MnOx-Promoted PdAg Alloy Nanoparticles for the Additive-Free Dehydrogenation of Formic Acid at Room Temperature

Article abstract of DOI:10.1021/acscatal.5b01121

Formic acid (HCOOH) has a great potential as a safe and a convenient hydrogen carrier for fuel cell applications. However, efficient and CO-free hydrogen production through the decomposition of formic acid at low temperatures (<363 K) in the absence of additives constitutes a major challenge. Herein, we present a new heterogeneous catalyst system composed of bimetallic PdAg alloy and MnO_x nanoparticles supported on amine-grafted silica facilitating the liberation of hydrogen at room temperature through the dehydrogenation of formic acid in the absence of any additives with remarkable activity (330 mol H₂·mol catalyst^-1·h^-1) and selectivity (>99%) at complete conversion (>99%). Moreover this new catalytic system enables facile catalyst recovery and very high stability against agglomeration, leaching, and CO poisoning. Through a comprehensive set of structural and functional characterization experiments, mechanistic origins of the unusually high catalytic activity, selectivity, and stability of this unique catalytic system are elucidated. Current heterogeneous catalytic architecture presents itself as an excellent contender for clean hydrogen production via room-temperature additive-free dehydrogenation of formic acid for on-board hydrogen fuel cell applications.

Full text of DOI:10.1021/acscatal.5b01121

Subscriber access provided by - Access paid by the | UCSB Libraries
Article
MnOx-Promoted PdAg Alloy Nanoparticles for the Additive-
Free Dehydrogenation of Formic Acid at Room Temperature
Ahmet Bulut, Mehmet Yurderi, Yasar Karatas, Zafer Say, Hilal Kivrak,
Murat Kaya, Mehmet GÜLCAN, Emrah Ozensoy, and Mehmet Zahmakiran
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b01121 • Publication Date (Web): 09 Sep 2015
Downloaded from http://pubs.acs.org on September 13, 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 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
MnOxꢀPromoted PdAg Alloy Nanoparticles for the
AdditiveꢀFree Dehydrogenation of Formic Acid at
Room Temperature
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
Ahmet Bulut,† Mehmet Yurderi, † Yasar Karatas, † Zafer Say,‡ Hilal Kivrak,§ Murat Kaya,¥
Mehmet Gulcan, † Emrah Ozensoy, ‡ Mehmet Zahmakiran*,†
† Nanomaterials and Catalysis (NanoMatCat) Research Laboratory, Department of Chemistry,
Yüzüncü Yıl University, 65080, Van, Turkey; ‡ Department of Chemistry, Bilkent University,
06800, Ankara, Turkey; § Department of Chemical Engineering, Yüzüncü Yıl University, 65080,
¥
Van, Turkey; Department of Chemical Engineering and Applied Chemistry, Atılım
University, 06836, Ankara, Turkey
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 39
1
2
3
4
ABSTRACT
5
6
7
8
9
Formic acid (HCOOH) has a great potential as a safe and a convenient hydrogen carrier for
fuel cell applications. However efficient and COꢀfree hydrogen production through the
decomposition of formic acid at low temperatures (< 363 K) in the absence of additives
constitutes a major challenge. Herein, we present a new heterogeneous catalyst system
comprised of bimetallic PdAg alloy and MnOx nanoparticles supported on amineꢀgrafted silica
facilitating the liberation of hydrogen at room temperature through the dehydrogenation of
formic acid in the absence of any additives with remarkable activity (330 mol H2.mol catalystꢀ1.hꢀ
1) and selectivity (> 99 %) at complete conversion (> 99 %). Moreover this new catalytic system
enables facile catalyst recovery and very high stability against agglomeration, leaching and CO
poisoning. Through a comprehensive set of structural and functional characterization
experiments, mechanistic origins of the unusually high catalytic activity, selectivity and stability
of this unique catalytic system are elucidated. Current heterogeneous catalytic architecture
presents itself as an excellent contender for clean hydrogen production via roomꢀtemperature
additiveꢀfree dehydrogenation of formic acid for onꢀboard hydrogen fuel cell applications.
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
Keywords: Formic Acid; Palladium; Silver; Alloy; Manganese; Dehydrogenation.
2
ACS Paragon Plus Environment

Page 3 of 39
ACS Catalysis
1
2
3
4
INTRODUCTION
5
6
7
8
9
Hydrogen (H2) is considered to be a promising energy carrier due to its high energy density
(142 MJ/kg), which is almost three times higher than that of natural gas (55 MJ/kg).1,2
Furthermore, hydrogen is also an environmentally friendly energy vector as the utilization of
hydrogen in proton exchange membrane fuel cells (PEMFC) generates only water as the
chemical product.3 However, controlled storage and release of hydrogen are still among the
critical technological barriers faced by the hydrogen economy.1ꢀ3 In this context, formic acid
(HCOOH, FA), which is one of the major stable and nonꢀtoxic products formed in biomass
processing, has recently attracted significant attention as a potential hydrogen carrier for fuel
cells designed towards portable use.4,5 In the presence of metal catalysts, FA can catalytically be
decomposed via dehydrogenation (1) and/or dehydration (2) pathways.
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
HCOOH → H2 + CO2 ꢁG°298K = ꢀ 35.0 kJ/mol (1)
HCOOH → H2O + CO ꢁG°298K = ꢀ 14.9 kJ/mol (2)
The selective dehydrogenation of FA is vital for the production of ultrapure H2, since toxic
carbon monoxide (CO) produced by the dehydration pathway significantly reduces the activity of
the precious metal catalysts in PEMFC.6 Recently, serious efforts have been focused on the
development of homogeneous catalysts for the selective dehydrogenation of FA.7ꢀ11 Even though
89
10
notable catalytic performances have been reported in some of these studies,12ꢀ14 the significant
13
challenges associated with the catalyst isolation and recovery processes substantially hinder the
practical use of such systems in onꢀboard applications. Along these lines, numerous recent
26
studies focused on the development of practical heterogeneous catalysts15 ꢀ
exhibiting
16171819
202122232425
significant activity under mild conditions with high selectivity and facile catalyst recovery
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 39
1
2
3
4
5
6
capabilities. In spite of these numerous former efforts, the majority of the heterogeneous
catalysts reported in the literature for FA dehydrogenation require elevated temperatures and the
7
8
9
utilization of extra additives (e.g. HCOONa, NR3, LiBF4 etc.,),27ꢀ29 while revealing low turnover
28
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
frequency (TOF) values and limited reusability.16ꢀ18,20ꢀ22 Until now, only a few Pd based
heterogeneous catalysts have been found to provide notable activities in the additiveꢀfree FA
dehydrogenation at low temperatures.19,23ꢀ25 In this regard, the development of highly active,
selective and reusable solid catalysts operating at low temperatures in the additiveꢀfree
dehydrogenation of FA bears an enormous technological importance. Of particular importance,
our recent study24 has showed that Pd and MnOx nanoparticles existing as a physical mixture on
the surface of amineꢀgrafted silica catalyze the additiveꢀfree FA dehydrogenation with 82 %
conversion and at room temperature. Moreover, the kinetic control experiments coupled with
spectroscopic and electrochemical studies have led to the following important insights; (i) MnOx
nanoparticles (NPs) act as COꢀsponges around the catalytically active Pd NPs and improve their
operational performance in the FA dehydrogenation, (ii) the surface grafted amine groups on
SiO2 support enhance the activity of Pd NPs by influencing the FA adsorption/storage process as
well as the nucleation and growth of the Pd and MnOx nanoparticles. These important findings
have fostered us to use MnOx and amineꢀgrafted SiO2 support for the development of new Pdꢀ
based heterogeneous catalytic systems for the additiveꢀfree dehydrogenation of FA.
Along these lines, in this study we present a facile synthetic route for obtaining PdAg alloy
and MnOx NPs supported on aminopropylꢀfunctionalized silica, which will hereafter be referred
to as PdAgꢀMnOx/NꢀSiO2. Furthermore, we show the remarkable catalytic performance of this
novel material architecture in the additiveꢀfree decomposition of FA at room temperature via
pathway (1) with high reusability performance. PdAgꢀMnOx/NꢀSiO2 catalyst was prepared with
4
ACS Paragon Plus Environment

