The physicochemical and catalytic properties of the Pt-CeO x system

Article abstract of DOI:10.1134/S003602440710010X

The properties of the Pt-CeO _x system prepared by the oxidation of the Pt₂Ce intermetallic compound were studied. The sample was characterized by X-ray diffraction in situ, thermogravimetry, scanning electron microscopy (with an accessory for energy dispersion analysis), transmission electron microscopy, and temperature-programmed reduction with hydrogen. The catalytic properties of the sample were studied in the model reaction of toluene hydrogenation. The oxidation of the intermetallic compound caused the appearance of metallic platinum and cerium oxide phases and high-dispersity platinum particles encapsulated in cerium oxide. Metallic platinum on the surface of the catalyst experienced rapid deactivation in the presence of hydrogen sulfide; high-dispersity platinum particles encapsulated in cerium oxide exhibited enhanced stability toward sulfur compounds.

Full text of DOI:10.1134/S003602440710010X

ISSN 0036-0244, Russian Journal of Physical Chemistry A, 2007, Vol. 81, No. 10, pp. 1601–1607. © Pleiades Publishing, Ltd., 2007.
Original Russian Text © P.A. Zosimova, A.V. Smirnov, V.V. Yushchenko, S.N. Nesterenko, I.I. Ivanova, 2007, published in Zhurnal Fizicheskoi Khimii, 2007, Vol. 81, No. 10,
pp. 1797–1803.
CHEMICAL KINETICS
AND CATALYSIS
The Physicochemical and Catalytic Properties
of the Pt–CeOx System
P. A. Zosimova, A. V. Smirnov, V. V. Yushchenko, S. N. Nesterenko, and I. I. Ivanova
Faculty of Chemistry, Moscow State University, Leninskie gory, Moscow, 119992 Russia
e-mail: PolinaZosimova@phys.chem.msu.ru
Received December 25, 2006
Abstract—The properties of the Pt–CeOx system prepared by the oxidation of the Pt2Ce intermetallic com-
pound were studied. The sample was characterized by X-ray diffraction in situ, thermogravimetry, scanning
electron microscopy (with an accessory for energy dispersion analysis), transmission electron microscopy, and
temperature-programmed reduction with hydrogen. The catalytic properties of the sample were studied in the
model reaction of toluene hydrogenation. The oxidation of the intermetallic compound caused the appearance
of metallic platinum and cerium oxide phases and high-dispersity platinum particles encapsulated in cerium
oxide. Metallic platinum on the surface of the catalyst experienced rapid deactivation in the presence of hydro-
gen sulfide; high-dispersity platinum particles encapsulated in cerium oxide exhibited enhanced stability
toward sulfur compounds.
DOI: 10.1134/S003602440710010X
INTRODUCTION
stability in the oxidation of CO when platinum particles
are fully coated with cerium oxide. The oxide film
thickness then reaches 10 layers [5]. The formation of
such a film increases the region of contact Pt–Ce inter-
actions, which increases catalytic activity. In addition,
ëÂéx films protect the surface of platinum from deac-
tivation by carbonaceous deposits and poisoning by
sulfur compounds. There is also the possibility of
hydrogen spillover from platinum to cerium oxide,
where hydrogen can participate in hydrogenation reac-
tions. For instance, the promoting action of cerium
oxide on platinum is observed in the hydrogenation of
diene and aromatic hydrocarbons [7].
The deactivation of noble metal-based catalysts,
including platinum catalysts, by sulfur compounds is a
serious problem in many industrial processes [1]. There
are several methods for increasing the resistance of
platinum-containing catalysts to sulfur compounds.
One of these is the introduction of modifiers that easily
form stable compounds with sulfur or change the elec-
tronic properties of platinum and thereby decrease the
strength of Pt–S bonds formed in the adsorption of sul-
fur-containing substances [2]. Similar problems arise
with catalytic systems based on platinum and rare-earth
metal oxides, which are extensively used in the cata-
lytic cracking of heavy oil fractions [3] and burning up
of exhaust gases [4]. Cerium oxide is used more often
than other rare-earth metal oxides. This oxide is an
effective modifier thanks to a special combination of its
acid-base and redox properties.
