Electrochemical CO2 Reduction to Formic Acid at Low Overpotential and with High Faradaic Efficiency on Carbon-Supported Bimetallic Pd-Pt Nanoparticles

Article abstract of DOI:10.1021/acscatal.5b00602

The electrochemical reduction of CO₂ has attracted significant interest recently, as it is a possible reaction for the storage of renewable energy. Here, we report on the synthesis of Pd_xPt_(100-x)/C nanoparticles and their electrocatalytic properties for the reduction of CO₂ to formic acid, compared with their activity for the reverse oxidation of formic acid to CO₂. We find that Pd_xPt_(100-x)/C nanoparticles have a very low onset potential for the reduction of CO₂ to formic acid of ca. 0 V vs RHE, which approaches the theoretical equilibrium potential of 0.02 V vs RHE for this reaction. Furthermore, the Pd₇₀Pt₃₀/C catalyst shows a faradaic efficiency of 88% toward formic acid after 1 h of electrolysis at -0.4 V vs RHE with an average current density of ～5 mA/cm². Therefore, this catalyst shows a competing or even better faradaic efficiency toward formic acid compared to recently reported catalysts, at a substantially lower overpotential, while avoiding a strong deactivation that was observed with previously reported Pd-based catalysts.

Full text of DOI:10.1021/acscatal.5b00602

Subscriber access provided by UNIVERSITY OF LEEDS
Article
Electrochemical CO2 reduction to formic acid at low overpotential and with
high faradaic efficiency on carbon supported bimetallic Pd-Pt nanoparticles
Ruud Kortlever, Ines Peters, Sander Koper, and Marc T.M. Koper
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b00602 • Publication Date (Web): 15 May 2015
Downloaded from http://pubs.acs.org on May 17, 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 21
ACS Catalysis
1
2
3
4
5
6
7
8
Electrochemical CO2 reduction to formic acid at low
overpotential and with high faradaic efficiency on carbon
supported bimetallic Pd-Pt nanoparticles
9
Ruud Kortlever, Ines Peters, Sander Koper, Marc T.M. Koper*
Leiden Institute of Chemistry, Leiden University,
P.O. Box 9502, 2300 RA Leiden, The Netherlands
Fax: (+)31 071 527 4451
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
Eꢀmail: m.koper@chem.leidenuniv.nl
Abstract
The electrochemical reduction of CO2 has attracted a significant interest recently, as it is a
possible reaction for the storage of renewable energy. Here we report on the synthesis of
PdxPt(100ꢀx)/C nanoparticles and their electrocatalytic properties for the reduction of CO2 to
formic acid, compared with their activity for the reverse oxidation of formic acid to CO2. We
find that PdxPt(100ꢀx)/C nanoparticles have a very low onset potential for the reduction of CO2
to formic acid of ca. 0 V vs. RHE, which approaches the theoretical equilibrium potential of
0.02 V vs. RHE for this reaction. Furthermore, the Pd70Pt30/C catalyst shows a faradaic
efficiency of 88% towards formic acid after 1 hour of electrolysis at ꢀ0.4 V vs. RHE with an
average current density of ~5 mA/cm2. Therefore this catalyst shows a competing or even
better faradaic efficiency towards formic acid compared to recently reported catalysts, at a
substantially lower overpotential, while avoiding a strong deactivation that was observed with
previously reported Pd based catalysts.
Keywords: Electrocatalysis, CO2 reduction, Palladiumꢀplatinum catalysts, Nanoparticles,
