2711
Communicationsto the Editor
which gives the isotope separation factor (a) as a function of time the form:
One important method for decomposition of ClON02 in the stratosphere is through its photolysis by ultraviolet radiation, presumably by one of the following mechanisms:6 ClONO2
Equation 4 applies to biphotonic processes too, except that K(1) are given by the equations derived by Leung.la2 Equations 1,2, and 4 show that at high light intensities, a(t) would differ from unity only if the spin levels being saturated differ in their chemical reactivity (i.e., different y). At low and intermediate intensities, large population differences between the spin levels together with large differences in the decay rates (pL# p j ) would lead to isotope enrichment even if the photochemical reactivity of the different spin levels is the same. Our calculations further show that, for the same system, monophotonic processes give more efficient separation than biphotonic ones. The above equations also indicate that the effect of microwaves is more pronounced the less the number of nuclear spin levels per electron spin level or if nuclear spin diffusion is fast. Statistical analysis shows that the effect of microwaves is reduced for molecules that contain more than one isotopic atom per molecule (e.g., di or tri instead of mono derivatives).
Kl2(1) and
Acknowledgment. The authors wish to thank Dr. Peter Esherick for valuable discussion of this problem. The financial support of the United States Energy Research and Development Administration is gratefully acknowledged. References and Notes (1) M. Leung and M. A. El-Sayed. J. Am. Chem. SOC.,97,669 (1975). ( 2 ) M. Leung, Thesis, Chemistry Department, UCLA, 1974. (3) Contribution No. 3704.
Department of Chemistry3 University of California 10s Angeles, California 90024
Tala1 Akasheh M. A. El-Sayed'
Stratospheric Formation and Photolysis of Chlorine Nitrate' Publication costs assisted by the U.S. Energy Research and Development Administration
Sir: Our current experiments and calculations indicate the possible existence in the mid-stratosphere of an additional chlorine-containing species, ClON02,3 which is usually identified by the name chlorine nitrate. Seven of the most important chemical reactions of chlorine in the stratosphere are summarized by the equilibria C
OH
03
l
m C1 e C 1 0
CHd,Hz,H02
0,NO
(1)
involving C1, C10, and HC1, and have been described in considerable detail earlier.4 The most probable method for formation of ClONOz involves the reaction of C10 with NO2, with stabilization by a third body, as in5 C10
+ NO2 + M
+
ClONO2
+M
(2)
-+
C10 + NO2
(34
C1+ NO3
(3b)
Chlorine nitrate does not react rapidly with most stable atmospheric species (N2,02, COz, S02, CH4, H2, NOz), but does react with NO, with oxygen atom^^,^ as in7
0
+ ClON02
+
C10
+ ON02
(4)
and perhaps by other radical reactions as well. The stratospheric chlorine equilibria of (1)can thus be extended to include ClON02, as in C10
3ClONOz
hu, 0
(5)
If the alternative pathway of (3b) is important then the pathways to equilibrium are diagrammatically slightly more complex, without any appreciable quantitative effect on the equilibria themselves. We report here the photochemical absorption cross sections for reaction 3. Our calculations of the stratospheric chlorine equilibria indicate that, if photochemical decomposition by reaction 3 is the only important removal mechanism for it, then ClONO2 should probably be detectable at altitudes from 20 to 30 km, although less abundant than HCl at all altitude^.^ Furthermore, the dependence of the equilibria of (1)and (5) on the intensity of solar ultraviolet radiation leads to the possibility of appreciable diurnal variations in CION02 concentration, with larger values in morning than evening.
+ +
c1+ 0 3
Received August 24, 1976
H
+ hv
c10 C10
0
c10 + 02
(6)
c1+ 0 2
(7)
NO 4 C1+ NO2
(8)
The concentrations of ClONO2 should diminish rapidly with increasing altitude above 30 km with the lowered pressure for its stabilization by M in (2), and with the rapidly increasing photodissociation rates for (3). Most of the C10,catalyzed removal of ozone by reactions 6 and 7 occurs above 30 km, with its peak near 40 km? and the presence of ClONO2 in the 20-30-km region would not have a major effect on estimates of the extent of reaction 7 above 35 km. The disappearance at night of 0 and NO with which C10 usually reacts in the daytime (reactions 7 and 8) may permit the conversion of most C10 to ClONO2 overnight, only to be photolyzed again in the morning. The detailed interaction between the C10 and NO, cycles is described in a separate paper.9 Chlorine nitrate is a moderately well-known laboratory chemical,10-21and both its ~ l t r a v i o l e t land ~ infrared spectra15JsJ9 have been reported earlier. It is a gaseous compound with a vapor pressure of 13 Torr at 220 K, and about 850 Torr near room temperature;l2Sz0when liquefied it is pale yellow. In the laboratory ClON02 has regularly been formed by reaction 2, and recent observations indicate that the rate of reaction of C10 with NO2 has a third-body rate coefficient (with Nz) of 1.5 X cm6 molecule-2 s-l at room temperature.22 The reported stability and reactivity of chlorine nitrate has varied with the conditions of the experiments, and caution is advised in handling. However, purified ClONO2 in our experiments is quite stable in the gas phase at room temperature (no observable change in infrared cells for 12 h), even in the presence of Nz, 0 2 , CH4, Ha, or SOz. No reaction was observed The Journal of Physical Chemistry, Vol. 80, No. 24, 1976
2712
Communications to the Editor
TABLE 11: Altitude and Zenith Angle Variation of Direct Solar Photodissociation Coefficients, J , for Chlorine Nitrate. CIONO?n
10-17
5 ABSORPTION CROSS-SECTIONS FOR CION02
lo-'*
a;
Photodissociation coefficient, s-l in units of
5
cm
Altitude, km 10 15 20 25 30 35 40 45 50 WAVELENGTH,
i
Figure 1. Ultraviolet spectrum of gaseous CION02 between 1860 and 4600 A: (0)this work; ( A ) data of ref 10, as read from published graph.
