A Compact, Mobile, Low-Threshold Squeezed Light Source
Jens Arnbak, Christian Scheffmann Jacobsen, Rayssa Bruzaca de Andrade, Xueshi Guo, Jonas Schou Neergaard-Nielsen, Ulrik Lund Andersen, Tobias Gehring
AA Compact, Mobile, Low-Threshold Squeezed Light Source
J. Arnbak, C. S. Jacobsen, R. B. Andrade, X. Guo, J. S. Neergaard-Nielsen, U. L. Andersen, and T. Gehring ∗ Center for Macroscopic Quantum States bigQ, Department of Physics, Technical University of Denmark (Dated: September 4, 2019)Strongly squeezed light finds many important applications within the fields of quantum metrology,quantum communication and quantum computation. However, due to the bulkiness and complexityof most squeezed light sources of today, they are still not a standard tool in quantum optics labs. Wehave taken the first steps in realizing a compact, high-performance 1550 nm squeezing source basedon commercially available fiber components combined with a free-space double-resonant parametricdown-conversion source. The whole setup, including single-pass second-harmonic generation in awaveguide, fits on a small breadboard and produces 9 . I. INTRODUCTION
Squeezed quantum states of light are a ubiquitous re-source in numerous applications associated with quantumsensing, quantum communication and quantum compu-tation [1–5]. One of the most celebrated examples is theapplication of squeezed light to improve the sensitivity ofgravitational wave interferometers, thereby extending thevolume in space within which gravitational events can beobserved. [6–8]. A recent impressive improvement in ob-servable volume is the eight-fold increase by the detectionof 6 dB squeezed light in the gravitational wave detec-tor GEO600 [9]. Quantum-enhanced sensitivity can alsobe achieved with squeezed light in tracking the motionand estimating bio-physical parameters of single livingcells [10–12].Apart from quantum sensing, squeezed light also hasapplications in quantum cryptography to extend the se-cure communication distance [13], to improve the crypto-graphic security [14] and to enable the implementation ofquantum secure basic cryptographic primitives [15]. Fi-nally, squeezed light has recently been found to be a vi-able resource for photonics continuous variable quantumcomputing due to development of new quantum error cor-recting codes [16, 17] and due to the inherent scalabilityof the squeezed light source [4, 18–21].All of the above mentioned applications would natu-rally benefit from a compact, mobile and robust squeezedlight source producing an appreciable amount of squeez-ing. However, in most experiments to date there havebeen a sharp trade-off between achieving high degrees ofsqueezing and the compactness (and thus robustness andtransportability) of the source.On one hand, quantum states have been significantlysqueezed by up to 15 dB (that is, a reduction of 97% ofthe vacuum noise) using a nonlinear crystal embedded inan optical cavity [22–27], but the associated experimen-tal setups have very large footprints and are not easily ∗ [email protected] transportable due to the need for multiple mode-cleaningcavities for the pump and local oscillator to maximize thesqueezing and cavity based second harmonic generationto supply the pump which easily exceeds 100 mW. Dueto this immobility, the squeezed light sources are oftenbuilt up around the application.On the other hand, compact and mobile squeezed lightsources have been constructed using either an opticalwaveguide in a single-pass configuration [28, 29] or asa cavity [30], using a micron-sized Silicon Nitride ringcavity [31, 32] or using a whispering gallery mode res-onator [33], but in these systems the squeezing degreehave been limited to maximum 3 dB. Using the opti-cal Kerr effect in fibers, the production of 2.4 dB pulsedsqueezed light was demonstrated on a mobile platform of0 . [34]. In all of these latter experiments, compact-ness has been traded with the squeezing degree.The trade-off has been partially settled in a couple ofrealizations: The free-space squeezed light source – anoptical parametric oscillator – used in the GEO600 grav-itational wave detector has a footprint of 1 . × . ×
60 cm free-space setupproducing 6 dB of two-mode squeezed light has recentlybeen demonstrated [36].
