Isothermal kinetic study of nitric oxide adsorption and decomposition on Pd(111) surfaces: Molecular beam experiments

Article abstract of DOI:10.1021/jp050813f

The kinetics of NO adsorption and dissociation on Pd(111) surfaces and the NO sticking coefficient (s_NO) were probed by isothermal kinetic measurements between 300 and 525 K using a molecular beam instrument. NO dissociation and N₂ productions were observed in the transient state from 425 K and above on Pd(111) surfaces with selective nitrogen production. Maximum nitrogen production was observed between 475 and 500 K. It was found that, at low temperatures, between 300 and 350 K, molecular adsorption occurs with a constant initial s_NO of 0.5 until the Pd(111) surface is covered to about 70-80% by NO. Then s_NO rapidly decreases with further increasing NO coverage, indicating typical precursor kinetics. The dynamic adsorption - desorption equilibrium on Pd(111) was probed in modulated beam experiments below 500 K. CO titration experiments after NO dosing indicate the diffusion of oxygen into the subsurface regions and beginning surface oxidation at a?￥475 K. Finally, we discuss the results with respect to the rate-limiting character of the different elementary steps of the reaction system. ? 2005 American Chemical Society.

Full text of DOI:10.1021/jp050813f

J. Phys. Chem. B 2005, 109, 13283-13290
13283
Isothermal Kinetic Study of Nitric Oxide Adsorption and Decomposition on Pd(111)
Surfaces: Molecular Beam Experiments
Kandasamy Thirunavukkarasu,† Krishnan Thirumoorthy,† Jo1rg Libuda,‡ and
Chinnakonda S. Gopinath*,†
Catalysis DiVision, National Chemical Laboratory, Dr. Homi Bhabha Road, Pune 411 008, India, and
Department of Chemical Physics, Fritz-Haber-Institut der Max-Planck-Gesellschaft, Faradayweg 4-6,
14195 Berlin, Germany
ReceiVed: February 16, 2005; In Final Form: May 30, 2005
The kinetics of NO adsorption and dissociation on Pd(111) surfaces and the NO sticking coefficient (sNO
)
were probed by isothermal kinetic measurements between 300 and 525 K using a molecular beam instrument.
NO dissociation and N2 productions were observed in the transient state from 425 K and above on Pd(111)
surfaces with selective nitrogen production. Maximum nitrogen production was observed between 475 and
500 K. It was found that, at low temperatures, between 300 and 350 K, molecular adsorption occurs with a
constant initial sNO of 0.5 until the Pd(111) surface is covered to about 70-80% by NO. Then sNO rapidly
decreases with further increasing NO coverage, indicating typical precursor kinetics. The dynamic adsorption
- desorption equilibrium on Pd(111) was probed in modulated beam experiments below 500 K. CO titration
experiments after NO dosing indicate the diffusion of oxygen into the subsurface regions and beginning
surface oxidation at g475 K. Finally, we discuss the results with respect to the rate-limiting character of the
different elementary steps of the reaction system.
1. Introduction
Palladium has been suggested as an alternative active element
for NO reduction due to its NO dissociation capacity and its
In the last 2 decades increasing governmental regulations all
over the world have stimulated tremendous growth in the
research on environmental catalysis and green chemistry.1-3 In
particular, the automotive catalyst technology has improved,
assisted by better petrochemical refining to yield cleaner fuels.
Improvements in internal combustion engines and the above
developments have led to fuel-efficient vehicles. The present
technology of three-way catalytic converters1,2 meets the
requirements of oxidation of CO. More problematic is the
oxidation of nonvolatile organic species to CO2 and, in
particular, the reduction of NO to N2, especially under the highly
oxidizing exhaust gas conditions of fuel-efficient lean-burn
engines.4 Although Rh is very active for NO reduction under
fuel-rich conditions, this is not the case under net oxidizing
conditions, since the excess oxygen inhibits the NO reduction
activity of Rh by oxidizing it. Pd has also been added to catalytic
converters as one of the active metals for NO reduction for the
past few years;5 however, the mechanistic details of Pd are not
fully understood yet. There were few new technologies6,7
introduced in the past decade, like NOx storage or zeolite based
catalysts, which provided different insights into the NO reduc-
tion. However, regarding long-term stability, low and high-
temperature NO reduction is still a challenge.4 The above points
clearly indicates that there is an urgent need to develop catalysts
for NO reduction under net oxidizing conditions. In this respect,
obtaining molecular-level insights into underlying catalysis
aspects of NO reduction is a major challenge for current
research.
stability under high temperature and oxidizing conditions.8-10
However, unlike Rh,11-15 Pd has not been subjected to intense
research, and its capability for NO reduction is not fully
understood. In the recent past, there have been numerous reports
on the NO reduction over Pd-based catalysts on different faces
of single crystals and a variety of supported systems.16-29
Despite that, there is a lack of fundamental understanding of
NO reduction on the Pd surfaces.
The present work of NO adsorption and dissociation on Pd-
(111) focuses on fundamental kinetic aspects under isothermal
conditions in a molecular beam instrument.30-34 It was suggested
that NO dissociates only at defect sites on Pd(111) without
significant inherent activity on Pd(111);17-20 nonetheless, there
are reports22-26,28,29,35 on steady-state NO dissociation to N2 and
N2O on Pd(111) as well as on Pd on supported oxides with
reductants, like CO and H2, indicating inherent catalytic activity
toward NO dissociation. Our present studies show that an
inherent NO dissociation activity exists on Pd(111) surfaces,
and the same can be observed clearly in the transient state. The
present work on the dissociation of NO on Pd(111) surfaces is
a part of our continuing study of the NO reduction reactions in
our laboratory, and the NO + CO reaction on Pd(111) is
reported in ref 35.
