Chemistry of Supported Palladium Nanoparticles during Methane Oxidation

Article abstract of DOI:10.1021/cs502036d

Time-resolved in situ, energy-dispersive X-ray absorption spectroscopy and mass spectrometry are used to correlate changes in the chemical state of alumina- and ceria-supported palladium nanoparticles with changes in activity and selectivity for methane oxidation. Specifically, modulation excitation spectroscopy experiments are carried out by periodically cycling between net-reducing and net-oxidizing reaction conditions. The XANES and EXAFS data show that the palladium nanoparticles are readily bulk-oxidized when exposed to oxygen, forming a PdO-like phase, and reduced back to a reduced (metal) phase when oxygen is removed from the feed. The difference between the two support materials is most noticeable at the switches between net-oxidizing and net-reducing reaction conditions. Here, a brief reduction in conversion is observed for the alumina-supported catalyst, but for the ceria-supported catalyst, this reduction in conversion is minor or not observed at all. This difference is attributed to differences in the oxidation kinetics and the oxygen storage capability of ceria.

Full text of DOI:10.1021/cs502036d

Subscriber access provided by UNIV TORONTO
Article
Chemistry of supported palladium nanoparticles during methane oxidation
Johan Nilsson, Per-Anders Carlsson, Sheedeh Fouladvand, Natalia M Martin, Johan
Gustafson, Mark A. Newton, Edvin Lundgren, Henrik Grönbeck, and Magnus Skoglundh
ACS Catal., Just Accepted Manuscript • Publication Date (Web): 06 Mar 2015
Downloaded from http://pubs.acs.org on March 7, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
Chemistry of supported palladium nanoparticles
during methane oxidation
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Johan Nilsson,∗,†,‡ Per-Anders Carlsson, Sheedeh Fouladvand, Natalia M.
†,‡
†,‡
†
,‡
¶
§
¶
Martin, Johan Gustafson, Mark A. Newton, Edvin Lundgren, Henrik
k,‡
†,‡
Gr ¨o nbeck, and Magnus Skoglundh
Applied Chemistry, Department of Chemistry and Chemical Engineering, Chalmers
University of Technology, SE-412 96 G ¨o teborg, Sweden, Competence Centre for Catalysis
(
KCK), Chalmers University of Technology, SE-412 96 G ¨o teborg, Sweden, Division of
Synchrotron Radiation Research, Lund University, Box 118, SE-221 00 Lund, Sweden,
ESRF, 71 Rue des Martyrs, 38000 Grenoble, France, and Department of Applied Physics,
Chalmers University of Technology, SE-41296 Goteborg, Sweden
E-mail: johan.nilsson@chalmers.se
Abstract
Time-resolved in situ energy-dispersive X-ray absorption spectroscopy and mass
spectrometry are used to correlate changes in chemical state of alumina and ceria
supported palladium nanoparticles with changes in activity and selectivity for methane
∗
To whom correspondence should be addressed
Applied Chemistry, Department of Chemistry and Chemical Engineering, Chalmers University of Tech-
†
nology, SE-412 96 G ¨o teborg, Sweden
‡Competence Centre for Catalysis (KCK), Chalmers University of Technology, SE-412 96 G ¨o teborg,
Sweden
¶
Division of Synchrotron Radiation Research, Lund University, Box 118, SE-221 00 Lund, Sweden
ESRF, 71 Rue des Martyrs, 38000 Grenoble, France
Department of Applied Physics, Chalmers University of Technology, SE-41296 Goteborg, Sweden
§
k
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
oxidation. Specifically, modulation excitation spectroscopy experiments are carried out
by periodically cycling between net-reducing and net-oxidizing reaction conditions.
The XANES and EXAFS data show that the palladium nanoparticles are readily bulk
oxidized when exposed to oxygen, forming a PdO-like phase, and reduced back to a
reduced (metal) phase when oxygen is removed from the feed. The difference between
the two support materials is most noticeable at the switches between net-oxidizing and
net-reducing reaction conditions. Here, a brief reduction in conversion is observed for
the alumina supported catalyst, but for the ceria this reduction in conversion is minor
or not observed at all. This difference is attributed to differences in the oxidation
kinetics and the oxygen storage capability of ceria.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
Introduction
Natural gas is available in abundant quantities and is widely utilized in energy conversion
processes. Combustion of natural gas, where the principal component is methane (CH ), pro-
4
duces low levels of pollutants such as nitrogen and sulfur oxides, and particulates compared
1,2
to other fossil fuels . Furthermore, less carbon dioxide (CO ) is formed per produced quan-
2
3,4
tity of energy when combusting methane compared to other hydrocarbons . There is also
potential to further reduce net CO emissions by substituting natural gas with methane-rich
2
5
6
biogas produced from biological feedstocks . Methane is, however, a strong greenhouse gas
and emissions from combustion processes must be carefully controlled to avoid detrimental
methane slip. Here we focus on the conversion of methane through catalytic total oxida-
tion, which is an important technique for emission abatement in, for example, automotive
applications.
Methane is the most difficult hydrocarbon to oxidize catalytically and thus relatively
high temperatures are required for the reaction to proceed with an acceptable rate. The
rate-limiting step in the reaction is generally considered to be the abstraction of the first
7
hydrogen in the dissociative adsorption of methane . Palladium has shown high activity for
2
ACS Paragon Plus Environment