II. SETUP
In this article, we present the construction of acompact squeezed light source with a footprint of30 cm ×
45 cm producing 9.3 dB squeezing. While thesource itself is a traditional double resonant paramet-ric down-conversion source, we reduced its size by re-placing bulky free-space optical components with less ef-fective, but more compact commercially available fiber-alternatives and operate without the use of filter cav-ities. Only the couplings of the pump beam and thecoherent control beam to the nonlinear cavity are ob-tained by free-space optics, as is the squeezed light out-put. In particular, the second-harmonic generator con-sists of a small single-pass waveguide module which pro- a r X i v : . [ qu a n t - ph ] S e p O P O DBSDBSHWP CollColl
BreadboardTo be boxed
HWP PDPDPD CircCircSHGEOM AOM PS1550 nm Laser
FunctionGenerator
100 MHz100 MHz40 MHz40 MHz80 MHzIso 1 B S Reference Cavity PS HomodyneDetection F li p PD PD PD P BS Iso 2
ElectronicsBox
300 mm mm AR HRR = 90% @1550 nm, 97.5% @775 nm F li p PD FIG. 1. Schematic representation of the experimental setup. The free-space part of the setup is placed on a breadboardmeasuring 30 cm ×
45 cm with the fiber components being placed around for convenience. The plan is to fit both the free-spacepart and the fiber components (marked by the teal rectangle) into a 19” box.The 775 nm light, used to generate the squeezed light, is generated in a single-pass LiNbO waveguide second-harmonic generator(SHG). A Pound-Drever-Hall lock in the pump path stabilizes the cavity. A coherent-locking scheme for locking the relativephase between pump and local oscillator utilizes a 40 MHz up-shifted pilot tone transmitted together with the squeezed light.The squeezed light is characterized by a balanced homodyne setup that utilizes a reference cavity to help mode-match thesqueezed light and local oscillator. Polarization sensitive fiber isolators are inserted along the local oscillator fiber path in orderto minimize polarization noise build-up along the fiber path.AOM: acousto-optic modulator, EOM: electro-optic modulator, BS: (50/50) beam-splitter, DBS: dichroic beam-splitter, OPO:double-resonant optical parametric oscillator, PD: photo-detector, HWP: half-wave plate, PS: phase-shifter, Iso: isolator, Circ:circulator, Coll: collimator, Flip: flip mirror duces enough light to saturate our source’s low pumppower threshold of 5 . . × × < . < . .
95 % @1550 nm,R > . . . ≈
66 MHz) and around 200 for the775 nm mode (FWHM ≈
17 MHz). The double-resonanceis achieved by tuning the temperature of the crystal andthe length of the cavity with a piezo-electric transducer.An NKT Photonics E15 BOOSTIK 1550 nm fiber lasersupplies the light to the setup. The light is divided intotwo paths; one serving as the local oscillator in the ho-modyne measurement setup. The other path is furthersplit into a pumping path and a path for the coherentcontrol beam [37], in the following called the pilot beam.The pumping path starts with a 1550 nm electro-opticmodulator (EOM) to modulate the phase at 100 MHz forthe OPO Pound-Drever-Hall (PDH) lock. The EOM hasa maximum input power of 300 mW and an insertion lossof 3 dB. This limits the input power into the LiNbO second-harmonic generator (SHG) wave-guide module FIG. 2. Overlapped Allan deviation (OADEV) and power-spectral density of polarization noise measurements in thelocal oscillator path. The data was taken at a 1 Hz samplingrate and is normalized to the mean and has mean subtracted.The blue trace is without any polarizing components. Theyellow trace is with one polarization sensitive isolator and thegreen trace is with two isolators. The red trace is polarizationinsensitive amplitude noise. (NTT Electronics WH-0775-000-F-B-C) to 150 mW, re-sulting in around 8 mW of 775 nm light to be used topump the OPO.The pilot path implements a coherent-locking schemefor locking the relative phase between the pump light andthe local oscillator by using a fiber acousto-optic modula-tor (AOM) to up-shift the frequency of a 1550 nm beamby 40 MHz. This pilot field enters the OPO throughthe HR side and interacts with the pump field throughdifference-frequency generation. The reflected light is de-tected and down-mixed with an 80 MHz tone to generatean error signal for locking the phase of the pilot field tothe pump field. The locked pilot field is transmitted withthe squeezing and beats with the local oscillator. Afterdetection in the homodyne detector, it is down-mixedwith a 40 MHz tone to provide an error-signal for lockingthe phase between the local oscillator and the squeezedfield.Both the pump and the pilot beams are coupled out ofthe fibers and collimated by fiber collimators. Using twomode-matching lenses and two steering mirrors, each ofthe free-space beams are coupled into the OPO. All free-space optics are placed on the 30 cm ×
45 cm breadboardwhich leaves enough space to place all fiber componentson it as well. In the actual experiment, the fiber compo-nents were not attached to the breadboard out of conve-nience, but will be placed in the box in the final versionof the device.The squeezed light is characterized in a balanced ho-modyne setup which is placed on a neighboring bread-board and uses a bright (10 mW) 1550 nm beam as alocal oscillator. The overlap between the squeezed light
FIG. 3. Graph showing experimentally obtained gain values(orange dots) as a function of input pump power. The blueline is a fit to eq. (1), and a threshold value of 5 . ± .