Formic acid.
1. Introduction
The electrochemical reduction of CO2 has attracted a lot of attention in recent years, since
there is a need for a carbonꢀneutral energy carrier for the storage of renewable power.1,2 By
the production of fuels from CO2 one could close the anthropogenic carbon cycle, which will
not only have the advantage of an easy implementation into the existing fuel infrastructure,
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 21
1
2
3
4
5
provided the fuel is a liquid, but could also have the advantage of lowering the possibly
harmful CO2 concentration in the atmosphere.
6
7
8
9
One of the possible products from electrocatalytic CO2 reduction is formic acid. Since
formic acid is used as fine chemical and can be used as a fuel in a direct formic acid fuel cell
(DFAFC),3 efficient production of formic acid from CO2 is a very relevant process. Currently,
most catalysts that have been reported to selectively make formic acid from CO2 are poor
metal based catalysts such as Cd, Sn, Pb and In.4,5 The main issue with this class of catalysts
is that in general they require very negative potentials to selectively produce formic acid,
although recent advances have decreased the overpotential significantly.6ꢀ8
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
Recently, we reported the electrochemical reduction of CO2 to formic acid on
electrodeposited palladium layers on a platinum substrate with considerably lower
overpotentials than previously reported catalysts.9 This catalyst material was based on the idea
that for a twoꢀelectron transfer reaction, such as the oxidation of formic acid, reversible
catalysts must exist:10 thus catalysts that are very active for the electrochemical oxidation of
formic acid should also be active for the reverse electrochemical reduction of CO2 to formic
acid. Since the best formic acid oxidation catalysts contain palladium and/or platinum,11,12
a
combination of these metals was chosen. It was shown that electrodeposited palladium layers
on a platinum substrate (PdꢀPt) exhibit a significant decrease in the onset potential for formic
acid production from CO2 compared to bulk palladium; a decrease from ꢀ1.2 V vs. RHE (for
bulk palladium) to ꢀ0.1 V vs. RHE (for PdꢀPt) was observed. An accompanying advantage of
the PdꢀPt catalyst was that since the electrode was already shown to be active for formic acid
oxidation, reversible reduction of CO2 and oxidation of formic acid is possible on this
material. Although PdꢀPt was shown to be able to produce formic acid from CO2 almost
selectively at low overpotentials, it was observed that the reduction of CO2 and oxidation of
formic acid on PdꢀPt was rapidly poisoned by the formation of CO on the surface of the
catalyst, thereby quickly diminishing the catalytic activity.9,13
In an effort to overcome this problem and to develop an electrocatalyst with low
overpotential, high faradaic efficiency and high stability, we will here explore the possibilities
of PdPt/C nanoparticles with different palladium to platinum ratios for both the reduction of
CO2 and the oxidation of formic acid. Since the oxidation of formic acid on PdPt
nanoparticles has already been studied extensively,14ꢀ16 we will use the systematic study of
formic acid oxidation on PdPt nanoalloys by Cai et al.16 as a guideline. The optimal PdꢀPt
2
ACS Paragon Plus Environment

