Bose-Einstein condensation of alkaline-earth atoms

Single atoms prepared in an excited state randomly return to the ground state with a probability approaching unity with exponential time dependence. However, if many atoms are excited, the decay characteristic dramatically changes. The first spontaneously emitted photon triggers adjacent atoms to emit in phase, such that a classical coherent light pulse emerges on a time-scale much shorter than the excited state lifetime, a phenomenon termed superradiance. The time, when such a pulse attains maximal intensity, is determined by the time when the first spontaneous photon emerges, and hence fluctuates from shot to shot. Therefore, the stochastic nature of spontaneous emission becomes directly visible in the easy to measure arrival time statistics of intense classical light pulses. We have studied this arrival time statistics in the superradiant emission of calcium atoms and found excellent quantitative agreement with an analytical model. Our work may be extended to build a superradiant calcium laser operating in a continuous mode with a spectral purity beyond what is presently possible with state of the art conventional lasers. Superradiant lasers could be useful for reading out atomic clocks with unprecedented precision.

Pulse Delay Time Statistics in a Superradiant Laser with Calcium Atoms
Torben Laske, Hannes Winter, and Andreas Hemmerich
Phys. Rev. Lett. 123, 103601 (2019)

To further increase phase space density in our cold atomic sample, we switch on a second dipole trap beam - crossed with respect to the first - in our experiment. This creates a more confined potential for the atoms. By ramping down the powers of the two dipole trap beams, we force evaporation in the dipole trap and reach degeneracy in about 1 s. Time-of-flight images for evaporation times of (a) 375, (b) 695, (c) 1015 and (d) 1115 ms (from top to bottom) are shown on the left.

http://pra.aps.org/pdf/PRA/v85/i3/e031603

We utilise the weak decay channel of the principal fluorescene line to the metastable 3P2 state to load atoms into a second MOT operating on the 3P2-3D3 transition. We now have atoms in the metastable triplet 3P2 state. We operate a dipole trap for ground state atoms, which can be loaded using light at 430 nm resonant with the [4s4p]3P2-[4p2]3P2 transition. The [4p2]3P2 state eventually decays to the ground state via the [4p2]3P2-[4s4p]3P1-1S0 transition. The depumping light is extremely weak, and has a very narrow beam waist. This ensures that only the slowest atoms of the second MOT stay long enough within the volume of the depumper and are de-excited to the ground state. Furthermore, the small volume of the depumping beam is adjusted well within the volume of the dipole trap. Thus, all depumped atoms are effectively captured in the dipole trap.

pra.aps.org/abstract/PRA/v76/i3/e033418

The challenge is to produce sufficiently dense and cold atoms in the metastable 3P2-state. We proceed in two steps. Atoms in their ground state are collected from a Zeeman cooler into a Doppler-limited standard magneto-optical trap using the strong 422 nm fluorescence line 34.6 MHz linewidth). The 3P2-state is continuously loaded from this MOT at a rate of 2*1010 atoms per second. A second MOT operating at 1978 nm is superimposed acting on the 3P2-state. Because of its narrow linewidth of 57 kHz this infrared transition provides a Doppler-temperature of only 1.4 μK. In addition, its Zeeman substructure implies the presence of polarization gradient cooling with the promise of temperatures approaching the recoil limit of 122 nK.

http://pra.aps.org/abstract/PRA/v65/i4/e041401

Magnetic Trapping

The 3P2 state is magnetic and it is possible to load 3P2 metastable atoms into a magnetic trap. Metastable atoms have been magnetically conveyed from the MOT into a miniaturized magnetic trap (QUIC geometry, 10 W heat dissipation). Inside the magnetic trap the atoms were Doppler-cooled by means of 1978 nm light.

Doppler-Cooling in the magnetic trap was used to selectively reduce the temperature with respect to the vertical axis. The elastic and inelastic collision rates of the metastable atoms in the magnetic trap were, however, determined to be exceptionally large and of the same order of magnitude, such that evaporative cooling is impossible in such a trap.

http://pra.aps.org/abstract/PRA/v67/i2/e021401
http://prl.aps.org/abstract/PRL/v96/i7/e073003