Observation of large atomic-recoil-induced asymmetries in cold atom spectroscopy

Abstract

The atomic recoil effect leads to large (25 %) asymmetries in simple spectroscopic investigations of Ca atoms that have been laser-cooled to 10 µK. Starting with spectra from the more familiar Doppler-broadened domain, we show how the fundamental asymmetry between absorption and stimulated emission of light manifests itself when shorter spectroscopic pulses lead to the Fourier transform regime. These effects occur on frequency scales much larger than the size of the recoil shift itself, and have not been observed before in saturation spectroscopy. These results are relevant to state-of-the-art optical atomic clocks based on freely expanding neutral atoms. 42.62.Fi When we recently started using much colder Ca atoms for our ultra-high resolution optical clock spectroscopy, we discovered unexpectedly large asymmetries in the resulting spectra, which we found to be the direct result of atomic recoil.[1] Most surprising was that these effects appeared on a frequency scale more than ten times larger than the size of the recoil effect itself. In this letter we use a simple saturation spectroscopic configuration to show how the small effect of atomic recoil can cause these large asymmetries in the spectra of samples of freely expanding cold atoms. Not only are these effects of interest from the point of view of basic physics, they also have important implications for future optical clocks based on laser-cooled neutral atoms. [1,2] The effects of atomic recoil have been of considerable interest to laser spectroscopists for more than 30 years [3] and are the foundation of laser cooling of atoms. The first experimental evidence that saturation absorption spectroscopy produced a recoil splitting in the Doppler-free spectra of atomic and molecular lines was demonstrated by Hall, Bordé, and Uehara using a highresolution laser spectrometer.[4] Since then there have been several beautiful demonstrations of recoil splitting via various forms of saturation spectroscopy (see for example refs. [5,6,7,8]). In all of these studies the closely spaced recoil components were superimposed on a broad Doppler background, thereby obscuring the recoilinduced asymmetries inherent in light-atom interactions. In the work presented here, the width of the Doppler background is only seven times that of the recoil splitting, thereby clearly exposing the asymmetric aspect of the recoil components in the Doppler-broadened regime (see Fig 1). Moreover, by gradually changing the resolution, we show how this leads to large spectral asymmetries in the Fourier-transform limit. These effects differ from asymmetries described in earlier papers, which resulted from spontaneous emission or collisional quenching. [3,4] As described elsewhere, the recoil doublet can be readily understood from conservation of momentum and energy. [3,4] When an atom initially at rest absorbs a photon of frequency ν, it recoils with a velocity, v r = FIG. 1: Two-pulse saturated absorption spectrum as a function of laser detuning from the Bohr frequency with a probe pulse dura...

Comb coherence-transfer and cavity ring-down saturation spectroscopy around 1.65 μm: kHz-accurate frequencies of transitions in the 2ν3 band of 12CH4

“…Finally, we have to consider physical sources of bias in the dip positions: the transverse relativistic (or second order) The two doublet components are associated to absorption and to stimulated emission by molecules in the lower and upper levels of the transition, respectively. 38 They have same intensity and equal and opposite recoil shifts relative to the transition frequency, so their overlap results in a small broadening with no shift of the observed dip. Concerning the cavity modes wave fronts, in our case their curvature radius is large since the cavity is close to confocal (with mirrors of 1 m radius).…”

Absolute frequency measurement of the neutral40Ca optical frequency standard at 657 nm based on microkelvin atoms

“…1j4T) while maintaining a high signal-to-noise ratio as shown in figure 3. The spectra were Fourier-transform limited, so all of the atoms in the sample contributed nearly equivalently to the signaL We note that the asymmetric envelope seen in figure 3(b) is a consequence of atomic recoil effects [25] and can lead to small shifts of the line ceutre [15,21].…”

Experimental progress in positronium laser physics

“…The recoil limit for Ps is larger than the Doppler limit, which is not true for other atomic systems, where it is difficult even to observe such recoil effects [673][674][675]. In the context of making a Ps BEC it is necessary to draw a distinction between Ps laser cooling in free space, and in a confined volume of some sort, which is generally considered to be necessary to maintain Ps densities.…”