$\mathrm{\Lambda }$-Enhanced Imaging of Molecules in an Optical Trap

We report on nondestructive imaging of optically trapped calcium monofluoride molecules using in situ $\mathrm{\Lambda }$-enhanced gray molasses cooling. 200 times more fluorescence is obtained compared to destructive on-resonance imaging, and the trapped molecules remain at a temperature of $20\mu \mathrm{K}$. The achieved number of scattered photons makes possible nondestructive single-shot detection of single molecules with high fidelity.

Figure 1

(a) Three-level system exhibiting velocity-dependent dark states. Two ground states $|a⟩$ and $|b⟩$ are addressed separately by two counterpropagating laser beams. (b) Specific scheme for $\mathrm{\Lambda }$ cooling of CaF. The cooling light consists of two components addressing the $|J=1/2,F=1⟩$ and $|3/2,2⟩$ hyperfine manifolds. The single-photon detuning for $|1/2,1⟩$ and two-photon detuning are denoted by $\mathrm{\Delta }$ and $\delta$, respectively. The hyperfine spacing between $|3/2,2⟩$ and $|1/2,1⟩$ is $h\times 73.160\mathrm{MHz}$ [40]. (c) Schematic of $\mathrm{\Lambda }$-cooling beams, overlaid with a fluorescence image of $\mathrm{\Lambda }$-cooled CaF molecules. Molecules in the optical dipole trap (ODT) appear as a bright spot surrounded by a larger cloud of untrapped molecules. [$\mathrm{\Lambda }$ imaging of trapped molecules is shown in Fig. 4.]

Figure 2

Dependence of $\mathrm{\Lambda }$ cooling in free space on various parameters. (a) Temperature versus intensity $I$ and two-photon detuning $\delta$ at fixed single-photon detuning ($\mathrm{\Delta }=2.9\mathrm{\Gamma }$) and hyperfine ratio ($R_{2,1}=0.92$). (b) Temperature versus $I$ and $R_{2,1}$ with $\mathrm{\Delta }=3.4\mathrm{\Gamma }$ and $\delta =0$. $R_{2,1}$ is shown on a logarithmic scale, with the horizontal axis being 10 times the base-10 logarithm of the power ratios. (c) Temperature versus $\mathrm{\Delta }$ and $I$ with $R_{2,1}=0.92$ at $\delta =0$. For all plots, $I$ was varied in steps of $I_0=6.8{\mathrm{mW}/\mathrm{cm}}^2$.

Figure 3

(a) Number of molecules transferred into the ODT versus two-photon detuning $\delta$ at various trap depths $V$. With the exception of $\delta$, $\mathrm{\Lambda }$-cooling parameters are set to the free-space optimum ($\mathrm{\Delta }=3.9\mathrm{\Gamma }$, $I=14{\mathrm{mW}/\mathrm{cm}}^2$, $R_{2,1}=0.92$). Transferred number for different depths are shown in purple circles (trap depth of $30\mu \mathrm{K}$), blue stars ($50\mu \mathrm{K}$), green upward triangles ($60\mu \mathrm{K}$), yellow diamonds ($90\mu \mathrm{K}$), orange downward triangles ($130\mu \mathrm{K}$), and red squares ($160\mu \mathrm{K}$). Dotted lines show fits to a skewed Gaussian curve. (b) Transferred number versus intensity $I$ ($I_0=6.8{\mathrm{mW}/\mathrm{cm}}^2$) at a trap depth of $V=k_B\times 130(10)\mu \mathrm{K}$. (c) Transferred number versus $\delta$ at a trap depth of $V=k_B\times 130(10)\mu \mathrm{K}$ and intensity of $I=31{\mathrm{mW}/\mathrm{cm}}^2$. For all plots, $\mathrm{\Delta }=3.9\mathrm{\Gamma }$ and $R_{2,1}=0.92$.

Figure 4

(a) Fraction of molecules remaining versus imaging time for $\mathrm{\Lambda }$ imaging shown (blue circles) and resonant imaging (red squares). (b) Total number of photons scattered versus imaging time. (c) In situ $\mathrm{\Lambda }$ imaging of trapped molecules ($\mathrm{\Delta }=3.9\mathrm{\Gamma }$, $\delta =2\pi \times 90\mathrm{kHz}$, $I=31{\mathrm{mW}/\mathrm{cm}}^2$, and $R_{2,1}=0.16$). The exposure time is 200 ms, and 50 individual images are averaged.