Epitaxial and Strong Support Interactions between Pt and LaFeO3Films Stabilize Pt Dispersion

Article abstract of DOI:10.1021/jacs.0c00138

The ability to stabilize very small Pt crystallites in supported-metal catalysts following harsh treatments is an important industrial problem. Here, we demonstrate that Pt particles can be maintained in the 1-to 2-nm range following multiple oxidation and reduction cycles at 1073 K when the particles are supported on 0.5-nm LaFeO3 films that have been deposited onto MgAl2O4 using atomic layer deposition. Characterization by scanning transmission electron microscopy suggests that when the catalyst is oxidized at 1073 K, the Pt crystallites are oriented with respect to the underlying LaFeO3. X-ray absorption spectroscopy also shows evidence of changes in the Pt environment. CO-oxidation rates for the reduced catalyst remain unchanged after five redox cycles at 1073 K. Epitaxial growth of Pt clusters and the consequent strong metal-support interaction between Pt and LaFeO3 are suggested to be the main reasons for the enhanced catalytic performances.

Full text of DOI:10.1021/jacs.0c00138

Subscriber access provided by Uppsala universitetsbibliotek
Article
Epitaxial and Strong Support Interactions between
Pt and LaFeO3 Films Stabilize Pt Dispersion
Xinyu Mao, Alexandre C. Foucher, Tiziano Montini, Eric A. Stach, Paolo Fornasiero, and Raymond J. Gorte
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.0c00138 • Publication Date (Web): 19 May 2020
Downloaded from pubs.acs.org on May 19, 2020
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 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 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.
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
Journal of the American Chemical Society
1
2
3
4
5
6
Epitaxial and Strong Support Interactions between Pt and LaFeO3 Films Stabilize Pt
Dispersion
7
8
9
Xinyu Maoa, Alexandre C. Foucherb, Tiziano Montinic, Eric A. Stachb, Paolo Fornasieroc, and
Raymond J. Gortea,b*
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
a
Department of Chemical and Biomolecular Engineering, University of Pennsylvania,
Philadelphia, PA 19104, USA.
b Department of Materials Science and Engineering, University of Pennsylvania, Philadelphia, PA
19104, USA.
c Department of Chemical and Pharmaceutical Sciences, University of Trieste, via L. Giorgieri 1,
34127, Trieste, Italy
*Corresponding Author: gorte@seas.upenn.edu
Abstract
The ability to stabilize very small Pt crystallites in supported-metal catalysts following
harsh treatments is an important industrial problem. Here, we demonstrate that Pt particles can be
maintained in the 1- to 2-nm range following multiple oxidation and reduction cycles at 1073 K
when the particles are supported on 0.5-nm LaFeO3 films that have been deposited onto MgAl2O4
using atomic layer deposition (ALD). Characterization by Scanning Transmission Electron
Microscopy (STEM) suggests that, when the catalyst is oxidized at 1073 K, the Pt crystallites are
oriented with respect to the underlying LaFeO3. X-Ray Absorption Spectroscopy (XAS) also
shows evidence for changes in the Pt environment. CO-oxidation rates for the reduced catalyst
remain unchanged after five redox cycles at 1073 K. Epitaxial growth of Pt clusters and the
consequent strong metal-support interaction between Pt and LaFeO3 are suggested to be main
reasons for the enhanced catalytic performances.
KEYWORDS
Pt catalyst; LaFeO3; Epitaxial; CO oxidation; Atomic Layer Deposition (ALD).
1
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 2 of 25
1
2
3
4
1. Introduction
5
6
7
8
Deactivation of supported-metal catalysts due to sintering of the metal particles is an
important problem for high-temperature catalysis applications, such as automotive emissions-
control. One proposed solution to this problem involves the use of mixed-oxide supports with the
perovskite structure because, in specific cases, catalytic metal atoms can reversibly enter and leave
the oxide lattice after high-temperature oxidation or reduction treatments 1-3. Reversible “ingress”
of large metal particles and “egress” of small particles could therefore allow re-dispersion of
deactivated catalysts. However, the concept has not been entirely successful. After high-
temperature treatments, the perovskites that are capable of ex-solving catalytic metals are usually
composed of large crystallites with low surface areas. In at least some cases, a significant fraction
of the metal particles that ex-solve from the perovskite lattice remain embedded in the bulk of the
perovskite crystallites, unavailable for catalyzing reactions 4-5. Large crystallites likely also
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
contribute to the fact that the kinetics of metal ingress are too slow to be practical 6-7
.
To avoid the issues associated with bulk perovskites, we have prepared the perovskite,
LaFeO3, as a thin film on high-surface-area MgAl2O4 by Atomic Layer Deposition (ALD) and
used this as the support for precious metals 8. A 0.5-nm, uniform film of LaFeO3 (density of 6.65
g/cm3) on a 120-m2/g substrate results in a material that is 28.5-wt% perovskite. Thus, we
hypothesize that even relatively high loadings of the perovskite lead to films that are so thin that
egress-ingress of large metal particles is not likely to occur. However, the present work will show
that strong chemical interactions between Pt and the LaFeO3 films are able to dramatically stabilize
small metal particles, even after multiple high-temperature, oxidation-reduction cycles. The
chemical bonding in the oxidized samples is observed directly through the formation of Pt particles
that are epitaxially aligned with the respect to LaFeO3 crystallites. The results suggest that
perovskite-supported metals may provide a solution to the sintering problem in some applications.
2
ACS Paragon Plus Environment