Page 5 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
high reproducibility through a simple impregnation route followed by subsequent sodium
borohydride (NaBH4) reduction in water at room temperature. The structural characterization of
PdAgꢀMnOx/NꢀSiO2 was performed by using a multitude of analytical techniques including
inductively coupled plasmaꢀoptical emission spectroscopy (ICPꢀOES), powder Xꢀray diffraction
(XRD), inꢀsitu Fourier transform infrared spectroscopy (inꢀsitu FTIR), Xꢀray photoelectron
spectroscopy (XPS), diffuse reflectance UVꢀVis (DRꢀUVꢀVis) spectroscopy, transmission
electron microscopy (TEM), high resolution transmission electron microscopy (HRTEM),
scanning transmission electron microscopyꢀenergy dispersive Xꢀray spectroscopy (STEMꢀEDX),
high angle annular dark fieldꢀscanning transmission electron microscopy (HAADFꢀSTEM),
linear sweep voltammetry (LSV), COꢀstripping voltammetry and N2ꢀadsorptionꢀdesorption
techniques. Comprehensive structural characterization efforts presented in the current
contribution indicate the presence of separate PdAg alloy and MnOx NPs, which nucleate on the
surface of the aminopropylꢀfunctionalized silica support surface and interact efficiently in a
synergistic manner. The resulting PdAgꢀMnOx/NꢀSiO2 material can catalyze dehydrogenation of
FA in the absence of additives with the high activity (turnover frequency (TOF) = 330 hꢀ1) even
at room temperature. Moreover, the exceptional durability of PdAgꢀMnOx/NꢀSiO2 against
agglomeration, leaching and CO poisoning renders this catalytic architecture an excellent
contender as a reusable heterogeneous catalyst in the hydrogen production from formic acid for
onꢀboard fuel cell applications.
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
EXPERIMENTAL SECTION
Materials. Palladium(II) nitrate dihydrate (Pd(NO3)2.2H2O) (~ 40 % Pd basis), manganese(II)
nitrate tetrahydrate (Mn(NO3)2.4H2O), silver(I) nitrate (AgNO3), aminopropyltriethoxysilane
(H2N(CH2)3Si(OC2H5)3, APTS), sodium borohydride (NaBH4), ninhydrin (C9H6O4), toluene
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 39
1
2
(C7H8) and sodium hydroxide (NaOH) were purchased from SigmaꢀAldrich®. Formic acid
(CH2O2, > 96 %), and silica gel (230ꢀ400 mesh) were purchased from Merck®. Toluene was
distilled over sodium and stored in a Labsconco nitrogenꢀatmosphere drybox (O2< 1 ppm).
Deionized water was distilled by water purification system (MilliꢀQ Water Purification System).
All glassware and teflonꢀcoated magnetic stirring bars were washed with acetone and copiously
rinsed with distilled water before drying in an oven at 423 K.
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
Characterization. Pd, Ag and Mn contents of the samples were determined by ICPꢀOES
(Leeman, Direct Reading Echelle) after each sample was completely dissolved in a mixture of
HNO3/HCl (1/3 v/v). Powder Xꢀray diffraction (XRD) patterns were recorded with a MAC
Science MXP 3TZ diffractometer using CuꢀKα radiation (wavelength 1.54 Å, 40 kV, 55 mA).
TEM, HRTEM, STEM, and HAADFꢀSTEM samples were prepared by dropwise addition of the
dilute catalyst suspension on a copperꢀcoated carbon TEM grid followed by the evaporation of
the solvent. The conventional TEM measurements were carried out on a JEOL JEMꢀ200CX
transmission electron microscope operating at 120 kV. HRTEM, STEM and HAADFꢀSTEM
analysis were performed using a JEOL JEMꢀ2010F transmission electron microscope operating
at 200 kV. Oxford EDX system and the Inca software were exploited to acquire and process
STEMꢀEDX data. The XPS measurements were employed via a Physical Electronics 5800 XP
spectrometer equipped with a hemispherical analyzer and a monochromatic AlꢀKα XꢀRay source
(1486.6 eV, 15 kV, 350 W, with pass energy of 23.5 eV). Gas phase decomposition products of
formic acid were analyzed by gas chromatography using FIDꢀ2014 and TCDꢀ2014GC analyzers
(Shimadzu). UV−Vis electronic absorption spectra were recorded on a Shimadzu UVꢀ2600
spectrophotometer.
6
ACS Paragon Plus Environment

