Isotope Exchange Reaction of Formate with Molecular Hydrogen on Ni(110) by IRAS

Article abstract of DOI:10.1021/jp9611986

Kinetics of the CH/CD isotope exchange reaction of formate (HCOO(a)/DCOO(a)) on Ni(110) with gaseous D2/H2 was investigated below the decomposition temperature (360 K) by infrared reflection absorption spectroscopy (IRAS).The apparent activation energies of HCOO(a) -> DCOO (a) and DCOO(a) -> HCOO(a) reactions were derived as 52 +/- 5 and 58 +/- 5 kJ*mol^-1, respectively.A half-order dependence of the reaction rate on hydrogen pressure was observed, suggesting that the rate-determining step was the reaction of the adsorbed formate with reversibly adsorbed hydrogen atom The reaction of HCOO(a) with D2 was 2.1 +/- 0.3 times faster than that of DCOO(a) with H2 at 300 K, and the origin of the isotope effect is discussed.

Full text of DOI:10.1021/jp9611986

J. Phys. Chem. 1996, 100, 18177-18182
18177
Isotope Exchange Reaction of Formate with Molecular Hydrogen on Ni(110) by IRAS
Akira Yamakata, Jun Kubota, Junko N. Kondo, Kazunari Domen,* and Chiaki Hirose
Research Laboratory of Resources Utilization, Tokyo Institute of Technology, 4259 Nagatsuta,
Midori-ku, Yokohama 226, Japan
ReceiVed: April 25, 1996; In Final Form: September 6, 1996X
Kinetics of the CH/CD isotope exchange reaction of formate (HCOO(a)/DCOO(a)) on Ni(110) with gaseous
D2/H2 was investigated below the decomposition temperature (360 K) by infrared reflection absorption
spectroscopy (IRAS). The apparent activation energies of HCOO(a) f DCOO (a) and DCOO(a) f HCOO-
(a) reactions were derived as 52 ( 5 and 58 ( 5 kJ‚mol-1, respectively. A half-order dependence of the
reaction rate on hydrogen pressure was observed, suggesting that the rate-determining step was the reaction
of the adsorbed formate with reversibly adsorbed hydrogen atom. The reaction of HCOO(a) with D2 was 2.1
( 0.3 times faster than that of DCOO(a) with H2 at 300 K, and the origin of the isotope effect is discussed.
1. Introduction
SCHEME 1: H-D Isotope Exchange Reaction of
Formate on Ni(110)
One of the promising approaches for understanding the details
of surface chemical reactions including heterogeneous catalysis
is in situ spectroscopic observation of adsorbed species under
the reaction conditions. Although a large number of studies
have been carried out to reveal the dynamic behavior of
adsorbed molecules on powder catalysts by transmission IR
spectroscopy,1,2 such studies are still limited on well-defined
single-crystal surfaces under steady-state reaction conditions.
Infrared reflection absorption spectroscopy (IRAS) is a powerful
technique which enables the observation of vibrational spectra
of the adsorbed species in the presence of molecules in the gas
phase.
2. Experimental Section
The experiment was carried out in an ultrahigh vacuum
(UHV) chamber equipped with quadruple mass analyzer (Q-
MS), LEED-AES optics. The base pressure of the chamber
was below 2 × 10-10 Torr (1 Torr ) 133 Pa). The Ni(110)
crystal (10 mm in diameter) was mounted on a manipulator by
0.3 mm tantalum wires spot-welded on the back of the crystal.
The surface was cleaned by argon ion sputtering and annealing
at 1023 K.
Formate has been observed as an important intermediate in
various heterogeneous catalytic reactions such as the water-
gas shift reaction and methanol synthesis reaction,3-5 and the
reactivity of the formate adsorbed on metal surface is a subject
worthy of careful study.