Page 3 of 25
Journal of the American Chemical Society
1
2
3
4
2. Experimental Methods
5
6
7
8
MgAl2O4 was prepared in our laboratory by co-precipitation of Mg and Al nitrate salts,
using (NH4)HCO3 solutions. After calcination to 1173 K, the powders were washed in
concentrated HNO3 and rinsed in distilled water to remove all but the MgAl2O4 phase. After this
treatment, the MgAl2O4 powder had a BET surface area of 120 m2/g and X-Ray Diffraction (XRD)
confirmed that only the MgAl2O4 phase existed. The LaFeO3 films were deposited onto the
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
8
MgAl2O4 support by ALD, using procedures described elsewhere . The ALD precursors were
La(TMHD)3 (TMHD = 2,2,6,6-tetramethyl-3,5-heptanedionato, Strem Chemicals, Inc.), and
ferrocene (Fe(Cp)2, Sigma-Aldrich). For each ALD cycle, the substrate was exposed to the
ferrocene or La(TMHD)3 vapors multiple times at 523 K to ensure surface saturation of the
precursor. The ferrocene was oxidized by exposing the sample to 1 atm of O2 in the ALD system
at 523 K for 300 s; however, the La(TMHD)3 could not be fully oxidized by these conditions. To
oxidize the adsorbed La(TMHD)3, the sample was placed in a muffle furnace at 873 K for 60 s.
The growth rates for the individual oxides were shown, gravimetrically, to be linear, with values
of 3.4×1017 La atoms/cycle·m2 (0.014 nm La2O3/cycle) and 1.0x1018 Fe atoms/cycle·m2 (0.025 nm
Fe2O3/cycle). Films with the correct perovskite stoichiometry were obtained by alternating
between 3 ALD cycles of La2O3 and 1 ALD cycles of Fe2O3. The final loading of LaFeO3 in this
study, 30±1.5 wt%, was confirmed by Inductively Coupled Plasma-Optical Emission
Spectrometry (ICP-OES). The BET surface area of LaFeO3/MgAl2O4 after calcination at 873 K
was 78 m2/g.
To ensure that there was good initial contact between the Pt and the LaFeO3 support, Pt
was added by ALD using one cycle of platinum (II) acetylacetonate (Strem Chemicals, Inc.).
The Pt was deposited onto the support prior to the final calcination step (i.e. The LaFeO3/MgAl2O4
sample had been calcined to only 873 K prior to adding Pt. After adding Pt, the catalyst was
oxidized with air in a muffle furnace at 573 K to remove the ligands. The Pt loading for the samples
was measured by ICP-OES to be 2.9±0.2 wt% (~1.1×1018 Pt/m2), which is comparable to the
9
previously reported growth rate of Pt for this precursor . Because our interest in the Pt-
LaFeO3/MgAl2O4 catalysts was in understanding the stability of these catalysts following high-
temperature oxidation and reduction, samples were exposed to five oxidation and reduction cycles
at 1073 K prior to the characterization shown in this paper. The oxidation cycles were carried out
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 25
1
2
3
4
5
6
in 10% O2-He for 1 h and the reduction cycles used flowing 10% H2-He mixtures for 1 h, and the
total flow rate was maintained at 100 mL/min.
The ALD-prepared samples were characterized initially by XRD using a Rigaku
Smartlab diffractometer equipped with a Cu Kα source (λ = 0.15416 nm); but additional
measurements were subsequently performed on a Philips X’Pert diffractometer equipped with a
Cu Kα source in the range 10 to 100° with a step of 0.02°. Rietveld analysis was performed using
the PowderCell 2.4 program. Crystallite sizes were calculated by applying Scherrer’s equation to
the main reflections of each crystalline phase in the region from 25 to 40°. For CO chemisorption
measurements, samples were reduced in 200 Torr H2 at 473 K before measuring CO uptakes at
room temperature. ICP-OES was performed on a Spectro Genesis spectrometer with a Mod
Lichte nebulizer. Initial Scanning Transmission Electron Microscopy (STEM) and Energy
Dispersive X-ray Spectra (EDS) measurements were performed with a JEOL JEM-F200 STEM,
operated at 200 kV; STEM images with high magnification were taken using a JEOL NEOARM
STEM. Powder specimens were diluted in isopropanol and put on lacey carbon films on copper
grids (Electron Microscopy Sciences).
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
Ex-situ XANES/EXAFS spectra were acquired on the SAMBA beamline at synchrotron
10
SOLEIL (Gif-sur-Yvette, France) at the Pt L3 edge (11563.7 eV) . Spectra were collected by
measuring the Kα fluorescence line with a 36 pixels germanium detector (Canberra), while
transmission data were simultaneously recorded with ionization chambers (IC-SPEC, FMB-
Oxford) for the sample and a Pt foil as standard. Data analysis was performed with the DEMETER
software suite 11.
Simultaneous Temperature Programmed Desorption/Thermogravimetric Analysis
(TPD/TGA) measurements were performed using equipment that is described in detail elsewhere
12. In these measurements, the sample was saturated with the vapor from liquid 2-propanol at room
temperature and then evacuated for 1 h to a base pressure of 10-8 Torr. The TPD-TGA
measurements were performed by ramping the sample temperature at 10 K/min while monitoring
the partial pressures with a residual-gas analyzer.
Differential, steady-state rates were measured over 0.1-g samples in a 1/4-inch, quartz,
tubular flow reactor. Products were detected with an on-line gas chromatograph (SRI8610C)
equipped with a Hayesep Q column and a TCD detector. Most CO-oxidation rates were measured
with the total gas flow rate maintained at 100 mL/min using a He carrier and partial pressures of
4
ACS Paragon Plus Environment