03 mWis extracted from the model. The error-bars assume a 5% erroron the power. and the local oscillator is optimized by coupling bothfields into a triangular reference cavity. This allows usto achieve a fringe visibility of around 99 %. The homo-dyne detector uses InGaAs photodiodes ( η QE >
99 %).The photo-electric signal is analyzed with a spectrum an-alyzer.
III. RESULTS
The setup utilizes polarization maintaining (PM) fibercomponents, and every time two PM fibers are combinedin a mating sleeve, a small polarization mismatch canappear due to an imperfect slow-axes alignment. Withmany components, such a mismatch can build up and re-sult in quite severe polarization noise. This problem canbe circumvented by using a polarization stabilizing feed-back loop, but a simpler solution is to insert polarizationfiltering components along the path, thus preventing po-larization mismatch to build up. The downside to thissimple solution is the introduction of some excess ampli-tude noise.In our setup, the EOM, AOM, SHG and circulators areall polarization filtering components thus preventing mis-match to build up in the pump and pilot paths. For thelocal oscillator, however, polarization noise was a prob-lem, and fiber isolators were therefore inserted into thefiber path.We characterize the effect of inserting polarizing com-ponents by inserting a power-meter after a polarizingbeam-splitter (PBS) measuring at a sampling rate of1 Hz. Fig. 2 shows a plot of the overlapped Allan de-viation and the power spectral density of these measure-ments. We measured without any isolators, with isolator1 and with two isolators, isolators 1 and 2. For theirpositions see the experimental schematic in Fig. 1. Asa baseline, we measure the polarization insensitive am-plitude noise. The data is normalized to the mean, andthe mean is subtracted in order to get the fluctuationsrelative to the mean.From the Allan deviation, we see a large increasein long-term stability of the power when isolator 1 ispresent. This can also be seen in the power spectraldensity, with a difference of almost two orders of magni-tude between having no polarizing components and hav-ing isolator 1. Between having one and two isolators,the difference is negligible for both graphs which is to beexpected as no additional fiber components are presentafter isolator 2. By comparing the green/orange traceswith the red trace, it seems that the amplitude noise withisolators is in general a bit higher than the intrinsic noisefrom the laser. This could be due to small polarizationmismatches being converted to amplitude noise by theisolators.The performance of our setup is first characterizedby estimating the pump threshold power via a classicalgain measurement of a 1550 nm field interacting with thepump field in the crystal. For this measurement, thefiber AOM is removed and the transmitted power of the1550 nm field is measured while varying the input powerof the 775 nm pump field. The gain is then estimated bycomparing with the transmitted 1550 nm light when thepump is blocked. The calculated gain values are plottedin Fig. 3. The gain is modelled as [38] g = 1 (cid:16) − (cid:113) P p P thr p (cid:17) , (1) where P p is the input pump power and P thr p is theOPO threshold power. From the fit of the experimentalgain values, we can extract an OPO threshold power of5 . ± .