For the IRAS measurement, a JEOL JIR-100 spectrometer
was used with either a MCT or InSb detector at a resolution of
4 cm-1. The incident angle of the infrared beam through a NaCl
window was about 84.5° from the surface-normal, and typically,
512 scans (12 min) were averaged for obtaining one spectrum.
In the TPD measurement, five different mass numbers were
The structure of adsorbed formic acid on Ni(110) was studied
in detail regarding the decomposition process.6-8 The adsorbed
species observed at around room temperature on Ni(110) was
assigned to the formate on the short-bridge site forming an
ordered c(2 × 2) layer by high-resolution electron energy loss
spectroscopy (HREELS) and low-energy electron diffraction
(LEED).9-12 The generation of formate from associatively
adsorbed formic acid on Ni(110) was also studied by IRAS,13
and the orientation of the formate with the C-H bond being
perpendicular to the surface was also confirmed. The decom-
position of formate was found to be induced by atomic hydrogen
on Ru(001),14,15 which suggests that the consideration of the
presence of atomic hydrogen is a key factor in elucidating the
decomposition mechanism of surface formate.
simultaneously monitored with a heating rate of 3 K‚s-1
.
Formic acid was introduced onto the Ni(110) through a
stainless steel nozzle (2 mm in diameter) located at 20 mm from
the sample. The formic acid (99% purity) was dried over copper
sulfate anhydride and purified by vacuum distillation and a
freeze-pump-thaw cycle. The surface with saturation cover-
age of formate was obtained by dosing about 1 L (1 L ) 1 ×
10-6 Torr‚s) of formic acid at 300 K, and the derived surface
displayed the c(2 × 2) structure. As recently reported by Haq
et al.,13 the coadsorbed CO was always observed at 2036 and
2073 cm-1 when formate was formed. The coverage of CO,
however, was less than 5% of the total number of surface Ni
atoms, and the surface prepared by the above method was used
as a formate full-covered one (θHCOO ) 1) in the following
experiments. The integrated intensity of CO bands did not
change, although they slightly broadened during the subse-
quently performed isotope exchange reaction. The surface was
exposed to D2 or H2 by back-filling the chamber under
continuous pumping. IRA spectra were recorded under evacu-
In this work, the isotope exchange reaction of formate
produced on Ni(110) under the atmosphere of hydrogen was
studied to elucidate the dynamic behavior of the adsorbed
species steadily interacting with H2 or D2. The hydrogen
exchange reaction of the formate on Ni(110) as depicted in
Scheme 1 was found to proceed at around room temperature in
the presence of gas-phase hydrogen without any side reaction
such as the decomposition reaction of formate itself.
X Abstract published in AdVance ACS Abstracts, October 15, 1996.
S0022-3654(96)01198-7 CCC: $12.00 © 1996 American Chemical Society

18178 J. Phys. Chem., Vol. 100, No. 46, 1996
Yamakata et al.
band. Therefore our assignment will be more reasonable. It
should be noted that the peaks of δ(CH), the out-of-plane
deformation band (π(CH)), and νa(OCO) were absent. Taking into
account of the surface selection rule of IRAS, the C-H bond
was regarded as being perpendicular to the surface.
The presence of gaseous or adsorbed hydrogen did not lead
to any noticeable change in the IRA spectrum of formate,
suggesting that the coadsorption of hydrogen on formate-covered
Ni(110) did not cause any significant change in IRA spectra.
Isotope Exchange Reaction of Formate on Ni(110). When
the HCOO-covered surface was exposed to D2 (7 × 10-6 Torr),
the intensity of the ν(CH) (2943 cm-1) and the νs(OCO) (1365
cm-1) bands commenced to decrease, and new bands assigned
to the CD stretching (ν(CD)) and the νs(OCO) bands of DCOO(a)
species appeared at 2188 and 1327 cm-1, respectively, as shown
in Figures 3 and 4. When the D2 gas was replaced by H2 after
the ν(CH) band almost disappeared, the reverse phenomenon was
observed; the ν(CH) band reappeared, and the initial peak intensity
of ν(CH) was recovered completely at the expense of the ν(CD)
band. The quantitative analysis of the process with normaliza-
tion taking into account of the absorption coefficients (ꢀCD/ꢀCH
) 2.1) is shown in Figure 5. The total coverage of formate
(HCOO(a) + DCOO(a)) was confirmed to be constant during
the isotope exchange reaction, i.e., no side reactions such as
decomposition or desorption occurred.
