Ionization of Excited Atoms in Thermal Collisions - The Atmosphere and Ionosphere - page 41

Tabl e 1. 4 Rate constants and activation energy U in binary collisions
of resonantly excited alkali atoms
Collision partners cm 3 s 1

U ,eV k ,

2Cs(6 2 P)

0.33 ˙ 0.04 (2 ˙ 0.2) 10-13, T D 425 K

2Rb(5 2 P) 0.20 ˙ 0.03 (3.2 ˙ 0.4)10-13, T D 470 K

2K(4 2 P)

0.1 (9 ˙ 2) 10-13, T D 500 K

2Na(3 2 P) - (3.8 ˙ 0.4) 10-11, T D 580 K

2Li(2 2 P) >0.5 10-15 10-16, Estimate

Tabl e 1. 5 Constants of the AI process in the binary collisions of the sodium atoms
in resonance states

k ,10 11 ,cm 3 s 1
(Bezuglov et al. 1987 ,
1989 )
Gas cell 3 2 P 1/2,3/2 T g 580 3.8
Gas cell 3 2 P 1/2,3/2 T g 650 1.1 *
Gas cell 3 2 P 1/2,3/2 T g 596 2.8
Single beam 3 2 P 3/2 T s 580 0.03 *
Crossing beams 3 2 P 3/2 T s 575 3.4 *
Laser cooling 3 2 P 3/2 T cool 0.75 10 3 1.1
Note :Here T g is the temperature in the gas cell, T s is the temperature of the source
beams, and T cool is the temperature of ultracold atoms in the laser trap. The asterisk
( * ) indicates constant values that were obtained self-consistently by Bezuglov et al.
( 1989 ) from the experimental data incorporating measurements in a vapor cell, a
single effusive beam, and two crossing beams

Experiment Excited states T , K

where

.T/ D .2T/ 1=2 =E

Here U 0 is the potential value corresponding to a point in terms of the avoided
crossing R D R c , is the value of the autoionization width at R max , V c is the relative
velocity of the particles, is the reduced mass, and T is the temperature expressed
in eV.
It became apparent that the accumulation of the experimental data on chemo-
ionization reveals differences between the results of the beam experiments and the
experiments carried out in a gas cell (plasma). The values are significantly higher
than the possible experimental errors (Table 1.5 ). The question arose about the
correctness of the published values of the rate constants (the “paradox of sodium”).
This situation has been analyzed by Bezuglov et al. ( 1989 ). It was shown that the
main reason for the scatter of the published experimental data is the peculiarities
of the distribution function of the excitation over atomic velocity F (¤) in different
experiments.
Calculations showed that F (¤) in the gas-cell plasma is significantly different
from the distribution function for a single beam. Intersecting beams with this respect
occupy an intermediate position. The function F (¤) at the same temperature of the