Light narrowing of magnetic resonances in ensembles of nitrogen-vacancy centers in diamond

We investigate optically detected magnetic resonance signals from an ensemble of nitrogen-vacancy centers in diamond. The signals are measured for different light powers and microwave powers, and the contrast and linewidth of the magnetic-resonance signals are extracted. For a wide range of experimental settings of the microwave and light powers, the linewidth decreases with increasing light power, and more than a factor of 2 “light narrowing” is observed. Furthermore, we identify that spin-spin interaction between nitrogen-vacancy centers and substitutional nitrogen atoms in the diamond leads to changes in the line shape and the linewidth of the optically detected magnetic resonance signals. Finally, the importance of the light-narrowing effect for optimizing the sensitivity of magnetic-field measurements is discussed.

Figure 1

(a) Two-level model of ODMR dynamics. Two states $|0\rangle$ and $|1\rangle$ are coupled with a MW field (Rabi frequency ${\Omega }_R$ and detuning $\Delta$); ${\Gamma }_P$ optical-pumping rate; ${\gamma }_1$ longitudinal spin-relaxation rate; ${\gamma }_2$ transverse spin-relaxation rate. (b) Level scheme of the NV${}^{-}$ center. (c) Schematic of the confocal setup. (d) Points: Measured red fluorescence $P_{fl}$ as a function of green pump power $P$. Solid line is a fit to a function of the form $P_{fl}=kP/(1+P/P_{\mathrm{sat}})$.

Figure 2

Examples of ODMR signals taken with the S5 (a)–(e) and S3 (f) diamond samples at room temperature. (a) ODMR signal with a wide scan range of the MW frequency. (b) Narrow scan range ODMR signal of the magnetic resonance centered around 2654 MHz. (c) Light power fixed and four different settings of the MW power. (d) MW power fixed at a relatively high value and four different settings of the light power. (e) Demonstration of light narrowing with the S5 sample. Notice the two $y$ axes. (f) Demonstration of light narrowing with the S3 sample. The bottom spectrum ($P=500$ mW setting) has been displaced vertically by $-0.001$.

Figure 3

Points: Experimental results of the FWHM and amplitude of ODMR signals as a function of Rabi frequency and light power. Solid lines: Global fits to the FWHM and the amplitude. The uncertainties on the experimental data points correspond to the standard deviations found from repeated measurements of the width and contrast.

Figure 4

Points: Fit parameter $a/{\gamma }_2^{\mathrm{intr}}$ for different light powers. The uncertainties on the points equal the 68$\%$ confidence intervals for the fit parameters $a\left(P\right)/{\gamma }_2^{\mathrm{intr}}$ calculated from the nonlinear least-squared fit of the measured widths to Eq. (9). Line: Weighted fit of $a\left(P\right)/{\gamma }_2^{\mathrm{intr}}$ to the function $a_{1}P/(1+P/b_1)+c_1$.

Figure 5

Contour plot of the experimental magnetic-field sensitivity $S_B$ as a function of light power $P$ and Rabi frequency $f_R$. Notice the logarithmic scales.

Figure 6

Five-level model of ODMR dynamics. The two ground states $|0\rangle$ and $|1\rangle$ are coupled with a MW field (Rabi frequency ${\Omega }_R$ and detuning $\Delta$). Laser light excite the states $|0\rangle$ and $|1\rangle$ with equal rate ${\tilde{\Gamma }}_P$ to the excited states $|e0\rangle$ and $|e1\rangle$, respectively. The excited states decay to the ground state with rate ${\Gamma }_0$. The state $|e1\rangle$ can also decay to the singlet state $|s\rangle$ with rate ${\Gamma }_f$. The singlet state $|s\rangle$ decays with a rate ${\Gamma }_s$ to either $|0\rangle$ or $|1\rangle$.