Page 7 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
Catalyst Preparation. The functionalization of the silica was carried out by adding a desired
amount of APTS to 30 mL of dry toluene containing 500 mg of silica. The resulting slurry was
stirred for 12 h. The white solid was filtered and washed repeatedly with toluene. The white
amineꢀfunctionalized silica (SiO2ꢀNH2) was dried in a vacuum oven (373 K and 10ꢀ1 Torr) and
used for further application. The presence of ꢀNH2 functionalities on the SiO2 support surface
was quantified by the colorimetric ninhydrin method.30 PdAgꢀMnOx/NꢀSiO2 catalyst was
obtained by the conventional impregnation and subsequent reduction steps.31 Typically, 5.0 mL
aqueous solution containing Pd(NO3).2H2O (6.24 mg, 23.4 ꢂmol Pd), AgNO3 (0.80 mg, 4.7
ꢂmol Ag), Mn(NO3)2.4H2O (4.7 mg, 18.7 ꢂmol Mn) and SiO2ꢀNH2 (100 mg, 100 ꢂmol NH2) is
mixed for 3 h. Then, the fresh 1.0 mL aqueous solution of NaBH4 (28 mg, 0.7 mmol) was added
to this mixture and the resulting solution was stirred for half an hour under ambient conditions.
After centrifugation (6000 rpm, 5 min), copious washing with water (3 x 20 mL), filtration, and
drying in oven at 373 K, PdAgꢀMnOx/NꢀSiO2 catalyst was obtained as a dark gray powder.
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
Activity Measurements. The catalytic activity of PdꢀMnOx/SiO2ꢀNH2 in the additiveꢀfree FA
dehydrogenation was determined by volumetric measurement of the rate of hydrogen evolution.
The volume of released gas during the reaction was monitored using a gas burette through water
displacement as described elsewhere.15ꢀ25 Before starting the catalytic activity tests, a jacketed
oneꢀnecked reaction flask (50.0 mL) containing a Teflonꢀcoated stirring bar was placed on a
magnetic stirrer (Heidolph MRꢀ3004) whose temperature was adjusted by circulating water
through its jacket from a constant temperature bath (Lab Companion RWꢀ0525). In a typical
catalytic activity test, PdAgꢀMnOx/NꢀSiO2 catalyst was weighed and transferred into the reaction
flask, and then 9.0 mL H2O was added into the reaction flask followed by rigorous stirring for 15
min. to achieve thermal equilibrium. Next, 0.5 mL aqueous FA solution (0.1 mL FA + 0.4 mL
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 39
1
2
3
4
5
6
H2O) was added into the reaction flask via its septum using a 1.0 mL gastight syringe and the
catalytic reaction was started (t = 0 min) by stirring the mixture at 600 rpm.
7
8
9
Catalytic Selectivity. The selectivity of PdAgꢀMnOx/NꢀSiO2 catalyst in the decomposition of
FA was investigated by GC analysis and NaOHꢀtrap experiments. The gas generated over PdAgꢀ
MnOx/SiO2ꢀNH2 catalyzed dehydrogenation of aqueous FA solution (10.0 mL of 0.3 M) was
collected in a GC analyzing balloon, which was then analyzed in GC by using pure CO, H2 and
CO2 as reference gases. NaOHꢀtrap experiments were performed to determine the molar ratio of
CO2 to H2 in the product mixture generated during the PdAgꢀMnOx/SiO2ꢀNH2 catalyzed
decomposition of aqueous FA solution (10 mL of 0.3 M).15ꢀ25 In these experiments, the trap (10.0
M NaOH solution) was placed between the jacketed reactor and gas burette. The generated gas
during the reaction was passed through the NaOH trap where CO2 was captured (3). Next, the
volume of the gas generated from the dehydrogenation of FA was monitored and compared to
those without the trap experiment. We observed that the volume of the generated gas decreased
by a factor of two in the presence of the NaOH trap. This result is indicative of the complete
adsorption of CO2 in NaOH solution (3) and the presence of equivalent molar amounts of CO2
and H2 (1.0:1.0) in the product mixture of the PdAgꢀMnOx/SiO2ꢀNH2 catalyzed additiveꢀfree FA
dehydrogenation.
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
2NaOH + CO2 → Na2CO3 + H2O (3)
Catalytic Stability. The recyclability of PdAgꢀMnOx/NꢀSiO2 in the additiveꢀfree
dehydrogenation of FA was determined by a series of experiments started with a 10.0 mL
aqueous FA solution (0.1 mL FA + 9.9 mL H2O) at 298 K. Immediately after the achievement of
complete conversion, another equivalent amount of FA was added to the reaction mixture,
8
ACS Paragon Plus Environment

Page 9 of 39
ACS Catalysis
1
2
leading further hydrogen evolution. The same procedure was repeated up to the 5th subsequent
catalytic recycle. In the reusability experiments the catalyst was isolated from the reaction
solution by centrifugation after the first catalytic run and washed with excess water and dried at
373 K. The dried catalyst was weighed and reused in the catalytic dehydrogenation of a 10.0 mL
aqueous FA solution (0.1 mL FA + 9.9 mL H2O) at 298 K. This identical catalyst was isolated
and reused up to 5 consecutive catalytic cycles.
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
Cyclic Voltammetry (CV) Measurements. CV measurements were carried out in a
conventional threeꢀelectrode cell with a Pt wire as the counter electrode and Ag/AgCl as the
reference electrode with a CHI 660E potentiostat. The working electrode was a glassy carbon
disk having a diameter of 3.0 mm held in a Teflon cylindrical housing. Before the CV
measurements, the surface of the glassy carbon electrode was polished with alumina to prepare
the surface of the electrode for the catalyst deposition process. For the electrode preparation,
typically 7ꢀ9 mg of catalyst was dispersed in a 1.0 mL 5% Nafion® medium (Aldrich) to obtain
a catalyst suspension. Then, 6.0 ꢂL of this suspension was dropꢀcast on the surface of the glassy
carbon electrode. Then, the electrode was dried at room temperature to remove the solvent. CV’s
were recorded between ꢀ0.23 V and 1.0 V with a scan rate of 10 mV s−1 in the presence of 0.5 M
H2SO4 and 1.0 M HCOOH. Prior to the CV experiments, the electrolyte was saturated with N2
and the electrode surface was activated via 0.5 M H2SO4.
Linear Sweep Voltammetry (LSV) Measurements. Linear sweep voltammetry (LSV)
technique was employed to investigate the CO adsorption on the PdAg/SiO2ꢀNH2 and PdAgꢀ
MnOx/SiO2ꢀNH2 catalyst surface, which can poison the surface during the decomposition of
formic acid.32,33 Before the LSV measurements, catalysts were exposed to a pretreatment process,
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 39
1
2
3
4
5
6
7
8
9
where the catalyst surfaces were maintained at a constant potential of 0.0 V for 180 s. After the
pretreatment, LSV measurements were carried out to determine the activity of the pretreated
surfaces in the electrocatalytic oxidation of FA. These measurements were conducted within the
bias range of ꢀ0.2 V to 0.9 V in a solution containing 0.5 M H2SO4 and 1.0 M HCOOH using a
10 mV s−1 scan rate.
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
CO-Stripping Voltammetry Measurements. 5 mg catalyst sample was dispersed in 1.0 mL
5% Nafion® solution (Aldrich) to obtain a catalyst suspension. Next, 5.0 ꢂL of this suspension
was dropꢀcast on the surface of the glassy carbon electrode. All electrolyte solutions were
deaerated with highꢀpurity nitrogen for 30 min prior to the measurements. For CO stripping
voltammetry, 0.5 M H2SO4 solution was first bubbled with pure nitrogen for 30 min in order to
remove the dissolved oxygen. CO was then purged into the solution for 20 min to allow
saturation of the electrocatalyst surface with adsorbed CO, while maintaining a constant potential
of 0 .0 V. Excess CO was then purged with nitrogen for 30 min.
In-Situ FTIR Spectroscopic Analyses. Detailed description of the instrumentation used in
the currently presented inꢀsitu FTIR experiments can be found elsewhere.34 Briefly, inꢀsitu FTIR
spectroscopic measurements were performed in transmission mode using a Bruker Tensor 27
FTIR spectrometer which was modified to house a batchꢀtype customꢀmade inꢀsitu spectroscopic
reactor. All catalysts were initially cleaned by annealing under vacuum (< 1x 10ꢀ3 Torr) at 400 K
for 2 h. This process is needed to thermally remove the adsorbed water on the material surfaces
and obtain materials that are relatively more transparent to the IR beam. CO adsorption
(poisoning) experiments were performed by introducing 20.0 Torr of CO (>99.995% purity, Air
Products) for 10 min at 323 K followed by evacuation at ~10ꢀ3 Torr. FA adsorption was
10
ACS Paragon Plus Environment

