Narrow-linewidth homogeneous optical emitters in diamond nanostructures via silicon ion implantation
Ruffin E. Evans, Alp Sipahigil, Denis D. Sukachev, Alexander S. Zibrov, Mikhail D. Lukin
NNarrow-linewidth homogeneous optical emittersin diamond nanostructures via silicon ion implantation
Ruffin E. Evans, ∗ Alp Sipahigil, ∗ Denis D. Sukachev,
1, 2
Alexander S. Zibrov, and Mikhail D. Lukin † Department of Physics, Harvard University, 17 Oxford St., Cambridge, MA 02138 Russian Quantum Center, Business-center “Ural”, 100A Novaya St., Skolkovo, Moscow 143025 (Dated: April 26, 2016)The negatively-charged silicon-vacancy (SiV − ) center in diamond is a bright source of indistin-guishable single photons and a useful resource in quantum information protocols. Until now, SiV − centers with narrow optical linewidths and small inhomogeneous distributions of SiV − transition fre-quencies have only been reported in samples doped with silicon during diamond growth. We presenta technique for producing implanted SiV − centers with nearly lifetime-limited optical linewidths anda small inhomogeneous distribution. These properties persist after nanofabrication, paving the wayfor incorporation of high-quality SiV − centers into nanophotonic devices. PACS numbers: 78.55.Ap, 81.05.Cy, 81.07.Gf, 42.50.ExKeywords: Silicon-Vacancy; ion implantation; diamond; photonics; quantum optics
I. INTRODUCTION
Coherent emitters of indistinguishable single pho-tons are a basic ingredient in many quantum infor-mation systems[1]. Atom-like emitters in the solidstate are a particularly appealing platform for prac-tical quantum information because they can be scal-ably integrated into nanophotonic devices. How-ever, no single solid-state system has yet combinedhigh brightness of narrowband emission and a lowinhomogeneous distribution of photon frequenciesfrom separate emitters (indistinguishability) withease of incorporation into nanophotonic structureson demand. For example, optically active semi-conductor quantum dots can be bright and inte-grable into nanostructures, but have a large inho-mogeneous distribution[2]. Nitrogen-vacancy (NV − )centers in bulk diamond[3] are bright and photo-stable, with a moderate inhomogeneous distribu-tion that allows straightforward tuning of multipleNV − centers into resonance. These properties allowproof-of-principle demonstrations of quantum infor-mation protocols such as remote spin-spin entangle-ment generation[4, 5] and quantum teleportation[6].Further progress towards developing NV − basedquantum devices has been hindered by low indistin-guishable photon generation rates associated withthe weak NV − zero-phonon line, a challenge thatcould be addressed by integrating NV − centers intonanophotonic structures. However, the optical tran-sition frequencies of NV − centers are very sensitiveto their local environment[7, 8], making integrationof spectrally stable emitters into nanophotonic struc- tures a major challenge[9].The negatively charged silicon-vacancy color cen-ter in diamond (SiV − ) has shown promise infulfilling the key criteria of high brightness[10],lifetime-limited optical linewidths[11], and a nar-row inhomogeneous distribution of optical transitionfrequencies[12]. The SiV − (Fig. 1) has electronicstates with strong dipole transitions where 70% ofthe emission is in the zero-phonon line (ZPL) at737 nm[10]. The inversion symmetry of the SiV − prevents first-order Stark shifts, suppressing spec-tral diffusion[11] and allowing indistinguishable pho-tons to be generated from separate emitters with-out the need for tuning or extensive pre-selection ofemitters[13]. When combined with a spin degree offreedom[14], the SiV − center’s bright narrowbandtransition, narrow inhomogeneous distribution, andspectral stability make it a promising candidate forapplications in quantum optics and quantum infor-mation science.Silicon-vacancy centers occur only rarely in natu-ral diamond[16], and are typically introduced duringCVD growth via deliberate doping with silane[17,18] or via silicon contamination[11, 12, 19–21].While these techniques typically result in a nar-row inhomogeneous distribution of SiV − fluores-cence wavelengths, these samples have a number ofdisadvantages. For example, the concentration ofSiV − centers can be difficult to control and localiza-tion of SiV − centers in three dimensions is impossi-ble.Ion implantation offers a promising solution tothese problems. By controlling the energy, quan-tity, and isotopic purity of the source ions, the a r X i v : . [ c ond - m a t . m e s - h a ll ] A p r .c. VC Si
736 737 738Fluorescence Wavelength50100150 Intensity (arb. u.) StrainedUnstrainedA B C DA’ D’ 0.20.40.60.8 -0.2 0 0.2200 MHzIntensity (kcps)Laser frequency offset (GHz)0.4 E g2 E u b. λλ gSOuSO A B C DSpin-orbit ɛ (strain) Spin-orbit+ StrainC’ D’A’ B’ d. B C
FIG. 1. Properties of the SiV − center. a. Atomicstructure of the SiV center. The V-Si-V axis lies alongthe (cid:104) (cid:105) lattice direction. The SiV − has D symmetry. b. Level structure of the SiV − center. The SiV − is asingle-hole system with double orbital and spin degen-eracy. This degeneracy is partially lifted by spin-orbitcoupling ( λ SOg = 47 GHz and λ SOu = 260 GHz[11, 15]).Lattice strain increases the splitting between these spin-orbit levels, shifting the transition frequencies. c. Flu-orescence spectra of the ZPLs of single SiV − centers inhigh-strain (blue, dashed) and low-strain (red) environ-ments at 9–15 K. Transitions B and C are less sensitiveto strain compared with transitions A and D becausethe ground and excited states shift in the same (oppo-site) directions for transitions B and C (A and D)[12].Unstrained spectrum offset and scaled vertically for clar-ity. d. Linewidth (FWHM) of representative implantedSiV − in bulk (unstructured) diamond measured by PLEspectroscopy (blue points: data; red line: Lorentzian fit).Inset: histogram of emitter linewidths in bulk diamond.Almost all emitters have a linewidth within a factor ofthree of the lifetime limit (94 MHz). depth, concentration, and isotope of the resultingimplanted ions can be controlled. Ion implanta-tion is widely commercially available. Targeted ionimplantation using a focused silicon ion beam isalso possible, allowing for placement of silicon de-fects in all three dimensions with precision on thescale of tens of nanometers[22]. Despite the advan-tages of ion implantation, there have been conflict-ing results[15, 22, 23] on the brightness and creationyield of SiV − centers produced using this methodand no systematic studies of the inhomogeneousdistribution of SiV − fluorescence wavelengths. Al- though there has been a single report of an im-planted SiV − with a linewidth roughly 10 times thelifetime limit[24], to the best of our knowledge therehas been up to now no consistent method for pro-ducing SiV − centers with bright, narrow-linewidthemission using ion implantation. These two criteriaof a low inhomogeneous distribution relative to thesingle-emitter linewidth and narrow single-emitterlinewidth relative to the lifetime limit are essentialfor quantum optics applications[1, 25].In this paper, we report the creation of SiV − centers in diamond using ion implantation. Im-plantation is followed by high-temperature high-vacuum annealing to facilitate SiV − formation andrepair implantation-induced damage to the lattice.The resulting emitters have narrow optical transi-tions within a factor of four of the lifetime lim-ited linewidth and a narrow inhomogeneous distri-bution such that half of the emitters have transi-tions that lie in a 15 GHz window. Finally, we incor-porate these SiV − centers into nanostructures anddemonstrate that their favorable optical propertiesare maintained even after fabrication. II. THE
SiV − CENTER IN DIAMOND
The silicon-vacancy color center is a point de-fect in diamond wherein a silicon atom occupiesan interstitial position between two vacancies (Fig.1a)[26]. The SiV − is a spin- system with ground( E g ) and excited ( E u ) states localized to the dia-mond bandgap[26–28]. Both states have double spinand orbital degeneracies partially lifted by the spin-orbit interaction (Fig. 1b) which splits each quar-tet into two degenerate doublets. The spin-orbitsplittings for the ground and excited states are 0.19and 1 .
08 meV (47 and 260 GHz), respectively (Fig.1c)[15, 27]. All transitions between the ground andexcited states are dipole-allowed with a ZPL energyof 1.68 eV ( λ = 737 nm) and an excited state life-time of under 1.7 ns[29]. These optical transitionscan have linewidths (Fig. 1d) comparable to the life-time limit of 94 MHz[11].The SiV − is sensitive to strain, which can bothshift the average energy (for axial strain) and in-crease the splitting (for transverse strain) in theground and excited state manifolds (Fig. 1b, lastcolumn)[14, 15]. Transitions B and C within theZPL are relatively insensitive to transverse strainbecause their ground and excited states shift in the2ame direction: both upward for transition B andboth downward for transition C (Fig. 1c)[12]. Tran-sition C is between the lowest energy ground and ex-cited states which are also isolated from the phononbath at low temperatures[29]. This transition istherefore the most suitable for applications in quan-tum information science. III. CREATING
SiV − CENTERS WITH IONIMPLANTATION
We create SiV − centers using the following proce-dure: First, we begin with a polished CVD diamond(Element Six Inc., [ N ] S < { } oriented topface). Previous work suggests that mechanical pol-ishing produces a strained and damaged layer closeto the surface that results in reduced mechanical sta-bility of nanofabricated structures[30]. We also ex-pect that the strain introduced by mechanical pol-ishing will lead to a larger inhomogeneous distribu-tion of SiV − wavelengths. We reduce this damage byremoving 5 µ m of diamond through reactive ion etch-ing, producing a smooth (under 1 nm RMS rough-ness) surface. More details on this technique can befound elsewhere[30, 31]. An otherwise identical con-trol sample was also put through the same implan-tation procedure but without this pre-etching step.We then implant Si + ions (Innovion Corporation)at a dose of 10 ions / cm and an energy of 150 keVresulting in an estimated depth of 100 ±
20 nm[32].After implantation, we clean the samples usingan an oxidative acid clean (boiling 1 : 1 : 1 perchlo-ric : nitric : sulfuric acid)[33] and then perform twohigh-temperature high-vacuum ( (cid:46) − Torr) an-neals. The first anneal is at 800 ◦ C for eight hoursafter a four-hour bake-out step at 400 ◦ C. At 800 ◦ C,vacancies are mobile[34–36] leading to the forma-tion of SiV − centers. The second anneal is thesame as the first, but with an additional step at1100 ◦ C with a two-hour dwell time. At this tem-perature, divacancies and other defects can alsoanneal out[37, 38]. For all annealing steps, weuse slow temperature ramps ( (cid:46) ◦ C per hour) tomaintain low pressures in our furnace. This an-nealing procedure, inspired by previous work withSiV − [20, 39] and NV − [31, 37, 40, 41] centers, bothaids in the formation of SiV − centers and also helpsremove damage to the crystal lattice, reducing lo-cal strain. The residual graphitic carbon producedduring these high-temperature anneals was removedby again performing the oxidative acid clean. Be- fore each annealing step, we use X-ray photoelec-tron spectroscopy to verify that the surface is free ofcontaminants. IV. RESULTS AND DISCUSSIONA.