2. Experimental Section
All kinetic and TPD experiments reported in this manuscript
were performed using an extension of the so-called King and
Wells collimated beam method. The home-built molecular beam
instrument (MBI) used consists a 12 L capacity stainless steel
ultrahigh vacuum (UHV) chamber evacuated with a 210 L/s
turbo-molecular drag pump (Pfeiffer, TMU261) to a base
pressure of about 3 × 10-10 Torr. The MBI is equipped with a
* Corresponding author. E-mail: cs.gopinath@ncl.res.in. Fax: 0091-20-
2589 3761.
† National Chemical Laboratory.
‡ Fritz-Haber-Institut der Max-Planck-Gesellschaft.
10.1021/jp050813f CCC: $30.25 © 2005 American Chemical Society
Published on Web 06/21/2005

13284 J. Phys. Chem. B, Vol. 109, No. 27, 2005
Thirunavukkarasu et al.
can be distinguished from common background signals like
12CO2 (amu 44). The second factor is related to the 13CO
employed to minimize the interference between background
12CO and the main product of N2 in the NO dissociation reaction.
However, 14NO (amu 30) present in the 15NO interferes with
the identification of 15N2 (amu 30), and most of the experiments
reported here are carried out with 14NO. However, cross-
reference experiments were carried out with 15NO too. Small
amounts of 13CO were present in the MBI chamber, adsorbing
especially below 400 K. However, the adsorbed 13CO does not
interfere our experiments since CO is displaced by NO in the
transient state of the subsequent experiments and can be detected
easily.35 Several test experiments have been carried out initially
to probe all possible products of NO decomposition, including
N2O and NO2 under a variety of experimental conditions. Apart
from nitrogen and nitrous oxide, no other nitrogen containing
products were produced under any conditions. The mass
spectrometer intensity was calibrated for NO by measuring the
uptake on clean Pd(111) at 300 K, assuming a saturation
coverage of 0.33 monolayers (ML) following Jacobi et al.17
Different parameters (sNO, ΘNO) were calculated by following
the procedures given in ref 33.
Figure 1. Schematic representation of molecular beam generation.
molecular beam doser setup (Figure 1), a quadrupole mass
spectrometer (Pfeiffer, Prisma QMS 200 M3), a sputter ion gun
(AG5000, VG Scientific), and sample preparation facilities. The
mass spectrometer is kept out of the line-of-sight of the sample
to avoid any angular desorption effects. The molecular beam
doser consists of a 13 mm disk multichannel array made up of
microcapillary glass tubes of 1 mm in length and 10 µm in
diameter each (Collimated Holes Inc.). The NO gas-flux in the
molecular beam (FNO) is determined and controlled by the
precision leak-valve opening and the backing gas pressure in
the gas-manifold. A laterally movable stainless steel shutter,
which is manually operated, was placed between the micro-
capillary doser and the Pd(111) crystal in order to interrupt the
beam when desired (Figure 1). The Pd single crystal (Metal
Crystals and Oxides Ltd., Cambridge, U.K.) was cut in the (111)
direction and polished using standard procedures. It was
mounted by spot-welding a 0.5 mm thick tantalum wire on the
periphery of the crystal and connected to a pair of copper rods,
which in turn were connected to a power-thermocouple feed-
through. A K-type thermocouple was welded on the backside
of the crystal and to the thermocouple leads of the feed-through.
The above feed-through is connected to liquid nitrogen reservoir.
The crystal can be cooled to about 100 K with liquid nitrogen
or resistively heated to 1373 K. For the present experiments,
the distance between the crystal and the beam doser is set to 5
mm.
The Pd(111) crystal surface was cleaned as described by
Ramsier et al.18 The procedure involves Ar+ sputtering at 1000
K in a partial pressure of oxygen (4 × 10-8 Torr) for 60 min,
followed by annealing for 30 min at 1000 K in 2 × 10-7 Torr
of oxygen, and, finally, flashing to 1200 K in a vacuum. This
cycle was repeated several times, testing the cleanliness of the
surface by monitoring CO and CO2 during high-temperature
flashing after oxygen adsorption at room temperature. Room-
temperature NO adsorption and subsequent TPD experiments
were also carried out, and the results (not shown) are in good
agreement with results reported earlier,17-19 ensuring surface
cleanliness of Pd(111). However with subsequent NO dosing
experiments, especially at high temperatures (above 500K),
small amounts of N2 and CO were observed during high
temperature flashing, possibly indicating the subsurface diffusion
of N and O atoms at high temperatures. Hence, after each set
of high-temperature experiments the surface was cleaned by
above Ar+ sputtering procedure. For the present manuscript NO/
Pd(111) studies between 300 and 525 K are being reported.
15NO (Air gas, 99% isotopically pure), 14NO, CO (99.9%),
and 13CO (Isotec) were used without any further purification.