Figure 1. Temperature-programmed desorption (TPD) spectra of
formic acid adsorbed on Ni(110) surface at 113 K.
Figure 2. IRAS spectrum of formate adsorbed on a Ni(110) surface
at 300 K.
Dipole-Dipole Interaction of the νs(OCO) Band. As
shown in Figures 3 and 4, the peak position of the νs(OCO) band
of HCOO(a) and DCOO(a) shifted as the reaction proceeded,
while those of the ν(CH) and ν(CD) bands stayed at the same
positions. Figure 6 is the plot against the coverage of the peak
positions of the νs(OCO) band of HCOO(a) and DCOO(a), where
the coverage was estimated from the intensities of the ν(CH) and
ν(CD) bands shown in Figure 3. The peak positions shifted
almost linearly with the coverage by about 7 and 10 cm-1 for
HCOO(a) and DCOO(a), respectively.
The frequency shift of the νs(OCO) mode of HCOO(a) and
DCOO(a) was attributed to the dipole-dipole coupling among
the same isotope species.21,22 The peak shift due to the dipole-
dipole coupling has been reported for the ν(CO) band of
CO(a),21-23 νs(OCO) band of formate,16 and ν(CO) band of
methoxy.24,25 The linear dependence of the presently observed
peak shift on the coverage seems to suggest that the isotope
exchange reaction on the Ni(110) surface proceeded homoge-
neously.21,22,26 In other word, islands of HCOO(a) and DCOO-
(a) were not formed during the exchange reaction.
The intensities of the νs(OCO) bands were much larger than
those of the ν(CH) and ν(CD) bands which have the linear
relationship between them in intensity (not shown), and the
distinct and linear shift of the νs(OCO) band with the coverage
was used to evaluate the coverage in the subsequent analysis,
as a more precise method than that calculated by the peak
intensities.
Kinetic Analysis of Isotope Exchange Reaction by IRAS.
The kinetics of the isotope exchange reaction is discussed on
the Ni(110) surface with full coverage of formate. The rates
of the isotope exchange reactions are defined as follows:
ation after dosing a certain amount of hydrogen. Since the
exchange reaction does not proceed in the absence of gaseous
hydrogen, the change of IRA spectra gives the reliable rate of
isotope exchange reaction.
3. Results and Discussion
TPD of Formic Acid on Ni(110). The TPD spectra of H2,
H2O, CO, CO2, and HCOOH from the formic acid adsorbed on
Ni(110) at 110 K are shown in Figure 1. The desorption peak
of HCOOH (m/e ) 46) around 185 K was assigned to the
desorption from the second layer or multilayer of formic acid
because this peak increased without saturation by increasing
the exposure. The other peaks of H2, H2O, CO, and CO2 at
185 K were assigned to the cracking of HCOOH.
A broad desorption peak of hydrogen was observed at
250∼320 K and was attributable to the dissociation of the
hydroxyl group of formic acid to form formate. The desorption
peaks of CO2 and H2 at around 360 K were due to the
dehydrogenation of formate as was reported previously, while
the desorption peak of CO at around 440 K was attributed to
the dehydration of formate.6-8 H2O should be produced
simultaneously in the dehydration but was not detected in our
experiment presumably because of the high background signal
of H2O from the wall of the chamber. The results indicated
that the decomposition of formate started at 330 K. Therefore
the CH/CD isotope exchange reaction was performed between
290 and 305 K.