SiV − centers in bulk diamond We confirm that the SiV − centers exhibit narrow-linewidth optical transitions by performing photo-luminescence excitation (PLE) spectroscopy after1100 ◦ C annealing. In this experiment, we scan thefrequency of a weak resonant laser (New Focus Ve-locity, linewidth ∆ f (cid:46)
25 MHz over the course ofthe experiment, stabilized with a High Finesse WS7wavemeter) across transition C and monitor the flu-orescence on the phonon-sideband (PSB). We in-tegrate over several scans to reconstruct the time-averaged shape and position of the SiV − ZPL (Fig.1d). We perform these measurements in a heliumflow cryostat at a sample stage temperature of 3 . − centers in bulk diamond have nar-row optical transitions with linewidths of Γ / π =320 ±
180 MHz (mean and standard deviation forN = 13 spatially resolved emitters). Almost all SiV − centers have a linewidth within a factor of three ofthe lifetime limit (Fig. 1d, inset). As defined here,these linewidths include the effects of phonon broad-ening and all spectral diffusion that happens at anytimescale during the course of the experiment (4–15minutes).We characterize the inhomogeneous distributionof the implanted SiV − fluorescence wavelengths af-ter each annealing step via photoluminescence spec-troscopy. To perform these measurements, we ex-cite the SiV − centers using off-resonant light froma 700 nm diode laser. Off-resonant excitation at520 nm is also possible. Using both of these wave-lengths together results in a superlinear enhance-ment in the observed count rate, suggesting that the520 nm laser may play a role in stabilizing the SiV − charge state. The resulting fluorescence is sent to aspectrometer (Horiba iHR550, 0 .
025 nm resolution).We perform these measurements at 9–15 K.After annealing at 800 ◦ C, the observed distribu-tion is broad, with about half of the emitter transi-3
Fluorescence Wavelength (nm)Probability Density
736 738 740 74210203040 a. b. d. Position (μm)
ZPLIntensity(kcps) e. Position (μm)
PSBIntensity(kcps)
735 736 737 738 73950100150200250300Total Fluorescence Intensity (arb. u.)
Fluorescence Wavelength (nm) c. FIG. 2. Inhomogeneous distribution of fluorescencewavelengths of implanted SiV − transitions. a. Kerneldensity estimation of distribution of bulk SiV − wave-lengths after 800 ◦ C (N = 19, red dashed curve) and1100 ◦ C (N = 13, blue solid curve) annealing. The distri-bution narrows from 3–4 nm (800 ◦ C anneal) to 0 .
03 nm(15 GHz, 1100 ◦ C anneal). b. Zoomed-in distribution(transition C) after 1100 ◦ C annealing. Note the smallerwavelength range on the horizontal axis. c. Sum ofspectra for different SiV − centers after 1100 ◦ C anneal-ing. The SiV − fine structure is clearly present, demon-strating that the inhomogeneous distribution is small. d , e. Spatial map of collected fluorescence (thousandsof counts per second) over a region of bulk diamond ex-citing off (d) and on (e) resonance. By comparing thedensities of emitters, we estimate that 30 ±
15% of theemitters are nearly resonant. These measurements weretaken at 9–15 K. tion wavelengths lying within a 3–4 nm range (Fig.2a, red dashed curve). Transition C was used whereunambiguous identification was possible; otherwise,the brightest transition (which should correspond totransition C[11, 12]) was used. After the 1100 ◦ C an-neal, the distribution becomes more than 100 times narrower, with about half of the 13 measured emit-ters (transition C) now lying in a 0 .