Page 3 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
methane oxidation and is commonly used in catalysts. High activity for methane oxidation
8,9
in palladium based catalysts has been attributed to reduced (metallic) palladium , metal
10
11,12
supported surface oxide , or bulk metal oxide
. These conflicting reports can be related
to the fact that the performance is sensitive to several catalyst parameters, such as morphol-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
3–15
ogy, support material, and pretreatment
. This shows that there is a clear need for in
situ characterization to better understand the behavior of palladium catalysts for methane
13,16
oxidation. Recent in situ studies have reported
that methane dissociates readily over
palladium catalysts when palladium is in the oxidized PdO phase, and also when palladium
is in its metallic state, after PdO has decomposed at high temperature. Previously rich-lean
cycling experiments have been performed for methane oxidation over Pd nanoparticles sup-
1
7
ported on alumina (Al O ) and ceria (CeO ) . The results from these experiments indicate
2
3
2
that the changes in methane conversion observed when dynamically changing inlet gas con-
ditions are strongly connected to oxidation and reduction of the active palladium phase. The
ceria support was found to promote the activity during rich-lean cycling and understanding
this behavior is relevant to, for example, the use of ceria as an oxygen storage material
18
in automotive catalysis . In the present study we aim to extend our knowledge of palla-
dium catalysts for methane oxidation during transient conditions by performing rich-lean
cycling experiments combined with in situ energy-dispersive X-ray absorption spectroscopy
(
ED-XAS) to monitor the state of the active Pd phase during the catalytic reaction.
X-ray absorption spectroscopy, also referred to as X-ray absorption fine structure (XAFS),
is a powerful technique for investigating the structure and chemical properties of condensed
matter. It is used to study many different types of materials, used in, for example, chemical
19
sensors, semiconductors, and catalysts . Information on the local geometry surrounding the
absorbing atom, including bond lengths, can be obtained by analyzing the extended X-ray
absorption fine structure (EXAFS) region in the spectra. The X-ray absorption near edge
structure (XANES) close to the absorption edge provides information about the electronic
structure of the absorbing atom, which can be used to monitor the chemical state of the
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
catalytically active phase.
2
0
Modulation excitation spectroscopy (MES) experiments are performed in the present
study by periodically changing the gas composition between net-reducing and net-oxidizing
conditions. This generates spectroscopic data suitable to be analyzed by the phase sensitive
detection (PSD) technique. PSD is a powerful technique through which a high degree of
differentiation between active species and so-called spectator species can be obtained. Species
that are involved in the catalytic reaction are expected to change reversibly and periodically
with time. The time-resolved spectra collected during the experiment are averaged over the
number of pulse sequences. This new set of time-resolved spectra has the duration τ which is
defined as the time required to complete one pulse sequence. The key feature of this method
is the use of the phase sensitive detection analysis described by the equation
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Z
τ
P SD
2
φ
PSD
A(e, t) sin(kωt + φk
Ak
(e) =
)dt
(1)
τ
0
P SD
φ
Ak
(e) is referred to as the phase-resolved or demodulated spectra associated with
the frequency kω (k = 1 corresponding to the fundamental frequency, i.e., to the frequency
of excitation). A(e, t) is the response of the system to the applied excitation at energy
PSD
e and at time t, and φk
is the phase angle selected for demodulation. Treating a set
of time-resolved spectra using eq 1 produces a new set of spectra called phase-resolved or
demodulated spectra at different demodulation phase angles. These demodulated spectra
2
1
have three advantageous properties . First, they contain only the contribution from species
that respond to the external stimulation with the same frequency, which gives a significant
increase in sensitivity to active species that respond to the stimulation compared to static
signals produced by spectator species. Second, the signal-to-noise ratio is vastly improved as
noise has a different response frequency and will be suppressed in the demodulated spectra.
20,21
Third, in principle, signals with different kinetics can be more clearly differentiated
.
4
ACS Paragon Plus Environment

Page 5 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
In the present study, rich-lean cycling experiments have been performed for methane
oxidation over Pd/Al O and Pd/CeO catalysts with in situ ED-XAS measurements. The
2
3
2
outlet concentrations from the sample cell were monitored with mass spectrometry. The
XAS spectra collected were analyzed using phase sensitive detection and EXAFS analysis to
monitor the chemical state of palladium during the catalytic reaction.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
Experimental Methods
The synthesis of the alumina and ceria supported Pd catalysts has been described else-
1
7
where . In brief 250 mg of either γ-alumina (Puralox SBa 200, Sasol) or ceria (99.5 H.S.A.
14, R ˆo ne-Poulenc) was added to 114 mg aqueous solution of tetraamminepalladium(II) ni-
trate (4.6 wt% (NH ) Pd(NO ) , Johnson Matthey) as the palladium precursor to yield a
5
3
4
3 2
Pd loading of 2%. The pH was adjusted to 11 by addition of diluted ammonia solution.
The resulting paste was continuously mixed for 15 min and then instantly frozen with liquid
nitrogen and freeze-dried. The freeze-dried powder sample was calcined in air for one hour
◦
◦
◦
at 550 C, with a heating rate of 5 C per minute from room temperature to 550 C.
Synchronous energy-dispersive in situ XAS and mass spectrometry measurements were
performed at beamline ID24 at the European Synchrotron Radiation Facility in Grenoble,
22
France. A modified Spectratech reaction cell was used to which gas was admitted using air-
actuated high-speed 4-way gas valves (Valco, VICI) to maintain constant flow when switching
between gas compositions. For Pd/Al O a sample cup with diameter 5 mm and depth
2
3
2.5 mm was used and it was loaded with 43 mg of catalyst powder. For Pd/CeO a smaller
2
sample cup with diameter 2.5 mm and depth 2.5 mm was used due to the higher attenuation
of X-rays in ceria compared to alumina. The sample cup was loaded with 15 mg catalyst
powder. The temperature in the reaction cell was controlled by a thermocouple inserted
in the top of the catalyst bed. ED-XAS was measured using an Si[311] polychromator in
23
Bragg configuration and a FReLoN detector to monitor the Pd K-edge at 24 350 eV. The
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
−
4
energy resolution can be estimated from ∆E/E which is about 1 × 10 and gives an energy
resolution of 2.5 eV. The X-ray spot size on the sample was about 150 µm (FWHM) in the
horizontal direction and in the vertical direction the beam was defocused from a standard
1
00 µm to about 300 µm. This method has been found to give good results for measurements
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
4
of spatially nonuniform samples at ID24 . The sampling time for ED-XAS data was 190 ms
for Pd/Al O and 1040 ms for Pd/CeO . A palladium foil (Goodfellow, 99.99% purity)
2
3
2
was used for energy calibration of the obtained ED-XAS spectra and reference spectra were
also recorded for PdO (Alfa Aeasar, 99.9% purity). The outlet stream from the reactor was
analyzed by mass spectrometry (Pfeiffer, Prisma) following the m/z 2 (H ), 4 (He), 15 (CH ),
2
4
18 (H O), 28 (CO), 32 (O ), and 44 (CO ). However, due to significant smearing it was not
2
2
2
possible to compare the response of the water signal with the corresponding responses of
the other reactants and reaction products. Energy calibration and normalization of the
25
XAS spectra were performed with the PrestoPronto software . For the EXAFS analysis all
processing steps, including energy calibration and normalization, were performed with the
26
27
Larch software using scattering paths calculated with FEFF6L .
◦
Oxygen pulse-response experiments were performed at 400, 350, and 300 C for Pd/Al O3
2
◦
and at 350, 300, and 250 C for Pd/CeO . Prior to the experiment, the sample was treated
2
◦
with 1.5% O and then exposed to 0.8% H at 400 C for 20 min. The temperature was
2
2
lowered to the temperature set for the experiment and the pulse-response experiment was
started by switching to a lean gas composition of 1.5% O /0.1% CH which was held for
2
4
5
min. The introduction of the first lean period triggered the recording of XAS spectra.
The feed gas was then switched to a rich gas composition of 0.1% CH which was held for
4
5
minutes. This 10 minute period with lean and rich feed gas compositions was then repeated
twice to give a total duration of the experiment of 30 min. Helium was used as the carrier gas
−
1
and the total flow was kept constant at 100 mL min for both the Pd/Al O and Pd/CeO
2
3
2
samples. For Pd/CeO a lower flow is needed to achieve the same residence time as for the
2
Pd/Al O sample, due to the smaller sample volume. This was, however, not possible to
2
3
6
ACS Paragon Plus Environment

