp, S.H¨ofling, and A. Fiore, Appl. Phys. Lett. , 111116(2013).[24] F. Najafi, J. Mower, N. Harris, F. Bellei, A. Dane, C.Lee, P. Kharel, F. Marsili, S. Assefa, K. K. Berggren,and D. Englund, arXiv:1405.4244 [physics.optics].[25] 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 Photonics Rev. , L42–L46(2014). [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, NaurePhoton. , 104–108 (2014).[27] K. D. J¨ons, U. Rengstl, M. Oster, F. Hargart, M. Held-maier, S. Bounouar, S. M. Ulrich, M. Jetter, and P. Mich-ler, arXiv:1403.7174 [quant-ph].[28] N. Prtljaga, R. J. Coles, J. OHara, B. Royall, E. Clarke,A. M. Fox, and M. S. Skolnick, Appl. Phys. Lett. ,231107 (2014).[29] S. Sohma, T. Watanabe, N. Ooba, M. Itoh, T. Shibata,and H. Takahashi, European Conference on Optical Com-munication, 1–2 (2006).[30] Y. Hibino, IEEE J. Sel. Top. Quantum Electron. ,1090–1101 (2002).[31] N. Matsuda, R. Shimizu, Y. Mitsumori, H. Kosaka, andK. Edamatsu, Naure Photon. , 95–98 (2009).[32] H. Nishi, T. Tsuchizawa, T. Watanabe, H. Shinojima, S.Park, R. Kou, K. Yamada, and S. Itabashi, Appl. Phys.Express , 102203 (2010).[33] M. Fiorentino, P. L. Voss, J. E. Sharping, and P. Kumar,IEEE Photon. Technol. Lett. , 983–985 (2002). [34] H. Takesue and K. Inoue, Phys. Rev. A , 031802(2004).[35] C. A. Husko, A. S. Clark, M. J. Collins, A. De Rossi, S.Combri´e, G. Lehoucq, I. H. Rey, T. F. Krauss, C. Xiong,and B. J. Eggleton, Sci. Rep. , 3087 (2013).[36] Q. Lin and G. P. Agrawal, Opt. Lett. , 3140–3142(2006).[37] S. Clemmen, A. Perret, J. Safioui, W. Bogaerts, R. Baets,S. Gorza, P. Emplit, and S. Massar, J. Opt. Soc. Am. B , 1977–1982 (2012).[38] R. Loudon, The Quantum Theory of Light , 3rd ed. (Ox-ford University, 2000).[39] T. Inagaki, N. Matsuda, O. Tadanaga, M. Asobe, and H.Takesue, Opt. Express , 23241–23249 (2013).[40] K. Yoshino, M. Fujiwara, A. Tanaka, S. Takahashi, Y.Nambu, A. Tomita, S. Miki, T. Yamashita, Z. Wang, M.Sasaki, and A. Tajima, Opt. Lett. , 223–225 (2012).[41] C. Schaeff, R. Polster, R. Lapkiewicz, R. Fickler, S.Ramelow, and A. Zeilinger, Opt. Express20