03 nm (15 GHz)window (Fig. 2a and 2b, blue solid curves). Inboth cases, we focus on transition C because it isthe brightest transition and relatively insensitive tostrain[12] and phononic decoherence[29]. The othertransitions are also much more narrowly distributedafter 1100 ◦ C annealing. In Fig. 2c, we plot a com-posite spectrum constructed by summing over all ofthe normalized 13 SiV − spectra taken after 1100 ◦ Cannealing. This composite spectrum is very similarto the spectrum of a single unstrained SiV − cen-ter (Fig. 1c) and shows the expected fine-structuresplitting, demonstrating that the inhomogeneousdistribution of SiV − transition wavelengths is smallcompared to the fine-structure splitting. This resultis comparable to reported inhomogeneous distribu-tions reported for SiV − centers created during CVDgrowth[11–13, 15]. It is possible that even highertemperature annealing could further reduce this in-homogeneous distribution[20, 41].To estimate the yield of conversion from implantedSi + ions to SiV − centers, we perform scanning confo-cal microscopy (Fig. 2d). Exciting with several milli-watts of off-resonant light (700 nm) gives around 10 counts per second (cps) into a single-mode fiber froma single SiV − in a 20 nm spectral range around theZPL. In the resulting microscope image, we countthe number of SiV − centers and estimate a den-sity of 0.5–1 /µ m . Based on our Si + implantationdensity of 100 /µ m , we estimate our SiV − creationyield after 800 ◦ C annealing to be 0.5–1%. There wasno clear difference in the yield after performing the1100 ◦ C anneal. Furthermore, the yield in the sam-ple that was not pre-etched was significantly higher(2–3%). The observations that higher-temperatureannealing did not increase the yield and that thesample with greater surface damage had a largeryield both support the model that SiV − formationis limited by the presence and diffusion of nearbyvacancies[38, 39]. This yield could be increased byelectron irradiating the sample to create a highervacancy density in a controllable way[18, 37, 39].To visualize the density of nearly resonant SiV − centers, we resonantly excited the SiV − centers witha Rabi frequency of several GHz using an external-cavity diode laser tuned to the center of the in-homogeneous distribution. We scan spatially overthe sample and collect fluorescence on the phononside-band (PSB). The resulting image taken in thesame region of the sample (Fig. 2e) has about4 factor of three fewer emitters compared to theimage taken with off-resonant excitation (N ∼ ∼ B. SiV − centers in nanostructures One major advantage of building quantum deviceswith solid-state emitters rather than trapped atomsor ions is that solid state systems are typically moreeasily integrated into nanofabricated electrical andoptical structures[42, 43]. The scalability of thesesystems is important for practical realization of evensimple quantum optical devices[44]. Unfortunately,many solid-state systems suffer serious deteriorationin their properties when incorporated into nanos-tructures. For example, the large permanent elec-tric dipole of NV − centers in diamond causes cou-pling of the NV − to nearby electric field noise, shift-ing its optical transition frequency as a function oftime. The SiV − is immune to this spectral diffusionto first order because of its inversion symmetry[13]and is therefore an ideal candidate for integrationinto diamond nanophotonic structures. Motivatedby these considerations, we fabricated an array ofdiamond nanophotonic waveguides (Fig. 3a) on thepre-etched sample characterized above using previ-ously reported methods[30, 45]. Each waveguide(Fig. 3a, inset) is 23 µ m long with approximatelyequilateral-triangle cross sections of side length 300–500 nm. After fabrication, we again performed thesame 1100 ◦ C annealing and acid cleaning procedure.Many SiV − centers are visible in a fluorescence im-age of the final structures (Fig. 3b). Photon correla-tion measurements (Appendix B) verify our abilityto create and image single SiV − centers.To characterize the optical coherence propertiesof SiV − centers in nanostructures, we again per-form PLE spectroscopy. SiV − centers in nanostruc-tures have narrow transitions with a full-width athalf-maximum (FWHM) of Γ n / π = 410 ±
160 MHz(mean and standard deviation for N = 10 emitters;see Fig. 3c inset for linewidth histogram), only a fac-tor of 4 . γ/ π = 94 MHz. The linewidths measured in nanos-tructures are comparable to those measured in bulk(unstructured) diamond (Γ b / π = 320 ±
180 MHz).The ratios Γ n /γ and Γ b /γ are much lower than thevalues for NV − centers, where the current stateof the art for typical implanted NV − centers in b.d.a. 20 µm Si C
ZPLIntensity(kcps)
Laser frequency offset(GHz)Time (minutes)20 40 60 800.00.20.4
240 MHz
Intensity(kcps)Laser frequency offset (GHz) c. FIG. 3. SiV − centers in nanostructures. a. Scanningelectron micrograph of six nanobeam waveguides. In-set: schematic of a triangular diamond nanobeam con-taining an SiV − center. b. Spatial map of ZPL fluo-rescence collected by scanning confocal microscopy withoff-resonant excitation. Multiple SiV − centers are visiblein each waveguide. c. Linewidth of representative im-planted SiV − inside a nano-waveguide measured by PLEspectroscopy (blue points: data; red line: Lorentzian fit).Inset: histogram of emitter linewidths in nanostructures.Most emitters have linewidths within a factor of four ofthe lifetime limit. d. Spectral diffusion of the emittermeasured in part c. The total spectral diffusion is un-der 400 MHz even after more than an hour of continuousmeasurement. This diffusion is quantified by measuringthe drift of the fitted center frequency of resonance fluo-rescence scans as a function of time. Error bars are sta-tistical error on the fitted center position. The lighteroutline is the FWHM of the fitted Lorentzian at eachtime point. nanostructures[9] and in bulk[31] is Γ n /γ (cid:38) b /γ (cid:38)
10 ( γ/ π = 13 MHz for NV − centers).It is possible for the lifetime in nanostructuresto be longer than the lifetime in the bulk sincethe local photonic density of states is generally re-duced inside such a structure[46, 47]. This poten-tial change in lifetime would change the lifetime-limited linewidth and can also provide indirect evi-dence of the SiV − quantum efficiency. To probe thiseffect, we measured the lifetime of nine SiV − centers.The lifetime measured in the nanobeam waveguides( τ = 1 . ± .
14 ns, N = 5) was not significantly dif-ferent from the lifetime measured in the bulk-likeanchors ( τ = 1 . ± .