Page 7 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
achieve without changing the concentration of the reactants and therefore the same total
flow and feed compositions were used in order to analyze the reaction kinetics in the same
−
1
concentration regime for both samples. The different space velocities, GHSV = 120 000 h
−
1
for Pd/Al O and 490 000 h for Pd/CeO , and the different amounts of catalyst used make
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2
3
2
it not possible to directly quantitatively compare the methane outlet concentrations from
Pd/Al O and Pd/CeO in the present experiment.
2
3
2
3
Results and Discussion
◦
28
The stable phase of palladium in air at temperatures lower than 790 C is PdO , whilst at
higher temperatures PdO decomposes to metallic Pd. Due to this, many previous studies
of methane oxidation over palladium catalysts have accessed the metallic Pd state in the
3
high conversion region at high temperature . The focus in the present study is on methane
oxidation at low temperatures. By performing in situ X-ray absorption spectroscopy mea-
surements during oxygen pulse-response experiments where the reaction conditions are pe-
riodically switched from net-reducing to net-oxidizing, we correlate the changes in chemical
state of the supported palladium nanoparticles with changes in activity and selectivity for
◦
methane oxidation. It can be worth to mention that at low temperatures (lower than 450 C)
29,30
water has been reported to inhibit methane oxidation over PdO catalysts
. In the present
study, however, water is only produced as a reaction product, which is expected to have a
minor effect on the methane oxidation kinetics as compared to addition of water in the feed.
◦
Figure 1 shows the evolution of the XAFS spectra for Pd/Al O at 400 C during the
2
3
rich-lean cycling experiment. Prior to the experiment, the catalyst is reduced in H and
2
when the first lean period is introduced the recording of the XAFS spectra starts. The left
panel of Figure 1 shows XAFS spectra recorded at the end of the lean and rich periods,
while the right panel shows the color coded intensities of the XAFS spectra together with
the outlet concentrations of oxygen and methane as functions of time during the experiment.
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
The XAFS spectra in the left panel are clearly different when they are recorded at the end of
a rich period compared to the end of lean period, and the spectra recorded at different rich or
lean periods are very similar to one another. The evolution of spectral features as a function
of time can be seen in the color coded intensity plot in the right panel and these can be
compared to the outlet concentrations plotted below. The outlet concentration of methane
correlates strongly to the spectral change observed when the feed gas is cycled between rich
and lean composition.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
24.2
(b)
(
a)
1
1
8
490
190
90
2
2
2
2
4.3
4.4
4.5
4.6
590
290
t (s)
24.7
O2
1
0
.2
.8
300
200
100
0
0.4
0
CH4
24.2 24.3 24.4 24.5 24.6
0
200
400
600
800
1000 1200 1400 1600
Energy (keV)
Time (s)
Figure 1: XAFS spectra recorded during consecutive 300 s lean (1.5% O and 0.1% CH ) and
2
4
◦
rich (0.1% CH ) periods over Pd/Al O at 400 C. (a) XAFS spectra recorded at the end of
4
2
3
the rich and lean periods. (b) Color coded intensities of XAFS spectra (blue: low intensity
and red: high intensity) and outlet concentrations of oxygen and methane as a function of
time recorded during the lean-rich cycling experiment.
The spectra recorded during the end of the rich and lean periods are similar to the spec-
tra of Pd foil and PdO powder, respectively, which are depicted in Figure 2. To analyze
8
ACS Paragon Plus Environment