IRAS Spectra of Formate on Ni(110). The IRA spectrum
of formate on Ni(110) surface at 300 K is shown in Figure 2.
Four peaks observed at 779, 1365, 2854, and 2943 cm-1 are
assigned to the OCO deformation band (δ(OCO)), the symmetric
CO stretching band (νs(OCO)), the combination band of the in-
plane CH deformation band (δ(CH)) and the asymmetric CO
stretching band (νa(OCO)), and the CH stretching band (ν(CH)),
respectively. The assignments are summarized in Table 1 along
with the values reported on other metal surfaces.16-20 The peaks
at 2840 and 2944 cm-1 have been assigned to the ν(CH) and the
combination band by Haq et al.,13 respectively, which is different
from the present one. The combination band enhanced by Fermi
resonance should not be more intense than the fundamental ν(CH)
dθHCOO
rHD ) -
rDH ) -
) k′′ θmHCOOPnD
(1a)
(1b)
HD
2
dt
dθDCOO
dt
) k′′ θmDCOO PnH
DH
2
where r, k′′, θHCOO, θDCOO, PD , and PH are the rate of the
2
2
exchange, apparent rate constant, coverage of HCOO(a) and
DCOO(a), and pressure of D2 and H2, respectively. The

Isotope Exchange Reaction of Formate with H2
J. Phys. Chem., Vol. 100, No. 46, 1996 18179
TABLE 1: Vibrational Frequencies of Formate on Various Metal Surfacesa
Ni(110)
2943
Ni(110)
2840
Cu(100)
2900
Cu(100)
Ru(001)
2939
Pt(110)
2950
Pt(111)
2950
ν(CH)
2910
2840
combination
νs(OCO)
δ(OCO)
2865
1365
779
2944
1352
770
2958
1358
-
2857b
1361
784
-
1340
785
-
1340
790
1330
760
method
ref
IRAS
c
IRAS
13
IRAS
16
HREELS
17
IRAS
18
HREELS
19
HREELS
20
a Frequencies are given in cm-1 b No explicit assignment by the author. c Present study.
.
Figure 3. Higher frequency region of IRA spectra of formate adsorbed
on a Ni(110) surface at 300 K under D2 flow at 7 × 10-6 Torr.
Figure 4. Lower frequency region of IRA spectra of formate adsorbed
on a Ni(110) surface at 300 K under D2 flow at 7 × 10-6 Torr.
subscripts of HD and DH are used to denote the reaction from
HCOO(a) to DCOO(a) and that from DCOO(a) to HCOO(a),
respectively, and the superscripts of m and n are the reaction
orders with respect to the coverage of formate and pressure of
D2 or H2, respectively.
Since the coverage of HCOO(a) (θHCOO) apparently follows
the exponential decay during the exchange reaction as shown
in Figure 5, the parameter m was estimated to be unity. Then
following equation was derived from (1a):
ln{θHCOO(t)/θHCOO(t ) 0)} ) -k′′ PnD t
(2)
DH
2
where t represents the reaction time, and θHCOO(t ) 0) and
HCOO(t) denote the coverage at t ) 0 and t, respectively. In
θ
Figure 5. Time course of the normalized peak intensity of the νCH
and νCD bands during the isotope exchange reaction of formate on Ni-
(110) surface; O, 0, and 4 represent the quantities of HCOO(a), DCOO-
(a), and HCOO(a) + DCOO(a), respectively.
