S. Chakraborty, S.K. Bhattacharya / Chemical Physics Letters 369 (2003) 662–667
667
dissociate after its typical lifetime in the following
way,
phase. However, the enrichment pattern observed
18
17
below 15 torr (equal enrichment in O and
O
WÞꢃ
with lower magnitude compared to that observed
at 15 torr) did not fit with the proposed hypothe-
sis. In that pressure range, the ozone production
was less than 0.05 lmol/min [14] and ozone de-
composition by interaction with the surface was
assumed to be significant. The results obtained in
the present experiment provide support for this
assumption.
O
3
þ Surface ðWÞ ! ðO3
3
1
þ
g
ꢃ
!
Oð PÞ þ O ðX R Þ þ W
2
ðR6Þ
The ozone complex being confined to the sur-
face has less degrees of freedom and can only have
restricted rotational and vibrational motions. This
reduces the density of states and makes the acti-
vated complex short-lived. We can then consider a
non-statistical Gao–Marcus model [12], which fa-
vors the asymmetric molecules for its intra-mo-
lecular energy redistribution. This reduces the
dissociation rate constant of the asymmetric mol-
ecules and results in mass independent isotopic
enrichment in the left-over ozone. The product
5. Conclusion
The isotopic behavior of ozone during dissoci-
ation on glass surface is anomalous and displays a
mass independent character. The product oxygen
is isotopically lighter similar to the case of visible
light photo-dissociation or thermal (<90 °C) de-
17
oxygen is correspondingly equally depleted in
O
18
17
18
and O. This explanation is at present speculative
and needs to be tested experimentally and theo-
retically.
To achieve larger fractionation, the ozone had
to go through the spiral many times. For example,
to get around 25& enriched ozone, the required
time was nearly 6 h and some amount of thermal
dissociation in this time cannot be avoided.
Therefore, the higher amount of fractionation
composition but the depletion in O and O are
17
18
equal resulting in a slope of unity in d O–d O
plot. As the experimental time increases, the effect
disappears due to the dominating effect of room
temperature thermal dissociation of ozone.
References
(
>25&) arises due to a combination of two pro-
[1] K. Mauersberger, P. L €a mmerzahl, D. Krankowsky, Geo-
phys. Res. Lett. 28 (2001) 3155.
cesses, i.e., surface interaction and gas phase dis-
sociation. The room temperature thermal
dissociation yields a mass dependent slope (ꢀ0.5)
in contrast to the surface dissociation (slope ꢀ1.0)
and the admixture of these two components pro-
duces a resultant slope with value less than one.
This explanation is supported by the thermal dis-
sociation data (circles in Fig. 1, representing the
situation when the ozone was kept within the spiral
for about 8 h at room temperature). They define a
slope of around a half, which is expected in case of
thermal decomposition at room temperature [8].
Recently Bhattacharya et al. [14] studied the
pressure dependency of isotopic enrichment in
ozone (produced from oxygen photolysis at dif-
ferent oxygen pressures ranging from 8 to 700
torr). They explained the observed enrichment
pattern for pressure upto 15 torr by considering
simultaneous formation and dissociation in gas
[
[
2] M.H. Thiemens, T. Jackson, Geophys. Res. Lett. 14 (1987)
24.
6
3] J. Morton, J. Barnes, B. Schueler, K. Mauersberger, J.
Geophys. Res. 95 (1990) 901.
[4] S.K. Bhattacharya, M.H. Thiemens, Geophys. Res. Lett.
5 (1988) 9.
1
[
[
5] J. Wen, M.H. Thiemens, J. Geophys. Res. 96 (1991) 10911.
6] A.R. Ravishankara, G. Hancock, M. Kawasaki, Y. Mats-
umi, Science 280 (1998) 60.
[7] S. Chakraborty, S.K. Bhattacharya, J. Chem. Phys. 118
(2003) in press.
[
[
8] J. Wen, M.H. Thiemens, Chem. Phys. Lett. 172 (1990) 416.
9] S. Kim, J. Yang, J. Astro. Space Sci. 14 (1997) 297.
[
10] S.J. Gregg, Surface Chemistry of Solids, Chapman and
Hall, London, 1961.
[11] E.R.S. Winter, Adv. Catal. 10 (1958) 196.
[12] Y.Q. Gao, R.A. Marcus, Science 293 (2001) 259.
[
13] K. Cristmann, Introduction to Surface Physical Chemistry,
Steinkopff Verlag Darmstadt Springer-Verlag, New York,
1
991.
[
14] S.K. Bhattacharya, Subrata Chakraborty, J. Savarino,
M.H. Thiemens, J. Geophys. Res. 107 (2002) 4675.