Page 9 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
the features of the time-resolved spectra we perform the demodulation described by eq 1 to
produce the demodulated spectra shown in Figure 2. These spectra contain only the signals
that change during the experiment and static signals like the absorption edge and the back-
ground disappear during the demodulation. A similar result can be obtained by subtracting
one XAFS spectrum from another, producing a difference spectrum. The difference spec-
trum between the XAFS spectra for Pd foil and PdO powder is shown in Figure 2 and it has
a second-derivative profile like the one found in the demodulated spectra. The demodulated
spectra are difference spectra, however, the signal-to-noise ratio is much improved due to the
phase sensitive detection technique. We use here the difference spectrum between Pd foil
and PdO as a reference spectrum for comparison with the demodulated spectra, to establish
a connection between the changes occurring in the XAFS spectra and the reference state
of metallic Pd and PdO, respectively. The difference spectrum corresponds to the situation
where palladium undergoes a straightforward oxidation-reduction process. Comparing the
difference spectrum with the demodulated spectra there is a high degree of correspondence
which shows that the changes in the XAFS spectra as a function of time are caused by an
oxidation-reduction process of the palladium nanoparticles. In the present experiments the
XANES region of the spectra changes considerably during the transition from the reduced
state to the oxidized state (Figure 1). Thus, the present analysis is not dependent on the
improved signal-to-noise ratio as achieved by the demodulation procedure. In principle phase
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
20
sensitive detection can also help to differentiate species appearing with different kinetics ,
however, in the present experiment the spectral changes are attributed to interconversion
between two states.
In the demodulated spectra the most intense feature is the peak corresponding to the
white line at 24.37 keV. Thus, we use the white line intensity to follow oxidation and re-
duction of Pd. Figure 3 and 4 show the outlet concentrations of reactants and reaction
products measured with mass spectrometry together with the white line intensity during the
pulse-response experiments for Pd/Al O and Pd/CeO , respectively, at different tempera-
2
3
2
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
0
0
0
.30
.20
.10
.00
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
−0.10
−0.20
24.30
24.35
24.40
24.45
24.50
Energy (keV)
◦
Figure 2: Demodulated spectra obtained from Pd/Al O at 400 C (φ
with scaled spectra of Pd foil (black) and PdO powder (gray). The red spectrum corresponds
PSD
k
◦
= 0–160 ) together
2
3
PSD
◦
to φk
= 0 and the blue spectrum is the scaled difference spectrum Pd -PdO.
foil
tures. The temporal response of the methane outlet concentration is qualitatively similar
for Pd/Al O and Pd/CeO during the first part of the lean period. When the lean pe-
2
3
2
riod is introduced, the outlet concentration of CH decreases rapidly in all measurements
4
◦
◦
except for Pd/Al O at 400 C and Pd/CeO at 250 C. Following the initial decrease the
2
3
2
CH concentration increases again, then drops, and later appears to approach steady state
4
for the duration of the lean period. The initial decrease in methane outlet concentration
when oxygen is introduced is most likely caused by an increase in methane oxidation due to
the increased availability of oxygen. Oxidation of the Pd nanoparticles starts when oxygen
is introduced, as indicated by the rapidly increasing white line intensity. The mechanism
for oxidation of Pd nanoparticles has been studied with high-pressure X-ray photoelectron
3
1
spectroscopy (XPS) where it was found that adsorbed oxygen initially creates a surface
oxide before larger domains of palladium oxide start growing. The growth of a bulk oxide
from the surface oxide has been associated with a strong kinetic hindrance in the oxidation
32
of the Pd(100) surface . For methane oxidation the PdO(101) facet has been identified as
16
highly active when a sufficiently thick film of oxide has been grown on Pd(100) . On the
basis of DFT calculations under-coordinated Pd sites in PdO(101) have been identified as
having a low energy barrier for hydrogen abstraction from CH when an oxygen atom is
4
1
0
ACS Paragon Plus Environment

Page 11 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
3
3
present directly below the adsorbing Pd atom , as is found in a two-layer PdO(101) film.
These observations can be connected to the methane conversion observed in the present
experiments; after the initial increase in methane conversion the conversion decreases which
indicates that the catalyst becomes less active during the initial phase of the oxidation when
the surface is covered by chemisorbed oxygen or a surface oxide. When the oxidation has
proceeded so that a sufficiently thick oxide is formed the activity of the catalyst increases
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
with PdO being the active phase. TEM images of PdO supported on θ-Al O indicate that
2
3
34
oxidation of Pd to PdO results in extensive surface roughening . It is not unlikely in our
experiments that the deep oxidation of the Pd nanoparticles results in volumetric expansion
and surface roughening which creates under-coordinated PdO sites which are highly active
for methane dissociation.
For the ceria supported catalyst a reduction in methane conversion can also be observed
initially after oxygen has been introduced, but it is much less pronounced compared to
the corresponding experiment for Pd/Al O . The introduction of oxygen leads to a rapid
2
3
increase of the white line intensity, corresponding to a shift towards an oxidized state of
palladium according to the phase sensitive detection analysis. For Pd/Al O the increase
2
3
in the white line intensity is linear for the first few seconds after introducing oxygen and
◦
after that the rate of increase declines. The profiles are similar at 400 and 350 C, but at
◦
3
00 C the increase in the white line intensity during the lean period is smaller than at higher
◦
temperatures. For Pd/CeO the increase in the white line intensity at 350 and 300 C is very
2
different compared to Pd/Al O , here the white line intensity increases almost linearly until
2
3
◦
the maximum value is reached. At 250 C the increase in the white line intensity is more
similar to that found for Pd/Al O . Since Pd is oxidized more rapidly when supported
2
3
on ceria compared to alumina the reduction of the conversion observed when oxygen is
introduced is minor on Pd/CeO compared to Pd/Al O . On Pd/CeO a stable oxide can
2
2
3
2
be established quickly so that chemisorbed oxygen on the Pd surface will have less impact
on the conversion. The more readily oxidation of Pd when supported on ceria compared
1
1
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
°
00 C
°
350 C
°
300 C
4
1
1
0
.6
.2
.8
1200
900
600
300
0
O2
0.4
CH4
0
6
4
2
000
000
000
0
CO2
C. B.
H2
1500
1000
500
CO
0
1
0
.05
1
.95
0
.9
0.85
600
800
1000
1200
600
800
1000
1200
600
800
1000
1200
Time (s)
Time (s)
Time (s)
◦
Figure 3: Oxidation of 0.1% CH over Pd/Al O at 400, 350, and 300 C where 1.5% O is
introduced for one period. The top panel shows the outlet concentrations of O and CH ,
the second panel shows the outlet concentration of CO and the carbon balance, defined as
CCH4 + CCO + CCO2 . The third panel shows the outlet concentrations of H and CO, the
4
2
3
2
2
4
2
2
bottommost panel shows the XAFS white line intensity at 24 372 eV.
1
2
ACS Paragon Plus Environment