08 ns, N = 4). Both values arein good agreement with the literature[24, 29].5y extracting the center frequency of each individ-ual scan, we also determine the rate of fluctuationof the ZPL frequency and therefore quantify spectraldiffusion (Fig. 3d). Optical transition frequencies inSiV − centers are stable throughout the course of ourexperiment, with spectral diffusion on the order ofthe lifetime-limited linewidth even after more thanan hour. Characterizing the inhomogeneous distri-bution of SiV − centers in nanostructures is challeng-ing because off-resonant excitation leads to strongbackground fluorescence, making exhaustive identi-fication of all SiV − centers in a given region difficult.Nevertheless, it is easy to find multiple SiV − centersin nanostructures at nearly the same resonance fre-quency: to find the above ten emitters, we scannedthe laser frequency over only a 20 GHz range.The residual broadening of the optical transitioncan result from a combination of second-order Starkshifts and phonon-induced broadening. The pres-ence of a strong static electric field would result in aninduced dipole that linearly couples to charge fluc-tuations, accounting for the slow diffusion. Finally,we expect that up to 50 MHz of additional broad-ening could arise from the hyperfine interaction[48]present due to our choice of Si ions. Determiningthe precise mechanisms limiting SiV − linewidths isan important topic of future study.To conclude, we have presented optical emis-sion from implanted SiV − centers with a narrowinhomogeneous distribution of SiV − optical tran-sition wavelengths and nearly lifetime-limited op-tical linewidths. These properties persist afternanofabrication, making the SiV − center uniquelysuited for integration into quantum nanophotonicdevices[49, 50]. Recent advances in diamond fab-rication technology[30, 51, 52] suggest the tanta-lizing possibility of scalably integrating these high-quality implanted SiV − centers into nanowire sin-gle photon sources[46] or nanocavities[53, 54]. Fur-thermore, combining our processing procedure withtargeted implantation of silicon using a focused ionbeam[22] either before or after fabrication[55] couldsignificantly improve photonic device yields and re-producibility by deterministically positioning indi-vidual SiV − centers in all three dimensions. Ourwork, combined with the promise of these future ad-vances, could make the SiV − center a new workhorsein solid-state quantum optics. ACKNOWLEDGMENTS
We thank D. J. Twitchen and M. Markham fromElement Six Inc. for providing the electronic gradediamond samples, A. Sushkov and S. Meesala forhelp with annealing, and N. P. de Leon and K. DeGreve for help with etching and sample processing.We also thank Y. Chu, B. J. Shields, K. D. Jahnke,L. J. Rogers, and F. Jelezko for discussions andvaluable insight. M. L. Goldman and C. T. Nguyenhelped develop some of the software used in the ex-periment. M. K. Bhaskar contributed to figure de-sign.Financial support was provided by the NSF, theCenter for Ultracold Atoms, the Air Force Officeof Scientific Research MURI “Multifunctional Light-Matter Interfaces based on Neutral Atoms & Solids”,the DARPA QuINESS program, and the ARL. R. E.was supported in part by the NSF Graduate Re-search Fellowship Program. This work was per-formed in part at the Center for Nanoscale Systems(CNS) of Harvard University which is supported un-der NSF award ECS-0335765.
APPENDIX A: EXPERIMENTAL SETUP
520 nm700 nm737 nmDual AxisScanningMirror Fiberto APDT:R RatioBeam Splitter740 nmBandpassFilter Long Pass Dichroic Beamsplittersf = 50 cmAchromaticLens0.95 NAObjective (T:R) 10:90 ZPLPSB4 K
FIG. 4. Confocal microscope design. The 520 nm and700 nm lasers are used to excite the SiV − off-resonantly.The 737 nm external-cavity diode laser is used to excitethe SiV − resonantly. Collection can be performed eitheron the ZPL (if the excitation is off-resonance) or the PSB(in either excitation scheme). The experiments were carried out using home-built scanning confocal microscopes as illustrated inFig. 4. The three lasers used for excitation (520 nmand 700 nm diode lasers used for off resonant excita-tion, 737 nm external-cavity diode laser used for res-onant excitation) are combined using dichroic beam-6plitters. A 760 nm long-pass dichroic beamsplit-ter separates the PSB fluorescence from the rest ofthe optical channels. An additional bandpass filter(740 ±
13 nm) is used on the ZPL channel. Singlephotons are detected using single photon countingmodules (Picoquant τ -SPAD and Excelitas SPCM-NIR). The cryogenic measurements were performedin 4 K helium flow cryostats. We used a 0.95 NAmicroscope objective (Nikon CFI LU Plan Apo Epi100 × ) in all experiments. During the cryogenicmeasurements, the objective was inside the vacuumchamber and the sample was clamped with an in-dium foil spacer to the cold finger of the cryostat.During the PLE measurements, the 520 nm laser ispulsed at a ∼
5% duty cycle to stabilize the chargestate of the SiV − center[9, 31]. The detectors aregated off during these pulses. APPENDIX B: FLUORESCENCEAUTOCORRELATION MEASUREMENTS -30 -20 -10 0 10 20 300.00.51.01.52.0 τ (ns) g ( ) ( τ ) g ( ) ( τ ) FIG. 5. Fluorescence autocorrelation measurement of aSiV − center inside a diamond nanobeam as described inthe text. Error bars are estimated assuming the noise onthe number of detected photons follows a Poisson distri-bution (shot noise). The extent of the dip at τ = 0is limited by finite detector bandwidth: we measure g (2) (0) = 0 .