It is to be noted that the 15NO contains about 1% of unlabeled
14NO. The purpose of using 13CO and 15NO is 2-fold: First it
allows a clear identification of the possible products such as
13CO2 (amu 45) and 15N2O (amu 46). In addition, these species
3. Results and Discussion
3.1. General Considerations. The experimental procedure
followed in a typical molecular beam dosing experiment is
explained in Figure 2. For all experiments, the clean Pd(111)
metal surface was kept at constant temperature and exposed to
an effusive 14NO molecular beam. All relevant gas-phase species
formed are detected by QMS as a function of time. Figure 2
shows a typical set of kinetic data recorded during 14NO
exposure of Pd(111) at 465 K: At time t ) 5 s, the molecular
beam with a NO beam was allowed to enter the UHV chamber
(Figure 1). With the shutter blocking the crystal, there is,
however, no direct exposure to the 14NO beam. Nonetheless, a
small amount of 14NO adsorption from the background cannot
be prevented. After few seconds, at t ) 10 s, the shutter was
removed to allow the beam to directly impinge on the Pd(111)
surface. An instantaneous decrease in the partial pressure of
14NO for few seconds can be noticed as a result of 14NO
adsorption from the beam. This adsorption in the transient state
continues until the surface is saturated with 14NO. The geo-
metrical arrangements in our MBI36 is such that about 25% of
the beam is intercepted by the crystal and, hence, the drop in
NO pressure upon shutter removal is relatively shallow; this is
mainly to ensure an uniform beam profile over the entire crystal
surface. A clear N2 and N2O production in the transient state
begins around t ) 15 s can be seen, indicating the dissociation
of 14NO on the Pd(111) surface. Nitrogen desorption is evident
at other temperatures also (Figure 4). However, the amount of
14N2O desorption is comparatively much lower than that of N2.
Amu 12 was also followed in all the experiments to ensure that
there is no contribution from CO and CO2.
The partial pressure of the 14NO increases up to a new steady-
state value, due to adsorption on UHV chamber walls;11
however, it does not represent any change in the molecular beam
characteristics. 14NO adsorption was continued for at least 60 s
before any steady-state measurements were made. Upon reach-
ing the steady-state, the beam was blocked for about 30 s.
During the beam blocking period in the steady state, an increase
in the partial pressure of 14NO (Figure 2) was observed clearly.
Moreover, the sudden increase in the 14NO partial pressure in
the steady state upon closing the shutter can be attributed to
the immediate desorption of 14NO, since the substrate temper-

Nitric Oxide Adsorption and Decomposition on Pd
J. Phys. Chem. B, Vol. 109, No. 27, 2005 13285
Figure 2. Experimental data from a typical isothermal kinetic test described in the text carried out with 14NO at 465 K and subsequently the
surface oxygen was titrated with 13CO beam at the same temperature. Different mass traces are deliberately shifted to show the features clearly.
activity does not sustain on its own. This is attributed to
poisoning by dissociated atomic species, mainly oxygen atoms.
It is also to be noted that the NO uptake after shutter removal
in the steady state is nearly the same as that of NO uptake on
clean Pd(111) in the transient state. This indicates that almost
all nitrogen and NO desorb during the beam blocking period
and the NO and nitrogen coverages are low.
A subsequent TPD shows no NO and a small nitrogen
coverage. This is further supported by TPD results (in Figures
6). After turning off the NO beam, the 13CO beam was switched
on at t ) 270 s. and constant temperature in order to titrate the
surface oxygen species produced by NO dissociation. A clear
production of 13CO2 indicates the NO dissociation on Pd(111).
The above CO dosing was carried out also to minimize the
diffusion of oxygen into the subsurface and the bulk region.
Finally, the 13CO beam was shut off, and the chamber
background pressure was allowed to return to normal before
TPD experiments were performed at a heating rate of 5 K/s.
The TPD results indicate a strongly temperature-dependent
coverage of NO and small amount of nitrogen (Figure 6). As
there is no desorption of nitrogen-related products above 550
K, the TPD experiments were stopped at 650 K. Subsequently,
the Pd(111) crystal was flashed to 1000 K before the next dosing
Figure 3. Temperature dependence of 14NO adsorption on Pd(111) in
experiment.
the transient state on clean metal surface. Scattered data points (open
3.2. Temperature-Dependent NO Adsorption and Dis-
sociation on Pd(111). Figure 3 shows the temperature depen-
dence of 14NO adsorption on Pd(111) between 300 and 525 K
at a constant FNO ) 0.05 ML/s. NO beam was turned on at t )
7 s followed by the shutter opening at t ) 10 s; there is a clear
decrease in the 14NO partial pressure due to 14NO adsorption
on the Pd(111) surface at the shutter removal point. A number
of reference experiments were made, in which the shutter was
not at all opened and the data were used as background for
those experiments reported in the manuscript for all the
circles) shown for 300 K are from the experiment carried out without
opening the shutter to use as standard for background subtraction.
Reference data has been shifted vertically for clarity.
ature is close to the NO desorption maximum of 500 K.17-20
This result shows that there is a substantial 14NO steady-state
coverage established throughout the reaction period. N2 and N2O
desorption stops completely within a minute of starting of NO
dosing and there is no change in the partial pressure of N2 and
N2O in the steady-state indicating that the NO dissociation

13286 J. Phys. Chem. B, Vol. 109, No. 27, 2005
Thirunavukkarasu et al.
Figure 4. (a) N2 and (b) N2O production from Pd(111) surface due to NO dosing at different temperatures. Note the maximum N2O production
at 475 K and N2 between 475 and 500 K. Solid arrows indicate the on-set of products desorption. Scattered data points shown for 475 K are from
the experiment carried out without opening the shutter.
calculation purposes. One such reference experiment is shown
in Figure 3 at 300 K. It is to be noted here that the above
procedure is reliable than fitting with an exponential or quadratic
equations. However, some background adsorption (about 5%)
cannot be avoided,33 and to reduce that contribution, a minimum
time gap (2-3 s) was kept between the beam on and the shutter
removal time. Further, the uniform beam profile over the crystal
surface also ensures the pressure on the crystal surface is much
higher than the background. At temperatures below 400 K, 14NO
was continuously adsorbed at the same rate for a few seconds,
before a gradual decrease occurs due to the surface reaching
the saturation NO coverage (ΘNO) around t ) 20 s. At and above
400 K, there is a gradually decreasing total NO adsorption
capacity compared to the experiments between 300 and 350 K.