Pt–ëÂé2 binary systems show high activity in the
oxidation of CO and reduction of NO [5, 6]. Suppos-
edly, there are two reasons for this. First, ëÂé2 can be
an additional source of oxygen in such systems. Next,
cerium oxide modifies platinum particles and increases
their stability [1, 7]. Metal clusters are firmly retained
on oxygen vacancies present on the surface of ëÂé2,
which prevents their aggregation and increases the sta-
bility of systems toward sintering in reducing media at
increased temperatures [8].
Multicomponent platinum catalysts are most often
deposited systems containing silica or alumina as carri-
ers. Note that none of the known methods for the prep-
aration of deposited multicomponent catalysts ensures
a homogeneous distribution of phases over the surface
of carriers and direct phase-carrier contacts. In addi-
tion, the ratio between the components is difficult to
vary over a wide range. Lastly, the interaction of active
components with carriers can substantially change the
catalytic properties of the active phases themselves,
which prevents us from studying the mutual influence
of the components.
In this work, we studied the Pt–ëÂé2 system syn-
thesized by the alloying of platinum and cerium metals
followed by alloy oxidation. This method allows sam-
The interaction of platinum with cerium oxide can ples with an arbitrary component ratio and a homoge-
also result in the formation of intermetallic phases on neous distribution of the components to be prepared. In
the surface of ëÂé2 [8, 9], and sequential redox treat- addition, according to the literature data [10], this
ments, in the encapsulation of Pt particles in cerium method for the preparation of Pt–Ce systems results in
oxide. Pt–ëÂé2 systems have the highest activity and the encapsulation of platinum particles in cerium oxide.
1601

1602
ZOSIMOVA et al.
EXPERIMENTAL
of the reactor was analyzed by gas-liquid chromatogra-
phy on an SE-30 capillary column of length 50 m. The
carrier gas used was helium. Analyses were performed
on a Kristall 2000M chromatograph with a flame-ion-
ization detector. The main reaction product was meth-
ylcyclohexane, selectivity with respect to it was no less
than 99.9%.
The Pt–Ce alloy with a Pt : Ce molar ratio of 2 : 1
corresponding to the Pt2ë intermetallic compound
was prepared in an arc furnace by alloying the pure
components on a copper electrode cooled with water in
an argon atmosphere at 50 kPa. The alloy was ground
in an agate mortar, and the fraction of 0.05–0.14 mm
particles was separated by screening.
The X-ray patterns of the initial and oxidized sam-
ples were taken on a Theta Bruker D-500 diffractome-
ter with CuKα radiation over the 2θ range 20°–80°. The
diffractometer was equipped with a Kevez Si(Li) solid-
state detector. Oxidation was studied by heating the
sample in a flow of synthetic air (20% O2 + 80% N2)
over the temperature range 20–600°C. The diffracto-
grams were taken in situ in steps of 100 K.
The images of the surface of the samples were
obtained by scanning electron microscopy on a 5900
LV JEOL instrument with an accessory for energy dis-
persion analysis. The detailed structure of the samples
was studied by transmission electron microscopy on a
JEOL 3000F instrument with a 300 kV source. For
obtaining images, the samples were placed into epoxide
resin to prepare their thin sections.
Thermogravimetric analysis was performed on a TA
SDT Q600 instrument. A sample (10–20 mg) was oxi-
dized in a flow of dry air (100 ml/min) while the tem-
perature increased to 700°C at a 10 K/min rate. The
accuracy of weighing was 10–7 g.
The reduction of oxidized samples was studied by
temperature-programmed reduction with H2. An inter-
metallic compound sample (1.5–15 mg) was thor-
oughly mixed with porous quartz powder, placed into a
quartz reactor, and oxidized in a flow of dry air
(20 ml/min) at 550°ë for 10 h. The sample was cooled
to room temperature and purged with Ar. After this, a
flow of Ar (20 ml/min) with 3.5 vol % H2 was passed
through the reactor. The temperature-programmed
reduction spectra were obtained over the temperature
range 25–1000°ë at a heating rate of 8 K/min. Changes
in the composition of the gas at the exit of the reactor
were recorded using a katharometer.