45; deconvolving the detector response yields g (2) (0) = 0 . To verify our ability to create single SiV − centers,we performed fluorescence autocorrelation measure-ments on SiV − centers inside diamond nanobeams.We performed this measurement by exciting theSiV − centers off resonantly as described above andsplitting the emission between two detectors in aHanbury-Brown–Twiss configuration. The relativearrival times of the photons on the two detectors were recorded using fast acquisition electronics (Pi-coQuant HydraHarp 400) with a resolution betterthan 128 ps. In this experiment, our total averagephoton count rate from this SiV − was 9 × countsper second.The relative photon detection times g (2) ( τ ) (nor-malized by defining g (2) ( ∞ ) = 1) from a represen-tative SiV − are displayed in Fig. 5. A value of g (2) (0) < . g (2) (0) = 0 . g (2) (0) = 0 .
15, indicating that the extent of our g (2) (0) dip is limited primarily by detector responseas expected. ∗ These authors contributed equally. † [email protected][1] J. L. O’Brien, A. Furusawa, and J. Vuˇckovi´c, Pho-tonic quantum technologies, Nat. Photonics , 687(2009).[2] P. Lodahl, S. Mahmoodian, and S. Stobbe, Inter-facing single photons and single quantum dots withphotonic nanostructures, Rev. Mod. Phys. , 347(2015).[3] M. W. Doherty, N. B. Manson, P. Delaney,F. Jelezko, J. Wrachtrup, and L. C. Hollenberg, Thenitrogen-vacancy colour centre in diamond, Phys.Rep. , 1 (2013).[4] H. Bernien, B. Hensen, W. Pfaff, G. Koolstra, M. S.Blok, L. Robledo, T. H. Taminiau, M. Markham,D. J. Twitchen, L. Childress, and R. Hanson, Her-alded entanglement between solid-state qubits sep-arated by three metres, Nature , 86 (2013).[5] B. Hensen, H. Bernien, A. Dr´eau, A. Reiserer,N. Kalb, M. Blok, J. Ruitenberg, R. Vermeulen,R. Schouten, C. Abell´an, et al. , Loophole-free Bellinequality violation using electron spins separatedby 1.3 kilometres, Nature , 682 (2015).[6] W. Pfaff, B. Hensen, H. Bernien, S. van Dam,M. Blok, T. Taminiau, M. Tiggelman, R. Schouten,M. Markham, D. J. Twitchen, et al. , Unconditionalquantum teleportation between distant solid-statequantum bits, Science , 532 (2014).[7] P. Tamarat, T. Gaebel, J. Rabeau, M. Khan,A. Greentree, H. Wilson, L. Hollenberg, S. Prawer,P. Hemmer, F. Jelezko, et al. , Stark shift control ofsingle optical centers in diamond, Phys. Rev. Lett. , 083002 (2006).[8] P. Siyushev, H. Pinto, M. V¨or¨os, A. Gali, F. Jelezko,and J. Wrachtrup, Optically controlled switching ofthe charge state of a single nitrogen-vacancy centerin diamond at cryogenic temperatures, Phys. Rev.Lett. , 167402 (2013).[9] A. Faraon, C. Santori, Z. Huang, V. M. Acosta, andR. G. Beausoleil, Coupling of nitrogen-vacancy cen-ters to photonic crystal cavities in monocrystallinediamond, Phys. Rev. Lett. , 033604 (2012).[10] E. Neu, D. Steinmetz, J. Riedrich-M¨oller, S. Gsell,M. Fischer, M. Schreck, and C. Becher, Single pho-ton emission from silicon-vacancy colour centres inchemical vapour deposition nano-diamonds on irid-ium, New J. Phys. , 025012 (2011).[11] L. J. Rogers, K. D. Jahnke, T. Teraji, L. Marseglia,C. M¨uller, B. Naydenov, H. Schauffert, C. Kranz,J. Isoya, L. P. McGuinness, et al. , Multiple intrin-sically identical single-photon emitters in the solidstate, Nat. Comm. , 4739 (2014).[12] H. Sternschulte, K. Thonke, R. Sauer, P. C.M¨unzinger, and P. Michler, 1.681-eV luminescencecenter in chemical-vapor-deposited homoepitaxialdiamond films, Phys. Rev. B , 14554 (1994).[13] A. Sipahigil, K. D. Jahnke, L. J. Rogers, T. Teraji,J. Isoya, A. S. Zibrov, F. Jelezko, and M. D. Lukin,Indistinguishable photons from separated silicon-vacancy centers in diamond, Phys. Rev. Lett. ,113602 (2014).[14] T. M¨uller, C. Hepp, B. Pingault, E. Neu, S. Gsell,M. Schreck, H. Sternschulte, D. Steinm¨uller-Nethl,C. Becher, and M. Atat¨ure, Optical signatures ofsilicon-vacancy spins in diamond, Nat. Comm. ,3328 (2014).[15] C. Hepp, T. M¨uller, V. Waselowski, J. N. Becker,B. Pingault, H. Sternschulte, D. Steinm¨uller-Nethl,A. Gali, J. R. Maze, M. Atat¨ure, et al. , ElectronicStructure of the Silicon Vacancy Color Center inDiamond, Phys. Rev. Lett. , 036405 (2014).