Page 5 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
CO and O2 maintained at 25 Torr and 12.5 Torr, respectively. To study the dependence of CO
oxidation rates on the partial pressure of CO, the total flow rate and flow rate of O2 were kept
constant at 100 mL/min and 1 mL/min, but PCO was varied by changing the flow rates of He and
CO. Before taking the measurements, the Pt catalysts were either oxidized or reduced in a 10%
gas mixture of either O2 or H2 in He at a total flow rate of 100 mL/min. Light-off reaction
measurements for CO oxidation were also measured over 0.1 g samples in a tubular flow reactor
in which the products could be continuously monitored using a residual-gas analyzer (SRS-RGA-
100). The flow rates of He, CO and O2 were the same as that used for differential, steady-state
rates. The temperature ramp rate was 10 K/min.
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. Results
3.1 TPD/STEM/XRD
Assuming the LaFeO3 uniformly covered the 120-m2/g MgAl2O4 support, a simple mass
balance indicates that a loading of 30-wt% corresponds to a film 0.56-nm thick if the LaFeO3 has
its bulk density of 6.65 g/cm3. Evidence for film uniformity came from multiple sources. First,
HAADF-STEM/EDS measurements were performed to determine the morphology of the
Pt/LaFeO3/MgAl2O4 samples. Figures 1a) and 1b) provide images, together with the EDS maps,
for an oxidized and reduced sample after the sample had already undergone five oxidation and
reduction cycles at 1073 K. In both cases, the lower resolution HAADF-STEM images in the upper
left-hand panels are indistinguishable from that of the MgAl2O4 support. More informative are the
EDS maps, which show that La, Fe, and Pt each uniformly cover the surface after either oxidation
or reduction. The EDS intensity variations for all three elements also overlap reasonably well with
intensity variations for Mg and Al from the support. STEM/EDS scans for the oxidized sample
with lower magnification are shown as Figure S1 and further confirm that La and Fe uniformly
cover the support surface.
Additional evidence for film uniformity came from TPD-TGA measurements of adsorbed
2-propanol, performed on the unmodified MgAl2O4 and the oxidized Pt/LaFeO3/MgAl2O4 that had
been treated with 5 redox cycles. These data are shown in Figure S2. MgAl2O4 exhibits Lewis-
acid character and most 2-propanol that is adsorbed at room temperature undergoes dehydration
to propene and water in a sharp reaction feature at 450 K in TPD. The MgAl2O4 feature is
completely absent in the TPD-TGA results for Pt/LaFeO3/MgAl2O4. Most of the adsorbed 2-
5
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 6 of 25
1
2
3
4
5
6
7
8
9
propanol also reacted during TPD on this sample but all of the reaction occurred at temperatures
higher than 450 K. Furthermore, a large fraction of the product formed from 2-propanol was
acetone. If any of the MgAl2O4 had been exposed, a feature at 450 K should have been observed.
Given that the average film thickness was only 0.56 nm, it is unlikely that there could be larger
perovskite crystallites with sufficient LaFeO3 remaining to cover the surface of the MgAl2O4.
While the uniform coverage of LaFeO3 over the MgAl2O4 surface had been established
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
8
previously for ALD preparation , the fact that these films are stable to multiple redox cycles at
1073 K is encouraging. It is also encouraging to observe well-dispersed Pt following either
oxidation or reduction at 1073 K. The higher-resolution images in the right-hand panels of Figures
1a) and 1b) indicate that Pt particles are present on both the oxidized and reduced samples, but
they are irregular in shape and only 1 to 2 nm in size. That the STEM images are representative of
the entire sample was demonstrated by the fact that the Pt content calculated from the EDS signals
varied from 2.7-wt% to 3.1-wt% on different regions with dispersed Pt, in good agreement with
the ICP measurement.
Figure 1a): High angle annular dark field STEM image (i) and EDS maps (ii-vi) of Mg, Al,
Pt, Fe and La on the oxidized Pt/LaFeO3/MgAl2O4 after 5 redox cycles, taken from the region
indicated by the green box. High angle annular dark field STEM image with larger
magnification (vii) on the same sample.
6
ACS Paragon Plus Environment

Page 7 of 25
Journal of the American Chemical Society
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
Figure 1b): High angle annular dark field STEM image (i) and EDS maps (ii-vi) of Mg, Al,
Pt, Fe and La on the reduced Pt/LaFeO3/MgAl2O4 after 5 redox cycles, taken from the region
indicated by the green box. High angle annular dark field STEM image with larger
magnification (vii) on the same sample.
To demonstrate that stabilization of the Pt particles on the Pt/LaFeO3/MgAl2O4 sample was
due to the presence of LaFeO3, HAADF-STEM measurements were also performed on a reduced
3.1-wt% Pt/MgAl2O4 sample that had undergone 5 redox cycles. The Pt on this sample was again
prepared by ALD and had an initial dispersion of 27%. As shown in Figure S3, the redox
treatments caused a dramatic growth of the Pt on this sample, with crystallite sizes ranging from
13
20 to 100 nm. Pt/MgAl2O4 has previously been reported to be relatively resistant to sintering ,
making the high dispersion of Pt/LaFeO3/MgAl2O4 sample even more noteworthy.
A closer inspection of the high-resolution STEM image of the oxidized sample, shown in
Figure 2, suggests that the small Pt particles on the oxidized sample are aligned with the underlying
LaFeO3 lattice. That these are indeed Pt crystallites was confirmed by the EDS mapping with the
signals of Pt M and Pt L emitted X-rays shown in Figure S4. The images show distinct registry of
the Pt atoms on top of the rows of LaFeO3. Other images of this sample, reported in Figure S5,
also show very small Pt clusters that are oriented along the same direction with the LaFeO3 lattice.
Higher-resolution STEM images of the reduced sample are shown in Figures 3 and S6. The range
of particle sizes is similar but slightly larger than that with the oxidized samples. What is different
is that at least some of the Pt crystallites are not oriented with respect to the underlying perovskite
lattice.
7
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 25
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
Figure 2: High angle annular dark field STEM image of the oxidized Pt/LaFeO3/MgAl2O4
after 5 redox cycles at 1073 K.
Figure 3: High angle annular dark field STEM image of the reduced Pt/LaFeO3/MgAl2O4
after 5 redox cycles at 1073 K.
8
ACS Paragon Plus Environment

Page 9 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
XRD measurements were performed to determine the structure of the deposited films, with
results reported in Figure 4. The bottom pattern, Figure 4i), is that of the MgAl2O4 support, while
the patterns for the oxidized (Figure 4ii)) and reduced (Figure 4iii)) samples with 2.9-wt% Pt and
30-wt% LaFeO3 are shown above it. The XRD patterns in this figure were obtained on the same
sample whose STEM images are shown in Figure 1. First, both the oxidized and reduced samples
show intense peaks for the perovskite phase, without major changes due to pretreatment, indicating
that the perovskite phase is reasonably stable. Based on the line widths (See Table 1), the crystallite
size of the perovskite phase is on the order of 10 nm, which appears to contradict the earlier
conclusion that the LaFeO3 exists in the form of a 0.56-nm LaFeO3 film. It is possible that a small
part of LaFeO3 films agglomerated into large crystallites that were not observed in STEM, leaving
sufficient LaFeO3 to cover the rest of the surface; but this would not explain the high intensity of
the perovskite features. Two-dimensional crystallites could exhibit intense diffraction peaks but
only in the directions perpendicular to the larger length scale 14. Therefore, a LaFeO3 film in the
form of two-dimensional crystallites, oriented in random directions, possibly with respect to the
underlying MgAl2O4 grains, would be expected to exhibit a standard XRD powder pattern.
The major difference in the patterns for the oxidized and reduced samples is in the presence
of very intense Pt features for the oxidized sample, which decrease dramatically in height after
reduction. This result is surprising for multiple reasons. First, the intense, narrow Pt feature appears
to be inconsistent with the observation of well-dispersed Pt in the STEM images. Second, high-
temperature reduction would normally increase metal particle size, not decrease particle size; and
this should cause the Pt peaks to sharpen and increase in height. Finally, it is important to recognize
that this sample had been oxidized and reduced five times at 1073 K prior to acquiring the
diffraction patterns; and the transition between the patterns for the oxidized and reduced samples
was completely reversible upon calcination or reduction at 1073 K. (XRD patterns for the
Pt/LaFeO3/MgAl2O4 after the first redox cycle are reported in Figure S7, and do not show any
significant difference from those shown in Figure 4.) Particle growth of Pt is usually irreversible.
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
9
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 10 of 25
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
Figure 4: XRD patterns of: (i) unmodified MgAl2O4; (ii) oxidized Pt/LaFeO3/MgAl2O4 after
5 redox cycles at 1073 K; (iii) reduced Pt/LaFeO3/MgAl2O4 after 5 redox cycles at 1073 K.
Perovskite features are marked with stars, metallic Pt with diamonds.
To better understand the diffraction patterns, Rietveld analysis was performed on a separate
sample batch prepared in the same manner. The diffraction patterns used in that analysis, shown
in the Supplemental Information Figure S8, are again identical to those shown in Figure 4.
Conclusions from the analysis are summarized in Table 1. First, the Rietveld analysis indicates
that the Pt content was approximately 4-wt% for both reduced and oxidized samples and that the
change in height and width of the Pt peaks upon reduction is due to an apparent decrease in Pt
crystallite size from 22 nm to 10 nm. Obviously, the estimate of crystallite size does not agree with
the STEM results in Figure 1 but suggests that the redox treatments induce re-dispersion under
reducing conditions and re-agglomeration under oxidizing condition. The estimates of the Pt
content on both the oxidized and reduced samples seems in reasonable agreement with the actual
10
ACS Paragon Plus Environment

