Coupling of a Single Nitrogen-Vacancy Center in Diamond to a Fiber-Based Microcavity

We report on the coupling of a single nitrogen-vacancy (NV) center in a nanodiamond to a fiber-based microcavity at room temperature. Investigating the very same NV center inside the cavity and in free space allows us to systematically explore a regime of phonon-assisted cavity feeding. Making use of the NV center’s strongly broadened emission, we realize a widely tunable, narrow band single photon source. A master equation model well reproduces our experimental results and predicts a transition into a Purcell-enhanced emission regime at low temperatures.

Figure 1

Cavity Setup: Fabry-Perot cavity consisting of a concave mirror on a fiber facet and a plane mirror onto which NDs have been spin coated. For further details see text.

Figure 2

(a) Room-temperature emission spectrum of a single NV center in a ND on a plane mirror. The spectrum has been fitted with 8 Lorentzian lines. (b) Model of the NV center with $n$ vibronic ground states $|g_i⟩$ and one excited state $|e⟩$. Broadening of the lines is due to spontaneous emission (${\gamma }_i$), pure dephasing (${\gamma }^{\star }$), and emission of phonons (${\gamma }_{i,i-1}$).

Figure 3

(a) Open squares: Intensity correlation $g^{(2)}(\tau )$ measurements at different excitation powers $P$ for a single NV center on a plane mirror outside the cavity. The curves are shifted vertically by 1 each for better visibility. $P=0.32\times$, $0.65\times$, $1.1\times$, $2.2\times$, $3.3\times$, $4.3\times$, $6.5\times$, $8.7\times$, $10.9\times$, $15.2\times P_{\mathrm{sat}}$ from bottom to top. Lines: Fit curves of $g^{(2)}(\tau )=1-(1+a)e^{-|\tau |/{\tau }_1}+a{e}^{-|\tau |/{\tau }_2}$. (b) Triangles: $g^{(2)}$ measurements at different excitation powers with the same NV center as in (a) inside the cavity. $P=0.25\times$, $0.51\times$, $0.84\times P_{\mathrm{sat}}$ from bottom to top. Lines: Fit curves with the same function as in (a). (c)–(e) Squares and triangles: Model parameters $a$, ${\tau }_1$, ${\tau }_2$ obtained from the fits in (a) and (b). Solid line: Theoretical dependence of model parameters on excitation power.

Figure 4

Cavity emission spectrum at effective cavity lengths (a) $l=3.5\mu \mathrm{m}$ and (b) $l=3.1\mu \mathrm{m}$. FSR denotes the free spectral range, ${\mathrm{TEM}}_{00}$ the fundamental transversal modes. The smaller modes at shorter wavelengths are higher order transversal modes.

Figure 5

(a) Intensity of ${\mathrm{TEM}}_{00}$ mode for fixed longitudinal mode number $n=17$ at different effective cavity lengths $l=n\times \lambda /2$. (b) Dots: Area of each ${\mathrm{TEM}}_{00}$ peak proportional to the photon count rates for $n=9–17$. Lines: Normalized count rates as obtained from the rate model.

Figure 6

(a) Simulated emission efficiency $P_{\mathrm{tot}}(\lambda )$ as a function of wavelength for two dephasing rates, ${\gamma }^{\star }=15\mathrm{THz}$ (dashed line) and ${\gamma }^{\star }=30\mathrm{GHz}$ (solid line), corresponding to room and cryogenic temperatures, respectively. (b) NV spontaneous emission lifetime and emission efficiency $P_{\mathrm{tot}}(\mathrm{ZPL})$ at the ZPL wavelength as a function of the pure dephasing rate ${\gamma }^{\star }$. For both (a) and (b) $\kappa =77\mathrm{GHz}$ ($\mathcal{F}=3500$) and $l=(9/2)\lambda$.