The catalytic activity of the samples was determined
in the model reaction of toluene hydrogenation on a
flow unit at atmospheric pressure and 150°ë. The reac-
tor was equipped with two thermocouples, one con-
trolled the temperature in the catalyst bed and the other,
furnace heating when the temperature deviated from
the required value. An intermetallic compound sample
(30–300 mg) oxidized at 550°ë in a flow of dry air was
mixed with porous quartz to decrease the hydrody-
namic resistance of the catalyst bed and temperature
gradients, because the process was exothermic.
The resistance of catalysts toward sulfur poisoning
was studied by periodically introducing pulses of a
mixture of N2 and H2S into the flow of hydrogen and
toluene supplied to the reactor (0.47 µg H2S per pulse).
RESULTS AND DISCUSSION
The Oxidation of the Pt2Ce Intermetallic Compound
The diffractogram obtained at 25°C for the initial
Pt–Ce alloy is shown in Fig. 1. The diffraction maxima
observed correspond to the cubic phase of the Pt2Ce
intermetallic compound with the lattice parameter a =
7.730 Å [11]. According to the thermogravimetric anal-
ysis data, the oxidation of Pt2Ce is accompanied by an
increase in sample weight, which begins at 300°C and
continues up to 550°C. The total increase in sample
weight after oxidation up to 700°C (6%) corresponds to
the complete oxidation of cerium to cerium oxide and
the formation of the Pt–CeO2 system.
The presence of metallic platinum and cerium oxide
in the oxidized Pt2Ce intermetallic compound was sub-
stantiated by the X-ray data. The diffractograms
obtained as the temperature of oxidation increased
sequentially are shown in Fig. 1. According to these
results, the formation of the CeO2 phase begins at as
low as 200°C, and metallic platinum reflections appear
at 300°C. The complete destruction of the initial Pt2Ce
phase is observed at 400°C. Above this temperature, the
phase composition of the sample does not change
noticeably. These results led us to select oxidation at
550°C as the standard procedure for sample activation.
At this temperature, the weight of samples ceased to
change, and the formation of the metal and oxide
phases was completed.
The morphology of both the initial intermetallic
compound sample and sample oxidized at 550°C was
examined using the scanning electron microscope. No
destruction of granules occurs during oxidation, but the
formation of tracks 50–100 nm wide is observed on the
surface of granules. According to the energy dispersion
analyzer data, these tracks and regions between them
largely contained metallic platinum and cerium oxide,
respectively. A more detailed examination of the oxi-
dized sample on the transmission electron microscope
showed that, along with separate metallic platinum and
cerium oxide regions, there were regions containing
The catalyst was then reduced in a flow of hydrogen both Pt and CeO2 (Fig. 2). According to the diffraction
at 450°ë for 1.5 h. The weight hourly space velocity of data, platinum and cerium atoms can isomorphously
toluene was varied from 0.70 to 3.75 h–1, and the flow substitute each other. On the other hand, high-disper-
of hydrogen through a bubbling reactor with toluene at sity Pt particles built into the cerium oxide CeO2 lattice
0°ë was changed. The composition of the gas at the exit are in all probability formed in these regions. Since Pt
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 81 No. 10 2007

THE PHYSICOCHEMICAL AND CATALYTIC PROPERTIES
1603
CeO2
Pt
Pt2Ce
*
*
600°C
500°C
400°C
300°C
200°C
100°C
25°C
*
*
*
*
*
20
30
40
50
60
70
80
2θ, deg
Fig. 1. Diffractograms obtained in situ during step heating of a Pt Ce sample in a flow of air.
2
and CeO2 form cubic face-centered lattices of the same with the data reported in [12], according to which the
type, this substitution should not cause noticeable dis- degree of reduction of pure ëÂé2 during heating to
tortions of the crystal lattice of CeO2.
1000°ë does not exceed 12% (ëÂé1.77).
To summarize, oxidation at 550°C results in the
destruction of the intermetallic phase and the formation
of the Pt and CeO2 phases and high-dispersity platinum
particles encapsulated in cerium oxide.
The consumption of hydrogen by an oxidized Pt2Ce
sample begins already at room temperature when the
sample is brought in contact with an H2 + Ar mixture.
The temperature-programmed reduction curve has two
maxima at 40 and 100°ë in the low-temperature region
(Fig. 3). The total area of the high-temperature peaks of
Pt2Ce decreases, and this decrease is not proportional to
the decrease of the fraction of cerium oxide in the Pt2Ce
sample studied compared with pure ëÂé2.
