The Possible Influence of the O(1D) + N2o isotope Exchange Reaction on the Isotopic Fractionation of N2O
|
|  Yüklə 80.5 Kb.
|
| The Possible Influence of the O(1D) + N2O Isotope Exchange Reaction on the Isotopic Fractionation of N2O
Yuk L. Yung1,2, Mao-Chang Liang1, Geoffrey A. Blake1,3 , Richard P. Muller4, and Charles E. Miller2 1 Division of Geological and Planetary Sciences, California Institute of Technology, Pasadena, CA 91125 2 Atmospheric Chemistry Element, Jet Propulsion Laboratory, Pasadena, CA. 91109 3 Division of Chemistry and Chemical Engineering, California Institute of Technology, Pasadena, CA 91125 4 Sandia National Laboratories, P.O. Box 5800, MS0196, Albuquerque, NM 87185-0196 Abstract. Recent experiments have shown that O(1D) + CO2 oxygen isotope exchange reaction can occur without quenching [Perri et al., 2003]. We propose an analogous O(1D) + N2O oxygen exchange reaction. We use ab initio quantum chemistry to investigate the structure and energetics of the potential energy surfaces on which the O(1D) + N2O reaction take place. Preliminary modeling results indicate we could account for the mass-dependent oxygen anomaly in N2O [Cliff and Thiemens, 1999] if the elastic exchange reaction were as high as 30-50% of the reactive channels. Introduction An understanding of the isotopic fractionation of nitrous oxide (N2O) is important for constraining the budget for its sources and sinks [see, e.g., Kim and Craig, 1993; Stein and Yung, 2003]. At present a satisfactory understanding has been achieved for explaining the mass-dependent fractionation [Blake et al., 2003; McLinden et al., 2003; Morgan et al., 2004]. However, there has not been a definitive explanation of the mass-independent fractionation (MIF) of the oxygen isotope anomaly of N2O. The MIF of N2O was discovered by Cliff and Thiemens [1997], whose notation we follow. See Thiemens et al. [2001] for an overview of the physical basis and applications of MIF to terrestrial and extraterrestrial environments. The mass dependent fractionation for oxygen isotope anomaly is given by 17O = 0.515 18O Define deviation from this as the residual from the above equation 17O = 17O 0.515 18O Cliff and Thiemens [1997] discovered that 17O 1 ‰ in the troposphere. Subsequent measurements confirmed and extended the result to the stratosphere [Cliff et al., 1999; Rockmann et al., 2001]. The latter reference gives 17O = 1.0 0.2 ‰ at 18O = 20.7 0.3 ‰. Rockmann et al. [2001] suggest a new source for N2O NH2 + NO2 N2O + H2O This source of N2O could have a 17O > 30 ‰, the ultimate source of the 17O being the reaction NO + O3 NO2 + O2 However, no quantitative modeling has been carried to evaluate the impact of this source on 17O. Recently, Perri et al. [2003] studied the isotopic exchange reactions for oxygen (O = 16O ; Q = 18O) and CO2
The rate coefficients are k2a = 6.710-11 and k2b = 4.910-11 cm3 s-1, summing up to k2 = 1.1610-10 cm3 s-1. The exchange reaction has not been measured. In this paper we use ab initio quantum chemistry to investigate the structure and energetics of the potential energy surfaces for the O(1D) + N2O reaction, so that an assessment can be made on the possibility of the elastic isotopic exchange reaction. Using the Caltech/JPL 2-D model, we carry out a modeling study to estimate the rate coefficient of the elastic isotopic exchange reaction that would be required to explain the observed 17O. Ab Initio Calculations To better understand the potential energy surfaces on which the NNO + O(1D) reaction takes place, we have used ab initio quantum chemistry to investigate the structures and energetics of various intermediates. We use density functional theory (Hohenberg 1964, Kohn 1965) with the B3LYP functional (Becke 1993) and the cc-pVTZ(-f) basis set (Dunning 1989). All calculations use the JAGUAR program suite, and we compute zero-point energy calculations on all structures. The O(1D) and O2(3) species have significant open-shell character. As a result, standard density functional theory techniques typically lead to large errors for these systems due to spin contamination. We therefore use the spin projection techniques of Wittbrodt and Schlegel (Wittbrodt 1996) to correct the energies for these species. At this level of theory, benchmark calculations have suggested an overall accuracy of ~3 kcal/mol. The results are summarized in Figure 