Page 13 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
°
50 C
°
300 C
°
250 C
3
1
1
0
.6
.2
.8
1200
900
600
300
0
O2
CH4
0.4
0
6
4
2
000
000
000
0
CO2
C. B.
1500
H2
1000
500
CO
0
1
0
.05
1
.95
0
.9
0.85
600
800
1000
1200
600
800
1000
1200
600
800
1000
1200
Time (s)
Time (s)
Time (s)
◦
Figure 4: Oxidation of 0.1% CH over Pd/CeO at 350, 300 and 250 C where 1.5% O is
introduced for one period. The top panel shows the outlet concentrations of O and CH ,
the second panel shows the outlet concentration of CO and the carbon balance, defined as
CCH4 + CCO + CCO2 . The third panel shows the outlet concentrations of H and CO, the
4
2
2
2
4
2
2
bottommost panel shows the XAFS white line intensity at 24 372 eV.
1
3
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
to alumina has also been observed previously in the reformation of PdO upon cooling after
35
35
decomposition at high temperature . In the study by Farrauto et al. it was found that the
hysteresis between heating and cooling for this process is significantly reduced for Pd/CeO2
compared to Pd/Al O . The difference between Pd/Al O and Pd/CeO in the present
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
2
3
2
3
2
study can be explained by the oxygen storage capability of ceria and oxygen transport to
the Pd particles which enhance the oxidation process. The difference in our measurements
◦
between Pd/CeO at 250 C and at higher temperatures indicates that the oxygen transport
2
from the support to Pd becomes slow at lower temperatures. Palladium interacts strongly
with the ceria support and the morphology and oxidation state of Pd nanoparticles can
vary depending on catalyst synthesis and pretreatment methods. High activity for methane
oxidation over Pd/CeO has been linked to cationic Pd species in close contact with the
2
3
6
37
support and species with Pd−O−Ce linkages .
When oxygen is introduced there is a sharp peak in the outlet concentration of CO which
2
likely is the result of the oxidation of carbonaceous species adsorbed on the surface during
the rich period. This process appears to have a longer duration over Pd/CeO compared to
2
Pd/Al O and following this peak the outlet CO concentration holds steady for both cata-
2
3
2
lysts during the remainder of the lean period. Also, the outlet concentration of CO increases
slightly at the start of the lean period. Comparing the outlet oxygen concentration between
Pd/Al O and Pd/CeO when oxygen is introduced there is a clear difference between the
2
3
2
two samples. For the ceria supported catalyst there is a shoulder on the rising edge while
for the alumina supported sample there is a smooth increase in oxygen concentration when
oxygen is introduced. This difference is likely connected to the differences in CO production
2
observed at this time.
At the start of the rich period there is a difference in behavior between the alumina and
ceria supported catalysts. The outlet methane concentration for Pd/Al O has a peak when
2
3
◦
◦
the oxygen supply is switched off at 350 and 300 C. The peak at 300 C is much stronger
◦
◦
than that at 350 C and it has a shoulder on the falling edge, but at 400 C this peak is not
1
4
ACS Paragon Plus Environment

Page 15 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
seen. This increase in concentration is likely caused by the rapidly decreasing concentration
of oxygen in the gas phase and the depletion of oxygen at PdO sites which are active for
methane oxidation during the lean period. However, when the reduction of the catalyst
has started, as indicated by the decreasing white line intensity, the conversion of methane
increases when metallic Pd sites become active for methane dissociation. After some time in
the rich period oxygen becomes the limiting reactant and thus partial oxidation of methane
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
increases as indicated by the increasing concentrations of CO and H , and the decreasing
2
concentration of CO2.
For Pd/CeO the outlet methane concentration decreases rapidly when oxygen is removed
2
◦
for the feed, although at 250 C a small peak can be seen before the concentration declines.
Simultaneously, the concentrations of H and CO increase which is different compared to
2
Pd/Al O where there is a delay between the removal of oxygen from the feed gas and the
2
3
increasing outlet concentrations of H and CO. It is possible that this difference is connected
2
to reduction of the ceria support which leads to increased oxygen mobility. Bozo et al. have
reported that in situ reduction of a Pd/CrZrO catalyst resulted in a strong increase in
2
18
methane conversion , this was attributed to the creation of oxygen vacancies in the support
improving the activity through electron transfer to the metal component and increased lattice
18
oxygen mobility . Haneda et al. reported high activity for a Pd/CeO /Al O catalyst with
2-x
2
3
a nonstochiometric cerium oxide support and this was attributed to a fast diffusion rate of
oxygen in the CeO2-x lattice and reverse spillover of oxygen to Pd. Reverse spillover could be
the cause of the increasing methane conversion for the Pd/CeO catalyst in the beginning of
2
the rich period since the increase in conversion appears not to be the result from a change in
the oxidation state of Pd. The increasing outlet concentrations of H and CO suggest that
2
partial oxidation is dominant at this time.
Based on the EXAFS analysis presented below we suggest that Pd/Al O at 400 and
2
3
◦
50 C during the lean period is oxidized from a metallic state to a bulk oxidized state. At
3
3
◦
00 C, however, both EXAFS and the white line intensity indicate that oxidation is far from
1
5
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
◦
complete. Initially during the rich period at 300 C the white line intensity decreases quite
slowly which indicates a limited reduction of the Pd nanoparticles, possibly only the removal
of oxygen adsorbed to Pd. After some time in the rich period the rate of reduction increases
until Pd is almost completely reduced. The white line intensity decreases more rapidly for
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Pd/Al O than for Pd/CeO when the oxygen supply is switched off which indicates that
2
3
2
the reduction of Pd in Pd/Al O is more rapid. Ceria is known to stabilize metals in an
2
3
4+
3+
oxidized state due to the high redox potential of the Ce /Ce couple (1.61 eV) compared
2
+
0 18
to e.g. Pd /Pd
.
Fouladvand et al. also found a lower conversion at the start of the rich period during
17
rich-lean cycling experiments with these types of catalyst systems . This was ascribed to
palladium in intermediate oxidation states being less active for methane oxidation, however,
the present XANES data indicate high activity when the catalyst is in between the oxidized
and reduced state. In the present study we interpret the lower conversion to be more of an
effect of the surface coverages rather than the overall oxidation state of the Pd nanoparticles.
38
39
Thevenin et al. have argued based on XPS studies, and also Kinnunen et al. based on
NH desorption experiments, that the simultaneous presence of reduced and oxidized Pd
3
is necessary for high activity for methane oxidation. For Pd/Al O the highest conversion
2
3
is observed at the beginning of the rich period during the reduction of oxidized palladium
to metallic palladium. This can be caused by a high activity for methane oxidation over
reduced Pd sites or by high activity over mixed Pd/PdO sites. It is difficult to compare the
conversion in the beginning of the rich period with the conversion in the later parts of the
rich period, when palladium is fully reduced, since oxygen is a limiting reactant.
To gain more information about the local structure surrounding the palladium atoms a
conventional EXAFS analysis was performed for the Pd/Al O catalyst, the measurements
2
3
for Pd/CeO had a much shorter post edge region not suitable for EXAFS analysis. The full
2
XAFS spectra are provided in Figures S1, S2, and S3 in Supporting Information. Figure 5
shows the Fourier transformed EXAFS spectra recorded during the end of the rich and lean
1
6
ACS Paragon Plus Environment

