[Experimental evaluation of the cross sections in Cs (6D(1/2)) + Cs(6S(1/2)) inelastic collisions].

At a Cs density higher than 9 x 10(14) cm(-3), cesium vapor was irradiated in a glass fluorescence cell with pulses of radiation from an YAG-laser-pumped OPO laser, populating 6D(5/2) state by two-photon absorption. Energy transfer in Cs6(D(5/2)) + Cs (6S) collisions was studied using methods of atomic fluorescence. At the different Cs densities, we have measured the time-integrated intensities of the components and fitted a three-state rate equation model to obtain the cross sections. The experimental points were fitted to a straight line very well. The authors converted the gradient and intercept into cross sections. The cross section for 6D(5/2)-->6D(3/2) transfer is (2.1 +/- 0.4) x 10(14) cm2. The cross section for excitation transfer out of the 6D doublet is sigmaQ = (1.6 +/- 0.4) x 10(-14) cm2. The cross section on contains information on reverse energy pooling collisions [i.e., Cs(6D(3/2)) + Cs (6S(1/2))-->Cs(6P) + 6Cs(P)] and contribution from mining in 6Dj-->7P(J'), This latter contribution could be subtracted out using the results of a second experiment. At a Cs density lower than 6.0 x 10(12) cm3, the laser was used to pump the 6D(3/2) and 6D(5/2) states, respectively. The resulting fluorescence included the direct component emitted in the decay of the 6D(J) state and the sensitized component arising from the collisionally populated 7P(J') state. Relative intensities of the components yielded the cross sections. The cross-sections for the processes Cs(6D(5/2)) + Cs(6S(1/2))-->Cs(7P(J')) + Cs(6S(1/2)) are (1.6 +/- 0.5) x 10(-15) cm2. for J'= 3/2 and (6.5 +/- 2.1) x 10(-16) cm2, for J' = 1/2, respectively. The cross-sections for the processes Cs(6D(3/2) + Cs(6S(1/2))-->Cs (7P(J')) + Cs(6S(1/2)) are (1.9 +/- 0.6) x 10(-15) cm2. for J' = 3/2 and (7.6 + 2.4) x 10(-16) cm2, for J' = 1/2, respectively. The 6D(J) -->7P(J'), energy transfer rate coefficient is small. The total quenching rate coefficient out of the 6D(J) state is much larger. Evidence suggests that the quenching of the 6D(J) state is caused predominantly by reverse energy-pooling process. The cross section for this process, i.e., Cs(6D(3/2))+Cs(6S(1/2))-->Cs(6P) + Cs(6P) is (1.3 +/- 0.4) x 10(-14) cm2.

[Collisional energy transfer between excited Rb atoms].

Energy pooling (EP) was observed in Rb vapor following pulsed optical excitation to the 5P1/2 state. The 5P3/2 state was populated by the energy transfer process: Rb(5P1/2)+Rb(5S1/2) --> Rb(5P3/2)+Rb(5Sl/2). The resulting densities of Rb atoms at the 5P1/2 level were obtained from the absorption of narrow spectral line from a Rb hollow cathode lamp, connecting the 5P1/2 state to 7S state. Since the effective lifetimes of the 5P1/2 and 5P3/2 states are approximately equal, the densities of the 5P3/2 level were obtained from the D2 to D1 fluorescence ratios where D1 and D2 are lines of the 5P1/2 --> 5S1/2 and 5P3/2 --> 5S1/2 transition. Because the time of the fine structure exchanging is much shorter than the lifetime of the 5D state, the fluorescence originating from the 5D state produced by the 5P1/2 + 5P3/2 and 5P3/2 + 5P3/2 processes follows the instantaneous production rate of the 5P1/2 + 5P1/2 process. It is clear that 5P/2 + 5P3/2 and 5P3/2 + 5P3/2 collisions can significantly influence the results obtained for the 5P1/2 + 5P1/2 rate since the energy defect for 5D state is much smaller for 5P1/2 + 5P3/2 and 5P3/2 + 5P3/2 collisions than for 5P1/2 + 5P1/2 collisions. Effective lifetimes of the 5P levels were calculated using radiation trapping theory. The time-integrated populations and signals were studied and analyzed. The resulting fluorescence included the direct component emitted in the decay of the optically excited 5P1/2 state and the sensitized component arising from the collisions for populating 5D state at different cell temperature. These relative intensities were combined with the measured excited atom densities to yield absolute energy-pooling rate coefficients. The cross sections (in units of 10(-14) cm2) for the energy-pooling collisions [i. e. , 5P1/2 + 5P1/2, 5P1/2 + 5P3/2, 5P3/2 + 5P3/2] are 0.78, 2.9 and 3.1, respectively. The dependence of the rates upon energy defect deltaE was examined, but the 5D3/2 level was approximately equally populated in 5P3/2 + 5P3/2 (deltaE = 68 cm(-1)) and 5P3/2 + 5P1/2 (deltaE = 306 cm(-1)) collisions. The 5P1/2 + 5P3/2 collisions are as efficient as 5P3/2 + 5P3/2 for populating 5D3/2 state.