Page 11 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
Pt content of 2.9-wt% even if, considering the complexity of the sample, the XRD pattern cannot
exclude the presence of a significant number of smaller Pt particles. Based on the size of the
calculated unit cell for Pt, 0.38980 nm compared to 0.39231 nm for pure Pt, the amount of reduced
Fe alloyed with Pt, if present, is very limited. There is no evidence for the formation of crystalline
PtO2 or of metallic Fe. The Rietveld analysis also suggests there may be an evolution of oxide
phases during redox cycling. In particular, the analysis indicates that contact between the LaFeO3
and MgAl2O4 may allow reversible solid-state reaction between these two phases. The cell
parameters determined for each oxide phase significantly differ from the theoretical values of the
parent phases, suggesting the presence of doping ions in significant amounts. Based on the ionic
radii of the various cations (Mg2+, Al3+, La3+ and Fe3+) 15, the trends observed in the cell parameters
are compatible with the mutual exchange between Mg2+ and La3+ and between Al3+ and Fe3+.
Notably, the relative amounts of the different phases differ depending on the thermo-chemical
treatment applied to the materials and a new phase (La3FeO6) appears in the reduced sample. For
all the oxide phases, the mean crystallite sizes are between 8 and 12 nm (Table 1). It should be
noted that any MgO or FeOx phases that must form if reaction occurs could not be distinguished,
possibly due to being amorphous. Finally, considering the complexity of the XRD patterns and the
high number of phases identified, a reliable determination of the exact composition of the different
oxide phases is not possible.
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
For comparison, XRD patterns and Rietveld analysis were also performed on the
Pt/MgAl2O4 sample after 5 redox cycles, with results shown in Table S1 and Figure S9. The
analysis indicates that the Pt particles supported on MgAl2O4 have an average size of ~20 nm, in
good agreement with the STEM images. Unlike the case for Pt/LaFeO3/MgAl2O4, no reversible
changes in particle size or composition were observed upon redox cycling over samples of
Pt/MgAl2O4.
11
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 12 of 25
1
2
3
4
5
Table 1: Results from Rietveld analysis of the XRD patterns of the oxidized and reduced Pt/LaFeO3/MgAl2O4, in comparison with the
cell parameters reported for the parent pure phases.
6
7
8
Parent
pure phase
Oxidized
Reduced
9
Phase
(Space
Group)
Cell
parameters
(nm)
Cell
parameters
(nm)
Cell
parameters
(nm)
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
Amount
(wt%)
Crystallite
size (nm)
Amount
(wt%)
Crystallite
size (nm)
Pt
4.4%
a = 0.39172
a = 0.80104
22
12
4.1%
a = 0.38980
a = 0.80517
10
10
a = 0.39231
a = 0.80806
(Fm m)
3
MgAl2O4
64.2%
66.4%
(Fd m)
3
a = 0.56182
b = 0.76394
c = 0.54441
a = 0.55398
b = 0.77324
c = 0.54456
a = 0.56619
b = 0.79447
c = 0.56008
LaFeO3
(Pnma)
16.0%
15.4%
----
8.7
14
6.2%
21.8%
1.5%
12
13
14
LaAlO3
a = 0.53967
c = 1.31445
a = 0.54109
c = 1.31508
a = 0.53655
c = 1.31120
(R c)
3
a = 0.97279
b = 1.21087
c =0.58461
a = 0.94443
b = 1.18767
c = 0.57373
La3FeO6
(Cmc )
21
----
----
12
ACS Paragon Plus Environment

Page 13 of 25
Journal of the American Chemical Society
1
2
3
4
3.2 XANES/EXAFS
5
6
7
8
To gain further insight into the nature of the interactions of Pt nanoparticles with the
LaFeO3, XAS spectra were acquired, with a focus on Pt. The XANES spectrum at the Pt L3 edge
of the reduced Pt/LaFeO3/MgAl2O4 sample, Figure S10a), is similar to that of a Pt foil, indicating
that all of the Pt atoms are in the zero-valent state. Differences between the reduced sample and
the Pt foil can be observed in the EXAFS part of the spectrum. The extracted EXAFS signal, Figure
S11a), resembles that obtained from a Pt foil but shows a lower intensity and a lower frequency.
Attempts to model the EXAFS signal with a single Pt-Pt distance did not provide a reasonable fit
of the experimental data. This was only achieved by introducing a Pt-Fe path at a distance slightly
shorter than Pt-Pt distance. The obtained fit (R-factor = 0.0060) is presented in Figure 5a) while
the contributions of each path to the FT-EXAFS signal are presented in Figure S11b). The
structural parameters extracted from this fitting are shown in Table 2. It is worth of noting the
absence of a relevant feature associated with a Pt-O distance near R = 0.20 nm (Reff = 0.16 nm),
as might be expected for Pt atoms on the surface of metal nanoparticles and interacting with the
support. This fact may suggest the formation of a thin layer of reduced Fe atoms at the interface
between the support and the small Pt nanoparticles. This is also consistent with there being an
important contribution from the Pt-Fe path (at a distance slightly shorted than Pt-Pt in Pt metal,
see Table 2 and Figure S11) and the lack of a direct interaction of Pt nanoparticles with the oxide
support.
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
Table 2: Fitting results from the ex situ EXAFS analysis at the Pt L3 edge for 1073 K oxidized and
reduced samples.
2 x 10-5
Coordination
number (CN)
Distance R
(nm)
Sample
Path
(nm2)
Pt-Fe
Pt-Pt
Pt-O
Pt-Pt
3.3 ± 0.9
4.7 ± 0.6
1.8 ± 0.4
8.3 ± 0.7
0.267 ± 0.001
0.273 ± 0.001
0.198 ± 0.001
0.276 ± 0.001
6 ± 2
4 ± 2
2 ± 2
5 ± 2
Reduced
Pt/LaFeO3/MgAl2O4
Oxidized
Pt/LaFeO3/MgAl2O4
13
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 14 of 25
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
Figure 5: Fit of the experimental k3-weighted FT-EXAFS signals for the (a) reduced and (b)
oxidized Pt/LaFeO3/MgAl2O4 samples after 5 redox cycles at 1073 K.
14
ACS Paragon Plus Environment