The decreasing saturation coverage with increasing temperature
is due to a reduction of the number of vacant sites, as a result
of NO dissociation to N and O atomic species. It should also
be noted that the initial NO adsorption probability remains
constant immediately after the shutter removal at reaction
temperatures between 300 and 425 K. The corresponding
changes in the amu 28 and 44 traces (due to 14N2 and 14N2O)
(Figure 4) provide a hint that, at least in part, the reduction in
Θ
NO above 400 K is due to NO dissociation. There is a decrease
in the initial and overall 14NO adsorption capacity above 450
K and no 14NO adsorption could be directly detected through
pressure change at 525 K.
As far as product desorption is concerned, two temperature
regimes can be observed in Figure 4: (i) NO adsorption without
formation of any nitrogen containing products below 400 K and
(ii) simultaneous N2 and N2O production at g 425 K. Above

Nitric Oxide Adsorption and Decomposition on Pd
J. Phys. Chem. B, Vol. 109, No. 27, 2005 13287
formation. These observations can be interpreted in terms of
an increasing NO dissociation probability, leading to a decreas-
ing time gap between shutter removal and the onset of N2 and
N2O desorption at high temperatures. The above observations
point toward the fact that N2 and N2O desorption could be
diffusion-controlled. In this case, the NO adsorption and
dissociation steps could be faster than the production of N2 and
N2O, under the above experimental conditions. Between 475
and 500 K, maximum nitrogen production can be clearly seen
after shutter removal, indicating the fast dissociation of 14NO
to a comparatively large extent. The total nitrogen formation
rate goes through a maximum between 475 and 500 K, and at
>500 K, a reduction in the reaction rate is observed. At 525 K,
the nitrogen desorption continues for little longer than a minute.
However a small amount of CO production is also inferred
(through amu 12), possibly due to the interaction between mobile
oxygen in the bulk/subsurface and carbon in bulk, along with
N2. At temperatures higher than 525 K, the CO production
increases significantly and, hence, reliable NO dissociation
experiments could not be performed. It is to be noted here that
the oxygen migration and hence considerable concentration of
oxygen in the subsurface reported recently by Leisenberger et
al.37 is in line with our observation.
Figure 5. Surface oxygen due to NO dissociation on Pd(111) is titrated
with 13CO beam at the same NO adsorption temperatures. Note the
decrease in oxygen coverage g 500 K is attributed to diffusion of
oxygen into subsurface.
A careful analysis of the products desorption traces in Figure
4 leads to the following four points: (a) There is a sudden rise
in the N2 and N2O production rate between 425 and 500 K, but
after a time delay; the decreasing time delay with increasing
temperature indicates a decrease in initial nitrogen buildup. (b)
The slope of the leading edge of the N2 (and N2O) traces
increases steeply with increasing temperature from 425 to 525
K. Also, a partial N2 production through an N2O intermediate
cannot be excluded on the basis of the experiments. (c) A clear
decrease in N2O production at >475 K with no N2O at 525 K
indicates, again a change in overall kinetics and mechanism.
(d) The selectivity for N2 production is 5-10 times higher than
for N2O, depending on the temperature. In this respect, the
presence of coadsorbed oxygen due to NO dissociation can be
expected to be an important factor. However, it is difficult to
quantify this effect, especially due to the onset of subsurface
diffusion and possible surface oxidation around 475 K (see next
section). Also on rhodium surfaces, it has been confirmed that
coadsorption of oxygen and nitrogen leads to N2 desorption at
lower temperatures as compared to N2 desorption from the
N-only containing Rh surfaces.38-40 The effect was attributed
to repulsive interactions between N and O atoms on the surface.
Indeed, oxygen pre-dosage was shown to completely inhibit NO
dissociation,16 indicating that similar effect may be operative
on Pd(111). Another interesting aspect is the dynamic NO
adsorption-desorption behavior observed on the Pd(111) surface
in the steady-state. Although N2 and N2O desorption stops in
the transient state, NO adsorption-desorption continues, as long
as the NO beam is maintained for the reaction temperatures
below 500 K. Remarkably, the extent of NO adsorption in the
steady-state after shutter removal at t ) 150 s is similar to that
of the NO uptake on clean Pd(111) surface at the beginning
(see Figure 2). Similar observations were made down to 350 K
(data not shown). This directly suggests that the oxygen and
nitrogen coverage due to NO dissociation does not affect the
NO adsorption in a significant manner, and hence, it might not
play any important role in the overall kinetics of NO dissocia-
tion, at least below 500 K. This is in sharp contrast to the NO
dissociation behavior on supported Pd particles studied re-
cently.28,29 The complete loss of the NO adsorption capacity in
the latter case at temperatures at >375 K supports the hypothesis
Figure 6. Representative results from the temperature-programmed
desorption at a heating rate of 5 K/s carried out after the experiments
described in Figures 3 and 4. NO dosing temperatures are given.
400 K, the formation of N2 and N2O production becomes evident
from the results shown in Figure 4. The reference experiments
mentioned above, without removing the shutter, showed no
considerable products formation is attributed to the low NO
coverage under these conditions. It is interesting to compare
the different kinetics of N2 desorption and NO uptake. Indeed,
a remarkable feature is the temporal behavior of the N2 and
N2O production between 425 and 500 K with a time delay
observed for the onset of N2 and N2O production, although NO
uptake is observed at t ) 10 s. This delay indicates the buildup
of some nitrogen coverage. A finite time delay of about 6-7 s
before the onset of N2 and N2O production at 425 K decreases
with increasing temperature (solid arrows in Figure 4). At 525
K, N2 formation appears almost simultaneously with NO uptake
upon shutter removal. Moreover, there is hardly any N2O

13288 J. Phys. Chem. B, Vol. 109, No. 27, 2005
Thirunavukkarasu et al.
that NO dissociation on the particles is strongly dominated by
defect sites.