03 mW. This shows that even though the systemonly has 8 mW of pump power available due to lossy fibercomponents, the double resonance allows the squeezer toutilize its full potential, since squeezing is best generatedbelow threshold.In order to characterize the squeezing performance,we first look at the 5 MHz side-band frequency using aresolution-bandwidth of 300 kHz and video-bandwidth of300 Hz. We measure for 200 ms and average each trace100 times. The pump power is varied in the range 0–4 mW, and the local oscillator phase is locked to squeez-ing and anti-squeezing, respectively. The measured val-ues as a function of pump power are shown in Fig. 4acorrected for electronic noise, which is about 22 dB be-low shot noise.The squeezing increases with power until a maximumof around 9 . . (cid:104) δ ˆ X − (cid:105) ( (cid:104) δ ˆ X (cid:105) )including phase noise can be modelled as (normalized toshot noise) [38] (cid:68) δ ˆ X ± (cid:69) ≈ η esc η opt V η QE ± cos ( φ ) 4 (cid:113) P p P thr p (cid:16) ∓ (cid:113) P p P thr p (cid:17) + 4 (cid:0) ωκ (cid:1) ∓ sin ( φ ) 4 (cid:113) P p P thr p (cid:16) ± (cid:113) P p P thr p (cid:17) + 4 (cid:0) ωκ (cid:1) , (2)where η esc η opt V η QE = η tot is the total efficiency, with η esc being the escape efficiency, η opt being the optical loss, V being the fringe visibility of the squeezing and the localoscillator, and η QE being the quantum efficiency of thephoto diodes. ω = 2 π × κ = 2 π ×
66 MHz is theFWHM bandwidth of the OPO. Finally, φ is the RMSvalue of the phase noise. We note that the equationis only valid for small values of φ . The solid lines inFig. 4a are a fit to Eq. 2. From the fit, we extract atotal efficiency η tot = 0 . ± .
01. This is in pretty goodagreement with our estimated efficiency η tot ≈ .
93 thatcomes from η esc ≈ .
97 (estimated from coating specifi-cations, ignoring scattering and absorption),
V ≈ . η QE ≈ .
99 and η opt ≈ . . The optical loss comesfrom 5 mirrors and 2 lenses, all of which have an esti-mated efficiency of 0 . ± FIG. 4. a) Squeezing and anti-squeezing relative to shot noise as a function of pump power at a side-band frequency of 5 MHz.The blue points are squeezing and the orange points are anti-squeezing. The theoretical model of Eq. 2 is fitted to the data andplotted in solid lines. The purple dashed line is the fitted squeezing model in the absence of phase noise. The electronic noise,which is 22 dB below shot noise, was subtracted from the data. The error-bars assume a 5% error on the power. b) Spectrum ofthe squeezing and anti-squeezing from 1 −
120 MHz for different pump powers. The traces are corrected for electronic noise andnormalized to the shot noise. The thin solid lines are fits of Eq. 2. Bands around 40 MHz, 80 MHz and 100 MHz are excludedfrom the fit as they contain the 40 MHz up-shifted pilot tone and electronic pick-up of modulation signals, and the 80 MHz and100 MHz peaks are removed from the figure. the way out beyond 100 MHz. Once again, for high pumppowers the squeezing starts to degrade at low frequenciesdue to phase noise, and from the fits of the curves, we ex-tract similar values for the efficiency and phase noise. Acurious exception is the 3 . . IV. SUMMARY
In summary, we have demonstrated that a squeezedlight source based on a traditional bulk squeezing cavitycan be built with a small footprint on an area smallerthan what is available in standard 19 inch boxes. Weachieved this by replacing most free-space optics withfiber optics except for the in- and out-coupling of theOPO. The result was in particular made possible bya single-pass waveguide SHG module which provided enough pump power for the low threshold of only a cou-ple of milliwatts, which in turn was due to the double-resonance of the OPO.Further engineering on setup packaging will enableeven smaller foot prints. A mobile turn-key squeezedlaser in a transportable box might make squeezing a stan-dard tool in quantum optics labs, which are ready to ex-plore many yet undiscovered applications.
