Quantum Gases of Ground State Molecules

The quest for ground state molecules at ultralow temperatures and high phase space densities marks one of the latest developments in the rapidly evolving field of ultracold quantum gases. Here, we pursue full molecular state control and in particular Bose-Einstein condensation of molecules in the rovibrational ground state. We combine quantum gas and molecular state control techniques to achieve this goal: The molecules are first efficiently associated on a Feshbach resonance out of an atomic Bose-Einstein condensate (BEC) loaded into an optical lattice. They are then transferred to the rovibrational ground state by using two successive two-photon transitions by means of the STIRAP technique. Only in this state collisional stability of the molecules can be assured. We expect that a molecular BEC will form in a Mott-insulator-to-superfluid-type phase transition when the lattice potential is ramped down after the transfer. Our technique combines high production and transfer efficiencies with high phase space densities inherited from the initial atomic sample. Bose-Einstein condensates of deeply bound molecules make up a new, more complex class of Bose-Einstein condensates with composite particles. We expect that they will serve as an ideal starting point for the investigation of fully state selective collisional processes in the zero-temperature limit and for the study of fully coherent chemical processes. They constitute bright and extremely narrow-band sources for precision molecular spectroscopy, and they will provide a testing ground for elaborate strategies to form condensates of more complex molecular systems. This project is intended to bridge the boundaries between atomic, molecular, and condensed matter physics with strong linking to the newly emerging field of ultracold (coherent) chemistry.

The field of ultracold quantum gases has undergone a truly spectacular development since the year 1995, when, for the first time, the phase transition from a classical atomic gas to a Bose-Einstein condensate (BEC), a macroscopic matter wave, was observed for a dilute laser-cooled sample of Rb atoms. Since then, many different atomic species have been put into the state of BEC, e.g. Na, Li, K, Cs, and He, to name a few, and this novel state of matter has found applications in numerous experiments on superfluidity, matter-wave interferometry, quantum many-body physics, quantum state control, quantum information, low-dimensional quantum physics, etc. An immediate question has been (and still is!) the following: Can more complex objects than atoms, e.g. molecules, also be put into the state of a BEC? If yes, many new opportunities would open up, largely due to the fact molecules, with their rich internal structure, show non-trivial interaction effects that are relevant in particular to the fields of coherent control, ultracold chemistry, precision metrology, and quantum-many body dynamics, with strong implications for the field of quantum simulation. The present project is aimed at pushing the frontier on Bose-Einstein condensation of ground-state molecules. Specifically, we work the Cs dimers as prototype molecules. Since the central preparation technique of modern atomic physics, namely laser cooling, cannot be readily applied to molecular samples, novel techniques such as Feshbach association and coherent ground-state transfer have been developed to take molecular samples to the point of BEC formation. In essence, molecules are formed out of quantum degenerate atomic samples, and the main challenge is to maintain phase-space density during the preparation and state-transfer processes. Within the present project we had to fight two limitations: Non-perfect state transfer efficiencies (<1) and losses due to collisional processes. We have been able to push the four-photon stimulated transfer efficiency from threshold to the rovibrational ground state of the Cs dimer to about 80%. This put us in the position to study collisional processes in a state selective way, i.e. with control of the molecular hyperfine state. A crucial question to address is the following: How stable is the molecular sample, even if all inelastic two-body processes are endothermic and hence strongly suppressed? Interestingly, all our measurements so far show that the sample is collisionally unstable. If this were the case, BEC of molecules would not be possible. We are thus presently trying to find out what might be the reason for the collisional instability. Could it be that molecules, in the simplest of all possible processes, namely in a two-body collision, form a comparatively long-lived two-body complex, eventually separating again, but temporarily hiding their presence while, in the meantime, being prone to three-body loss? We have to admit that we have not yet solved this riddle and expect to have exciting new results in the near future.