[16] C. Lo, Natural Colorless Type IaB Diamond withSilicon-Vacancy Defect Center, Gems and Gemology L , 293 (2014).[17] A. M. Edmonds, M. E. Newton, P. M. Martineau,D. J. Twitchen, and S. D. Williams, Electron para-magnetic resonance studies of silicon-related defectsin diamond, Phys. Rev. B , 245205 (2008).[18] U. F. S. D’Haenens-Johansson, A. M. Edmonds,B. L. Green, M. E. Newton, G. Davies, P. M. Mar-tineau, R. U. A. Khan, and D. J. Twitchen, Opticalproperties of the neutral silicon split-vacancy centerin diamond, Phys. Rev. B , 245208 (2011).[19] E. Neu, C. Hepp, M. Hauschild, S. Gsell, M. Fischer,H. Sternschulte, D. Steinm¨uller-Nethl, M. Schreck,and C. Becher, Low-temperature investigations ofsingle silicon vacancy colour centres in diamond,New J. Phys. , 043005 (2013).[20] C. D. Clark, H. Kanda, I. Kiflawi, and G. Sittas,Silicon defects in diamond, Phys. Rev. B , 16681(1995). [21] J. L. Zhang, H. Ishiwata, T. M. Babinec, M. Radu-laski, K. Mller, K. G. Lagoudakis, C. Dory, J. Dahl,R. Edgington, V. Soulire, G. Ferro, A. A. Fokin,P. R. Schreiner, Z.-X. Shen, N. A. Melosh, andJ. Vukovi, Hybrid Group IV Nanophotonic Struc-tures Incorporating Diamond Silicon-Vacancy ColorCenters, Nano Letters , 212 (2016).[22] S. Tamura, G. Koike, A. Komatsubara, T. Teraji,S. Onoda, L. P. McGuinness, L. Rogers, B. Nay-denov, E. Wu, L. Yan, et al. , Array of brightsilicon-vacancy centers in diamond fabricated bylow-energy focused ion beam implantation, Appl.Phys. Express , 115201 (2014).[23] C. Wang, C. Kurtsiefer, H. Weinfurter, and B. Bur-chard, Single photon emission from SiV centres indiamond produced by ion implantation, J. Phys. B , 37 (2006).[24] B. Pingault, J. N. Becker, C. H. H. Schulte,C. Arend, C. Hepp, T. Godde, A. I. Tartakovskii,M. Markham, C. Becher, and M. Atat¨ure, All-optical formation of coherent dark states of silicon-vacancy spins in diamond, Phys. Rev. Lett. ,263601 (2014).[25] I. Aharonovich, S. Castelletto, D. Simpson, C. Su,A. Greentree, and S. Prawer, Diamond-basedsingle-photon emitters, Rep. Prog. Phys. , 076501(2011).[26] J. P. Goss, R. Jones, S. J. Breuer, P. R. Briddon,and S. ¨Oberg, The twelve-line 1.682 eV lumines-cence center in diamond and the vacancy-siliconcomplex, Phys. Rev. Lett. , 3041 (1996).[27] L. J. Rogers, K. D. Jahnke, M. W. Doherty, A. Di-etrich, L. P. McGuinness, C. M¨uller, T. Teraji,H. Sumiya, J. Isoya, N. B. Manson, et al. , Electronicstructure of the negatively charged silicon-vacancycenter in diamond, Phys. Rev. B , 235101 (2014).[28] A. Gali and J. R. Maze, Ab initio study of the splitsilicon-vacancy defect in diamond: Electronic struc-ture and related properties, Phys. Rev. B , 235205(2013).[29] K. D. Jahnke, A. Sipahigil, J. M. Binder, M. W.Doherty, M. Metsch, L. J. Rogers, N. B. Manson,M. D. Lukin, and F. Jelezko, Electron-phonon pro-cesses of the silicon-vacancy centre in diamond, NewJ. Phys. , 043011 (2015).[30] M. J. Burek, N. P. de Leon, B. J. Shields, B. J.Hausmann, Y. Chu, Q. Quan, A. S. Zibrov, H. Park,M. D. Lukin, and M. Lonˇcar, Free-standing me-chanical and photonic nanostructures in single-crystal diamond, Nano Lett. , 6084 (2012).[31] Y. Chu, N. P. de Leon, B. J. Shields, B. Hausmann,R. Evans, E. Togan, M. J. Burek, M. Markham,A. Stacey, A. S. Zibrov, A. Yacoby, D. J. Twitchen,M. Lonˇcar, H. Park, P. Maletinsky, and M. D.Lukin, Coherent Optical Transitions in ImplantedNitrogen Vacancy Centers, Nano Lett. , 1982(2014).[32] J. F. Ziegler, M. D. Ziegler, and J. P. Biersack,SRIM–The stopping and range of ions in matter , 1818(2010).[33] M. Hauf, B. Grotz, B. Naydenov, M. Dankerl,S. Pezzagna, J. Meijer, F. Jelezko, J. Wrachtrup,M. Stutzmann, F. Reinhard, et al. , Chemical con-trol of the charge state of nitrogen-vacancy centersin diamond, Phys. Rev. B , 081304 (2011).[34] G. Davies, S. C. Lawson, A. T. Collins, A. Main-wood, and S. J. Sharp, Vacancy-related centers indiamond, Phys. Rev. B , 13157 (1992).