At g500 K on Pd(111), the steady-state NO coverage in the
dynamic adsorption-desorption equilibrium decreases drasti-
cally, indicating fast desorption. Furthermore, it is likely that
the substantial drop in N2O production above 500 K is related
to the rapidly decreasing NO coverage in this temperature region
(see section 3.6).
3.3. Surface Oxygen Coverage and Oxygen Diffusion into
Subsurface/Bulk. Figure 5 shows the results on the surface
oxygen titration with 13CO beam after the NO adsorption
experiments, as explained in Figure 2. CO dosing was carried
out at the same reaction temperature as that of NO adsorption.
13CO2 production was observed at temperatures between 400
and 475 K. However, CO2 desorption decreases sharply at 500
K and no CO2 was observed at 525 K, despite nitrogen
desorption observed in the transient state, shown in Figure 4.
This is attributed mainly to subsurface diffusion of oxygen and
surface oxidation of Pd(111). As noted earlier, there is a small
amount of 12CO production along with N2 at 525 K in the NO
adsorption experiments. Further, there is some carbon oxide
formation while flashing to 1000 K, after reaction. Above
observations supports the hypothesis of oxygen diffusion into
the subsurface and surface oxide formation. Desorption of no
or very small amount of molecular oxygen above 700 K in
TPD20,21 due to NO dissociation on a variety of Pd surfaces,
including Pd(311) where NO dissociation is largest, unambigu-
ously hints the disappearance of oxygen from the surface and
it has been attributed to oxygen diffusion and supports our above
conclusion. Leisenberger et al.37 observed oxygen diffusion into
the near surface region at 523 K and in good agreement with
our results.
Figure 7. Temperature dependence of ΘNO coverage and sticking
coefficient calculated from the NO uptake data in Figure 3. Note the
same initial sticking coefficient is maintained between 300 and 425 K
up to about 80% saturation coverage for adsorption temperatures at
300 and 325 K
experiments indicate NO uptake on the surface at adsorption
temperature <500 K. It may be recalled here that the NO uptake
on clean Pd(111) at the beginning of the experiment as well as
in the steady-state after temporary blocking is almost the same
(Figure 2 - note the change in x-axis scale after the break).
This shows that there is complete desorption of NO with no
beam at temperatures >425 K on the time scale of the
experiment.
TPD results display a decrease in nitrogen desorption,
suggesting that the atomic nitrogen coverage decreases with
increasing temperature. However, the nitrogen coverage seen
on Pd(111) in the TPD is e0.06 ML and comparable to that of
nitrogen coverage observed after NO + CO reaction on Pd-
(111).35 Nonetheless, the nitrogen coverage is much lower than
that observed on Rh(111)11,46 after NO and/or NO + CO
reactions.
3.5. Sticking Coefficient of NO on Pd(111). Figure 7 shows
the temperature and ΘNO dependence of sNO on Pd(111), by
adopting the method employed in refs 33 and 47-49. Briefly,
sNO is directly proportional to the change in pressure due to
NO uptake on Pd(111) after shutter removal with respect to
the total pressure if there was no adsorption. Furthermore, sNO
(0.5) is inversely proportional to the fraction of the beam
intercepted, which is 0.25 in the present case. Estimated error
margin of sNO values from the uptake measurements is up to
5% and should be considered for any purpose. A time evolution
of sNO was calculated first from the experimental data, given in
Figure 3, and converted in terms of increasing ΘNO. The initial
sNO is constant between 300 and 425 K and decreases at
temperatures higher than 425 K. A maximum of sNO ) 0.5 was
observed at low temperatures between 300 and 350 K on clean
Pd(111) up to 70-80% of the saturation coverage. This behavior
is typical of precursor-mediated adsorption. At high temperature,
the plateau becomes smaller due to faster desorption and
phenomenologically a description of sNO on the basis of a
Langmuir model becomes more appropriate. Very similar trend
Recently, surface oxide formation on Pd(111) has been
investigated experimentally in great detail.41-45 Also, the
oxidation of Pd surfaces has been investigated theoretically44
and shows that subsurface oxygen is just a precursor to surface
oxidation. Decreasing the CO2 signal at 500 K in our results
may also be due to a reduced reaction probability of CO and
the inability to strongly adsorb CO due to the oxidized palladium
surface. It is to be noted that the more oxidized Pd surface and
PdO are unreactive toward CO, as found by Zhang and
Altmann.42 Indeed the amount of CO2 desorption at 475 K is
somewhat lower than at 450 K, indicating that surface oxidation
may begin around 475 K, in accordance with adsorption studies
on Pd(111) at elevated temperatures.41,42 The complex interac-
tion of the Pd surfaces with oxygen significantly complicates
the kinetic analysis, especially at temperatures above 475 K.
However, sustainable CO2 production in NO + CO reactions
on Pd(111) up to 625 K23,35 suggests that surface oxide
formation and subsurface diffusion may be strongly suppressed
in the presence of CO as a reducing agent.
3.4. Temperature-Programmed Desorption Studies of NO/
Pd(111). Figure 6 shows the TPD of NO and N2 species
recorded after the above dosing experiments. No significant N2O
was observed. NO desorbs between 450 and 550 K, in good
agreement with the earlier TPD results.16-19 Gradual decrease
in the quantity of NO desorbed is observed in the TPD from
the NO dosing experiments conducted above 300 K (not shown).
