Laser cooling and trapping of neutral atoms

C.S. Adams, E. Riis
1997 Progress in Quantum Electronics  
The ability to cool, manipulate, and trap atoms using laser light has allowed a new, rapidly expanding field to emerge. Current research focuses on improving existing cooling techniques, and the development of cold atoms as a source for applications ranging from atomic clocks to studies of quantum degeneracy. This review explains the basic mechanisms used in laser cooling and trapping, and illustrates the development of the field by describing a selection of key experiments. Copyright @ 1997
more » ... evier Science Ltd. All rights reserved 6 C. S. ADAMS & E. RIIS those presented here, except that ions can be trapped for much longer, thanks to their stronger interaction with electric and magnetic fields. The penalty, however, is that only a few ions can be trapped simultaneously due to their mutual Coulomb repulsion. The topic of ion trapping is outside the scope of this review and the interested reader is referred e1sewhere. (48) 2. HISTORICAL BACKGROUND Introduction The idea that light can exert pressure has been around for a long time. For instance, in the 17th century, Kepler speculated that the repulsion of comet tails from the sun may be due to light pressure. It was later realised that other processes were more important, but the hypothesis did identify a significant astrophysical effect, and stimulated further work on understanding its origin. However, it was not until Maxwell formulated his electromagnetic theory of light in 1 873(49) that a proper theoretical basis for the concept was established. He showed that an electromagnetic field exerts a pressure equal to its energy per unit volume. For light from the sun, or from a thermal source, the radiation pressure is extremely small. Even so, the value predicted by Maxwell was verified experimentally by Lebedev,(50' and Nichols and H~ll('~' around the turn of the century. An important step towards our present understanding of radiation pressure, and indeed the basis for the most intuitive model of the effect, came with the introduction of a quantum mechanical view of light. In 1917 Einstein showed that a quantum of light, or photon, with energy hv, carries a momentum, hvlc = h/h, where h is Planck's constant, c, v, and h are the speed, frequency, and wavelength of the light respective1y.'52' The basic mechanism for radiation pressure is the conservation of momentum during the absorption and emission of light. The photon momentum, directed along the propagation direction of the light, is expressed in terms of the wave vector k (k = 27-r/h) as Ak, where A = h/2rr. Striking evidence for the particle-like nature of radiation, as well as an excellent demonstration of the effect of conservation of both momentum and energy in the interaction of radiation with matter, were obtained shortly after the theory was developed. The Compton effect, discovered in the early 1920~,(~~) IS a manifestation of these conservation laws in the scattering of X-rays by electrons. The wavelength of the scattered X-rays was observed to increase by an amount known as the Compton wavelength, h, = h/m,c = 2.4 pm, where m, is the mass of an electron. The corresponding energy is transferred to the recoiling electron. Although the recoil of an atom produced by scattering a single photon is substantially smaller, the radiation pressure on atoms can be much larger due to the resonant nuture of the process. For an allowed optical transition, the resonant scattering rate is typically lo7 photons per second corresponding to an atomic acceleration of order 105g. If an atom of mass m absorbs a photon, the energy hv is almost entirely converted into internal energy, i.e. the atom ends up in an excited state. The momentum, however, causes the atom to recoil in the direction of the incoming light and change its velocity v by an amount Rk/m. The atom soon returns to the ground state by spontaneously emitting a photon. The conservation of momentum in this process causes the atom to recoil again. However, this time, the direction is opposite to that of the emitted photon. As spontaneous emission is a random process with a symmetric distribution given by the appropriate dipole radiation pattern, it does not contribute to the net change in momentum when averaged over many absorption/spontaneous emission cycles or a large sample of atoms. Figure 2 illustrates how an atom, on average, changes its velocity by an amount Rk/m each time it runs through this cycle. Laser cooling and trapping of neutral atoms The first experimental demonstration of this effect in an atomic system was reported by Frisch in 1933.(54) A well collimated thermal Na beam with a mean velocity of 900 ms-' was resonantly excited from the side with a Na lamp. The recoil (for the yellow Na resonance line) corresponds to a velocity change of 3 ems-' per scattered photon. Frisch observed this as a slight deflection away from the lamp. The results were consistent with an estimate that only one third of the atoms were excited. The low excitation rate was a fundamental limitation, which would persist until the development of narrow-band tunable lasers. The high spectral brightness of laser sources dramatically increases the rate of absorption-fluorescence cycles, resulting in a substantial force. Ashkin showed(55* %) that under realistic experimental conditions, the radiation pressure force produced by a laser resonant with a strong optical transition, such as the resonance transition of an alkali atom, could be used for isotope separation, velocity analysis, and atom trapping. The radiation pressure force is given by,t5@
doi:10.1016/s0079-6727(96)00006-7 fatcat:nxbhbdzhljaybjwbvgrdljtfvu