Deterministic Enhancement of Coherent Photon Generation from a Nitrogen-Vacancy Center in Ultrapure Diamond

The nitrogen-vacancy (NV) center in diamond has an optically addressable, highly coherent spin. However, a NV center even in high-quality single-crystalline material is a very poor source of single photons: Extraction out of the high-index diamond is inefficient, the emission of coherent photons represents just a few percent of the total emission, and the decay time is large. In principle, all three problems can be addressed with a resonant microcavity. In practice, it has proved difficult to implement this concept: Photonic engineering hinges on nanofabrication, yet it is notoriously difficult to process diamond without degrading the NV centers. Here, we present a microcavity scheme that uses minimally processed diamond, thereby preserving the high quality of the starting material and a tunable microcavity platform. We demonstrate a clear change in the lifetime for multiple individual NV centers on tuning both the cavity frequency and antinode position, a Purcell effect. The overall Purcell factor $F_P=2.0$ translates to a Purcell factor for the zero phonon line (ZPL) of $F_P^{\mathrm{ZPL}}\sim 30$ and an increase in the ZPL emission probability from about 3% to 46%. By making a step change in the NV’s optical properties in a deterministic way, these results pave the way for much enhanced spin-photon and spin-spin entanglement rates.

Popular Summary

Atom-sized defects known as “color centers” in the crystal structure of diamonds give these gemstones their unique colors. One type of color center, called a nitrogen-vacancy (NV) center, is also widely used as an experimental tool, as in recent seminal experiments proving the weirdness of quantum physics (specifically, remote entanglement). Applications in the burgeoning field of quantum information, however, demand that widely separated NV centers are efficiently coupled with each other. In principle, each NV emits photons that can be used to create entanglements between NV centers, provided the photons cannot be distinguished from each other. In practice, the coupling rate is low because of the small probability that photons emitted by different NVs are indistinguishable. We show a simple and flexible means of generating high entanglement rates by employing a tunable, miniaturized Fabry-Pérot microcavity.

While there has been progress in coupling NV centers to nanofabricated optical cavities, extensive nanofabrication tends to further degrade the quality of the emitted photons. Our device consists of a high-quality, minimally processed, single-crystalline diamond membrane bonded to a planar mirror; the microcavity is completed with a second, concave mirror. We achieve spectral and spatial tunability of the cavity mode using piezopositioners. For several NVs, we demonstrate a step change in the fraction of light suitable for remote entanglement from 3% to nearly 50%.

On account of the generic design, our approach is immediately applicable to other types of color centers and could constitute a key building block in many future quantum technologies.

Figure 1

(a) Schematic of the tunable microcavity containing a thin diamond membrane. Both the antinode location and resonance frequency of the microcavity mode can be tuned in situ. (b) PLE scans of near-surface (68 nm) NV centers in unprocessed diamond (blue symbols; the lower panel: resonant laser power 3 nW and scan speed $340\mathrm{MHz}/\mathrm{s}$) and in microstructured ($t_{\mathrm{d}}\le 1\mu \mathrm{m}$) diamond (red symbols; the central panel: resonant laser power 10 nW and scan speed $7.8\mathrm{GHz}/\mathrm{s}$), yielding averaged zero-phonon line widths of about 100 MHz and 1 GHz, respectively (upper panel). (c) Detaching a $20\times 20\mu {\mathrm{m}}^2$ membrane using a micromanipulator. (d) Images (recorded with a wavelength outside the DBR stop band) of the diamond microcavity demonstrating the in situ control of the lateral position. The arrows indicate the position of the concave top mirror.

Figure 2

(a) PL spectra around the ZPL transition for different air-gap detunings $\mathrm{\Delta }L$ about $L=1.96\mu \mathrm{m}$. Each resonance corresponds to the ZPL emission of a single NV center, as labeled. The PL is excited using a pulsed supercontinuum laser source ($P\sim 10\mathrm{mW}$, $\lambda =560\pm 20\mathrm{nm}$) and detected by a grating spectrometer. (b) Integrated PL versus $L$ for ZPL2 along with a Voigt fit (FWHM Lorentzian contribution ${\mathrm{\Gamma }}_L=60.6\mathrm{pm}$). Together with the cavity dispersion in (a), ${\mathrm{\Gamma }}_L$ determines the cavity $Q$-factor, $Q=\lambda /{\mathrm{\Gamma }}_{\lambda }=\lambda /({\mathrm{\Gamma }}_L·\mathrm{d}\lambda /\mathrm{d}L)=58500$. (c) Photon autocorrelation [$g^{(2)}(t)$] measurement on ZPL2 exhibiting a clear single-photon emission $g^{(2)}(0)=0.27$. The ZPL emission is filtered ($637\pm 7\mathrm{nm}$) and analyzed via a Hanbury Brown-Twiss setup (integration time: 45 000 s).

Figure 3

(a) PL decay curves of ZPL2 following pulsed excitation as a function of cavity-length detuning $\mathrm{\Delta }L$ for an acquisition time of 1000 s. The data for delays larger than 3 ns are fitted to a single-exponential convoluted with the instrumental response. The inset shows the normalized decay curves highlighting the clear change of the decay rate on changing $\mathrm{\Delta }L$. (b) Recombination rate ${\gamma }_{\mathrm{R}}$ versus $\mathrm{\Delta }L$ for a fixed lateral position. For each ZPL resonance, ${\gamma }_{\mathrm{R}}$ exhibits a Purcell effect. The experimental data are fitted to Lorentzian curves with FWHM ${\mathrm{\Gamma }}_L=(0.32\pm 0.05)\mathrm{nm}$. The color of the symbols matches the decay curves in panel (a). (c) Recombination rate versus lateral position detuning $\mathrm{\Delta }x$ on ZPL6 for zero spectral detuning. The experimental data are fitted to a Gaussian with FWHM $0.80\mu \mathrm{m}$.

Figure 4

(a) Left panel: Measured PL spectra on tuning the microcavity length $L$ over a wide range. The inset highlights the higher-order lateral modes. Right panel: Calculated mode dispersion. (b) The layer structure of the microcavity along with the refractive index dependence on $z$. The vacuum electric field $E_{\mathrm{vac}}$ is plotted against $z$ for $(x,y)=(0,0)$ for the lowest attainable fundamental microcavity mode. Parameters: Diamond thickness $t_d=0.77\mu \mathrm{m}$, air-gap thickness $L=1.96\mu \mathrm{m}$, diamond refractive index $n_d=2.41$, refractive indices of Bragg mirror $(n_H,n_L)=(2.06,1.46)$, and radius of curvature of top mirror $R=16\mu \mathrm{m}$.