Symmetry Breaking Between Two Dynamic Attractors in the Parametrically-Driven Magneto-Optical Trap

Kihwan Kim,Myoung-Sun Heo,Kihwan Lee,Noh, Heung-Ryoul
DOI: https://doi.org/10.1109/iqec.2005.1560828
2005-01-01
Abstract:We have observed spontaneous symmetry breaking between populations of dynamic double wells produced by parametrically-driven magneto-optical trap. We found that population difference between both states mainly varied by total number of atoms in both wells. After Maxwell's Demon thought experiment, when there occur a symmetry breaking (SB) of temperature or population between two separated compartments filled with gas, it has been natural to search out underlying mechanism not explicitly revealed in the system. It is widespread in nature with examples from cosmology to biology and numerous studies about the symmetry breaking have been studied. Through the thought experiment, the second law of thermodynamics were developed and clearly understood with the concept of entropy. Recently in a vibro-fluidized granular gas, spontaneous SB of temperature and population between two boxes connected by a hole of certain height, were reported and understood by density dependent inelastic collision rate and employed to rachet. Nowadays there have been lots of studies about fluctuation-induced transitions in equilibrium and far from equilibrium states. The double well structure of those systems is very similar to that of the box separated into two compartments of the same section. When there are diffusions which come from thermal noise or spontaneous emissions, the populations of both states are nearly same except some fluctuations. We have studied the transitions between the dynamic double and triple well which were produced in the parametrically driven magneto-optical trap (MOT) system. Surprisingly, we have observed the population symmetry between dynamic double well in phase space was broken at certain experimental conditions. In MOT the atoms are almost non-interacting ideal gas, which is much different from the granular particles that have density dependent inelastic collision mechanism. Since inelastic collisions between atoms can be neglected in describing dynamics, the population symmetry breaking of our system is very strange and interesting phenomena. In this summary we report the experimental investigations of the SB. We have found the most important factor to cause SB was the total number of atoms in both states. There was the critical number of atoms above which population equality between dynamic states was broken. The critical number has been measured and studied at various experimental parameters such as modulation frequencies and amplitudes. Our experimental setup is nearly same to that reported in previous papers. At before studies we have observed two or three clouds (two oscillating states and one stable state) of atoms in MOT were produced by the parametric excitation, which were understood by double or triple wells in rotational phase space. Due to fluctuations from spontaneous emissions there were transitions between double or triple well, which makes the populations of two dynamic wells nearly equal, even though removing one population of two dynamic clouds. The transition rates were controlled by changing the modulation frequencies (f) and amplitudes (h), which were related to the depth of potentials in parametric space. We have observed SB of populations between two dynamic states at quite large parameter region of f and h, from super-critical bifurcation to sub-critical bifurcation. Figure 1(a) shows typical images of SB in populations. Figure 1(a) shows the CCD images at symmetry state (Left) and asymmetry state (Right). Figure 1(b) was obtained from the absorptions of resonant probe laser simultaneously. The probe laser propagated vertically to the oscillation axis of limit cycle (LC) through 3 mm left to the center of LC motion and had an intensity of 2 μW/cm and 1 mm width times 10 mm height. The area of each peak in the Fig. 1(b) was proportional to the number of atoms as explained in our previous works. These experiments were done at Bz-field gradient b = 14 G/cm, cooling laser detuning δ = -2.6 Γ, and intensity Iz = 0.39 Is, where Is is the averaged saturation intensity, 3.78 mW/ cm. The intensity of the other axes was 10 times larger than that of z-axis. Here the measured trap frequency was 43.6 QTuG1-6
What problem does this paper attempt to address?