The HF† infrared chemiluminescence from the reactions of fluorine atoms with PH3, SiH4, H2O, H2O2, H2S, NH3, N2H4, and three additional carbon compounds, (CH3)2O, (CH3)2S, and (CH3)3N, was studied in the same way as described in the preceding paper. With the exceptions of the N2H4 reaction, a large fraction of the energy, ≳ 45%, was released as vibrational energy of the HF† product, and population inversions were frequently found. The strong chemical interaction between HF† and NH3, (CH3)3N, N2H4, and PH3 may have led to efficient vibrational relaxation for these reactions; however, the redissociation of the adducts may give mainly HF† (v = 0), and modification of the vibrational distribution may not be severe. The results can be used to establish upper limits to some uncertain bond energies: D(SiH3–H)≤ 87 kcal mole−1; D(PH2–H)≤ 78 kcal mole−1, D(NH2–H)≤ 110 kcal mole−1 and D(N2H3–H)≤ 85 kcal mole−1. Since these F atom reactions tend to populate vibrational levels up to the maximum allowed by the thermochemistry, these limits with the exception of D(H–N2H3), which has a substantial uncertainty, probably are within 5 kcal mole−1 of the actual bond energy. Although rotational relaxation was only partially arrested, a sizeable amount of HF† rotational energy was found from the (CH3)2O, (CH3)2S, PH3, and SiH4 reactions. The (CH3)2O and (CH3)2S reactions partition more rotational energy and less vibrational energy to HF† than the reactions with most other primary C–H bonds.

No abstract

No abstract

Articles you may be interested inCollisional excitation of CN(X2Σ+) by para-and ortho-H2: Fine-structure resolved transitionsThe A 2Π-X 2Σ+ red and B 2Σ+-X 2Σ+ violet systems of the CN radical: Accurate multireference configuration interaction calculations of the radiative transition probabilities J. Chem. Phys. 89, 7334 (1988); 10.1063/1.455264Stateresolved study of collisional energy transfer between A 2Πv=7 and X 2Σ+ v=11 rotational levels of CN Rotational relaxation of CN(B 2~+) formed by interaction of metastable argon atoms with CNBr was studied. The CN(B.2~+) initially is formed only in a few perturbed rotational levels for some vibrational states. Transfer out of the initially formed rotational levels via collision with argon took place with about the same rate for all rotational levels studied, although the magnitude of the rotational level spacings differed for the various cases that were investigated. Transition probability models were developed to simulate the observed pressure dependence of the spectrum. The best fit was obtained with a model having down transitions given by Pi~i= (0.2 1 ;-;1+0.05) (2j+1) /(2i+1). For all models, it was essential to include a large probability of multiquantum (I i-j I >4) transitions in order to fit the data.

In considering complex chemical kinetics either in the laboratory or in the atmosphere one may make numerical computations including a large number of reactions. On the other hand• one'may simplify the problem, set up a shortened set of differential rate equations, and carry out an analytic solution to the problem. These two approaches are supplementary, and both should be carried out. The first method yields quantitative results even for highly complex systems, but it is often difficult to interpret the detailed numerical results. The mathematical equations of the second method may demonstrate the important variables and reveal the essence of the problem, but the simplifications required to obtain analytic solutions may make the approach unsuitable for quantitative analyses. The article by Duewer et al. [1977] is a study of the first kind, and we have no quarrel with the desirability of this numerical study nor with the computed results. However, we believe that the authors have made omissions and mistakes in their interpretation of the chemistry. Specifically, we think that they have failed to consider pertinent laboratory data that greatly reduce the probability of their cases B and C, and we think that they have classified ozone-destroying processes with respect to Ox, NOx, and HO• in a confusing and undesirable way. In this comment we seek to present chemical, algebraic, and numerical considerations that provide additional insight into the meaning of their study. FAILURE TO CONSIDER PERTINENT LABORATORY DATA Garyin and Harnpson [1975] presented tables of chemical kinetic rate constants with references to laboratory studies and to other reviews. For a large number of reactions they gave recommended values k• and an estimate of 'reliability' expressed as +iX log k•. Duewer et al. [1977] defined the three sets of rate constants A, B, and C with subsets for A and C. In essence, sets A and A' are the two sets of rate constants recommended by the Climatic Impact Assessment Program (CIAP). Set B was constructed from ktwhere the plus or minus sign was selected for each reaction so as to maximize the calculated effect of NO• on stratospheric ozone. Sets C were similarly constructed so as to maximize the calculated effect of HO•(H, HO, HOO) and of O•(O, O'D, Os) on stratospheric ozone and so as to minimize the calculated effect of NOx on stratospheric ozone. In one important case involving two of the four rate constants varied to give case their study does not correctly express the status of laboratory studies or the recommendations of Garvin and Hampson [1975].A simplified mechanism, which can be solved analytically, for the purely HOx destruction of ozone by H20 is, in the notation of Table 1 of the article by Duewet et al. [1977], Copyright ¸ 1977 by the American Geophysical Union. k•o, kss, k•, k•:, k•4, k•g, k•8, Js, and k::. To the long-chain approximation the steady state rate expression is -d In [Os] dt where ratios of rate constants are defined as R, = k,d(k•,) '/: (2) R: = k,,k,g/k,:k,, (3) Rs = k,o/kss (4)...