Figure 7, the values of ln{θHCOO(t)/θHCOO(t ) 0) are plotted as
a function of the reaction time when the pressure of the hydrogen
was kept constant at 7 × 10-6 Torr and the temperature at 300
K. A good linear relationship between ln{θHCOO(t)/θHCOO(t )
0) and t was obtained, which confirms that the reaction rate is
in first order for θHCOO. The first-order dependence of the
reaction rate on θHCOO again suggests that the reaction occurs
homogeneously on the surface.
temperature and θHCOO, the parameter n is obtained by varying
the D2 pressure. The D2 pressure was varied from 5 × 10-6 to
1.4 × 10-5 Torr at 300 K, and n was calculated to be 0.5 ( 0.1
as shown in Figure 8. Then, the following eqs 4a,b are derived
from eq 1:
To determine the dependence of the reaction rate on the
hydrogen pressure, parameter n was substituted into eq 1a
leading the relation
0.5
rHD ) k′′ θ
P
(4a)
(4b)
HD HCOO D2
ln rHD ) ln k′′ + ln θHCOO + n ln PD
(3)
HD
2
0.5
rDH ) k′′ θ
P
DH DCOO H2
When the rate of the reaction is measured at a constant

18180 J. Phys. Chem., Vol. 100, No. 46, 1996
Yamakata et al.
k′′
HD
rHD
)
θ
θD
HCOO
KD
) kHDθHCOOθD
(7a)
(7b)
k′′
DH
rDH
)
θ
θH
DCOO
KH
) kDHθDCOOθH
In eq 7, rHD or rDH and kHD or kDH represent the rate and rate
constant of the surface reaction indicated by the suffixes (rate-
determining step). kHD and kDH are related to the apparent rate
Figure 6. Coverage-dependent feature of the peak shift of νs(OCO) bands
of formate on a Ni(110) surface during the isotope exchange reaction;
O and 0 represent HCOO and DCOO, respectively.
constants k′′ and k′D′H as follows:
HD
k′′
HD
kHD
)
)
(8a)
(8b)
KD
k′′
DH
kDH
KH
Denoting the apparent activation energy and the activation
energy of the rate-determining step by EqHD′′ and EHq D, left-hand
and right-hand sides of eq 8a are rewritten as follows:
EqHD′′
k′′ ) A′′ exp -
(9a)
(9b)
(
)
HD
RT
Figure 7. Coverage dependence of the isotope exchange reaction on
formate. The initial coverage of formate was set to be 1.
∆GD
K ) exp -
(
)
D
RT
EqHD′′
∆GD
kHD ) A′′ exp -
exp -
(
)
(
)
RT
RT
EqHD′′ - ∆HD
) A exp -
(9c)
(9d)
(
)
RT
EqHD ) EqHD′′ - ∆HD
Here, ∆GD or ∆HD and A′′ or A represent the change of free
energy on adsorption or heat of adsorption of D(a) and apparent
pre-exponential factor or pre-exponential factor of the rate-
determining step including the entropy term of the adsorption
of hydrogen, respectively. The energy diagram of the isotope
exchange reaction of formate is represented as depicted in
Scheme 2. Equations 4a and 9a lead to the relation for the
apparent activation energy EHD′′ expressed as follows:
Figure 8. Pressure dependence of the isotope exchange reaction of
HCOO(a) on the deuterium. The initial coverage of formate was set to
be 1, and the temperature of the surface was kept constant at 300 K.
The above-derived value 0.5 ( 0.1 of parameter n and unity of
m indicate that the rate-determining step is the reaction between
dissociatively adsorbed hydrogen and surface formate. Inci-
dentally, for the activation of molecular hydrogen to be the rate-
determining step, the order n should be unity, and a null value
of n should lead to the saturation in the amount of dissociatively
adsorbed hydrogen under the reaction condition.
EqHD′′
ln rHD ) -
+ (ln A + ln θHCOO + ln P0D.5) (10)
2
RT
In eq 10 the last three terms on the right-hand side are constant
at constant values of PD and θHCOO. The apparent activation
2
The adsorbed deuterium (D(a)) and deuterium in the gas phase
(D2(g)) are under equilibrium condition as expressed as follows:
energy was obtained by measuring the rate of the reaction at 7
× 10-6 Torr of H2 or D2 in the temperature range from 290 to
305 K. The Arrhenius plots (the plots of ln(rHD) and ln(rDH
)
1/2D2(g) a D(a)
(5)
(6)
versus (1/T) are shown in Figure 9, and the apparent activation
energies of the HCOO(a) to DCOO(a) and the DCOO(a) to
HCOO(a) reactions were estimated to be 52 ( 5 and 58 ( 5
kJ‚mol-1, respectively. The temperature of the gas-phase H2
or D2 was not changed on the grounds that the dissociated
hydrogen atoms play a key role as described above.