[Reactive and nonreactive energy transfer in Cs(6D(J))+ (H2, He) collisions].

Cs vapor, mixed with a gas was irradiated in a glass fluorescence cell with pulses of 886nm radiation from a YAG-laser-pumped OPO laser, populating 6D3/2 state by two-photon absorption. Cross sections for 6D3/2 --> 6D5/2 transition induced by collisions with various H(e) atoms and H2 molecules were determined using methods of atomic fluorescence. The resulting fluorescence included a direct component emitted in the decay of the optically excited state and a sensitized component arising from the collisionally populated state. At the different densities, we have measured the relative time-integrated intensities of the components and fitted a three-state rate equation model to obtain the cross sections for 6D3/2 --> 6D5/2 transfer: sigma = (55 +/- 13) x 10(-16) and (16 +/- 4) x 10(-16) cm2 for H2 and H(e), respectively. The cross sections for the effective quenching of the 6D5/2 state were also determined. The total transfer rate coefficients from the 6D5/2 state for H(e) is small [1.2 x 10(-10) cm3 x s(-1)]. The total quenching rate coefficient of the 6D5/2 state is larger for H2 [6.7 x 10(-10) cm3 z s(-1)]. For H2 case, the quenching rate coefficient corresponds to reaction and nonreactive energy transfer. Evidence suggests that the nonreactive energy transfer rate coefficient is [6.3 x 10(-10) cm3 x s(-1)]. Hence the authors estimated the cross section (2.0 +/- 0.8) x 10(-16) cm2 for reactive process Cs(6D5/2) + H2 --> CsH + H. Using the dependence on the pressure of H2 (or H(e)) of the integrated fluorescence monitored at the 6D5/2 --> 6P3/2 transition the cross section (4.0 +/- 1.6) x 10(-16) cm2 for Cs (6D3/2) + H2 --> CsH + H was obtained. Thus, the relative reactivity with H2 follows an order of Cs (6D3/2) > Cs (6D5/2).

[Spectroscopic investigation of retrofluorescence in a pure optically thick Rb vapour near the surface].

A low-power tunable laser was used to populate the Rb(5P(3/2)) hyperfine-structure levels in a pure optically thick vapour in the presence of a dissipative surface. The retrofluorescence intensities and spectrum profile for the 780 nm (5P(3/2)--> 5S(1/2)) and 795 nm (5P(1/2)-->5S(1/2)) lines were measured and analyzed. The glass-vapor interface was considered as two distinct regions, a wavelength-thickness vapor layer adjacent to the surface and a more remote vapor region The first region was analyzed as a spectral filter that annihilated the absorbed photons and the second one as a rich spectral light source. The authors discussed two possible mechanisms for the 5P(1/2) population in the cell [i.e., mechanism(1): collisions Rb(5P(3/2))+Rb(5S(1/2))-->Rb(5P(1/2))+Rb(5S(1/2)); mechanism(2): collisions Rb(5D)+Rb(5S)-->Rb(5P)+Rb(5P)]. For each one of the possible mechanisms considered, the authors gave the theoretical formulation of the retrofluorescence integrated signal associated with 795 nm(5P(1/2)-->5S(1/2)), which was compared with experiment. Two important characteristic aspects of retrofluorescence spectra must be taken into account when dealing with retrofluorescence signals for atomic process investigation: the retrofluorescence intensity dependence on laser power and sensitized laser retrofluorescence line shapes. When the laser frequency is scanned through the hyperfine resonance line, the sensitized retrofluorescence spectra signal corresponding to the 795 nm line has a profile similar to the profile of the retrofluorescence signal at the 780 nm. The authors have pointed out that mechanism(1) gives the linear dependence of the trtrofluorescence as a function of laser power and the spectrum profile. The population of the 5P(1/2) atomic level in an optically thick vapour can be principally explained by the fine-structure excitation transfer process [mechanism(1)]. It appears from our experimental and theoretical investigations that, the spectral properties of the laser-induced Rb 795 nm sensitized retrofluorescence in a pure optically thick vapour near a dissipative surface cannot be explained by the mechanism(2).