Raman cooling imaging: Detecting single atoms near their ground state of motion

We demonstrate imaging of neutral atoms via the light scattered during continuous Raman sideband cooling. We detect single atoms trapped in optical tweezers while maintaining a significant motional ground-state fraction. The techniques presented provide a framework for single-atom resolved imaging of a broad class of atomic species.

Figure 1

(a) Schematic of the experimental setup from above (top view) and in the plane perpendicular to $\widehat{z}$ (axial view). The Raman beams (RB1,2,3,4; red), optical pumping beams (OP; blue), polarization-gradient cooling beams (PGC; orange), and the quantization axis (B; black) are indicated in relation to the optical tweezer potential. The dashed lines indicate the beam polarizations. The light scattered by the atom is collected by the same objective lens used to generate the optical tweezers and is imaged using an $f=1$ m achromatic doublet. (b) Raman cooling cycle in 1D. Starting in the state $|F,m_F;n_{\mathrm{vib}}\rangle =|2,2;n\rangle$, a coherent Raman sideband transition to the $|1,1;n-1\rangle$ state is performed using a pair of Raman beams $(\mathrm{RB}1+\mathrm{RB}2,3,4)$. Subsequently, a spontaneous Raman transition optically pumps the atom back to the $|2,2;n-1\rangle$ state. By repeating this process, atom population will accumulate in the $|2,2;0\rangle$ state, which is ideally a dark state. (c) Example of an image generated via continuous Raman cooling. The scale bar (white) indicates separation between tweezer traps for imaging.

Figure 2

Photon detection histograms. The red (blue) bars on the right (left) indicate images taken with (without) an atom in the trap, as determined by the preselection image. The dashed (solid) black Gaussian curve approximates the expected distribution for the atomic signal (background). (a) Fluorescence signal from 25 ms of PGC imaging. (b) Signal collected during 1 s of continuous Raman cooling. The presence of an atom is triggered using the pre- and postselection PGC images. (c) Same as in (b), but now applying a sinusoidal parametric drive with $\Delta U/U\sim 0.055$, resonant with the radial trap frequency, to the optical tweezer potential.

Figure 3

(a) Varying the amplitude of the parametric drive applied to the optical tweezer potential during 1 s of Raman cooling imaging. The green diamonds are the atom survival probability, as determined using the pre- and postselection images, and the orange dashed line is the measured vacuum lifetime limit. The red circles (blue squares) are the signal collected with (without) an atom in the trap during the entire imaging period. The blue line indicates the average background value of 16.5 detected photons. The inset displays the equilibrium radial ground-state fraction $(P_{\mathrm{GS}}$ radial) measured via Raman sideband spectroscopy immediately following a period of Raman cooling imaging [23]. (b) Signal as a function of the parametric drive length, for fixed total imaging length (1 s) and drive amplitude $(\Delta U/U\sim 0.04)$. The red line is a linear fit to the data. All error bars indicate the standard error of the measurement.