Page 15 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
The XANES/EXAFS results for the oxidized Pt/LaFeO3/MgAl2O4 were significantly
different. The white-line intensity in the XANES spectrum, Figure S10b), was intermediate
between that observed for Pt metal and PtO2. Linear Combination Analysis indicated that 69% of
Pt is in the zero-valent state, while 31% was present as oxidized Pt ions. The EXAFS signal at the
Pt L3 edge in the oxidized Pt/LaFeO3/MgAl2O4 sample, Figure S12a), significantly differs from
that of the reduced sample, becoming more similar to (and being dominated by) a contribution
from metallic Pt. Two contributions related to Pt-O and metal Pt-Pt paths can be clearly observed
in the FT-EXAFS signal, Figure S12b). Modeling of the EXAFS signal included these 2 paths,
with the result presented Figure 5b) (R-factor = 0.0070) and the structural parameters reported in
Table 2. The combination of XANES and EXAFS results on the oxidized Pt/LaFeO3/MgAl2O4
indicates that the Pt exists primarily in the reduced state, even if some re-oxidation occurs by
passivation of the surface and/or at the interface with the support. It is noteworthy that the Pt-Fe
path is lost after re-oxidation of the sample, consistent with changes in the Fe-containing phases
observed by XRD analysis.
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
One possible way to understand the EXAFS results is that a layer of oxygen atoms between
Pt and the LaFeO3 support is removed upon reduction so that Pt contacts Fe in the reduced sample.
The oxygen atoms are then inserted back after oxidation, leading to a strong bond between Pt and
LaFeO3. This bonding is could be responsible for the epitaxial registration and for maintaining
high Pt dispersion. It is noteworthy that the calculated coordination number for Pt in the oxidized
sample was 8.3, suggesting a Pt particle size in the same range observed in STEM, 1 to 2 nm
assuming a cuboctahedral morphology 16.
There is an obvious discrepancy between the Pt particle sizes obtained from XRD, from
STEM, and from XAS. This can only be rationalized by considering differences in what the
techniques measure. XRD is only sensitive to that fraction of the sample which has a high degree
of crystallinity, and small particles would not contribute to the pattern. If there is a distribution of
Pt particle sizes, XRD could therefore overestimate the actual mean crystallite size. Furthermore,
preferential orientation of particles could modify the intensity and width of the various reflections
14. EXAFS on the other hand is particularly sensitive to smaller nanoparticles and the coordination
number can change dramatically for even minor changes in size 16. This can result in the crystallite
size being underestimated. While the STEM measurements in this study showed only very small
15
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 16 of 25
1
2
3
4
5
6
particles, it must be acknowledged that the images sample only a fraction of the sample and may
not be representative of the overall composition.
Several different scenarios can be used to explain the results. The first is that the Pt particles
exist in a bimodal distribution. In this picture, the Pt XRD peaks for the oxidized sample are due
mainly to the fraction of larger Pt particles, while the STEM and EXAFS over-represent smaller
particles. This model is compatible with the epitaxial registration observed for the small Pt
nanoparticles but would require reversible sintering of the Pt upon oxidation and spreading of Pt
particles upon reduction. The Pt coordination number of 8.3 observed for oxidized sample would
be an average of that for small and large particles. Indeed, if only 69% of oxidized catalyst has Pt
in the metallic form, the actual CN of Pt in the metal nanoparticles is close to 12, the value for
bulk Pt. The scenario would require 69% of the Pt to be present in the form of large metal particles
and 31% in the form of isolated Pt atoms, neither of which was observed in HAADF-STEM. A
second possible model is that the intense Pt diffraction feature observed in XRD on the oxidized
sample is due to scattering from collections of small Pt clusters, oriented with respect to each other,
on much larger, two-dimensional perovskite crystallites. After reduction, the epitaxy is partially
lost, accounting for the increased width and decreased height of the Pt peaks in the XRD patterns
of the reduced sample. We attempted to introduce preferential orientations for both the metallic Pt
and the LaFeO3- and LaAlO3-related phases in the Rietveld refinement of the XRD patterns;
however, the quality of the fit did not improve.
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
16
ACS Paragon Plus Environment