Page 11 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
performed on CO preꢀadsorbed (i.e poisoned) catalyst surfaces where the FTIR spectra for the
COꢀpoisoned surfaces were used as the corresponding background spectra. In these set of
experiments, after 20.0 Torr of CO adsorption at 323 K for 10 min, 5.0 Torr of FA was dosed on
the poisoned catalysts for 5 minutes. Next, the reactor was evacuated (P < 10ꢀ2 Torr) and FTIR
spectra were acquired. All FTIR spectra in this study were collected at 323 K.
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
RESULTS AND DISCUSSION
Catalyst Preparation and Characterization. PdAgꢀMnOx/NꢀSiO2 catalyst can simply and
reproducibly be prepared by following the procedure comprising of the conventional
impregnation31 of Pd2+, Ag+ and Mn2+ onto amineꢀfunctionalized silica and their borohydride
reduction at room temperature. After centrifugation, copious washing with water, PdAgꢀ
MnOx/NꢀSiO2 catalyst was isolated as gray powder and characterized by multi pronged
techniques. The elemental composition of the asꢀprepared PdAgꢀMnOx/SiO2ꢀNH2 catalyst was
found to be Pd0.44Ag0.19Mn0.37 (i.e. 1.27 wt% Pd, 0.57 wt% Ag, and 0.54 wt% Mn;
corresponding to relative molar amounts of 11.9 µmol Pd, 5.3 µmol Ag, 9.8 µmol Mn and 0.98
mmol NH2/g SiO2) by ICPꢀOES analyses and ninhydrin method.30 The survey XPS spectrum of
PdAgꢀMnOx/SiO2ꢀNH2 catalyst (Figure S1) shows the existence of Pd, Ag and Mn with the
elements of support material (Si, O, C and N). Pd 3d, Ag 3d and Mn 2p XPS spectra of the
PdAgꢀMnOx/SiO2ꢀNH2 catalyst together with their deconvoluted chemical states are given in
Figure 1 (a), (c) and (e). These results reveal the presence of metallic (i.e. Pd0 with Pd 3d5/2 at
334.5 eV and Pd 3d3/2 at 339.8 eV), as well as oxidic (i.e. Pd2+ with Pd 3d5/2 at 336.9 eV and Pd
3d3/2 at 341.9 eV) Pd states.35 Ag was also found be in both metallic (i.e. Ag0 with Ag 3d5/2 366.8
eV and Ag 3d3/2 372.8 eV), and oxidic (i.e. Ag+ with Ag 3d5/2 366.5 eV; Ag 3d3/2 372.9 eV)
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 39
1
2
states.36 On the other hand, Mn was found to exist in the form of Mn2+ (Mn 2p3/2 at 640.3 eV),
Mn3+ (Mn 2p3/2 at 641.6 eV) and Mn4+ (Mn 2p3/2 at 642.8 eV)37,38 exhibiting an additional highꢀ
binding energy shakeꢀup peak feature at 645.5 eV.38 XPS spectra presented in Figure 1 (b) and
(d) suggest that after Ar+(g) sputtering for 5 min, XPS features associated with Pd2+ and Ag+
could be removed where Pd0 and Ag0 exist as the predominant features. However, oxidic forms
of Mn continue to exist even after Ar+(g) sputtering (Figure 1(f)) without any major alteration.
Existence of oxidic states of Mn is not unexpected as the high oxophilicity of Mn makes Mn0
NPs highly reactive upon exposure to air leading to the generation of various forms of MnOx
states with high stability.39,40
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
12
ACS Paragon Plus Environment

Page 13 of 39
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
Figure 1. Pd 3d XP spectra (a) before, (b) after (5 min.) Ar+(g) sputtering, Ag 3d XPS spectra (c)
before, (d) after (5 min.) Ar+(g) sputtering, Mn 2p XP spectra (e) before, (f) after (5 min.) Ar+(g)
sputtering for PdAgꢀMnOx/NꢀSiO2 catalyst.
TEM, HRTEM, STEMꢀEDX and HAADFꢀSTEMꢀmapping investigations were performed to
examine the size, morphology and the composition of the PdAgꢀMnOx/NꢀSiO2 catalyst. TEM
images of PdAgꢀMnOx/NꢀSiO2 given in Figure 2 (a) and (b) reveal the presence of PdAg and
MnOx nanoparticles (for comparison TEM image of metalꢀfree see NꢀSiO2 is given in Figure S2).
The mean particle size for the images given in Figure 2 (aꢀb) was found to be ca. 6.4 nm using
the NIH image program,41 which included the particle size analysis for > 100 nonꢀtouching
particles (Figure 2 (b) inset). STEMꢀEDX analysis of a large number of different domains on the
PdAgꢀMnOx/NꢀSiO2 surface confirmed the presence of Pd, Ag and Mn in the analyzed regions
(Figure 2 (c)). HRTEM image of PdAgꢀMnOx/NꢀSiO2 is given in Figure 2(d) displaying the
highly crystalline nature of the NPs on the PdAgꢀMnOx/NꢀSiO2 surface. Three different
crystalline fringe distances (0.20, 0.23 and 0.35 nm) were measured for three individual NPs.
42
43
The fringe distances of 0.20 and 0.35 nm can be assigned to MnO2 and Mn2O3 phases,
respectively. On the other hand, the fringe distance of 0.23 nm, which is between the (111)
lattice spacing of faceꢀcentered cubic (fcc) Ag (0.24 nm) and fcc Pd (0.22 nm), can be attributed
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 39
1
2
3
4
5
6
7
8
9
to a PdAg alloy structure. In addition to that, XRD pattern of the asꢀsynthesized catalyst (Figure
S3) exhibits a diffraction peak located between the characteristic Pd (111) and Ag (111)
diffraction features. DRꢀUVꢀVis spectrum taken from solid powders of PdAgꢀMnOx/NꢀSiO2
(Figure S4), shows a very weak surface plasmon resonance (SPR) feature, whereas Ag/SiO2ꢀNH2
and AgꢀMnOx/NꢀSiO2 samples reveal much stronger SPR bands at ca. 440 nm. It is worth
mentioning that, quenching of the SPR bands due to alloying was also observed for oleylamineꢀ
stabilized PdAg alloy NPs.23
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
14
ACS Paragon Plus Environment

