on the surface. The observed sticking probability is therefore
the same as the intrinsic sticking probability of reaction (1),
S . As multilayer adsorption can occur at this temperature,
1
the sticking probability remains at 1 indeÐnitely.
Summary
Chloroform adsorbs molecularly on Cu(111) at 100 K with a
sticking probability of 1. On heating the monolayer surface
formed at 100 K to 170 K, the chloroform reacts to form che-
misorbed chlorine and adsorbed ethyne. On further heating,
the ethyne product desorbs at just above room temperature
with Ðrst order kinetics, an activation energy of 77 ^ 6 kJ
mol~1 and a pre-exponential factor of 1011B1 s~1, leaving a
Fig. 9 Schematic showing how the potential energy curve of physi-
sorbed chloroform at zero coverage crosses the curve for adsorption
of CHCl and Cl to give activated adsorption.
2
(J3 ] J3)R30¡-Cl surface. Chloroform adsorption on clean
Using a value of E [ E \ 3.5 ( ^ 0.7) kJ mol~1 we obtain
3
2
Cu(111) at room temperature and above is activated, with an
activation energy of 3.5 ^ 0.7 kJ mol~1 at zero coverage, with
a sticking probability of 0.23 ^ 0.04 at 320 K. For chloroform
adsorption below room temperature the initial sticking prob-
ability is higher than expected and this is thought to be due to
clustering of the adsorbed chloroform under the action of
intermolecular attractive interactions, which causes the
adsorption process to become less activated and hence
increases the sticking probability. For situations where the
ethyne product remains on the surface during chloroform
adsorption, the sticking probability increases to greater than
theory values for
0
S
of 0.13 ^ 0.04, 0.21 ^ 0.04 and
0
.29 ^ 0.04 for 216, 320 and 480 K respectively. The last two
agree reasonably well with the measured values of 0.23 ^ 0.04
and 0.26 ^ 0.04, but the measured value of 0.35 for 216 K is
signiÐcantly higher than the theory value of 0.13. We suggest
that at the higher temperatures of 320È480 K physisorbed
chloroform is dispersed across the surface as individual mol-
ecules, and thus E is the activation energy for desorption of a
2
single isolated molecule on the Cu(111) surface. However, at
216 K, at the lowest initial coverages obtainable in our experi-
ment, the chloroform has already formed clusters on the
surface in which there are attractive interactions between the
molecules. Such attractive interactions between adsorbed
species will lower one or possibly both of the curves in Fig. 9,
causing the crossing point to move towards zero energy, and
hence increasing the sticking probability. If we use the experi-
mental value of S at 216 K in eqn. (10) we obtain E [ E \
0.5. This is thought to be due to attractive interactions
between the adsorbed ethyne and the physisorbed chloroform,
the e†ect of which is to convert the activated adsorption
process at zero coverage into a non-activated process at Ðnite
coverage.
References
0
3
2
1
3
.1 kJ mol~1, which is lower than the zero coverage value for
20È480 K. This supposition is consistent with the annealing
1
2
3
R. G. Jones and C. J. Fisher, Surf. Sci., 1999, 424, 127.
D. A. King and M. G. Wells, Surf. Sci., 1972, 29, 454.
J.-L. Lin and B. E. Bent, J. Phys. Chem., 1993, 97, 9713, and
references therein.
experiments carried out on a monolayer of chloroform
adsorbed molecularly at 100 K. It was found that the mono-
layer did not desorb molecularly, but cracked on the surface.
Such behaviour is indicative of a non-activated adsorption
process, and shows that for the monolayer covered surface,
the crossing point of the two curves in Fig. 9 has moved below
zero, making the process non-activated adsorption with disso-
ciation rather than molecular desorption the dominant
process.
If we now consider the ethyne desorption proÐle versus
coverage for adsorption at 480 K, Fig. 3D, we Ðnd that it
matches the sticking probability in shape. This means that, for
each chloroform molecule adsorbed, an ethyne molecule is
promptly desorbed. It follows that the sticking probability at
4
M. Xi and B. E. Bent, L angmuir, 1994, 10, 505, and references
therein.