Page 17 of 25
Journal of the American Chemical Society
1
2
3
4
3.3 CO Oxidation Rates
5
6
7
8
Differential CO-oxidation rates were measured in 25 Torr CO and 12.5 Torr O2 on
Pt/LaFeO3/MgAl2O4 (2.9-wt% Pt, 30-wt% LaFeO3), Pt/FeOx/MgAl2O4 (3.0±0.2-wt% Pt, 10±1-wt%
Fe2O3), Pt/MgAl2O4 (3.1±0.2-wt% Pt) samples. All of the samples were prepared by ALD, with
the Pt loadings achieved by the amount of Pt added in one ALD cycle. To determine the effect of
pretreatment conditions, the initial measurements were performed on each catalyst prior to any
redox treatments. After ALD deposition of Pt, the catalysts were calcined to only 573 K to remove
the ligands on the Pt precursor before measuring the “as-prepared” rates. Rate measurements were
then obtained on each of the catalysts after oxidation at 1073 K, after reduction at 1073 K, and
after five redox cycles. Arrhenius plots for the Pt/LaFeO3/MgAl2O4 are shown in Figure 6, while
the analogous plots for Pt/FeOx/MgAl2O4 and Pt/MgAl2O4 are reported in Figure S13. Light-off
profiles for Pt/LaFeO3/MgAl2O4 are shown in Figure S14 and are consistent with data in Figure 6.
Rates on the as-prepared Pt/LaFeO3/MgAl2O4 were high and it was necessary to obtain the data at
relatively low temperatures in order to achieve differential conditions for the catalyst loading and
flowrates that were used. When not promoted by a reducible oxide 17, CO-oxidation rates on Pt are
dependent only on metal surface area 18. Using these rates from the literature 19, the Pt dispersion
of the as-prepared catalyst is estimated to be 26%, in good agreement with room-temperature CO
chemisorption measurements, as shown in Table 3. When this catalyst was calcined to 1073 K, the
rates decreased a factor of about 25; but the rates returned to their initial value after reduction at
1073 K. Repeated oxidation and reduction cycles at 1073 K caused the catalyst to go between the
active and inactive states.
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
17
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 18 of 25
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
Figure 6: Steady-state, differential reaction rates for CO oxidation with 25 Torr of CO and
12.5 Torr O2 for 0.1-g sample of Pt/LaFeO3/MgAl2O4. Rates measured on the as-prepared
sample are marked in solid squares. Rates measured on the sample that after oxidation in 10%
O2-He at 1073 K for 1 h are marked in circles; rates measured on the sample that after
reduction in 10% H2-He at 1073 K for 1 h are marked in diamonds. Black symbols denote the
first redox cycle, while red symbols denote the fifth cycle.
Table 3: Summary of Pt dispersion (%) measured by CO Chemisorption or estimated based on
dispersion and activity of Pt (111) in the literature for catalysts after 5th reduction and oxidation
pretreatments at 1073 K.
As-prepared
1073 K Oxidized
1073 K Reduced
measured estimated measured estimated measured estimated
Pt/LaFeO3/MgAl2O4
Pt/FeOx/MgAl2O4
23
24
27
26
-
0.2
0.3
0.1
-
30
0.6
0.4
>100
23.5
0.9
0.1
1.1
0.1
20 Pt/MgAl2O4
-: Not able to be measured.
Activation and deactivation of the Pt/LaFeO3/MgAl2O4 cannot be due simply to oxidation
or reduction of the Pt. First, as shown in Figure S15, reduction at 773 K did not restore activity in
a catalyst that had been oxidized at 1073 K and oxidation at 773 K did not deactivate a catalyst
18
ACS Paragon Plus Environment

Page 19 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
that had been reduced at 1073 K. Furthermore, deactivation of the oxidized catalyst cannot be due
to reversible formation of Pt single atoms, as has been reported in other studies 21-22, since that
would be inconsistent with the STEM and XRD data. That the reduced Pt/LaFeO3/MgAl2O4
exhibited superior activity supports the fact that the Pt is likely in the dispersed form, since the CO
oxidation rate is directly proportional to the surface metal dispersion. There was also a major
change in the catalyst after the first high-temperature oxidation cycle in that we were not able to
measure dispersion by CO chemisorption. As shown in Table 3, adsorption of CO at room
temperature was negligible after high-temperature treatments, even though dispersions estimated
from the reaction rates were very high. This suggests that the CO-oxidation reaction on
Pt/LaFeO3/MgAl2O4 is significantly different from that which occurs on normal Pt catalysts. In
agreement with this, reaction orders for CO on the Pt/LaFeO3/MgAl2O4, shown in Figure S16,
were almost zero, whereas the reaction is normally inverse-first-order on conventional Pt 23. Also,
the activation energy for CO oxidation over reduced Pt/LaFeO3/MgAl2O4, 60±5 kJ/mol, was
significantly lower than the activation energies calculated for Pt/MgAl2O4, 100±5 kJ/mol.
The failure to observe CO chemisorption, and the weaker dependence of the reaction on
CO partial pressure over the Pt/LaFeO3/MgAl2O4, suggests that the reaction mechanism is
different. It is likely that the reaction occurs at the interface between Pt and the LaFeO3 support,
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
24
similar to what has been observed on Pt-CeO2 catalysts . In addition to the change in reaction
order, reaction at the metal-oxide interface could also explain the decreased activation energy
compared to the conventional Pt catalyst.
To ensure that the reaction results on Pt/LaFeO3/MgAl2O4 are not due to the ALD
preparation or the presence of FeOx, reaction measurements were also performed on the
Pt/MgAl2O4 20 and Pt/FeOx/MgAl2O4 catalysts, with results shown in Figure S13. The as-prepared
Pt/MgAl2O4 exhibited similar rates to those observed on as-prepared Pt/LaFeO3/MgAl2O4. The
rates again decreased dramatically upon high-temperature calcination but were not restored by
high-temperature reduction. As shown in Table 3, the dispersions estimated from the specific
reaction rates agreed well with dispersions determined by CO chemisorption and the STEM
measurements reported earlier (Figure S3). Rates on the as-prepared Pt/FeOx/MgAl2O4 were very
high. Promotion of CO-oxidation rates on Pd catalyst due to contact with FeOx has been reported
previously 25 and is likely responsible for the high rates reported here. However, high-temperature
redox cycling again resulted in a dramatic decrease in rates that could not be restored. STEM
19
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 20 of 25
1
2
3
4
5
6
images of the Pt/FeOx/MgAl2O4 catalyst, shown in Figure S17, demonstrated a large growth in the
Pt particles and segregation of the FeOx. Finally, a previous report of Pt/La2O3/MgAl2O4 also
showed strong deactivation of the catalyst following redox cycling 20.
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
4. Discussion
Maintaining high metal dispersions in supported-metal catalysts upon high-temperature
redox cycling is extremely challenging and could have important implications for catalysts
operating under harsh conditions, such as those experience by automotive emissions-control
catalysts. The present work demonstrates thin LaFeO3 films may be capable of achieving this
important goal. This could have important commercial consequences by allowing the use of
reduced amounts of precious metals in these applications.
The observation that metal particles are oriented with respect to the perovskite film
indicates that there must be strong bonding interactions between the metal and perovskite. It is
likely that similar bonding interactions are present with bulk perovskites and that these are at least
partially responsible for the ex-solution phenomenon observed in some bulk perovskites. These
bonding interactions are specific to the particular metal and perovskite. For example, it has been
reported that Pt can enter the perovskite lattice for both the CaTiO3 and LaFeO3, while Pd can only
3
enter the LaFeO3 lattice, not that of CaTiO3 . Related to this, recent work also found evidence for
strong support interactions with Pt, not with Pd, on thin CaTiO3 films 26. The perovskite structure
must be important since we did not observe similar stabilities for Pt particles when La2O3 or FeOx
were deposited individually. Finally, the strength of the bonding interactions varies with the
specific metal and perovskite. While there was evidence for strong support interactions for Pt on
thin CaTiO3 films, relatively large Pt particles were still formed upon high-temperature redox
cycling.
The properties of the Pt are clearly affected by interactions with the LaFeO3 film, as
demonstrated by the fact that CO adsorption is suppressed and both the reaction order and
activation energy for CO oxidation are affected. Since electronic interactions can extend a few
atomic distances, this may be related to the fact that a significant fraction of Pt particles are very
small in the present work, so that the Pt atoms at the surface of the particles must be close to the
support interface. However, we cannot rule out the possibility that Pt can go beneath the surface
of the perovskite, especially following oxidation. This could help explain the large difference in
20
ACS Paragon Plus Environment