The equilibrium constant KD for the above is defined as
θD
KD )
P0D.5
The activation energy of the rate-determining step (Eq) can
be estimated by the eq 9d. Because there are no data for ∆HD
on formate-covered Ni(110), the value of the clean surface was
used as a first approximation. The heat of adsorption of H2 on
2
where θD represent the coverage of D(a). Then eq 7 follows
from eqs 4 and 6

Isotope Exchange Reaction of Formate with H2
J. Phys. Chem., Vol. 100, No. 46, 1996 18181
SCHEME 3: Mechanism of the Isotope Exchange
Reaction
Figure 9. Arrhenius plots for the isotope exchange reaction of formate;
O and 0 represent the reactions of HCOO(a) f DCOO(a) and DCOO-
(a) f HCOO(a), respectively. The pressure of D2 or H2 was kept
constant at 7 × 10-6 Torr.
SCHEME 2: Energy Diagram of the Isotope Exchange
Reaction
that the transition state of the hydrogenation does not differ too
much from the dehydrogenated state,33 and it will be a reactant-
like transition state. The KIE in hydrogenation is, therefore,
expected to be very small in dehydrogenation reactions. If the
rate-determining step of the exchange reaction in Scheme 3a
or 3b is the C-H (C-D) bond cleavage, the value of the KIE
should be larger than unity, i.e., kHD/kDH > 1.
On the contrary, assuming that the rate-determining step is
the C-D or C-H bond formation in Scheme 3a or 3b, the KIE
appears to be very small, which could account for the present
result. However, in Scheme 3a, it is unlikely that the intermedi-
ate OCO species exist stably; it would rather desorb as CO2.
The absence of the decomposition of formate species during
the isotope exchange reaction in the present study may exclude
the possibility of Scheme 3a. Because the intermediate in
Scheme 3b is known to exist as oxymethylene on oxide
surfaces,34 one may be tempted to take the “addition-abstrac-
tion” mechanism as a likely one. The recent theoretical
calculation, however, has shown that the oxymethylene on Cu
surface is unstable.35 In this respect, the Scheme 3b might be
questionable.
clean Ni(110) was reported to be about 45 kJ‚mol-1 (per H
atom),27 and the activation energy of the rate-determining step
of the isotope exchange reaction would be about 100 [)52∼58
+ 45] kJ‚mol-1
.
Isotope Effect. The rate of the reaction from CH to CD (rHD
)
is 2.1 ( 0.3 times faster than that from CD to CH (rDH) at 300
K with PD ) PH . The rate-determining step was found to be
2
2
the reaction between adsorbed hydrogen and formate. Referring
to eq 8, it consists of the thermodynamical and kinetic isotope
effects (KIEs) as expressed by the following relation:
rHD KD kHD
)
(11)
rDH KH kDH
The thermodynamical isotope effect (KD/KH) is evaluated by
the difference of the zero-point energies between molecular
hydrogen and an adsorbed one, which is equivalent to the
difference of the heat of adsorption of a D atom from that of a
H atom. It was reported that the heat of adsorption of a D atom
is 1.3∼2.1 kJ‚mol-1 (per H and D atom) larger than that of H
atom on Ni film.28,29 Thus KD/KH is estimated by eqs 6 and 9b
as 1.7∼2.3 at 300 K. This estimation indicates a very small
KIE in the isotope exchange reaction, i.e., kHD/kDH Z 1.