The Reduction of the Pt–CeO2 System
Immediately before catalytic experiments, oxidized
intermetallic compound samples were reduced in a flow
of hydrogen at 450°ë. Preliminarily, we studied the
influence of such treatment on oxidized samples by the
method of temperature-programmed reduction with H2
(Fig. 3). For comparison, the temperature-programmed
reduction profile obtained for pure cerium oxide is also
shown in Fig. 3. The reduction of pure ëÂé2 is accom-
panied by the appearance of two peaks at 485 and
790°ë. According to the literature data, the first peak
corresponds to the reduction of ëÂé2 on the surface,
and the second peak, to the partial reduction of the bulk
Such a shape of reduction curves for cerium oxide in
the presence of platinum was described in [12, 14] for
deposited Pt/CeO2 systems. The authors explain the
appearance of the low-temperature peak at 120–250°ë
and a decrease in the intensity of the peak at 450°ë by
the reduction of CeO2 on the surface at a lower temper-
ature because of the spillover of hydrogen activated on
platinum. In addition, a contribution to the low-temper-
ature hydrogen absorption peak can be made by the
reduction of platinum oxide [15]. The presence of plat-
inum oxide is not substantiated by X-ray measurements
but is quite probable. This complicates quantitative
estimates of the degree of reduction of cerium oxide.
ëÂé2 phase (ëÂ4+
ëÂ3+) [12–14]. The total amount
of H2 consumed in reduction corresponds to the forma-
tion of the ëÂé1.83 phase. This result closely agrees
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 81 No. 10 2007

1604
ZOSIMOVA et al.
the reaction during the first several minutes [16].
CeO2[111]
According to the thermodynamic data, equilibrium
shifts toward the initial reagents starting with 150°ë.
For this reason, comparatively low toluene conversion
is observed at the initial stage of the reaction. The con-
version of toluene reaches maximum values after the
minimization of temperature gradients when the tem-
perature of the system is stabilized at 150°ë. A subse-
quent decrease in activity (see Fig. 6) can be related to
the gradual deactivation of the platinum catalyst caused
by the formation of condensed aromatic compounds on
its surface [17].
Pt[200]
Pt[111]
Pt
The Pt2Ce intermetallic compound was almost inac-
tive in the hydrogenation of toluene compared with
metallic platinum. The absence of catalytic activity in
spite of a high content of platinum in this compound
can be caused by changes in the electronic and struc-
tural properties of platinum as a result of alloy forma-
tion. Another reason for low activity can be related to
size effects. For instance, a decrease in the size of plat-
inum particles to 3 nm or less sharply decreases its cat-
alytic activity [18]. Because of the distribution of plati-
num in the bulk of the Pt2Ce intermetallic compound,
the number of closely spaced platinum atoms can be
insufficient for the formation of active centers.
CeO2
5 nm
The conversion of toluene on the Pt–CeOx system
after oxidation followed by reduction is higher than on
metallic platinum (Fig. 4). The activity of the catalyst
increases during the reaction, and the conversion
reaches a constant level of 80% in 2 h. The second
redox cycle increases the conversion insignificantly, to
83%, and no activity increase during the initial reaction
stage is observed. The third and subsequent cycles give
similar results. As distinct from metallic platinum, no
noticeable deactivation of the Pt–CeOx system was
observed when the reaction continued for four hours. In
addition, effects related to superheating at the initial
stage were weaker.
Fig. 2. Transmission electron microscope images of Pt Ce
2
oxidized at 550°ë.
It is, however, clear that the presence of platinum in
oxidized Pt2Ce samples substantially facilitates reduc-
tion. The sample subjected to oxidation and reduction
will be denoted by Pt–CeOx.
The Hydrogenation of Toluene on the Pt–CeOx System
The catalytic properties of the samples were studied
in the hydrogenation of toluene. A powder of metallic
platinum containing particles of the size equal to that of
Pt–CeOx particles was used as a reference sample. After
each catalytic experiment, samples were subjected to
oxidative treatment in a flow of air at 550°ë followed
by reduction in hydrogen at 450°ë and then again used
in the reaction (Fig. 4).