Between 300 and 325 K, no change in NO coverage was
observed in TPD (as well as NO uptake in dosing) experiments,
suggesting that the γ-saturation coverage16-19 is retained.
However, the NO yield in TPD decreases linearly from 0.33 to
0 ML between 300 and 325 and 450 K dosing temperatures,
respectively. Nonetheless, the temperature-dependent NO dosing

Nitric Oxide Adsorption and Decomposition on Pd
J. Phys. Chem. B, Vol. 109, No. 27, 2005 13289
in sNO on Pd(110) surface was reported by Sharpe and Bowker21
and indicates the lifetime of chemisorbed NO species decreases
linearly with increasing temperature on Pd(111) and Pd(110).
3.6. Reaction Mechanism. The isothermal experimental
results in a molecular beam setup provide some new insights
into the mechanism for the thermal dissociation of NO on Pd-
(111) surfaces. Earlier reports18,25 suggested the main elementary
steps of the reaction in this system are
results of a time gap between NO uptake and nitrogen formation
and the repulsive interactions between the atomic species has
been observed and reported on Pd(110),21 Rh(110),39 and Rh-
(111)11,39,46 surfaces. Furthermore, on the basis of the observa-
tion of a significant NO coverage and nitrogen buildup it can
also not be excluded that the N2 formation, at least in part, might
be through an N2O intermediate also. Indeed, a similar slope in
the leading edges of N2 and N2O desorption kinetics supports
this view; nonetheless, further support is necessary. The small
nitrogen coverage observed in the TPD after reactions above
425 K indicates that the oxygen could exert repulsive interac-
tions with nitrogen, providing an additional driving force for
N2 desorption. Nevertheless, the high rate of subsurface diffusion
of oxygen and surface oxide formation around 525 K may
change the nature of the surface and affect the overall kinetics.
As the nitrogen coverage decreases drastically >500K along
with an increase in NO desorption rate, both nitrogen recom-
bination and adsorption are not expected to contribute substan-
tially to the rate control. However, NO dissociation might play
a significant role toward the overall kinetics at temperatures
higher than 525 K.
Indeed, fast CO2 production below 525 K in NO + CO
reactions on Pd(111) supports the relatively fast NO dissocia-
tion.35 The complete stopping of NO dissociation in the steady-
state is attributed to the sizable oxygen coverage observed
through CO titration and it is in correspondence with the
observation of Schmick et al.16 and Henry et al.25 that coad-
sorbed oxygen eliminates NO dissociation. Also on supported
Pd particles, we have recently observed large oxygen coverages
after NO dissociation, together with small amounts of N
adsorbed at specific sites on the particles.28,29
NO(g) T NO(ads)
(1)
(2)
(3)
(4)
NO(ads) f N(ads) + O(ads)
N(ads) + N(ads) f N2(g)
N(ads) + NO(ads) f N2O(g)
O(ads) >475 K8 O(sub-surface) + O(surface-oxide)
(5)
Although the above mechanism may appear simple, there are
few complications that have not been resolved until now. In
particular, TPD experiments show weakly and strongly adsorbed
nitrogen peaks at 450 K and between 545 and 595 K,
respectively, by Goodman et al.;23 however only a minor N2
peak at 500 K was observed by several other groups.16,18,22 TPD
results of Nakamura et al.20 and Wickham et al.50 show no
evidence of NO dissociation on Pd(111). In contrast, pronounced
N2 desorption was observed at 670 K by Conrad et al.17 Sugai
et al.19 also shows through Auger electron spectroscopy that
surface-bound N remains on the surface as high as 700 K.
Although present isothermal experiments may not be compared
directly to the earlier TPD results due to the different reaction
conditions in a molecular beam setup, certain valid conclusions
may be derived as discussed in the following.
4. Conclusions
Our present results shows that, indeed, there is NO dissocia-
tion activity on Pd(111), which leads to dominant production
of N2 and minor N2O production between 400 and 500 K and
exclusive N2 production above 500 K. Nevertheless, some NO
dissociation activity is possible from the defect sites (step sites)
and cannot be ruled out. A substantial NO coverage on Pd-
(111) is observed between 400 and 500 K in the dynamic
adsorption-desorption equilibrium and it is not strongly affected
by the presence of N and O atomic species. This result is in
general agreement with the observation of comparable coverages
of NO coverage via high-resolution electron energy loss
spectroscopy measurements by Ramsier et al.18 This suggest
that the NO adsorption [eq 1] cannot be the rate-determining
step (RDS), at least below 500 K. Post-mortem TPD experi-
ments shows that there is only one N2 desorption peak with a
peak maximum around 500-530 K and in good agreement with
majority of earlier reports.16,18,22 New observation of nitrogen
desorption show some delay with respect to NO uptake,
indicating that atomic nitrogen recombination to gaseous N2
could have large degree of rate control. The delay in the nitrogen
desorption below 500 K may also be explained in terms of the
necessity of a critical surface N concentration before their
average interatomic distances are short enough, so that they can
interact to form N2 molecule. Nonetheless, repulsions among
the dissociated atomic species lowers desorption barrier as well
as the strength of adsorption and hence it should also be taken
into account. However, as the temperature increases the diffusion
rate of atomic nitrogen also increases and hence a decrease in
the initial delay as well as N-buildup and instant N2 production
can be observed at 500-525 K. In fact, the common leading
edge of the isothermal kinetic runs, below 525 K, is expected
for a two-dimensional diffusion-limited reactions.46 Similar
A simple molecular beam instrument fabricated for gas-phase
surface reaction studies on metal surfaces is described. The
instrument is employed to probe NO adsorption and its
dissociation on Pd(111) surface through isothermal kinetic
experiments between 300 and 525 K. Molecular NO adsorption
was observed at e400 K. The adsorption kinetics indicate that
NO dissociation starts on Pd(111) surfaces at around 400 K.