ACKNOWLEDGEMENTS
The authors acknowledge support from the InnovationFund Denmark through the Quantum Innovation Cen-ter, Qubiz as well as support from the Danish NationalResearch Foundation, Center for Macroscopic QuantumStates (bigQ, DNRF142). Furthermore the authors ac-knowledge the EU project CiViQ (grant agreement no.820466) and the EU project UNIQORN (grant agreementno. 820474). [1] U. L. Andersen, T. Gehring, C. Marquardt, andG. Leuchs, 30 Years of Squeezed Light Generation, Phys-ica Scripta , 053001 (2016), arXiv:1511.03250.[2] U. L. Andersen, J. S. Neergaard-Nielsen, P. van Loock,and A. Furusawa, Hybrid discrete- and continuous-variable quantum information, Nature Physics , 713(2015), arXiv:1409.3719 [quant-ph].[3] A. I. Lvovsky, Squeezed light, in Photonics Volume 1:Fundamentals of Photonics and Physics , edited by D. An-drews (Wiley, West Sussex, United Kingdom, 2015) pp.121–164.[4] C. Weedbrook, S. Pirandola, R. Garc´ıa-Patr´on, N. J.Cerf, T. C. Ralph, J. H. Shapiro, and S. Lloyd, Gaus- sian quantum information, Reviews of Modern Physics , 621 (2012), arXiv:1110.3234.[5] S. Pirandola, B. R. Bardhan, T. Gehring, C. Weedbrook,and S. Lloyd, Advances in photonic quantum sensing,Nature Photonics , 724 (2018).[6] R. Schnabel, N. Mavalvala, D. E. McClelland, and P. K.Lam, Quantum metrology for gravitational wave astron-omy, Nature Communications , 121 (2010).[7] The LIGO Scientific Collaboration, A gravitational waveobservatory operating beyond the quantum shot-noiselimit, Nature Physics , 962 (2011).[8] The LIGO Scientific Collaboration, Enhanced sensitiv-ity of the LIGO gravitational wave detector by using squeezed states of light, Nature Photonics , 613 (2013).[9] GEO600, Gravitational-wave detectors begin thirdobservation run, url: (2019).[10] M. A. Taylor and W. P. Bowen, Quantum metrology andits application in biology, Physics Reports , 1 (2016),arXiv:1409.0950.[11] M. A. Taylor, J. Janousek, V. Daria, J. Knittel, B. Hage,H.-A. Bachor, and W. P. Bowen, Biological measure-ment beyond the quantum limit, Nature Photonics ,229 (2013).[12] M. A. Taylor, J. Janousek, V. Daria, J. Knittel, B. Hage,H.-A. Bachor, and W. P. Bowen, Subdiffraction-LimitedQuantum Imaging within a Living Cell, Physical ReviewX , 011017 (2014).[13] L. Madsen, V. Usenko, M. Lassen, R. Filip, and U. An-dersen, Continuous variable quantum key distributionwith modulated entangled states., Nature Communica-tions , 1083 (2012), article number:1083.[14] T. Gehring, V. Handchen, J. Duhme, F. Furrer, T. Franz,C. Pacher, R. F. Werner, and R. Schnabel, Implemen-tation of continuous-variable quantum key distributionwith composable and one-sided-device-independent secu-rity against coherent attacks, Nature Communications ,8795 (2015).[15] F. Furrer, T. Gehring, C. Schaffner, C. Pacher, R. Schn-abel, and S. Wehner, Continuous-variable protocol foroblivious transfer in the noisy-storage model, NatureCommunications , 10.1038/s41467-018-03729-4 (2018).[16] D. Gottesman, A. Kitaev, and J. Preskill, Encoding aqubit in an oscillator, Phys. Rev. A , 012310 (2001).[17] N. C. Menicucci, Fault-tolerant measurement-basedquantum computing with continuous-variable clusterstates, Phys. Rev. Lett. , 120504 (2014).[18] K. Marshall, C. S. Jacobsen, C. Sch¨afermeier,T. Gehring, C. Weedbrook, and U. L. Andersen,Continuous-variable quantum computing on encrypteddata, Nature Communications , 13795 (2016).[19] S. Yokoyama, R. Ukai, S. C. Armstrong, C. Sornphiphat-phong, T. Kaji, S. Suzuki, J.-I. Yoshikawa, H. Yonezawa,N. C. Menicucci, and A. Furusawa, Ultra-large-scalecontinuous-variable cluster states multiplexed in the timedomain, Nature Photonics , 982 (2013), arXiv:1306.3366[quant-ph].[20] M. V. Larsen, X. Guo, C. R. Breum, J. S. Neergaard-Nielsen, and U. L. Andersen, Deterministic generationof a two-dimensional cluster state for universal quan-tum computing, arXiv e-prints , arXiv:1906.08709 (2019),arXiv:1906.08709 [quant-ph].