Page 15 of 39
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
Figure 2. (aꢀb) TEM images and particle size distribution (inset in (b)), (c) TEMꢀEDX spectrum,
(d) HRTEM image of the PdAgꢀMnOx/NꢀSiO2 catalyst; (e) and (f) HAADFꢀSTEM images,
HAADFꢀSTEM elemental mapping for Pd, Ag and Mn of the PdAgꢀMnOx/NꢀSiO2 catalyst.
The compositional analyses of the PdAgꢀMnOx/NꢀSiO2 catalyst were also performed by
detailed HAADFꢀSTEM measurements. Figure 2 also presents HAADFꢀSTEM images (Figure 2
(e) and (f)) as well as elemental mapping analysis of Pd, Ag and Mn for two independent
particles existing on the PdAgꢀMnOx/NꢀSiO2 catalyst surface. While HAADFꢀSTEMꢀelemental
mapping suggests the existence of a PdAg alloy for the first particle in the absence of any Mn
signal, the second particle reveals the exclusive presence of Mn lacking appreciable Pd or Ag
signals. These results support the presence of individual PdAg alloy NPs that are separate from
the MnOx NPs on the NꢀSiO2 support surface. On the other hand, as it is difficult to establish a
true statistical analysis using microscopic probes, presence of a minor amount of overlapping
domains of PdAg and MnOx cannot be excluded.
Catalytic Performance Results. For a comprehensive elucidation of the catalytic activity of
the PdAgꢀMnOx/NꢀSiO2 catalytic architecture towards additiveꢀfree dehydrogenation of FA; we
followed a systematic approach, where the individual catalytic performance of each structural
subcomponent (i.e. singleꢀcomponent samples such as Pd1.0/NꢀSiO2, Ag1.0/NꢀSiO2, Mn1.0/Nꢀ
SiO2) was studied (Figure 3 (a)ꢀ(d)) in addition to binary combinations of these subcomponents
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 39
1
2
3
4
5
6
7
8
9
(i.e. Pd0.45ꢀMn0.55/NꢀSiO2, Pd0.42Ag0.58/NꢀSiO2, Ag0.49ꢀMn0.51/NꢀSiO2) as well as their ternary
counterparts (i.e. Pd0.40Ag0.12ꢀMn0.48/NꢀSiO2, Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2, Pd0.48Ag0.27ꢀMn0.25/Nꢀ
SiO2, Pd0.55Ag0.35ꢀMn0.10/NꢀSiO2). Evidently, Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 catalyst provides the
best activity compared to all other investigated catalysts. Comparison of Figure 3 (a) and Figure
3 (b) strongly implies that Pd has a critical role in the FA decomposition, as the catalytic activity
is completely lost in the absence of Pd (e.g. Ag1.0/NꢀSiO2, Mn1.0/NꢀSiO2, and Ag0.49ꢀMn0.51/Nꢀ
SiO2). On the other hand, the initially high activity of the singleꢀcomponent Pd1.0/NꢀSiO2 catalyst
is not sustainable as the active sites can readily be deactivated due to the generation of a
poisonous byproduct (i.e. CO), demonstrating the paramount challenges associated with the use
of a singleꢀcomponent catalytic architecture in formic acid decomposition lacking additional
promoters such as Ag and Mn.4,5
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
16
ACS Paragon Plus Environment

Page 17 of 39
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
Figure 3. The volume of generated gas (CO2 + H2) (mL) versus time (min.) graphs of (a)
monometallic, (b) bimetallic, (c) trimetallic catalysts, (d) TOF values versus catalysts in different
[Pd]:[Ag]:[Mn] ratios for the additiveꢀfree FA dehydrogenation (0.25 M in 10.0 mL H2O) at 313
K.
Although, binary systems such as Pd0.45ꢀMn0.55/NꢀSiO2 and Pd0.42Ag0.58/NꢀSiO2 show better
activities than Pd1.0/NꢀSiO2, their performances are still far inferior to the ternary Pd0.44Ag0.19
ꢀ
Mn0.37/NꢀSiO2 system (Figure 3). Morphological investigation by TEM (Figure S5) reveals that
NPs of the singleꢀcomponent and binary systems have smaller particle sizes with respect to the
ternary Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 system. The uniqueness of the Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2
catalyst structure was further supported by two control experiments where physical mixture of (i)
Pd/NꢀSiO2, Ag/NꢀSiO2, Mn/NꢀSiO2 (with a Pd:Ag:Mn molar ratio of 0.45:0.21:0.34) and (ii)
Pd0.41Ag0.59/NꢀSiO2 and Mn0.37/NꢀSiO2 exhibited a lower activity than that of Pd0.44Ag0.19
ꢀ
Mn0.37/NꢀSiO2 catalyst in FA dehydrogenation in the absence of additives under identical
conditions (Figure S6). These control experiments demonstrate a proximity requirement
associated with the synergistic structural components of the Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 catalyst,
which interact in an efficient manner during additiveꢀfree dehydrogenation of FA. The generated
gas obtained via Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2ꢀcatalyzed FA dehydrogenation was analyzed by gas
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 39
1
2
3
4
5
6
7
8
9
chromatography (GC) in the presence or absence of the NaOH trap (Figure S7). These
experiments revealed that the generated gas is a mixture of H2 and CO2 with a H2:CO2 molar
ratio of 1.0:1.0 where CO was below the detection limit (i.e. < 10 ppm). In other words, these
experiments point to the important fact that COꢀfree H2 generation can be achieved in the
absence of additives from an aqueous FA solution for fuel cell applications6 at ambient
conditions by utilizing a Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 catalyst.
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 (a) shows the volume of generated gas (CO2 + H2) versus the reaction time for
Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 catalyzed additiveꢀfree FA dehydrogenation at different temperatures.
It is apparent that Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 catalyst provides initial TOF values of 330, 530,
700, 1430 mol H2.mol catalystꢀ1.hꢀ1 at 298, 303, 308 and 313 K, respectively (see Supporting
Information for details of the calculation of initial TOF values). The initial TOF value of 330
mol H2.mol catalystꢀ1.hꢀ1 (corresponding to 750 mol H2.mol Pdꢀ1.hꢀ1; 524 mol H2.mol (Pd+Ag)ꢀ
1.hꢀ1 assuming that the active sites are only provided by Pd or PdAg NPs; respectively) measured
at 298 K is one of the most remarkable TOF values reported for FA dehydrogenation at room
temperature using a heterogeneous catalyst without utilizing any additives (Table 1).
18
ACS Paragon Plus Environment