1. All intermediates and products are predicted to be exothermic with respect to the reactants, in agreement with the (barrierless) kinetic results (DeMore 1997). For O(1D) attack on the N atom, there is a van der Waals entrance channel that leads to a trans-ONNO molecule that lies only ~20 kcal/mole above the 2NO products. Lying close in energy to the trans-ONNO structure is a cyclic variant in which one of the O-atoms is bound to both N-atoms. None of these structures can lead directly to O-atom exchange, and no barriers are predicted along this reaction pathway. For O-atom attack, however, there is a C2v intermediate that lies ~27 kcal/mole below the entrance channel. There is a ~16 kcal/mole barrier that leads directly to the N2 + O2 products at the level of theory employed. Though this barrier lies below the reactants, quantum mechanically there can still be reflections off this barrier. If the C2v intermediate is sufficiently long-lived (of order the vibrational period(s), or a few picoseconds), its symmetry should make O-atom exchange possible on the singlet surface. Quenching to the triplet surface may also lead to O-atom exchange. The degree to which these processes occurs will be sensitive not only to the height and shape of the pontential barrier that leads to N2 + O2, but also to the coupling of this path on the potential energy surface to other barriers that connect to the cyclic ONNO structure and thereby to the NO product channel. In principle, the degree of exchange could be calculated by multi-dimensional reactive scattering calculations, but a much finer sampling of the potential energy surface would be required. Such calculations are beyond the scope of this work, and so the ab initio calculations serve mainly to illustrate the plausibility of the reaction scheme under consideration and as a motivation for further experimental study of the O(1D) + N2O system. Modeling Results and Discussion The Caltech/JPL two-dimensional (2 -D) model has been described elsewhere [Morgan et al., 2004]. Four new reactions are added to the model. They are reaction (1) and where P = 17O. O(1D) + N2Q Q(1D) + N2O (3) P(1D) + N2O O(1D) + N2P (4) O(1D) + N2P P(1D) + N2O (5) For simplicity, we assume k1 = k3 = k4 = k5 = k2, is an unknown constant. One may ask why the addition of these four reactions could result in producing an MIF for N2O. The reason is simple. Suppose N2O were in photochemical equilibrium with O(1D), then the reactions (1), (3), (4) and (5) imply N2Q/N2O = Q(1D)/O(1D) N2P/N2O = P(1D)/O(1D) In other words, N2O equilibrate isotopically with O(1D). Since O(1D) is known to have MIF, its MIF is transferred to N2O, much in the same way as in the case of CO2 [Yung et al., 1991, 1997]. Mixing in the atmosphere tends to dilute the MIF, and the net effect is much less than that given by the isotopic equilibrium. Because Q P 100 ‰, the induced MIF may be non-trivial even after dilution by mixing. The results of our model for the case = 0.5 and Q = P= 100 ‰ are summarized in Figure 2. The solid line is the baseline case when the four new reactions were turned off. The dash-dot line shows the impact of the new exchange reactions. For comparison, the dashed line is a least squares fit through all the data points. It is most urgent to measure the elastic exchange rate coefficient. What we have shown is clearly the upper limit by assuming that all the MIF in N2O is derived from O(1D). A smaller exchange may still be able to account for part of the MIF. Acknowledgements. We thank R.-L. Shia for helping to run the 2-D model for N2O and M. Gerstell and J. Kaiser for helpful discussions. Y. L. Yung, Division of Geological and Planetary Sciences, Caltech, Pasadena, CA, 91125 (yly@gps.caltech.edu) Received: Revised: Accepted: References
Thiemens MH, Savarino J, Farquhar J, et al. Mass-independent isotopic compositions in terrestrial and extraterrestrial solids and their applications ACCOUNTS CHEM RES 34 (8): 645-652 AUG 2001 Figure 1. A summary of the ab initio density functional calculations on the O(1D) + N2O singlet reaction surface. All energies are in kcal/mole. Transition states for the 2NO product channel are connected by the dotted curve as a guide to the eye, the N2 + O2 product channel is highlighted by the solid curve. Experimental values for the reactant and product energetics are summarized at lower left. 
|