[21] M. Chen, N. C. Menicucci, and O. Pfister, Experimentalrealization of multipartite entanglement of 60 modes of aquantum optical frequency comb, Phys. Rev. Lett. ,120505 (2014).[22] H. Vahlbruch, M. Mehmet, K. Danzmann, and R. Schn-abel, Detection of 15 dB Squeezed States of Light andtheir Application for the Absolute Calibration of Pho-toelectric Quantum Efficiency, Physical Review Letters , 1 (2016), arXiv:arXiv:1011.1669v3.[23] Y. Takeno, M. Yukawa, H. Yonezawa, and A. Furusawa,Observation of -9 dB quadrature squeezing with improve-ment of phase stability in homodyne measurement, Op-tics Express , 4321 (2007). [24] T. Eberle, S. Steinlechner, J. Bauchrowitz, V. H¨andchen,H. Vahlbruch, M. Mehmet, H. M¨uller-Ebhardt, andR. Schnabel, Quantum Enhancement of the Zero-AreaSagnac Interferometer Topology for Gravitational WaveDetection, Physical Review Letters , 251102 (2010).[25] M. Mehmet, S. Ast, T. Eberle, S. Steinlechner,H. Vahlbruch, and R. Schnabel, Squeezed light at 1550nm with a quantum noise reduction of 12.3 dB, OpticsExpress , 25763 (2011).[26] S. Shi, Y. Wang, W. Yang, Y. Zheng, and K. Peng, De-tection and perfect fitting of 13.2 dB squeezed vacuumstates by considering green-light-induced infrared absorp-tion, Optics Letters , 5411 (2018).[27] A. Sch¨onbeck, F. Thies, and R. Schnabel, 13 dB squeezedvacuum states at 1550 nm from 12 mW external pumppower at 775 nm, Optics Letters , 110 (2018).[28] F. Mondain, T. Lunghi, A. Zavatta, ´E. Gouzien,F. Doutre, M. De Micheli, S. Tanzilli, and V. D’Auria,Chip-based squeezing at a telecom wavelength, PhotonicsResearch , 36 (2019), arXiv:1811.02097.[29] N. Takanashi, T. Kashiwazaki, T. Kazama, K. Enbutsu,R. Kasahara, T. Umeki, and A. Furusawa, Detectionof 3-dB continuous-wave squeezing at 1.55 um from afiber-coupled single-pass PPLN ridge waveguide, arXiv ,1906.09749 (2019), arXiv:1906.09749.[30] M. Stefszky, R. Ricken, C. Eigner, V. Quiring, H. Her-rmann, and C. Silberhorn, A waveguide cavity resonatorsource of squeezing, Physical Review Applied , 044026(2017), arXiv:1702.02855.[31] A. Dutt, S. Miller, K. Luke, J. Cardenas, A. L. Gaeta,P. Nussenzveig, and M. Lipson, Tunable Squeezing Us-ing Coupled Ring Resonators on a Silicon Nitride Chip,Optics Letters , 223 (2015), arXiv:1506.03791.[32] V. D. Vaidya, B. Morrison, L. G. Helt, R. Shahrokhshahi,D. H. Mahler, M. J. Collins, K. Tan, J. Lavoie, A. Re-pingon, M. Menotti, N. Quesada, R. C. Pooser, A. E.Lita, T. Gerrits, S. W. Nam, and Z. Vernon, Broadbandquadrature-squeezed vacuum and nonclassical photonnumber correlations from a nanophotonic device, arXive-prints , arXiv:1904.07833 (2019), arXiv:1904.07833[quant-ph].[33] A. Otterpohl, F. Sedlmeir, G. Shafiee, T. Dirmeier,U. Vogl, G. Schunk, D. V. Strekalov, H. G. Schwe-fel, T. Gehring, U. L. Andersen, G. Leuchs, andC. Marquardt, Squeezed vacuum states from a whisper-ing gallery mode resonator, arXiv , 1905.07955 (2019),arXiv:arXiv:1905.07955v1.[34] C. Peuntinger, B. Heim, C. R. M¨uller, C. Gabriel,C. Marquardt, and G. Leuchs, Distribution of squeezedstates through an atmospheric channel, Phys. Rev. Lett. , 060502 (2014).[35] H. Vahlbruch, A. Khalaidovski, N. Lastzka, C. Gr¨af,K. Danzmann, and R. Schnabel, The GEO 600 squeezedlight source, Classical and Quantum Gravity ,10.1088/0264-9381/27/8/084027 (2010).[36] Y.-J. Wang, W.-H. Yang, Y.-H. Zheng, and K.-C.Peng, A compact einstein–podolsky–rosen entangledlight source, Chinese Physics B , 070303 (2015).[37] H. Vahlbruch, S. Chelkowski, B. Hage, A. Franzen,K. Danzmann, and R. Schnabel, Coherent control of vac-uum squeezing in the gravitational-wave detection band,Phys. Rev. Lett. , 011101 (2006). [38] T. Aoki, G. Takahashi, and A. Furusawa, Squeezing at946nm with periodically poled ktiopo4, Opt. Express14