Reactive and Non-reactive Quenching of O(1D) by the Potent Greenhouse Gases SO2F2, NF3, and SF5CF3

Laser flash photolysis of O3/Reactant/He mixtures at 248 nm has been coupled with time-resolved detection of ground state oxygen atoms by atomic resonance fluorescence spectroscopy to investigate the reactive and non-reactive quenching of O(1D) by the potent greenhouse gases SO2F2, NF3, and SF5CF3. Our results for the O(1D) + SO2F2 reaction are in good agreement with one previously published study [Dillion et al., Atmos. Chem. Phys., 8, 1547, 2008]. Deactivation of O(1D) occurs on essentially every collision with SO2F2 and the yield for reactive quenching, i.e., quenching that results in destruction of SO2F2, is a little greater than 50% over the temperature range 200-350 K. For the O(1D) + NF3 reaction, deactivation of O(1D) is about a factor of 6 slower than for SO2F2 but the yield for reactive quenching is close to 100%; our results suggest that removal of NF3 by O(1D) in the atmosphere is more than a factor of 2 faster than suggested by results reported in one previously published study [Sorokin et al., J. Chem. Phys., 108, 8995, 1998]. No reaction of O(1D) with SF5CF3 is observed. The upper limit rate coefficient for total removal of O(1D) by SF5CF3 is estimated to be 2 x 10-13 cm3 molecule-1 s-1, i.e., deactivation of O(1D) requires more than one thousand collisions with SF5CF3. To our knowledge, there are no kinetic data in the literature for the O(1D) + SF5CF3 reaction. As part of this study, we also obtained the first available kinetic information about O(1D) reactions with SO2 and CS2 at 298 K. Deactivation of O(1D) by both SO2 and CS2 is nearly gas kinetic; reactive quenching yields are 76% and 94% for SO2 and CS2, respectively. Reaction with O(1D) appears to be only a minor atmospheric loss pathway for SO2F2 and SF5CF3 but a very important loss pathway for NF3. Model calculations by Prather and Hsu [Geophys. Res. Lett., 35, L12810, 2008] suggest that the atmospheric lifetime of NF3 is reduced from 640 years to 550 years when the O(1D) + NF3 reaction is included as a loss pathway with an assumed rate coefficient for reactive quenching of 1.0 x 10-11 cm3 molecule-1 s-1 (as suggested by Sorokin et al.). Our results suggest that the atmospheric lifetime of NF3 is significantly shorter than 550 years.