Page 19 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
Figure 4. (a) The volume of the generated gas versus time plots for Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2
(2.69 mM) catalyzed additiveꢀfree dehydrogenation of FA (0.25 M FA in 10.0 mL H2O) at
different temperatures, (b) analogous data for 0.25 M FA in 10.0 mL H2O at 298 K
corresponding to varying catalyst (Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2) concentrations.
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
More importantly, FA dehydrogenation can be completed in less than an hour with > 99 %
conversion. The values of observed rate constants kobs determined from the linear portions of the
volume of generated gas (CO2 + H2) versus reaction time plots at four different temperatures are
used to obtain Arrhenius and Eyring plots (Figure S8 and Figure S9) to calculate activation
parameters. Using these plots, apparent activation energy (Ea), apparent activation enthalpy
≠
≠
(ꢁHa ) and apparent activation entropy (ꢁSa ) values were calculated to be 72.4 kJ/mol, 69.1
kJ/mol and 37.5 J/mol.K, respectively. The positive magnitude of the apparent activation entropy
implies the presence of a dissociative mechanism in the transition state (vide infra).44
Table 1. Comparison of the catalytic performance data for the currently studied PdꢀMnOx/SiO2ꢀ
NH2 catalyst with the prior best heterogeneous catalyst systems reported for the dehydrogenation
FA in the absence of any additives at low temperatures
Catalyst
Temp. (K)
Conv. (%) Activity (h-1)a
Ref.
Ag@Pd
AgPd
293
293
298
298
298
298
298
298
36
10
89
91
51
96
28
52
63
72
98
37
45
85
41
180
18
18
19
20
21
21
22
23
Au@Pd
CoAuPd/C
CoAuPd/GO
CoAuPd/DNA
AuPd
AgPd
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 39
1
2
3
4
5
6
7
8
9
PdꢀMnOx
PdAuꢀMnOx/MOFꢀGraphene
PdAuCr/NꢀSiO2
298
298
298
298
298
63
94
99
99
99
150
382b
730
24
25
26
PdAgꢀMnOx/NꢀSiO2
PdAgꢀMnOx/NꢀSiO2
524c
this study
this study
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
330
a
TOF = mol H2/mol total metal × h and these TOF values are not corrected for the number of
exposed surface atoms; that is, the values given are lower limits; b Based on Au and Pd atoms;
c
Based on Ag and Pd atoms
In addition to the effect of temperature, we also investigated the influence of the catalyst
concentration on the rate of the additiveꢀfree dehydrogenation of FA by performing the catalytic
reaction starting with different Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 concentrations at 298 K (Figure 4 (b)).
The reaction rates for each catalyst concentration were calculated from the linear portion of each
plot given in Figure 4(b). The logarithmic plot of the hydrogen generation rate versus catalyst
concentration (Figure S10) gives a line with a slope of 1.18, which indicates that Pd0.44Ag0.19
ꢀ
Mn0.37/NꢀSiO2 catalyzed additiveꢀfree FA dehydrogenation, is close to firstꢀorder with respect to
the catalyst concentration within the investigated concentration window.
The catalytic stability of the PdAgꢀMnOx/NꢀSiO2 catalyst in the additiveꢀfree
dehydrogenation of FA was investigated by performing recycling and reusability experiments.
When all of FA was converted to CO2 and H2 in a particular cycle, more FA was added into the
solution and the reaction was continued up to five consecutive catalytic cycles. It was found that
PdAgꢀMnOx/NꢀSiO2 catalyst (i.e. Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2) retains 86 % of its initial activity
and provides 98 % of conversion without CO generation after the 5th consecutive cycle (Figure
5(a)). Ease of isolation and reusability characteristics of PdAgꢀMnOx/NꢀSiO2 were also tested in
the FA dehydrogenation under identical conditions. After the complete dehydrogenation of FA,
20
ACS Paragon Plus Environment

Page 21 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
PdAgꢀMnOx/NꢀSiO2 catalyst was isolated as a dark gray powder and bottled under nitrogen
atmosphere. Then, the isolated PdAgꢀMnOx/NꢀSiO2 catalyst was reꢀdispersed in the aqueous FA
solution. This reꢀdispersed catalyst preserved 80 % of its initial activity with a 95 % conversion
of FA to CO2 and H2 even after the 5th catalytic reuse (Figure 5(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
Figure 5. (a) Activity and conversion versus number of catalytic recycles for Pd0.44Ag0.19
ꢀ
Mn0.37/NꢀSiO2ꢀcatalyzed FA dehydrogenation in the absence of additives at 298 K, (b) TEM
image of Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 recovered after the 5th catalytic recycle (inset shows the
particle size distribution of PdAgꢀMnOx NPs), (c) Activity and conversion versus number of
catalytic reuse at 298 K, (d) TEM image of Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 catalyst recovered after
the 5th catalytic reuse with PdAgꢀMnOx NPs size distribution (inset).
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 39
1
2
TEM analysis of PdAgꢀMnOx/NꢀSiO2 samples recovered after 5th consecutive catalytic run of
the recycling and reusability experiments (Figure 5 (b) and Figure 5 (d)) show a slight increase in
the average particle size of PdAgꢀMnOx NPs (6.8 and 7.9 nm, respectively) consistent with the
minor decrease in the activity at the end of these experiments. Moreover, ICPꢀOES and
elemental analysis of used and isolated PdAgꢀMnOx/NꢀSiO2 catalyst samples and reaction
solutions indicated that (i) metal and ꢀNH2 contents of the catalysts remained intact after use and
(ii) leaching of metals and/or surface grafted amines into the reaction solution was not observed.
Additionally, removing the PdAgꢀMnOx/NꢀSiO2 catalyst from the reaction solution can
completely stop the dehydrogenation of FA. These results are indicative of the high stability of
PdAgꢀMnOx NPs against agglomeration and leaching throughout the catalytic runs.
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
Effect of Surface Grafted Amine Groups on the Catalytic Activity. In a series of
additional experiments, we also compared the catalytic activities of PdAgꢀMnOx/NꢀSiO2, which
were functionalized with different amounts of amine groups, in the additiveꢀfree
dehydrogenation of FA under identical reaction conditions in order to understand the effect of
the surfaceꢀgrafted amine functionalities on the catalytic reactivity (Figure S11). We found that
amineꢀfree SiO2 supported PdAgꢀMnOx catalyst (Pd0.41Ag0.20ꢀMn0.39/SiO2) provides the lowest
gas generation rate (6.35 mL/min) and the catalytically optimum amine loading is 0.98 mmol
NH2/g, where the maximum gas generation rate can be achieved by Pd0.44Ag0.19ꢀMn0.37 NPs (30.8
mL/min).
The limited reactivity of Pd0.41Ag0.20ꢀMn0.39 NPs supported on amineꢀfree SiO2 can be
explained by the absence of ꢀNH2 functionalities on the support material, which may have a
direct impact on the FA adsorption/storage process as well as the nucleation and growth of the
22
ACS Paragon Plus Environment

