All-Optical Atomic Bose-Einstein Condensates

Abstract

Given the tremendous impact of BEC research in last 7 years and the continued growth of the field, it is important to explore different methods for reaching BEC, particularly methods that offer new capabilities, simplicity, or speed. We have recently demonstrated such a method by creating a Bose condensate of Rb atoms directly in a crossed-beam optical dipole force trap using tightly focused CO2 gas laser beams [1]. In the broader scope of research with ultracold degenerate gases, our system stands out for several reasons. First, all-optical BEC provides the first new path to achieving BEC since the first pioneering demonstrations [2-4], and it is surprising simple and an order of magnitude faster than standard BEC experiments. Also, optical trapping potentials are essentially spin-independent and hence are well suited for studying the formation and dynamics of spinor condensates. Finally, we can engineer a rich variety of spatial confinements, including large period oneand threedimensional lattices that offer the possibility of optically resolving individual lattice sites. All-optical methods of reaching the BEC phase transition have been pursued since the early days of laser cooling. Despite many impressive developments beyond the limits set by Doppler cooling, the best previous efforts yielded atomic phase space densities a factor of 3 away from the BEC transition [5, 6]. Hence, optical traps have played only a supporting role in BEC experiments. The MIT group used a magnetic trap with an ‘optical dimple’ to reversibly condense a magnetically confined cloud of atoms evaporatively cooled to just above the phase transition [7]. Additionally, Bose condensates created in magnetic traps have been successfully transferred to shallow optical traps for further study [8]. In all these cases, however, magnetic traps provided the principle increase of phase space density (by factors up to 10) to the BEC transition. Evaporative cooling in optical traps was first demonstrated in 1994, where, starting with only 5000 atoms, a phase space density increase of a factor of ~30 was realized [9]. Whereas the first demonstrations of evaporative cooling of alkali atoms in magnetic traps lead quickly to the observation of BEC, the progress in optical traps was slower. A principle challenge faced by all-optical traps is that the small