Removal of the potent greenhouse gas NF3 by reactions with the atmospheric oxidants O(D-1), OH and O-3

Nitrogen trifluoride, NF3, a trace gas of purely anthropogenic origin with a large global warming potential is accumulating in the Earth's atmosphere. Large uncertainties are however associated with its atmospheric removal rate. In this work, experimental and theoretical kinetic tools were used to study the reactions of NF3 with three of the principal gas-phase atmospheric oxidants: O(D-1), OH and O-3. For reaction (R2) with O(D-1), rate coefficients of k(2)(212-356 K) = (2.0 +/- 0.3) x 10(-11) cm(3) molecule(-1) s(-1) were obtained in direct competitive kinetics experiments, and experimental and theoretical evidence was obtained for F-atom product formation. These results indicate that whilst photolysis in the stratosphere remains the principal fate of NF3, reaction with O(D-1) is significant and was previously underestimated in atmospheric lifetime calculations. Experimental evidence of F-atom production from 248 nm photolysis of NF3 was also obtained, indicating that quantum yields for NF3 destruction remain significant throughout the UV. No evidence was found for reaction (R3) of NF3 with OH indicating that this process makes little or no contribution to NF3 removal from the atmosphere. An upper-limit of k(3)(298 K) < 4 x 10(-16) cm(3) molecule(-1) s(-1) was obtained experimentally; theoretical analysis suggests that the true rate coefficient is more than ten orders of magnitude smaller. An upper-limit of k(4)(296 K) < 3 x 10(-25) cm(3) molecule(-1) s(-1) was obtained in experiments to investigate O-3 + NF3 (R4). Altogether these results underpin calculations of a long (several hundred year) lifetime for NF3. In the course of this work rate coefficients (in units of 10(-11) cm 3 molecule(-1) s(-1)) for removal of O(1 D) by n-C5H12, k(6) = (50 +/- 5) and by N-2, k(7) = (3.1 +/- 0.2) were obtained. Uncertainties quoted throughout are 2 sigma precision only.