Page 23 of 39
ACS Catalysis
1
2
PdAg and MnOx NPs on the support surface (Figure S12). In a recent study Yamashita et al.45
reported that a resin bearing ꢀN(CH3)2 acted as a significantly more efficient organic support
material in the catalytic decomposition of FA than those bearing ꢀSO3H, ꢀCOOH and ꢀOH for Pd
or Ag@Pd NPs. Their mechanistic studies revealed that OꢀH bond cleavage in FA is facilitated
by the ꢀN(CH3)2 functionalities leading to the formation of metalꢀbound formate species along
with a ꢀ[N(CH3)2H]+ species, followed by the dehydrogenation of the metalꢀbound formate,
producing H2 and CO2. In the light of these results, it is reasonable to propose that the existence
of surfaceꢀgrafted amine functionalities in the currently exploited support material efficiently
acts as a proton scavenger providing a basic environment around the PdAgꢀMnOx NPs. It is
feasible that amineꢀgrafted SiO2 surface may provide an adsorption reservoir where the
generated formate species on the support surface can spillover on the active PdAgꢀMnOx
domains. Dissociative adsorption of FA and the –NH2ꢀfacilitated OꢀH bond scission is followed
by the consecutive CꢀH bond cleavage from the metalꢀbound formate intermediate.
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
On the other hand, the lower activity of the catalysts in Figure S11 (see Supporting
Information) with high amine loadings (i.e. > 1 mmol NH2/g) can be attributed to two main
factors namely, the decreasing particle size of PdAgꢀMnOx NPs on the support surfaces and the
poisoning of the PdAgꢀMnOx NPs by the excessive amount of surface ꢀNH2 functionalities
covering these NPs. As shown by the TEM images given in Figure S12, PdAgꢀMnOx NP size
distribution can be fineꢀtuned by changing the coverage of the surfaceꢀgrafted amine groups.
Figure S12 clearly indicates that the particle size of PdAgꢀMnOx NPs decreases with the increase
in amine concentration. Decreasing particle size may increase the surface concentration of point
defects (e.g. corner atoms, kinks etc.) as well as extended defects (e.g. steps) on the PdAgꢀMnOx
NPs which may in turn create coordinativelyꢀunsaturated sites with a high affinity towards
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 39
1
2
3
4
5
6
catalytic poisons such as CO. Similarly, such defect sites may also adsorb other reactants and/or
products in an adversely strong manner disfavoring the Sabatier principle.46
7
8
9
The influence of the type of the support material was also investigated by synthetically
replacing NꢀSiO2 with some of the ubiquitous support materials used in catalysis such as Al2O3,
TiO2 and C. The catalytic activities of Pd0.37Ag0.17ꢀMn0.46/Al2O3, Pd0.38Ag0.20ꢀMn0.32/TiO2,
Pd0.40Ag0.20ꢀMn0.40/C and Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 catalysts prepared by the same method were
investigated in the additiveꢀfree FA dehydrogenation under identical conditions (Figure S13).
The activity was observed to decrease in the following order: Pd0.44Ag0.19ꢀMn0.37/NꢀSiO2 (30.8
mL/min) > Pd0.40Ag0.20ꢀMn0.40/C (10.7 mL/min) > Pd0.37Ag0.17ꢀMn0.46/Al2O3 (4.32 mL/min) >
Pd0.38Ag0.20ꢀMn0.32/TiO2 (2.1 mL/min). The formation of largeꢀsized PdAgꢀMnOx NPs (i.e.
sintering) and the lack of –NH2 functionalities could be responsible for the significantly lower
performances of such catalysts (Figure S14).
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
Influence of MnOx Nanoparticles on Poisoning Resistivity of PdAg Alloy Nanoparticles.
The enhancement of Pd activity in FA dehydrogenation through Ag incorporation has already
been reported for Ag@Pd18 and PdAg23 NPs, in which the activity increase has been attributed to
a synergic effect.47 In order to shed some light on the effect of MnOx NPs, we conducted linear
sweep voltammetry (LSV), CO stripping voltammetry and inꢀsitu FTIR analyses.
(i) Linear Sweep Voltammetry (LSV) Analyses. We first performed LSV measurements on
PdAg/NꢀSiO2 and PdAgꢀMnOx/NꢀSiO2 catalysts to explore their stability in the electroꢀoxidation
of formic acid. As described in detail in the experimental section, the catalysts were
electrochemically preꢀtreated before the LSV measurements by using a bias potential of 0.0 V
for 0ꢀ180 s, where CO and OH intermediates form and bind to the active surface sites of
24
ACS Paragon Plus Environment

Page 25 of 39
ACS Catalysis
1
2
50
catalysts.48ꢀ
Afterwards, LSV measurements were conducted on the poisoned catalysts
3
4
49
5
6
7
8
surfaces, in which the maximum current versus potential values was recorded as a function of the
poisoning time (Figure 6 (a) and Figure 6 (b)). It is clearly seen that PdAgꢀMnOx/NꢀSiO2 catalyst
has characteristically higher activity than PdAg/NꢀSiO2 toward formic acid oxidation in terms of
peak current values. Additionally, investigation of the relative peak current versus poisoning
time (Figure 6 (c)) suggests that although PdAg/NꢀSiO2 almost completely loses its initial
activity upon 180 s of poisoning; PdAgꢀMnOx/NꢀSiO2 catalyst retains > 85 % of its initial
activity after an identical poisoning treatment.
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
The recent comprehensive 13C NMR and ATRꢀIR studies51 using formate as a surface probe
for metal nanoparticles have shown that (i) adsorption of formic acid on metal nanoparticles
during FA decomposition gives three modes of adsorbed formates; (bridging, linear and
multilinear), and (ii) formate with OCO group, binds to surface metal atoms in a manner very
similar to carbon monoxide. For this reason, as in the case of previously reported works,52 in
which FA was selectively dehydrogenated to CO2 + H2 and no CO was generated, we performed
CO stripping voltammetry and inꢀsitu FTIR analyses by using CO as a probe molecule to
understand the influence of MnOx on the poisoning resistivity of PdAg alloy nanoparticles.
(ii) COꢀStripping Voltammetry Analyses. Figure 6 (d) gives COꢀstripping voltammograms for
PdAg/NꢀSiO2 and PdAgꢀMnOx/NꢀSiO2 catalysts (for clarity only 1st scans were compared, see
Figure S15 and S16 for complete voltammograms). The onset potentials were found to be 0.30
and 0.23 V for PdAg/NꢀSiO2 and PdAgꢀMnOx/NꢀSiO2 (ꢁVCOads = 70 mV), respectively. By
52 57
5354
5556
considering various previous studies, ꢀ where even smaller energy differences (ꢁVCOads ≤ 70
mV) were reported, this result clearly demonstrates that the CO poisoning resistance of PdAg
alloy NPs can be significantly enhanced by the promotional effect of MnOx NPs. Another
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 39
1
2
3
4
5
6
significant feature of the COꢀstripping voltammograms given in Figure 6(d) is the fact that
MnOxꢀpromoted catalyst releases a significantly greater amount of CO during the voltage sweep.
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. LSV measurements on (a) PdAg/NꢀSiO2 and (b) PdAgꢀMnOx/NꢀSiO2 catalysts in 0.5
M H2SO4 + 1M HCOOH solution with a 10 mV s−1 scan rate, (c) The relative maximum peak
current vs time graph for PdAg/NꢀSiO2 and PdAgꢀMnOx/NꢀSiO2 catalysts, (d) COꢀstripping
voltammograms for PdAg/NꢀSiO2 and PdAgꢀMnOx/NꢀSiO2 catalysts in H2SO4 solution with a 10
mVs−1 scan rate.
As shown in the Table S1 (see Supporting Information), specific surface area is not the
primary factor in CO storage, as MnOx incorporation does not significantly alter it. This
observation is in perfect agreement with the current inꢀsitu FTIR experiments that will be
26
ACS Paragon Plus Environment