TABLE I: Photochemical Absorption Cross Sections, cr, for Chlorine Nitrate, ClONOz A, 8,
1860 1900 1950 2000 2050 2100 2150 2200 2250 2300 2350 2400 2450 2500 2550 2600 2650 2700 2750 2800 2850 2900 2950
u, 10-20 cm2
995 690 502 372 344 348 375 376 307 231 159 118 85.4 65.7 50.9 40.7 32.8 26.1 20.2 14.5 10.5 7.34 5.12
A, 8,
3000 3050 3100 3150 3200 3250 3300 3350 3400 3450 3500 3600 3700 3800 3900 4000 4100 4200 4300 4400 4500 4600
u, 10-20
cm2
3.91 2.79 2.03 1.45 1.07 0.79 0.61 0.48 0.38 0.34 0.29 0.23 0.19 0.15 0.11
0.085 0.059 0.042 0.028 0.019 0.013 0.008
with NO2 in 2 h, while NO reacted with an apparent bimolecular rate constant of about cm3 molecule-l s-l. For each of these molecules, the bimolecular reaction rates are sufficiently slow that at atmospheric concentrations none is important in the removal of C ~ O N O Z . ~ ~ We have measured the ultraviolet spectrum of ClONOz with carefully purified samples, as shown in Figure 1and Table I. Repeated purifications with several preparations (by two different routes) have confirmed that the weak absorption in the 3500-4600-A region is attributable to ClON02, and not to impurities. Our measurements are in excellent agreement with the earlier measurementslZ in the region of strong absorption, as shown in Figure 1.We have combined these absorption cross sections with solar intensities4 to calculate the The Journal of Physical Chemistry, Vol. SO?No. 24, 1976
Oo
7.6 7.6 8.0 9.2
Solar zenith angle: 20' 40° 60' 7.5 7.5
7.8 8.9
7.1 7.1 7.4
12
12
8.3 11
20 39 70 95
19 37 68 94
17 33 62 90
6.5 6.5 6.7 7.3 8.6 12
24 49 81
80' 5.2 5.2 5.4 5.7 6.4 7.6 11
24 50
Additional photodecomposition (-6 X s-l at all altitudes) arises from multiply scattered radiation between 3000 and 4600 A. direct photodissociation coefficients, J , for ClONOz shown in Table 11. Molecules with appreciable absorption a t wavelengths longer than 3000 A also have important (and variable) dissociation from multiply scattered radiation as well. Estimates of the intensity of such scattered radiation indicate approximately a factor of 2 increase in J value for that part of the dissociation of ClONOz lying between 3000 and 4600 A.24,25This is equivalent to the addition on the average of about 6 X 10-5 s-l to the J value a t all altitudes. The precise increment for scattered radiation varies with zenith angle and assumed albedo of the lower layers of the atmosphere. The rate of photodissociation of ClONO2 is much faster than that of the known stratospheric species, HON02, which has very small absorption cross sections beyond 3000 A. The infrared spectrum of C10N0219shows several intense absorption peaks (e.g., 780.2 cm-l) suitable for identification if its stratospheric concentrations are high enough. References a n d Notes (1) This research has been supported by ERDA Contract No. AT(04-3)-34, P.A. 126, and by NASA Contract No. NSG-7208. (2) Alfred P. Sloan Foundation Fellow. (3) Abstracts of the 12th International Symposium on Free Radicals, Laguna Beach, Calif., Jan 1976. (4) F. S. Rowland and M. J. Molina, Rev. Geophys. Space Phys;, 13, 1 (1975). (5) The alternative bimolecular reactions leading to ClOO NO, OClO NO, or CI NO3 are endothermic by 10.5, 13.7, and 13.7 kcal/mol, respectively. (6) The quantum yield for photodecomposition of CION02 has not yet been measured, but is assumed to be 1.0 because of the uv spectral characteristics, and the large excesses of available energy over that needed for decomposition. (7) L. T. Molina, J. E. Spencer, and M. J. Molina, Chem. Phys. Lett., in press. The observed secondary reactions in these experiments strongly indicate that the reaction products from (4) are CIO and NO3 as shown. (8) A. R. Ravishankara, G. Smith, G. Tesi, and D. D. Davis, 12th informal Conference on Photochemistry, Gaithersburg, Md., June 1976. (9) F. S. Rowland, J. E. Spencer, and M. J. Molina, J. Phys. Chem., following paper in this issue. (10) H. Martin and T. Jacobsen, Angew. Chem., 67, 524 (1955). (1 1) H. Martin and R. Gareis, Z. Elektrochem., 60, 959 (1956). (12) H. Martin, Angew. Chem., 70, 97 (1958). (13) H. Martin, W.Meise, and E. Engelmann, Z.Phys. Chem. (FrankfurtamMain), 24, 285 (1960). (14) M. Schmeisser, W. Fink, and K. Brandle, Angew. Chem., 69, 780 (1957). (15) K. Brandle, M. Schmeisser, and W. Luttke, Chem. Ber., 93, 2300 (1960). (16) M. Schmeisser and K. Brandle, Angew. Chem., 73, 388 (1961). (17) L. F. R. Cafferata, J. E. Sicre, and H. J. Schumacher. Z. Phys. Chem. (Frankfurt am Main), 29, 188 (1961).