Page 17 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
5
Data
Fit (Pd−Pd)
4
3
2
1
0
0
−5
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4
6
8
−
1
k (Å
)
0
1
2
3
4
5
6
R (Å)
5
Data
Fit (Pd−O)
5
4
3
2
1
0
0
−5
4
6
8
−1
k (Å
)
0
1
2
3
4
5
6
R (Å)
◦
Figure 5: Fourier transformed EXAFS spectra from Pd/Al O at 350 C. 40 spectra were
averaged and analyzed with a k range from 2.7 to 9 A . (top) Spectra recorded at the end
of a rich period (1182–1190 s) fitted to the shortest Pd-Pd scattering path in bulk Pd, fit
range is 1.2 to 3.0 A. (bottom) Spectra recorded at the end of a lean period (882–890 s)
2
3
−
1
˚
˚
˚
fitted to the shortest Pd-O scattering path in bulk PdO, fit range is 1.0 to 2.0 A.
5
Data
Fit
3
0
2
1
0
.5
2
−5
4
6
8
−1
k (Å
)
.5
1
.5
0
0
1
2
3
4
5
6
R (Å)
◦
Figure 6: Fourier transformed EXAFS spectrum from Pd/Al O at 300 C. 40 spectra were
averaged and analyzed with a k range from 2.7 to 9 A . The spectra were recorded at the
2
3
−
1
˚
end of the lean period 882–890 s and fitted to a Pd-O and Pd-Pd scattering path with fit
˚
range 1.0 to 3.0 A.
1
7
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
◦
periods at 350 C. The spectra recorded during the rich period feature at double peak at
˚
1.7 and 2.5 A which is attributed to a single Pd neighbor. This assignment has also been
40
made by Reimann et al. based on the rationalization that: (i) the amplitude ratio of the
two peaks does not change with different k-weights as would be expected for contributions
of lighter backscattering atoms, (ii) the two peaks can be fitted with a good result to a path
of the first Pd shell of bulk Pd, (iii) the double peak feature is found also in the EXAFS
analysis of a pure Pd foil if the same k-range is used for Fourier transformation. The fit
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
˚
using a path for the first Pd shell in bulk Pd yields a bond distance of 2.70(1) A for our data,
˚
which is close to the first shell Pd−Pd distance in bulk palladium at 2.75 A. Comparing the
spectra recorded during the lean period with the spectra recorded during the rich period it
can be seen that the main peak has moved to a much shorter distance. This peak was fitted
˚
to a Pd−O scattering path with a bond distance of 2.02(1) A, which can be compared to
41
2
.02 for the shortest Pd−O distance in bulk PdO . The EXAFS analysis shows that during
the lean period the Pd nanoparticles are in an oxidized state similar to bulk PdO but when
the switch is made to the rich period the PdO phase is reduced to a state which is closer
◦
to metallic Pd. For Pd/Al O at 300 C the oxidation of Pd during the lean periods is not
2
3
as complete as is found at higher temperatures. Figure 6 shows the Fourier transformed
◦
EXAFS spectra at 300 C recorded at the same time as the EXAFS spectra from the lean
◦
period at 350 C shown in Figure 5. Here we clearly see a much stronger contribution from
Pd-Pd scattering compared to the higher temperature which indicates that Pd is not fully
oxidized. The EXAFS is fitted to a Pd-O scattering path and a Pd-Pd scattering path which
˚
gives a Pd-O distance of 2.00(2) and Pd-Pd distance 2.71(2) A. To show how the spectrum
changes between the rich and lean periods the time evolution of the Fourier transformed
EXAFS spectrum is shown in Figure 7. When the switch is made between the rich to lean
periods at 600 s the double peak from Pd-Pd scattering diminishes and simultaneously the
Pd-O scattering peak increases in intensity. When the switch is made back to a rich period
at 900 s the process is reversed. The increase in intensity of the Pd-O scattering peak at
1
8
ACS Paragon Plus Environment