NO dissociation to N2 and N2O was observed clearly in the
transient state at temperatures g425 K. Major amounts of
nitrogen desorption were found along with a minor amount of
N2O desorption, in agreement with earlier results from TPD.
The decrease in the time delay between nitrogen desorption
onset and the shutter removal at increasing temperature suggests
that nitrogen formation occurs through a diffusion-controlled
mechanism and it could be partly rate limiting, at least below
500 K. The rate of N2 formation does not follow that of NO
uptake. The adsorption-desorption equilibrium of NO is
established in the presence of NO beam, indicating substantial
NO coverages below 500 K. Oxygen diffusion into the
subsurface sites and surface oxide formation starts around 475
K and increases significantly above 525 K. Sticking coefficient
measurements show s ) 0.5 for NO adsorption between 300
and 350 K up to high coverage, indicating the precursor
mediated adsorption. sNO decreases with increasing coverage
and temperature. The present molecular beam experiments on
NO dissociation throw more light than the earlier conventional
TPD experiments.
Acknowledgment. The authors are grateful to Professor H.
-J. Freund, FHI, Berlin, for his keen interest, and support for
the present work. We thank the reviewers for constructive

13290 J. Phys. Chem. B, Vol. 109, No. 27, 2005
Thirunavukkarasu et al.
(20) Nakamura, I.; Fujitani, T.; Hamada, H. Surf. Sci. 2002, 514, 409.
(21) Sharpe, R. G.; Bowker, M. Surf. Sci. 1996, 360, 21.
(22) de Wolf, C. A.; Nieuwenhuys, B. E. Surf. Sci. 2002, 196, 469.
(23) Rainer, D. R.; Vesecky, S. M.; Koranne, M.; Oh, W. S.; Goodman,
D. W. J. Catal. 1997, 167, 234. Ozensoy, E.; Hess, C.; Goodman, D. W.
J. Am. Chem. Soc. 2002, 124, 8524.
(24) Piccolo, L.; Henry, C. R. Appl. Surf. Sci. 2000, 162-163, 670.
(25) Piccolo, L.; Henry, C. R. Surf. Sci. 2000, 452, 198. Prevot, G.;
Henry, C. R. J. Phys. Chem. B. 2002, 106, 12191.
(26) Seldmair, C.; Seshan, K.; Jentys, A.; Lercher, J. A. J. Catal. 2003,
214, 308.
(27) Hirsimaki, M.; Valden, M. J. Chem. Phys. 2001, 114, 2345.
(28) Joha’nek, V.; Schauermann, S.; Laurin, M.; Libuda, J.; Freund, H.-
J. Angew. Chem., Int. Ed. 2003, 42, 3035.
(29) Johanek, V.; Schauermann, S.; Laurin, M.; Gopinath, C. S.; Libuda.
J.; Freund, H.-J. J. Phys. Chem. B 2004, 108, 14244.
(30) D’Evelyn, M. P.; Madix, R. J. Surf. Sci. Rep. 1984, 3, 413.
(31) Asscher, M.; Somorjai, G. A. In Atomic and Molecular Beam
Methods; Scoles, G., Ed.; Oxford University Press: Oxford, U.K., 1988;
Vol. 2, p 489.
suggestions. C.S.G. thanks Dr. S. Sivaram, Director, NCL, for
his constant encouragement. Financial support to this project
by the Volkswagen Foundation (Program of Partnership, “The
Mechanism and Kinetics of NO Reduction Reactions on Noble
Metal Surfaces-From Single Crystal Surfaces to Supported
model Catalysts”) is gratefully acknowledged. C.S.G. thanks
the Alexander von Humboldt (AvH) foundation for partial
financial support during the molecular beam facility fabrication
at NCL, Pune, India, under the instrument donation program to
the former AvH fellows. C.S.G. also thanks the NCL for an
in-house project (MLP 004226). K.T. thanks CSIR, New Delhi,
for senior research fellowship.
References and Notes
(1) Taylor, K. C. Catal. ReV. Sci. Eng. 1993, 3, 457.
(2) Shelef, M.; Graham, G. W. Catal. ReV. Sci. Eng. 1994, 36, 433.
(3) Kreuzer, T.; Lox, S. E.; Lindner, D.; Leyrer, J. Catal. Today 1996,
29, 17.
(32) Rettner, C. T.; Auerbach, D. J.; Tully, J. C.; Kleyn, A. W. J. Phys.
Chem. 1996, 100, 13021.
(33) Liu, J.; Xu, M.; Nordmeyer, T.; Zaera, F. J. Phys. Chem. 1995,
99, 6167.
(34) Libuda, J.; Meusel, I.; Hartmann, J.; Freund, H.-J. ReV. Sci. Instrum.
2000, 71, 4395.
(4) Jobson, E. Top. Catal. 2004, 28, 191.
(5) Tagliaferri, S.; Ko˜ppel, R. A.; Baiker, A. Stud. Surf. Sci. Catal.
1998, 116, 61. van Yperen, R.; Lindner, D.; Mussmann, L.; Lox, E. S.;
Kreuzer, T. Stud. Surf. Sci. Catal. 1998, 116, 51.
(6) Berndt, H.; Schutze, F. W.; Richter, M.; Sowade, T.; Grunert, W.