To summarize, the interaction of platinum with
cerium oxide changes its catalytic activity and
increases the stability of catalyst functioning compared
with metallic platinum.
Catalyst Deactivation by Hydrogen Sulfide
The conversion of toluene on metallic platinum was
40% in one hour on a powder not subjected to prelimi-
nary treatment and 50% after its oxidative activation
(Fig. 6). The second cycle of redox treatment increased
conversion to 70%. In subsequent cycles, catalytic
activity remained virtually unchanged. An increase in
activity after the first two cycles of redox treatment is
likely related to the formation of microcracks on the
surface of platinum, which increases the active surface
area accessible to the reagents.
The experimental data on the deactivation of cata-
lysts by the pulsed introduction of hydrogen sulfide into
the flow of the reagents are shown in Fig. 5. The con-
version of toluene on metallic platinum decreases
directly proportionally to the total amount of hydrogen
sulfide introduced. The linear dependence obtained is
evidence that the poisoning of platinum with sulfur is a
homogeneous deactivation [19] when all poisoning
molecules are uniformly distributed over active catalyst
Considerable catalyst superheatings caused by the sites and block the access of reagents to them. Note that
high exothermic effect of toluene hydrogenation platinum fully restores its catalytic activity after regen-
(∆rH298 = –205 kJ/mol) were observed for platinum in eration.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 81 No. 10 2007

THE PHYSICOCHEMICAL AND CATALYTIC PROPERTIES
1605
Pt–CeO2
CeO2
0
200
400
600
800
°C
Fig. 3. Temperature-programmed spectra of reduction with H of cerium(IV) oxide and Pt Ce oxidized at 550°ë.
2
2
y, %
100
y, %
3'
2'
80
60
40
20
0
80
3
60
40
20
2
2
1
1'
2'
1'
0
1
2
3
4
τ, h
1
100
300
500
[H2S], µg/g Pt
Fig. 4. Dependences of toluene conversion (y) on the dura-
tion of reaction on (1–3) metallic platinum and (1'–3') Pt Ce
2
samples: (1, 1') initial samples; (2, 2') samples subjected to
redox treatment, cycle 1; and (3, 3') samples subjected to
Fig. 5. Dependences of toluene conversion on the amount of
–1
H S introduced during the hydrogenation of toluene on (1, 2)
redox treatment, cycle 2; conditions: 0.7 h , 150°ë, and
2
–1
Pt metal and (1'–2') Pt–CeO at 0.7 h , 150°ë, and 1 atm.
1 atm.
x
gram of platinum, the slope of the deactivation curve
changes, the activity of the sample decreases much
more slowly in spite of continuing hydrogen sulfide
The deactivation curves of the Pt–CeOx system can
be conventionally divided into two linear portions. The
conversion of toluene sharply decreases along the first
curve portion. After the introduction of 100 µg H2S per introduction. Such a shape of the deactivation curve can
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 81 No. 10 2007

1606
ZOSIMOVA et al.
4 µm
4 µm
(‡)
(b)
(c)
(d)
Fig. 6. (a) Scanning electron microscope and (b–d) energy dispersion analyzer images of Pt–CeO samples after poisoning with
x
H S. White regions in energy dispersion analyzer images correspond to (b) Ce, (c) Pt, and (d) S.
2
be explained by the presence of two active platinum cerium oxide are more stable toward poisoning with
centers of different natures with different tolerance
toward poisoning with sulfur. We believe that the first
curve portion corresponds to the fast deactivation of
metallic platinum particles on the surface of catalyst
granules. The second curve portion corresponds to slow
poisoning of platinum particles that are in contact with
CeOx and, in the limiting case, encapsulated in cerium
oxide.
sulfur.
CONCLUSIONS
To summarize, the calcining of the Pt2Ce intermetal-
lic compound obtained by alloying the metals in a flow
of air at 550°ë is accompanied by the destruction of the
initial intermetallic phase and formation of metallic
platinum and cerium oxide particles and high-disper-
sity platinum particles encapsulated in cerium oxide.
As distinct from the initial intermetallic compound,
oxidized samples possess catalytic activity in the
hydrogenation of toluene; their activity is higher than
that of metallic platinum.