5
6
S. Y. Chan, S. Turton and R. G. Jones, Surf. Sci., 1999, 234, 433.
M. F. Kadodwala, A. A. Davis, G. Scragg, B. C. C. Cowie, M.
Kerkar, D. P. Woodru† and R. G. Jones, Surf. Sci., 1997, 392,
1
99.
7
A. Bah, T. Ceva, B. Croset, N. Dupont-Pavlovsky and E.
Ressouche, Surf. Sci., 1998, 395, 307.
8
9
J. M. Chen, S. C. Yang and Y. C. Liu, Surf. Sci., 1997, 391, 278.
C. H. Chu and M. H. Hon, Diamond and Relat. Mater., 1993, 2,
3
11.
1
1
1
0
1
2
J. E. Scha† and J. T. Roberts, J. Phys. Chem., 1996, 100, 14151.
J. E. Scha† and J. T. Roberts, Surf. Sci., 1999, 426, 384.
K. D. Schiebaum, A. Gerlach, W. Gopel, W. M. Muller, F.
Vogtle, A. Dominik and H. J. Roth, Fresenius J. Anal. Chem.,
1994, 349, 372.
480 K is that for chloroform on a surface consisting of an
increasing coverage of chemisorbed chlorine. For adsorption
at 320 K, Fig. 3C, desorption of ethyne is delayed until almost
half coverage. It is noticeable that this adsorbed ethyne has
the e†ect of increasing the sticking probability during the Ðrst
half of the adsorption. The e†ect is even more pronounced at
13
14
15
St. J. Dixon-Warren, E. T. Jensen and J. C. Polanyi, J. Chem.
Phys., 1993, 98, 5938.
St. J. Dixon-Warren, D. V. Heyd, E. T. Jensen and J. C. Polanyi,
J. Chem. Phys., 1993, 98, 5954.
M. Kadodwala and R. G. Jones, J. V ac. Sci. T echnol. A, 1993, 11,
216 K, Fig. 3B, where all the ethyne product is retained by the
2019.
surface, and the sticking probability soars to 0.7. These results
may again be rationalised using attractive interactions
between molecules. If physisorbed chloroform, or possibly the
16 S. Turton, M. Kadodwala and R. G. Jones, Surf. Sci., accepted.
1
1
7
8
N. K. Singh and R. G. Jones, Surf. Sci., 1990, 232, 229.
R. G. Jones, S. Turton and R. Ithnin, Chem. Phys. L ett., 1996,
261, 539.
initial dissociation products, CHCl (ads) and Cl(chem), or
1
2
9
0
D. Westphal and A. Goldmann, Surf. Sci. 1983, 131, 113.
W. K. Walter, D. E. Manolopoulos and R. G. Jones, Surf. Sci.,
1996, 348, 115.
2
both, are stabilised by the presence of ethyne on the surface, it
will again cause, the crossing point of the two curves in Fig. 9
to move towards zero, and eventually go below it, making the
adsorption process non-activated and increasing the sticking
probability above 0.5. Clearly the true interaction potential
will be multidimensional, but for the results presented here a
simple one dimensional potential is sufficient.
For adsorption at 100 K the measured sticking probability
was 1.0. At this temperature reaction (1) is the only observable
reaction, reactions (2)È(5) being too slow to be observable,
and hence we are simply trapping intact chloroform molecules
21 G. Comsa and R. David, Surf. Sci. Rep., 1985, 5, 145.
2
2
2
2
3
4
X.-Y. Zhu, Annu. Rev. Phys. Chem., 1994, 45, 113.
J. E. Bartmess and R. M. Georgiadis, V acuum, 1983, 33, 149.
Handbook of Chemistry and Physics, CRC Press, Boca Raton, FL,
5
8th edn., 1978.
2
5
K. Kimura, S. Katsumata, Y. Yamazaki and S. Iwata, Handbook
of He I Photoelectron Spectra of Fundamental Organic Molecules,
Japan ScientiÐc Soc. Press, Tokyo, 1981.
Paper Paper 9/05989J
5228
Phys. Chem. Chem. Phys., 1999, 1, 5223È5228