+
+
+
2713
Communications to the Editor (18) A. J. Arvia, L. F.R. Cafferata, and H. J. Schumacher, Chem. Ber., 96, 1187 (1963). (19) R . H. Miller, D. L. Bernitt, and I. C. Hisatsune. Spectrochim. Acta, PartA, 23, 223 (1967). (20) C. J. Schack, Inorg. Chem., 6, 1938 (1967). (21) M. Schmeisser, Inorg. Syn., 9, 127 (1967). (22) Unpublished experiments by W. B. DeMore et al., F. Kaufman et al., and J. Birks et al. The temperature dependence is approximately r3, leading to a rate constant for formation about a factor of 2 to 3 higher at stratospheric temperatures. (23) Reference 16 also reports a rapid reaction of CION02 with SOn and NO2, but we have not observed such reactions at room temperature. (24) D. Wuebbies and F. Luther, unpublished calculations. (25) P. Crutzen, Joint lAOC/ICACGP Symposium on Atmospheric Ozone, Dresden, GDR, Aug 1976.
Department of Chemistry University of California Irvine, California 927 17 Received June 3, 1976
F. S. Rowland” John E. Spencer Mario J. Moiina2
Estimated Relative Abundance of Chlorine Nitrate among Stratospheric Chlorine Compounds Publication costs assisted by the Administration
U.S.Energy Research and Development
Sir: The stratospheric cross combination of the radicals C10 and NO2 can lead to the formation of chlorine nitrate, ClONOZ, as in1-3 C10
+ NO2 + M
ClONOz + M
-+
(1)
The chief process for removal of chlorine nitrate from the stratosphere is solar photolysis, as in ClONOz
+ hv
-
C10 + NO2
(2a)
C1+ NO3
(2b)
+
which effectively reverses (1).Although photodissociation is theoretically possible with radiation in the infrared to about 11000 A, chlorine nitrate is transparent at wavelengths longer than about 4600 A.3 At altitudes above about 35 km the most important wavelengths for photolysis of chlorine nitrate are in the 2000-2200-A range for which it has absorption cross sections of 300-400 X cm2. Radiation of such wavelengths, however, is strongly absorbed by both 0 2 and 0 3 (although not as strongly as radiation with X 2200 A by 0 3 ) . Below 30 km ClONOz photolysis occurs primarily with wavelengths longer than 3000 A. A process of secondary importance for stratospheric removal of ClONOz is its attack by O(3P) atoms, as
0 + ClON02
-
C10
+ NO3
(3)
While very complete models of the stratosphere with tens of interacting species, a hundred or more chemical reactions, and diffusion of species in one or more dimensions are necessary for detailed understanding of stratospheric chemistry, the very complexity of the models forces approximations of varying crudeness to avoid computer calculations of prohibitive length. By contrast, reasonable estimates of the expected stratospheric ratio of (ClONOZ)/(ClO) vs. altitude can be obtained simply from the measured concentrations of N02,6 known rate constant for (1),7 and calculated photodissociation coefficient, J , for reaction 2: based on the measured laboratory absorption cross sections vs. wavelength. The (ClON02) /(ClO) ratios can then be coupled with (ClO)/(HCl)ratios from
more complex calculations for estimates of the stratospheric distribution of C1 among the chlorine species, C10, HC1, ClON02, and C1 itself. The calculation is simple if the mechanism for photolysis is the reverse of the formation reaction, Le., (2a). If part or all of the photolysis goes by (2b) the C1 atoms thus released react immediately (