Page 21 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
activity between the oxidized and reduced samples. The fact that such high temperatures are
required to cycle the perovskite-containing catalysts between “activated” and “deactivated” states
suggests that oxidation and reduction cause a major restructuring of the catalyst, beyond simply
oxidizing and reducing the Pt surface.
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
There is obviously still much to learn about these perovskite-supported metals. The
catalytic properties of these materials are clearly very different from that of conventional
supported-metal catalysts. The fact that they can exist in a deactivated state following high-
temperature oxidation also implies that they can be pretreated and used in an appropriate manner.
However, the fact that one can maintain very high dispersions under very harsh conditions makes
the materials very intriguing and worth additional consideration.
5. Conclusion
In this work, we successfully prepared a high-surface-area LaFeO3 thin film on MgAl2O4
support using ALD. We demonstrated that, following high-temperature reduction, Pt on the
LaFeO3/MgAl2O4 was able to exhibit superior activity for CO oxidation, even following harsh
pretreatments. Pt/LaFeO3/MgAl2O4 was found to switch between “active” and “inactive” states for
CO oxidation after high-temperature reduction or oxidation. Stabilization of Pt particle size results
from strong interactions between the Pt and the support, as evidenced by the appearance of Pt
particles that are epitaxially aligned with respect to the perovskite support.
Acknowledgements
This work was funded by the Department of Energy, Office of Basic Energy Sciences,
Chemical Sciences, Geosciences and Biosciences Division, Grant No. DE-FG02-13ER16380.
A.C.F. would like to acknowledge the Vagelos Institute for Energy Science and Technology at
the University of Pennsylvania for a graduate/postdoctoral fellowship. The authors acknowledge
SOLEIL for provision of synchrotron radiation facilities and thank Dr. E. Fonda for technical
assistance in using beamline SAMBA. T.M. and P.F. acknowledge University of Trieste, MIUR
(through FFABR 2017 project), INSTM Consortium and ICCOM-CNR for financial support.
21
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 22 of 25
1
2
3
4
Supporting Information
5
6
7
8
Figure S1, STEM image and EDS mappings for the oxidized Pt/LaFeO3/MgAl2O4; Figure S2,
TPD/TGA for 2-propanol on MgAl2O4 and Pt/LaFeO3/MgAl2O4; Figure S3, STEM images of
Pt/MgAl2O4; Figure S4, high-resolution EDS mappings with line scan signals of the oxidized
Pt/LaFeO3/MgAl2O4; Figure S5-6, high-resolution STEM images of Pt/LaFeO3/MgAl2O4 samples;
Figure S7, XRD patterns of Pt/LaFeO3/MgAl2O4 after 1st redox cycle; Figure S8, Rietveld
refinements of Pt/LaFeO3/MgAl2O4 samples; Table S1, results from Rietveld analysis of the XRD
patterns of the oxidized and reduced Pt/ MgAl2O4; Figure S9, XRD and Rietveld refinements of
Pt/MgAl2O4 after 5 redox cycles; Figure S10-12, XANES spectra, and extracted and Fourier-
transformed EXAFS spectra of Pt/LaFeO3/MgAl2O4 samples; Figure S13, Arrhenius plots of CO
oxidation rates on samples of Pt/MgAl2O4 and Pt/FeOx/MgAl2O4; Figure S14, light-off profiles
for CO oxidation on Pt/LaFeO3/MgAl2O4; Figure S15, Arrhenius plots of CO oxidation rates on
samples of Pt/LaFeO3/MgAl2O4 after different pretreatments; Figure S16, dependence of CO
oxidation rates on P(CO) over the reduced Pt/LaFeO3/MgAl2O4; Figure S17, STEM and EDS
mappings of the oxidized Pt/FeOx/MgAl2O4.
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
22
ACS Paragon Plus Environment

