화학공학소재연구정보센터
Journal of Physical Chemistry, Vol.99, No.43, 15919-15924, 1995
Collision-Induced Radiative Transitions B(1)Sigma(+)(G)-)A(1)Delta(+)(G), B(1)Sigma(+)(G)-)X(3)Sigma(-)(G), and A(1)Delta(G)-)X(3)Sigma(-)(G) of O2
The quantum yields of the b(1) Sigma(g)(+) --> a(1) Delta(g) and the b(1) Sigma(g)(+) --> X(3) Sigma(g)(-) emissions of O-2 have been determined for the first time in solution. In dry CCl4, where the lifetime of O-2(b(1) Sigma(g)(+)) amounts to tau(Sigma) = 130 +/- 10 ns, we obtained Q(b-a) = (4.5 +/- 1.6) x 10(-4) and Q(b-X) = (5.2 +/- 1.8) x 10(-8), corresponding to rate constants of the respective radiative processes of k(b-a) = (3.4 +/- 1.1) x 10(3) s(-1) and k(b-X) = 0.40 +/- 0.13 s(-1). The rate constant of the phosphorescence b --> X is only weakly enhanced by collisions with solvent molecules, in contrast to the b --> a fluorescence rate constant, which is by about 6 orders of magnitude larger in liquid CCl4 compared with the isolated O-2 molecule. For the a(1) Delta(g) --> X(3) Sigma(g)(-) phosphorescence we determined in CCl4 k(a-X) = 1.1 +/- 0.4 s(-1). These measurements proved to be key experiments for the understanding of the mechanism of the collision-induced radiative transitions b --> a and a --> X. Our results confirm predictions of a theoretical perturbation model developed by Minaev, according to which the ratio k(a-X)(c)/k(b-a)(c) of the bimolecular rate constants of the respective collision-induced radiative transitions should be constant. The comparative analysis of k(b-a)(c) obtained in liquid CCl4 with literature data of k(b-a)(c) determined in the gas phase reveals an almost quadratic dependence of k(b-a)(c) on the molar refraction R of the collider. This relation demonstrates that the electric dipole moment induced into the b --> a transition is practically directly proportional to the molecular polarizability of the collider. Because of intensity borrowing of the a --> X transition from the collision-induced b --> a transition, k(a-X)(c) depends in the same way on R as k(b-a)(c). This result also explains the up to now quantitatively not understood solvent effect on k(a-X)(c).