Page 19 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
the rich-lean switch is very similar to the increase in white line intensity seen in Figure 3
showing that both XANES and EXAFS give the same description of the oxidation process.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3
2
1
0
1
000
6
8
00
4
6
00
2
400
t (s)
0
R (Å)
◦
Figure 7: Fourier transformed EXAFS spectra as a function of time for Pd/Al O at 350 C
during consecutive 300 s lean (1.5% O and 0.1% CH ) and rich (0.1% CH ) periods.
2
3
2
4
4
Comparing the present results for methane oxidation over supported palladium catalysts
42,43
with methane oxidation over supported platinum catalysts
it is found that exposure to
net-oxidizing conditions does not have a severe detrimental effect on the methane conversion.
For platinum catalysts periodic cycling has been suggested as a method to reduce the effects
of oxygen poisoning and to increase the average conversion. However, for the Pd/Al O3
2
catalyst in the present study rapid changes in the gas composition was not found to be
beneficial to increase the methane conversion. On the other hand, for the Pd/CeO catalyst
2
the interactions between the active metal phase and the support is found to be beneficial
for maintaining high conversion when the gas phase changes and periodic operation can
potentially have a positive effect on the overall conversion.
4
Conclusions
Using the phase sensitive detection technique we show that the spectral changes observed in
the XAFS spectra during methane oxidation are caused by the oxidation and the reduction
of the palladium nanoparticles in the Pd/Al O and Pd/CeO samples. The phase sensitive
2
3
2
1
9
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
detection analysis adds much clarity and unambiguity when it comes to understanding the
observed spectral changes, and also provides a clear rationale for using the white line intensity
as a measure of the palladium oxidation-reduction process. In combination with the EXAFS
analysis this allows us to obtain a clear picture of the oxidation-reduction process of the
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
active palladium phase. The investigations of rich-lean cycling over Pd/Al O and Pd/CeO
2
2
3
reveal that the major difference between the two support material can be observed at the
switches between net-reducing and net-oxidizing feed gas compositions. When a pulse of
oxygen is introduced to the system a brief decrease of methane conversion is observed,
however, this effect is stronger for Pd/Al O than for Pd/CeO . The decrease in conversion
2
3
2
is attributed to chemisorbed oxygen or surface oxides on the palladium surface and for
Pd/CeO the more rapid oxidation of Pd compared to Pd/Al O reduces the effects of
2
2
3
this surface composition. Also when oxygen is removed from the feed it is possible to
observe a brief decrease in the methane conversion over Pd/Al O . This is not observed for
2
3
Pd/CeO where the highest conversion is found just after oxygen is removed from the feed
2
which is attributed to reverse spillover of oxygen from the support to the active Pd sites.
For Pd/Al O the highest conversion during the pulse response experiments is observed in
2
3
the beginning of the rich periods during the reduction of oxidized palladium to metallic
palladium, which indicates that reduced Pd sites or mixed Pd/PdO sites have high activity
for methane oxidation. Compared to methane oxidation over platinum catalysts no strong
effects of oxygen poisoning are observed for the palladium catalysts in the present study.
However, due to the lower conversion observed when the feed gas composition is changed
periodic cycling can potentially have a negative effect on the conversion over time.
Acknowledgement
Dr. S.J.A. Figueroa is thanked for fruitful discussions concerning the experiments. The
authors thank the European Synchrotron Radiation Facility (ESRF), Grenoble, France, for
providing the beamtime. This work has been performed within the project ”Unravelling cat-
2
0
ACS Paragon Plus Environment

Page 21 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
alytically active sites with X-ray absorption spectroscopy”, Swedish Research Council (621-
˚
2011-5009), the R ¨o ntgen-Angstr o¨ m Collaboration ”Catalysis on the atomic scale”, Swedish
Research Council (349-2011-6491), ”Time-resolved in situ methods for design of catalytic
sites within sustainable chemistry”, Swedish Research Council (349-2013-567), and partly
within the Competence Centre for Catalysis, which is hosted by Chalmers University of
Technology and financially supported by the Swedish Energy Agency and the member com-
panies AB Volvo, ECAPS AB, Haldor Topsøe A/S, Scania CV AB, Volvo Car Corporation
AB and W ¨a rtsil ¨a Finland Oy.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information Available
EXAFS analysis details.
This material is available free of charge via the Internet at
http://pubs.acs.org/.
References
(
1) de Gouw, J. A.; Parrish, D. D.; Frost, G. J.; Trainer, M. Earth’s Future 2014, 2, 75–82.
2) Engerer, H.; Horn, M. Energy Policy 2010, 38, 1017–1029.
(
(3) Ciuparu, D.; Lyubovsky, M. R.; Altman, E.; Pfefferle, L. D.; Datye, A. Catal. Rev.
002, 44, 593–649.
2
(4) Han, J.; Zemlyanov, D. Y.; Ribeiro, F. H. Catal. Today 2006, 117, 506–513.
(5) Weiland, P. Appl. Microbiol. Biotechnol. 2010, 85, 849–860.
(6) Myhre, G.; Shindell, D.; Br ´e on, F.-M.; Collins, W.; Fuglestvedt, J.; Huang, J.; Koch, D.;
Lamarque, J.-F.; Lee, D.; Mendoza, B.; Nakajima, T.; Robock, A.; Stephens, G.; Take-
mura, T.; Zhang, H. In Climate Change 2013: The Physical Science Basis. Contri-
bution of Working Group I to the Fifth Assessment Report of the Intergovernmental
Panel on Climate Change; Stocker, T., Qin, D., Plattner, G.-K., Tignor, M., Allen, S.,
2
1
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Boschung, J., Nauels, A., Xia, Y., Bex, V., Midgley, P., Eds.; Cambridge University
Press: Cambridge, United Kingdom and New York, NY, USA, 2013; pp 659–740.
(
(
(
7) Burch, R.; Hayes, M. J. J. Mol. Catal. A: Chem. 1995, 100, 13–33.
8) Hicks, R. F.; Qi, H.; Young, M. L.; Lee, R. G. J. Catal. 1990, 122, 280–294.
9) Lyubovsky, M.; Pfefferle, L. Catal. Today 1999, 47, 29–44.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
10) Oh, S. H.; Mitchell, P. J.; Siewert, R. M. J. Catal. 1991, 132, 287–301.
(11) Burch, R.; Loader, P. K.; Urbano, F. J. Catal. Today 1996, 27, 243–248.
(12) McCarty, J. G. Catal. Today 1995, 26, 283–293.
(13) Matam, S. K.; Aguirre, M. H.; Weidenkaff, A.; Ferri, D. J. Phys. Chem. C 2010, 114,
9439–9443.
(14) Yoshida, H.; Nakajima, T.; Yazawa, Y.; Hattori, T. Appl. Catal., B 2007, 71, 70–79.
(
(
15) Burch, R.; Urbano, F. J. Appl. Catal., A 1995, 124, 121–138.
16) Hellman, A.; Resta, A.; Martin, N. M.; Gustafson, J.; Trinchero, A.; Carlsson, P.-A.;
Balmes, O.; Felici, R.; van Rijn, R.; Frenken, J. W. M.; Andersen, J. N.; Lundgren, E.;
Gr ¨o nbeck, H. J. Phys. Chem. Lett. 2012, 3, 678–682.
(
17) Fouladvand, S.; Schernich, S.; Libuda, J.; Gr ¨o nbeck, H.; Pingel, T.; Olsson, E.;
Skoglundh, M.; Carlsson, P.-A. Top. Catal. 2013, 56, 410–415.
(
(
18) Bozo, C.; Guilhaume, N.; Herrmann, J.-M. J. Catal. 2001, 203, 393–406.
19) Iwasawa, Y. In X-ray Absorption Fine Structure for Catalysts and Surfaces; Iwa-
sawa, Y., Ed.; World Scientific Publishing: Singapore, 1996; p V.
(
20) Urakawa, A.; B u¨ rgi, T.; Baiker, A. Chem. Eng. Sci. 2008, 63, 4902–4909.
2
2
ACS Paragon Plus Environment