Page 23 of 25
Journal of the American Chemical Society
1
2
3
4
5
6
Reference
1. Nishihata, Y.; Mizuki, J.; Akao, T.; Tanaka, H.; Uenishi, M.; Kimura, M.; Okamoto, T.;
Hamada, N., Self-regeneration of a Pd-perovskite catalyst for automotive emissions control.
Nature 2002, 418 (6894), 164-167.
7
2.
Tanaka, H.; Tan, I.; Uenishi, M.; Kimura, M.; Dohmae, K., Regeneration of palladium
8
9
subsequent to solid solution and segregation in a perovskite catalyst: an intelligent catalyst. Top.
Catal. 2001, 16 (1-4), 63-70.
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.
Tanaka, H.; Taniguchi, M.; Uenishi, M.; Kajita, N.; Tan, I.; Nishihata, Y.; Mizuki, J.;
Narita, K.; Kimura, M.; Kaneko, K., Self-regenerating Rh- and Pt-based perovskite catalysts for
automotive-emissions control. Angew. Chem., Int. Ed. 2006, 45 (36), 5998-6002.
4.
Pan, X., Reversible precipitation/dissolution of precious-metal clusters in perovskite-based
catalyst materials: Bulk versus surface re-dispersion. J. Catal. 2012, 293, 145-148.
5.
Oh, T.-S.; Jang, H. M., Facet-Dependent in situ Growth of Nanoparticles in Epitaxial Thin Films:
The Role of Interfacial Energy. J. Am. Chem. Soc. 2019, 141 (18), 7509-7517.
6.
CaTiO3 Catalysts during Reduction and Oxidation Treatments by Transmission Electron
Microscopy. ACS. Catal. 2017, 7 (3), 1579-1582.
7.
Self-Regeneration of Pd-LaFeO3 Catalysts: New Insight from Atomic-Resolution Electron
Microscopy. J. Am. Chem. Soc. 2011, 133 (45), 18090-18093.
8.
Fornasiero, P.; Gorte, R. J., Smart Pd Catalyst with Improved Thermal Stability Supported on
High-Surface-Area LaFeO3 Prepared by Atomic Layer Deposition. J. Am. Chem. Soc. 2018, 140
(14), 4841-4848.
Katz, M. B.; Zhang, S.; Duan, Y.; Wang, H.; Fang, M.; Zhang, K.; Li, B.; Graham, G. W.;
Kim, K. J.; Han, H.; Defferriere, T.; Yoon, D.; Na, S.; Kim, S. J.; Dayaghi, A. M.; Son, J.;
Dai, S.; Zhang, S. Y.; Katz, M. B.; Graham, G. W.; Pan, X. Q., In Situ Observation of Rh-
Katz, M. B.; Graham, G. W.; Duan, Y. W.; Liu, H.; Adamo, C.; Schlom, D. G.; Pan, X. Q.,
Onn, T. M.; Monai, M.; Dai, S.; Fonda, E.; Montini, T.; Pan, X. Q.; Graham, G. W.;
9.
Dasgupta, N. P.; Liu, C.; Andrews, S.; Prinz, F. B.; Yang, P., Atomic layer deposition of
platinum catalysts on nanowire surfaces for photoelectrochemical water reduction. J. Am. Chem.
Soc. 2013, 135 (35), 12932-12935.
10.
Villain, F. In SAMBA: The 4–40 keV X-ray absorption spectroscopy beamline at SOLEIL, UVX
2010-10e Colloque sur les Sources Cohérentes et Incohérentes UV, VUV et X; Applications et
Développements Récents, EDP Sciences: 2011; pp 41-47.
11.
absorption spectroscopy using IFEFFIT. Journal of synchrotron radiation 2005, 12 (4), 537-541.
12. Yeh, Y.-H.; Zhu, S.; Staiber, P.; Lobo, R. F.; Gorte, R. J., Zn-promoted H-ZSM-5 for
endothermic reforming of n-hexane at high pressures. Ind. Eng. Chem. Res. 2016, 55 (14), 3930-
3938.
13.
nanoparticles on specific MgAl2O4 spinel facets at high temperatures in oxidizing atmospheres.
Nat. commun. 2013, 4 (1), 1-8.
14.
Nanoscale Materials. ACS Nano 2019, 13(7), 7359-7365
15. Shannon, R. T.; Prewitt, C. T., Effective ionic radii in oxides and fluorides. Acta
Briois, V.; Fonda, E.; Belin, S.; Barthe, L.; La Fontaine, C.; Langlois, F.; Ribbens, M.;
Ravel, B.; Newville, M., ATHENA, ARTEMIS, HEPHAESTUS: data analysis for X-ray
Li, W.-Z.; Kovarik, L.; Mei, D.; Liu, J.; Wang, Y.; Peden, C. H., Stable platinum
Holder, C. F.; Schaak, R. E., Tutorial on Powder X-ray Diffraction for Characterizing
Crystallographica Section B: Structural Crystallography and Crystal Chemistry 1969, 25 (5), 925-
946.
23
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 24 of 25
1
2
3
4
5
6
7
8
9
16.
Vaarkamp, M., Obtaining reliable structural parameters from EXAFS. Catal. Today 1998,
39 (4), 271-279.
17.
Bera, P.; Gayen, A.; Hegde, M. S.; Lalla, N. P.; Spadaro, L.; Frusteri, F.; Arena, F.,
Promoting effect of CeO2 in combustion synthesized Pt/CeO2 catalyst for CO oxidation. J. Phys.
Chem. B 2003, 107 (25), 6122-6130.
18.
Haneda, M.; Watanabe, T.; Kamiuchi, N.; Ozawa, M., Effect of platinum dispersion on the
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
catalytic activity of Pt/Al2O3 for the oxidation of carbon monoxide and propene. Appl. Catal., B
2013, 142, 8-14.
19.
Metal-Surfaces. Surf. Sci. Rep. 1991, 14 (1-2), 1-107.
20. Mao, X.; Foucher, A. C.; Stach, E. A.; Gorte, R. J., “Intelligent” Pt Catalysts Based on
Thin LaCoO3 Films Prepared by Atomic Layer Deposition. Inorganics 2019, 7 (9), 113.
21. Jones, J.; Xiong, H.; DeLaRiva, A. T.; Peterson, E. J.; Pham, H.; Challa, S. R.; Qi, G.; Oh,
Rodriguez, J. A.; Goodman, D. W., High-Pressure Catalytic Reactions over Single-Crystal
S.; Wiebenga, M. H.; Hernández, X. I. P., Thermally stable single-atom platinum-on-ceria
catalysts via atom trapping. Science 2016, 353 (6295), 150-154.
22.
Gänzler, A. M.; Casapu, M.; Vernoux, P.; Loridant, S.; Cadete Santos Aires, F. J.; Epicier,
T.; Betz, B.; Hoyer, R.; Grunwaldt, J. D., Tuning the structure of platinum particles on ceria in situ
for enhancing the catalytic performance of exhaust gas catalysts. Angew. Chem., Int. Ed. 2017, 56
(42), 13078-13082.
23.
Kale, M. J.; Christopher, P., Utilizing quantitative in situ FTIR spectroscopy to identify
well-coordinated Pt atoms as the active site for CO oxidation on Al2O3-supported Pt catalysts. ACS
Catal. 2016, 6 (8), 5599-5609.
24.
Cargnello, M.; Doan-Nguyen, V. V.; Gordon, T. R.; Diaz, R. E.; Stach, E. A.; Gorte, R. J.;
Fornasiero, P.; Murray, C. B., Control of metal nanocrystal size reveals metal-support interface
role for ceria catalysts. Science 2013, 341 (6147), 771-773.
25.
CO Oxidation over Pd Catalysts on ALD-Modified Al2O3. Catal. Lett. 2019, 149 (4), 905-915.
26. Lin, C.; Foucher, A. C.; Ji, Y.; Curran, C. D.; Stach, E. A.; McIntosh, S.; Gorte, R. J.,
Mao, X.; Foucher, A.; Stach, E. A.; Gorte, R. J., A Study of Support Effects for CH4 and
“Intelligent” Pt Catalysts Studied on High-Surface-Area CaTiO3 Films. ACS Catal. 2019, 9 (8),
7318-7327.
For Table of Contents only
24
ACS Paragon Plus Environment

Page 25 of 25
Journal of the American Chemical Society
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
25
ACS Paragon Plus Environment