The regeneration of Pt–CeOx fully restores its initial
activity. When the same sample was again deactivated
with hydrogen sulfide, the conversion level starting
from which the conversion of toluene ceased to
decrease increased from 5 to 15%. This could be related
to an increase in the number of platinum particles inter-
acting with cerium oxide as a result of redox treat-
ments.
Studies of catalyst poisoning with hydrogen sulfide
showed that the Pt–CeOx system contained two types of
Pt–CeOx systems. After the introduction of hydrogen active centers that were differently deactivated in the
An additional experiment was performed to sub-
stantiate the suggested mechanism of deactivation of
sulfide in an amount of 100 µg H2S/g Pt, which corre-
sponded to the inflection point of the deactivation
curve, the sample was studied by scanning electron
microscopy with energy dispersion analysis (Fig. 6).
The images clearly show that sulfur is largely accumu-
lated on platinum tracks. This substantiates the sugges-
tion that platinum metal predominantly experiences
presence of hydrogen sulfide. Centers of the first type
were metallic platinum particles on the surface of cata-
lyst granules, they were rapidly deactivated with hydro-
gen sulfide. Centers of the second type were high-dis-
persity platinum particles incorporated into the struc-
ture of cerium oxide; they possessed enhanced stability
poisoning, whereas platinum centers encapsulated in toward the action of hydrogen sulfide.
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 81 No. 10 2007

THE PHYSICOCHEMICAL AND CATALYTIC PROPERTIES
1607
8. S. Bernal, J. J. Calvino, M. A. Cauqui, et al., Catal.
ACKNOWLEDGMENTS
Today 50, 175 (1999).
This work was financially supported by the Russian
Foundation for Basic Research (project nos. 06-03-
32830, 06-03-32914, 05-03-32760, and 05-03-32045).
P.A. Zosimova thanks LG-Chem company for personal
stipend for young scientists and postgraduates (2006).
9. S. Bernal, J. J. Calvino, J. M. Gatica, et al., J. Catal. 169,
510 (1997).
10. Ch. Hardacre, T. Rayment, and R. M. Lambert, J. Catal.
158, 102 (1996).
11. G. K. Boreskov, Heterogeneous Catalysis (Nauka, Mos-
cow, 1988) [in Russian].
12. C. D. Leitenburg, A. Trovarelli, and J. Kaspar, J. Catal.
REFERENCES
166, 98 (1997).
1. D. J. Fullerton, A. V. K. Westwood, R. Brydson, et al.,
13. F. Fajardie, J.-F. Tempere, J.-M. Manoli, et al., J. Chem.
Catal. Today 81, 659 (2003).
Soc., Faraday Trans. 94, 3727 (1998).
14. S. E. Golunski, H. A. Hatcher, R. R. Rajaram, and
2. A. Corma, A. Martinez, and S. V. Martinez, J. Catal. 169
T. J. Truex, Appl. Catal., B 5, 367 (1995).
(2), 480 (1997).
15. R. W. McCabe, C. Wong, and H. S. Woo, J. Catal. 114,
3. S. Damyanova and J. M. C. Buenob, Appl. Catal., A 253,
354 (1988).
135 (2003).
16. D. R. Stull, E. F. Westrum, and G. S. Sinke, The Chemi-
cal Thermodynamics of Organic Compounds (Wiley,
New York, 1969; Mir, Moscow, 1971).
4. E. Rogemond, N. Essayem, R. Frety, et al., J. Catal. 166,
229 (1997).
5. C. Hardacre, G. M. Roe, and R. M. Lambert, Surf. Sci.
17. J. Chupin, N. S. Gnepa, S. Lacombeb, and M. Guisnet,
326, 1 (1995).
Appl. Catal., A 206, 43 (2001).
6. P. Bera, K. C. Patil, V. Jayaram, et al., J. Catal. 196, 293
18. W. F. Graydon and M. D. Langan, J. Catal. 69, 180
(2000).
(1981).
7. R. D. S. Monteiro, F. B. Noronha, L. C. Dieguez, and 19. R. Hughes, Deactivation of Catalysts (Academic, Lon-
M. Schmal, Appl. Catal., A 131, 89 (1995). don, 1984; Khimiya, Moscow, 1989).
RUSSIAN JOURNAL OF PHYSICAL CHEMISTRY A Vol. 81 No. 10 2007