Page 23 of 25
ACS Catalysis
1
2
3
4
5
6
7
8
9
(
(
(
21) Ferri, D.; Newton, M. A.; Nachtegaal, M. Top. Catal. 2011, 54, 1070–1078.
22) Newton, M. A. Top. Catal. 2009, 52, 1410–1424.
23) Labiche, J.-C.; Mathon, O.; Pascarelli, S.; Newton, M. A.; Ferre, G. G.; Curfs, C.;
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Vaughan, G.; Homs, A.; Carreiras, D. F. Rev. Sci. Instrum. 2007, 78, 091301.
(
24) Newton, M. A. J. Synchrotron Radiat. 2007, 14, 372–381.
(
25) Prestipino, C.; Figueroa, S. prestopronto – Software for the analysis of Quick Ex-
afs and Dispersive XAFS data. http://code.google.com/p/prestopronto/ (accessed
September 8, 2014).
(
26) Newville, M. J. Phys.: Conf. Ser. 2013, 430, 012007.
(27) Rehr, J. J.; Mustre de Leon, J.; Zabinsky, S. I.; Albers, R. C. J. Am. Chem. Soc. 1991,
13, 5135–5140.
1
(28) Mallika, C.; Sreedharan, O. M.; Gnanamoorthy, J. B. J. Less-Common Met. 1983, 95,
213–220.
(29) Ciuparu, D.; Katsikis, N.; Pfefferle, L. Appl. Catal., A 2001, 216, 209–215.
(30) Burch, R.; Urbano, F. J.; Loader, P. K. Appl. Catal., A 1995, 123, 173–184.
(
31) Westerstr ¨o m, R.; Messing, M. E.; Blomberg, S.; Hellman, A.; Gr ¨o nbeck, H.;
Gustafson, J.; Martin, N. M.; Balmes, O.; van Rijn, R.; Andersen, J. N.; Deppert, K.;
Bluhm, H.; Liu, Z.; Grass, M. E.; H ¨a vecker, M.; Lundgren, E. Phys. Rev. B: Condens.
Matter Mater. Phys. 2011, 83, 115440.
(32) Lundgren, E.; Gustafson, J.; Mikkelsen, a.; Andersen, J.; Stierle, a.; Dosch, H.; Todor-
ova, M.; Rogal, J.; Reuter, K.; Scheffler, M. Phys. Rev. Lett. 2004, 92, 046101.
2
3
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 25
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
(
33) Martin, N. M.; Van den Bossche, M.; Hellman, A.; Gr ¨o nbeck, H.; Hakanoglu, C.;
Gustafson, J.; Blomberg, S.; Johansson, N.; Liu, Z.; Axnanda, S.; Weaver, J. F.; Lund-
gren, E. ACS Catal. 2014, 4, 3330–3334.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
34) Datye, A. K.; Bravo, J.; Nelson, T. R.; Atanasova, P.; Lyubovsky, M.; Pfefferle, L.
Appl. Catal., A 2000, 198, 179–196.
(
35) Farrauto, R. J.; Lampert, J. K.; Hobson, M. C.; Waterman, E. M. Appl. Catal., B
1
995, 6, 263–270.
(
36) Xiao, L.-h.; Sun, K.-p.; Xu, X.-l.; Li, X.-n. Catal. Commun. 2005, 6, 796–801.
(
37) Colussi, S.; Gayen, A.; Farnesi Camellone, M.; Boaro, M.; Llorca, J.; Fabris, S.;
Trovarelli, A. Angew. Chem., Int. Ed. 2009, 48, 8481–8484.
(
(
(
(
38) Thevenin, P.; Pocoroba, E.; Pettersson, L. J.; Karhu, H.; V ¨a yrynen, I. J.; J a¨ r ˚a s, S. J.
Catal. 2002, 207, 139–149.
39) Kinnunen, N. M.; Hirvi, J. T.; Ven ¨a l ¨a inen, T.; Suvanto, M.; Pakkanen, T. A. Appl.
Catal., A 2011, 397, 54–61.
40) Reimann, S.; St ¨o tzel, J.; Frahm, R.; Kleist, W.; Grunwaldt, J.-D.; Baiker, A. J. Am.
Chem. Soc. 2011, 133, 3921–3930.
41) Waser, J.; Levy, H. A.; Peterson, S. W. Acta Crystallogr. 1953, 6, 661–663.
(42) Becker, E.; Carlsson, P.-A.; Gr ¨o nbeck, H.; Skoglundh, M. J. Catal. 2007, 252, 11–17.
(43) Becker, E.; Carlsson, P.-A.; Kylhammar, L.; Newton, M. A.; Skoglundh, M. J. Phys.
Chem. C 2011, 115, 944–951.
2
4
ACS Paragon Plus Environment

Page 25 of 25 ACS Catalysis
1
2
3
4
5
6
ACS Paragon Plus Environment