The microscopic reaction mechanism is then discussed on
the basis of the rate-determining step with negligible KIE. The
isotope exchange reaction is usually considered to proceed either
through an “abstraction-addition” mechanism in which a C-H
bond of formate dissociates first and the resulting complex
accepts a D atom (Scheme 3a) or through an “addition-
abstraction” mechanism (Scheme 3b).30 The two mechanisms
depict the exchange reaction as consisting of two steps occurring
in a different order; C-H (C-D) bond cleavage and C-D (C-
H) bond formation. The KIE of C-H (C-D) bond cleavage
was studied for the decomposition of the adsorbed species by
means of TPD.31,32 For example, the C-H bond is cleaved
4.4 or 5.5 times faster than the C-D is at 300 K in the case of
formate31 or methoxy.32 These results indicate that the transition
state involved has very little C-H bond character, and the KIE
is dominated by the difference of the zero-point energies
between the reactants (product-like transition state). It implies
As another possibility, the exchange reaction does not
necessarily proceed by two steps but by one step. In this case,
the absence of the KIE implies that the difference of the zero-
point energies between the reactants and the transition state are
almost the same. The present study leads us to propose a
concerted mechanism (Scheme 3c) in which both abstraction
and addition proceed simultaneously in the transition state.
At the present stage, we prefer Scheme 3c to Scheme 3b as
a reaction mechanism of the isotope exchange reaction. For
more detailed discussions, further experiments and theoretical
calculations would be required.
4. Conclusion
The isotope exchange reaction of formate (HCOO(a)/DCOO-
(a)) on Ni(110) with gaseous D2/H2 was investigated below the
decomposition temperature (330 K) by means of IRAS. When
D2 was admitted on the HCOO(a)-covered surface at 300 K,
the peak intensity of the ν(CH) band was decreased and that of
the ν(CD) band was increased with the D2 dosing time. The
feature reversed when D2 was replaced by H2 after HCOO(a)
was exchanged completely to DCOO(a). In the course of the
exchange reaction, any side reactions such as decomposition
or desorption were not observed. The apparent activation
energies for the HCOO(a) to DCOO(a) and the DCOO(a) to
HCOO(a) reactions were obtained as 52 ( 5 and 58 ( 5

18182 J. Phys. Chem., Vol. 100, No. 46, 1996
Yamakata et al.
kJ‚mol-1, respectively. The half-order dependence of the
reaction on hydrogen (deuterium) pressure suggested that the
rate-determining step was the reaction of the formate with
reversibly adsorbed atomic hydrogen. The activation energies
of the rate-determining step were roughly estimated to be ca.
100 kJ‚mol-1 by taking account of the heat of adsorption of
hydrogen.
(12) Jones, T. S.; Ashton, M. R.; Richardson, N. V. J. Chem. Phys.
1989, 90, 7564.
(13) Haq, S.; Love, J. G.; Sanders, H. E.; King, D. A. Surf. Sci. 1995,
325, 230.
(14) Sun, Y. K.; Weinberg, W. H. J. Chem. Phys. 1991, 94, 4587.
(15) Xie, J.; Mitchell, W. J.; Lyons, K. J.; Weinberg, W. H. J. Chem.
Phys. Lett. 1994, 101, 9195.
(16) Hayden, B. E.; Prince, K.; Woodruff, D. P.; Bradshow, A. M. Surf.
Sci. 1983, 133 , 589.
The observation that the reaction rate of HCOO(a) with D2
was 2.1 ( 0.3 times faster than that of DCOO(a) with H2 at
300 K was attributed to the difference of the amounts of
adsorption of deuterium and hydrogen. The kinetic isotope
effect was very small in the present exchange reaction, and a
possible reaction mechanism was proposed.
(17) Sexton, B. A. Surf. Sci. 1979, 88, 319.