Appl. Catal., B 2003, 40, 51. Olsson, L.; Persson, H.; Fridell, E.; Skoglundh,
M. J. Phys. Chem. B 2001, 105, 6895.
(35) Thirunavukkarasu, K.; Thirumoorthy, K.; Libuda, J.; Gopinath, C.
S. J. Phys. Chem. B 2005, 109, 13272.
(36) Thirunavukkarasu, K.; Gopinath, C. S. Unpublished work.
(37) Leisenberger, F. P.; Koller, G.; Sock, M.; Surnev, S.; Ramsey, M.
G.; Netzer, F. P.; Klo¨tzer, B.; Hayek, K. Surf. Sci. 2000, 445, 380.
(38) Belton, D. N.; DiMaggio, C. L.; Ng, K. Y. S. J. Catal. 1993, 144,
273 and references therein.
(7) Takahashi, N.; Shinjoh, H.; Iijima, T.; Suzuki, T.; Yamazaki, K.;
Yokota, K.; Suzuki, H.; Miyoshi, N.; Matsumoto, S.; Tanizawa, T.; Tanaka,
T.; Tateishi, S.; Kasahara, K. Catal. Today 1996, 27, 63.
(8) Salasc, S.; Skoglundh, M.; Fridell, E. Appl. Catal., B 2002, 36,
145.
(9) Holles, J. H.; Davis, R. J.; Murray, T. M.; Howe, J. M. J. Catal.
2000, 195, 193.
(39) Zaera. F.; Gopinath, C. S. J. Chem. Phys. 2001, 116, 1128 and
references therein.
(10) Martinez-Arias, A.; Fernandez-Garcia, M.; Hungria, A. B.; Iglesias-
Juez, A.; Duncan, K.; Smith, R.; Anderson, J. A.; Conesa. J. C.; Soria, J.
J. Catal. 2001, 204, 238.
(11) Gopinath. C. S.; Zaera, F. J. Catal. 1999, 186, 387. Zaera. F.;
Gopinath, C. S. J. Chem. Phys. 1999, 111, 8088.
(12) Zhdanov. V. P.; Kasemo, B. Surf. Sci. Rep. 1997, 29, 31.
(13) Peden, C. H. F.; Goodman, D. W.; Blair, D. S.; Berlowitz, P. J.;
Fisher, G. B.; Oh, S. H. J. Phys. Chem. 1988, 92, 1563.
(14) Permana, H.; Ng, K. Y. S.; Peden, C. H. F.; Schmieg, S. J.; Lambert,
D. K.; Belton, D. N. J. Catal. 1996, 164, 194.
(15) Borg, H. J.; Reijerse, J. F. C.-J. M.; van Santen, R. A.; Niemants-
verdriet, J. W. J. Chem. Phys. 1994, 101, 10052.
(16) Schmick, H.-D.; Wassmuth, H.-W. Surf. Sci. 1982, 123, 471.
(17) Conrad, H.; Ertl, G.; Ku¨ppers, J.; Latta, E. E. Surf. Sci. 1977, 65,
235. Bertolo, M.; Jacobi, K.; Nettesheim, S.; Wolf, M.; Hasselbrink, E.
Vacuum 1990, 76. Bertolo, M.; Jacobi, K. Surf. Sci. 1990, 226, 207.
(18) Ramsier, R. D.; Gao, Q.; Neergaard Waltenburg, H.; Lee, K.-W.;
Nooij, O. W.; Leerts, L. Yates, J. T., Jr. Surf. Sci. 1994, 320, 209. Ramsier,
R. D.; Gao, Q.; Neergaard Waltenburg, H.; Yates, Jr., J. T. J. Chem. Phys.
1994, 100, 6837.
(40) Bowker, M.; Guo. Q.; Joyner, R. W. Surf. Sci. 1991, 257, 33.
(41) Klo¨tzer, B.; Hayek, K.; Konvicka, C.; Lundgren, E.; Varga, P. Surf.
Sci. 2001, 482, 237.
(42) Zheng, G.; Altman, E. I. Surf. Sci. 2000, 462, 151. Zheng, G.;
Altman, E. I. Surf. Sci. 2002, 504, 253. Zheng, G.; Altmann, E. I. J. Phys.
Chem. B 2002, 106, 1048.
(43) Bondzie, V. A.; Kleban, P. H.; Dwyer, D. J. Surf. Sci. 2000, 465,
266.
(44) Todorova, M.; Reuter, K.; Scheffler, M. J. Phys. Chem. B 2004,
108, 14477. Lundgren, E.; Gustafson, J.; Mikkelsen, A.; Andersen, J. N.;
Stierle, A.; Dosch, H.; Todorova, M.; Rogal, J.; Reuter, K.; Scheffler, M.
Phys. ReV. Lett. 2004, 92, 046101.
(45) Teschner, D.; Pestryakov, A.; Kleimenov, E.; Ha¨vecker, M.; Bluhm,
H.; Sauer, H.; Knop-Gericke, A.; Schlo¨gl, J. Catal. 2005, 230, 186.
(46) Aryafar, M.; Zaera, F. J. Catal. 1998, 175, 316.
(47) Madey, T. E. Surf. Sci. 1972, 33, 355.
(48) Jiang, L. Q.; Koel, B. E. J. Phys. Chem. 1992, 96, 8694.
(49) Campbell, C. T.; Valone, S. M. J. Vac. Sci. Technol. A 1985, 3,
408.
(19) Sugai, S.; Watanabe, H.; Kioka, T.; Miki, H.; Kawasaki, K. Surf.
Sci. 1991, 259, 109.
(50) Wickham, D. T.; Banse, B. A.; Koel, B. E. Surf. Sci. 1991, 243,
83.