Page 27 of 39
ACS Catalysis
1
2
3
4
5
6
7
8
9
discussed in the forthcoming section suggesting that MnOx domains function as sacrificial sites,
which can efficiently store CO in the form of carbonates, preventing the catalytic poisoning of
the PdAg active sites. It is worth mentioning that in the presence of MnOx, although a
significantly greater quantity of catalytic poisoning species (i.e. CO) is captured by the catalyst
surface, these antagonistic species can be very readily and reversibly removed from the catalyst
surface, evident by the lower COꢀoxidation onset potential for PdAgꢀMnOx/NꢀSiO2 catalysts with
respect to that of PdAg/NꢀSiO2 (Figure 6(d)).
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
(iii) InꢀSitu FTIR Spectroscopy. Inꢀsitu FTIR experiments were performed (Figure 7) to
investigate the relative CO poisoning characteristics of different adsorption sites on the PdAgꢀ
MnOx/NꢀSiO2 catalyst at the molecular level. It is well known that Ag and Pd can dissolve in
each other to form bimetallic alloys with a wide range of compositions.58 Former comprehensive
FTIR spectroscopic studies on supported PdAg bimetallic NPs and PdAg ultrathin films
demonstrated that CO interacts very weakly with the Ag adsorption sites (irrespective of the Ag
content of the bimetallic NP) leading to vibrational features with intensities that are typically
below the experimental detection limit;59 ꢀ63 while the interaction of CO with the Pd adsorption
sites is significantly strong. CO vibrational signatures on monometallic Ag NPs deposited on
oxide surfaces typically appear at ≥ 2169 cmꢀ1.64 Before starting to inꢀsitu FTIR analyses, the
catalytic reactivity of PdAgꢀMnOx/NꢀSiO2 catalyst annealed under vacuum (< 1x 10ꢀ3 Torr) at
400 K for 2 h was checked in the catalytic dehydrogenation of FA in order to show this
pretreatment does not significantly affect the catalytic nature of PdAgꢀMnOx/NꢀSiO2 catalyst.
The result of this control experiment revealed that the catalytic activity of PdAgꢀMnOx/NꢀSiO2
catalyst is not affected by this pretreatment as annealed PdAgꢀMnOx/NꢀSiO2 catalyst gave almost
6061
62
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 39
1
2
3
4
5
6
the same activity and conversion values with that of fresh PdAgꢀMnOx/NꢀSiO2 catalyst (Figure
S17).
7
8
9
CO adsorption on PdAg reveals unique vibrational features dissimilar to monometallic Pd
surfaces. Pd and PdAg nanoparticles exhibit a truncatedꢀcuboꢀoctahedral geometry (Wulff
Polyhedron) exposing (111) facets with a minor contribution from (100) facets.65,66,67
CO/Pd(111) yields νCO at c.a. 2110, 1960 and 1895ꢀ1810 cmꢀ1 corresponding to atop (linear),
bridging and threefold adsorption geometries, respectively.66ꢀ68 CO/Pd(100) leads to exclusively
bridging adsorption (1997ꢀ1807 cmꢀ1).66ꢀ67 Figure 7(a) presents inꢀsitu FTIR spectra recorded
after the saturation of PdAg/NꢀSiO2, PdAgꢀMnOx/SiO2 and MnOx/SiO2 surfaces with CO at 323
K. Ag incorporation into the Pd structure strongly alters the CO adsorption on the atop and
bridging Pd sites evident by νCO at 2045 and 1887 cmꢀ1 (Figure 7(a)) with a characteristic red
shift (ꢁνCO = 65ꢀ73 cmꢀ1) compared to that of Pd(111) (i.e. 2110 and 1960 and cmꢀ1,
respectively[66]). This is in perfect agreement with a former study,69 demonstrating the correlation
between the redꢀshift in νCO upon Ag incorporation into the Pd lattice and weakening of the CO
adsorption strength. Similarity in the magnitude of the red shift for atop and bridging Pd
adsorption sites suggests that Ag is rather uniformly distributed in the bimetallic NP, influencing
atop and bridging Pd adsorption sites alike.
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
28
ACS Paragon Plus Environment

Page 29 of 39
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
Figure 7. (a) Inꢀsitu FTIR spectra acquired upon CO adsorption (20.0 Torr of CO exposure for
10 min at 323 K) on PdAg/NꢀSiO2, PdAgꢀMnOx/NꢀSiO2 and MnOx/NꢀSiO2. (b) Inꢀsitu FTIR
spectra acquired upon formic acid adsorption (5.0 Torr of FA exposure for 5 min at 323 K) on
CO preꢀpoisoned PdAg/NꢀSiO2, PdAgꢀMnOx/NꢀSiO2 and MnOx/NꢀSiO2. All of the spectra were
acquired at 323 K in vacuum (10ꢀ3 Torr).
Weakening of the CO adsorption strength is very critical for FA dehydrogenation, as it
directly indicates the increased tolerance against CO poisoning. Strongly adsorbed CO on the
defective monometallic Pd particles may lead to “adsorbateꢀinduced surface reconstruction”
resulting in the disintegration (i.e. leaching) of the monometallic Pd particles.59 Decreasing the
CO adsorption strength via alloying with Ag increases the stability and catalytic lifetime as
demonstrated by the stability and reusability tests (videꢀinfra). Lack of νCO at ≥ 2169 cmꢀ1 in
Figure 7(a) suggests that Ag sites are either not exposed or CO adsorption on Ag is weak.
Furthermore, 1887 cmꢀ1 signal of PdAgꢀMnOx/NꢀSiO2 is much stronger than that of the PdAg/Nꢀ
29
ACS Paragon Plus Environment