[35] P. De´ak, B. Aradi, M. Kaviani, T. Frauenheim,and A. Gali, Formation of NV centers in diamond:A theoretical study based on calculated transitionsand migration of nitrogen and vacancy related de-fects, Phys. Rev. B , 075203 (2014).[36] A. M. Zaitsev, Optical properties of diamond: adata handbook (Springer Science & Business Media,2001).[37] V. Acosta, E. Bauch, M. Ledbetter, C. Santori, K.-M. Fu, P. Barclay, R. Beausoleil, H. Linget, J. Roch,F. Treussart, et al. , Diamonds with a high densityof nitrogen-vacancy centers for magnetometry ap-plications, Phys. Rev. B , 115202 (2009).[38] T. Yamamoto, T. Umeda, K. Watanabe, S. On-oda, M. Markham, D. J. Twitchen, B. Naydenov,L. McGuinness, T. Teraji, S. Koizumi, et al. , Ex-tending spin coherence times of diamond qubitsby high-temperature annealing, Phys. Rev. B ,075206 (2013).[39] C. D. Clark and C. Dickerson, The 1.681 eV centrein polycrystalline diamond, Surf. Coat. Tech. ,336 (1991).[40] S. Pezzagna, B. Naydenov, F. Jelezko, J. Wrachtrup,and J. Meijer, Creation efficiency of nitrogen-vacancy centres in diamond, New Journal of Physics , 065017 (2010).[41] J. Orwa, C. Santori, K. Fu, B. Gibson, D. Simp-son, I. Aharonovich, A. Stacey, A. Cimmino, P. Ba-log, M. Markham, et al. , Engineering of nitrogen-vacancy color centers in high purity diamond byion implantation and annealing, J. Appl. Phys. ,083530 (2011).[42] T. D. Ladd, F. Jelezko, R. Laflamme, Y. Nakamura,C. Monroe, and J. L. O’Brien, Quantum computers,Nature , 45 (2010).[43] K. J. Vahala, Optical microcavities, Nature ,839 (2003).[44] Y. Li, P. C. Humphreys, G. J. Mendoza, and S. C.Benjamin, Resource Costs for Fault-Tolerant Lin-ear Optical Quantum Computing, Phys. Rev. X ,041007 (2015). [45] B. J. Hausmann, B. J. Shields, Q. Quan, Y. Chu,N. P. de Leon, R. Evans, M. J. Burek, A. S. Zibrov,M. Markham, D. J. Twitchen, et al. , Coupling of NVcenters to photonic crystal nanobeams in diamond,Nano Lett. , 5791 (2013).[46] T. M. Babinec, B. J. Hausmann, M. Khan,Y. Zhang, J. R. Maze, P. R. Hemmer, andM. Lonˇcar, A diamond nanowire single-photonsource, Nat. Nanotech. , 195 (2010).[47] Y. Chu and M. D. Lukin, Quantum optics withnitrogen-vacancy centers in diamond, arXiv preprintarXiv:1504.05990 (2015).[48] L. J. Rogers, K. D. Jahnke, M. H. Metsch,A. Sipahigil, J. M. Binder, T. Teraji, H. Sumiya,J. Isoya, M. D. Lukin, P. Hemmer, et al. , All-opticalinitialization, readout, and coherent preparation ofsingle silicon-vacancy spins in diamond, Phys. Rev.Lett. , 263602 (2014).[49] I. Aharonovich, A. D. Greentree, and S. Prawer,Diamond photonics, Nat. Photonics , 397 (2011).[50] B. J. Hausmann, B. Shields, Q. Quan,P. Maletinsky, M. McCutcheon, J. T. Choy,T. M. Babinec, A. Kubanek, A. Yacoby, M. D.Lukin, et al. , Integrated diamond networks forquantum nanophotonics, Nano Lett. , 1578(2012).[51] J. Riedrich-M¨oller, L. Kipfstuhl, C. Hepp, E. Neu,C. Pauly, F. M¨ucklich, A. Baur, M. Wandt, S. Wolff,M. Fischer, et al. , One-and two-dimensional pho-tonic crystal microcavities in single crystal diamond,Nat. Nanotech. , 69 (2012).[52] M. J. Burek, Y. Chu, M. S. Liddy, P. Patel,J. Rochman, S. Meesala, W. Hong, Q. Quan, M. D.Lukin, and M. Lonˇcar, High quality-factor opti-cal nanocavities in bulk single-crystal diamond, Nat.Comm. , 5718 (2014).[53] J. C. Lee, I. Aharonovich, A. P. Magyar, F. Rol,and E. L. Hu, Coupling of silicon-vacancy centersto a single crystal diamond cavity, Opt. Express ,8891 (2012).[54] J. Riedrich-M¨oller, C. Arend, C. Pauly, F. Mucklich,M. Fischer, S. Gsell, M. Schreck, and C. Becher, De-terministic coupling of a single silicon-vacancy colorcenter to a photonic crystal cavity in diamond, NanoLett. , 5281 (2014).[55] A. Sipahigil, R. E. Evans, D. D. Sukachev, M. J.Burek, C. T. Nguyen, J. Borregaard, M. K.Bhaskar, J. L. Pacheco, H. Atikian, R. M. Cama-cho, F. Jelezko, E. Bielejec, H. Park, M. Lonˇcar,and M. D. Lukin, (unpublished)., 5281 (2014).[55] A. Sipahigil, R. E. Evans, D. D. Sukachev, M. J.Burek, C. T. Nguyen, J. Borregaard, M. K.Bhaskar, J. L. Pacheco, H. Atikian, R. M. Cama-cho, F. Jelezko, E. Bielejec, H. Park, M. Lonˇcar,and M. D. Lukin, (unpublished).