(18) Weisel, M. D.; Chen, J. G.; Hoffmann, F. M.; Sun, Y.-K,; Weinberg,
W. H. J. Chem. Phys. 1992, 97, 9396.
(19) Hoffmann, P.; Bare, S. R.; Richardson, N. V.; King, D. A. Surf.
Sci. 1983, 133, L459.
(20) Columbia, M. R.; Crabtree, A. M.; Thiel, P. A. J. Am. Chem. Soc.
1992, 114, 1231.
(21) Hoffmann, F. M. Surf. Sci. Rep. 1983, 8, 211.
(22) Chabal, Y. J. Surf. Sci. Rep. 1988, 3, 107.
(23) Ortega, A. Thesis, Freie Universitat Berlin, 1983.
(24) Ryberg, R. Phys. ReV. Lett.. 1982, 49, 1579.
(25) Bare, S. R.; Stroscio, J. A.; Ho, W. Surf. Sci. 1985, 150 , 399.
(26) Burrows, V. A.; Sundaresan, S.; Chabal, Y. J.; Christman, S. B.
Surf. Sci. 1985, 160, 122.
Acknowledgment. A.Y. is grateful to Dr. K. Onda and T.
Ohtani for valuable discussions and suggestions.
References and Notes
(27) Christmann, K.; Schober, O.; Ertl, G. ; Neumann, M. J. Chem.
Phys. 1974, 60, 4528.
(28) Weldler, G.; Broker, F. J.; Fisch, G.; Schroll, G. Z. Phys. Chem.:
N.F. 1971, 76, 212.
(29) Ozaki, A. Isotopic Studies of Heterogenious Catalysis; Kodansha
(1) Urban, M. W. Vibrational Spectroscopy of Molecules and Macro-
molecules on Surfaces; John Wiley and Sons: New York, 1993.
(2) Anderson, J. R.; Boudart, M. Catalysis Science and Technology;
Springer-Verlag: Berlin, Heidelberg, New York, Tokyo, 1984.
(3) Campbell, I. M. Catalysis at surfaces; Chapman and Hall: London,
1988.
Academic Press: Tokyo, 1977.
(4) Keim, W. Catalysis in C1 Chemistry; D. Reidel Publishing Co.:
Dordrect, The Netherlands, 1983.
(5) Rodrigusz, J. A.; Goodmann, D. W. Surf. Sci. Rep. 1991, 14, 1.
(6) McCarty, J. G.; Falconer, J. L.; Madix, R. J. J. Catal. 1973, 30,
235.
(7) Benziger, J. B.; Madix, R. J. Surf. Sci. 1979, 79, 394.
(8) Falconer, J. L.; Madix, R. J. Surf. Sci. 1974, 46, 473.
(9) Madix, R. J.; Gland, J. L.; Mitchell, G. E.; Sexton, B. A. Surf. Sci.
1983, 125, 481.
(10) Jones, T. S.; Richardson, N. V.; Joshi, A. W. Surf. Sci. 1988, L948,
207.
(30) King, D. A.; Woodruff, D. P. The chemical Physics of Solid Surfaces
and Heterogenious Catalysis: Elsevier: Amsterdam, 1982; Vol. 4., Chapter
8.
(31) Madix, R. J.; Telford, S. G. Surf. Sci. 1992, 277, 246.
(32) Madix, R. J.; Telford, S. G. Surf. Sci. 1995, 328, L576.
(33) Campbell, C. T.; Campbell, J. M.; Dalton, P. J.; Henn, F. C.;
Rodriguez, J. A.; Seimanides, S. G. J. Phys. Chem. 1989, 93, 806.
(34) Busca, G.; Lamotte, J.; Lavalley, J.C.; Lorenzelli, V. J. Am. Chem.
Soc.. 1987, 109, 5197.
(35) Kakumoto, T. Energy ConVers. Manage. 1995, 661, 36(6-9).
(11) Jones, T. S.; Richardson, N. V. Surf. Sci. 1989, 211/212, 377.
JP9611986