Page 3 of 21
ACS Catalysis
1
2
3
4
5
ratio for the reduction of CO2 will be studied, including their corresponding faradaic
efficiency, and compared to the optimal ratio for formic acid oxidation.
6
7
8
9
2. Experimental
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.1 Materials
Cylindrical highly ordered pyrolitic graphite electrodes (HOPG), purchased at GE Quartz
Europe GmbH, with a diameter of 5 mm and 10 mm connected to platinum wire were used as
working electrodes. Prior to every experiment the working electrodes were polished
mechanically to a mirrorlike finish using P2500 sandpaper. After this, the electrodes were
sonicated in ultraꢀpure water.
Electrolytes were made from ultraꢀpure water (Millipore MilliQ gradient A10 system,
18 mꢁ cm) and high purity reagents (Merck Suprapur, Sigma Aldrich TraceSelect). Before
each experiment the electrolytes were first purged with Argon (Air Products, 5.7) for 15
minutes to remove air from the solution. In the case of CO2 reduction experiments the
electrolyte was subsequently purged with CO2 (Linde, 4.5) for at least 30 minutes to saturate
the solution.
2.2 Preparation of PdxPt(100-x)/C Nanoparticles
The procedure followed for the synthesis was similar to the one reported by Cai et al..16
However, we extended this method to also make platinum enriched nanoparticles. As was
reported by Cai et al., this method is not particularly suitable for the synthesis of Pt/C
nanoparticles. Since nanoparticles with high Pt content showed poor electrocatalytic
properties compared to Pd rich nanoparticle, Pt/C nanoparticles were not included in this
study. Stock solutions of 0.05 M K2PdCl4, 0.05 M K2PtCl4 and 0.1 M
ethylenediaminetetraacetic acid (EDTA) were made and added in the appropriate amounts
(see Table S1) to 26 ml of ultraꢀpure water in 100 ml round bottom flask. The flask was
heated to 60 oC and kept at this temperature for 40 minutes under vigorous stirring. After 40
minutes the solution was allowed to cool to room temperature and the pH of the solution was
adjusted to ca. 9ꢀ10 with a 0.3 M NaOH solution. Afterwards, 32 mg of Vulcan XCꢀ72 (Cabot
Corporation) was added as carbon support and this suspension was sonicated for 30 minutes.
Next, 6 ml 0.05 M Na2CO3 solution containing 18 mg NaBH4 was added dropwise over the
course of 20 minutes. After this addition, the mixture was stirred at room temperature for 2
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 21
1
2
3
4
5
hours. The nanoparticles were collected by centrifugation at 4500 rpm for 20 minutes, and
were washed three times with ultraꢀpure water.
6
7
8
9
After synthesis the nanoparticles were dried in air and analyzed with Xꢀray powder
diffraction, inductively coupled plasma optical emission spectroscopy (ICPꢀOES) and
transmission electron microscopy (TEM). Xꢀray powder diffraction analysis was performed
on a Philips X’pert diffractometer, equipped with an X’celerator, using CuΚα radiation. The
data were recorded in the 2θꢀrange between 10o and 100o. For ICPꢀOES, nanoparticle samples
were first allowed to dissolve in aqua regia (3:1 HCl/HNO3) for at least 48h. The samples
were then filtered and analyzed with a Varian VistaꢀMPX CCD Simultaneous ICPꢀOES. TEM
images were obtained with a JEOL TEM 1010.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.3 Electrochemical Measurements
Cyclic voltammetry measurements were carried out in a standard electrochemical cell using a
threeꢀelectrode assembly at room temperature. The cell and all other glassware were first
cleaned by boiling in a 1:1 mixture of concentrated sulfuric and nitric acid and were cleaned
before every experiment by boiling in ultraꢀpure water. A highly ordered pyrolitic graphite
electrode with a diameter of 5 mm was used as working electrode, on which 10 ꢂl of a 5ꢀwt%
nafion solution containing approximately 10 mg/ml of PdPt/C nanoparticles was dropcasted.
A coiled gold wire was used as counter electrode. All potentials are reported versus the
reversible hydrogen electrode (RHE) as a reference electrode in a separate compartment filled
with the same electrolyte, at the same pH as the electrolyte in the electrochemical cell. The
voltammograms were recorded on an Ivium A06075 potentiostat at a scan rate of 50 mV sꢀ1.
Faradaic efficiency measurements were carried out in a customꢀmade Hꢀtype cell with
a cathodic and an anodic compartment separated by a Nafion 115 membrane (see Figure S6 in
the Supporting Information). The cell was cleaned in a similar way as the cell used for cyclic
voltammetry. A highly ordered pyrolitic graphite electrode with a diameter of 10 mm was
used as working electrode, on which 50 ꢂl of a 5ꢀwt% nafion solution containing
approximately 10 mg/ml of PdxPt(1ꢀx)/C nanoparticles was dropcasted. A platinum flag
electrode was used as counter electrode and a Ag/AgCl reference electrode was used. The
resistance of the cell was determined before every experiment by electrochemical impedance
spectroscopy (EIS) on an Ivium A06075 potentiostat. For the chronoamperometry
measurements an Autolab PGSTAT101 potentiostat was used, which compensated for the
4
ACS Paragon Plus Environment

Page 5 of 21
ACS Catalysis
1
2
3
4
5
ohmic drop that was measured with EIS. After the measurements the applied potential was
further corrected, if necessary, to give the real electrode potential vs. RHE.
6
7
2.4 Detection and Quantification of Products
8
9
For online detection of products dissolved in the electrolyte as a function of applied potential
online High Performance Liquid Chromatography (HPLC) was used.17 While changing the
potential from 0.0 V to the required potential, samples were collected with an open tip
positioned close (~10 ꢂm) to the electrode. Sampling was done at a rate of 60 ꢂL minꢀ1 and
each sample had a volume of 60 ꢂL. Since the potential was changed at 1 mV sꢀ1, each sample
contained the products averaged over a potential change of 60 mV. After voltammetry, these
samples were analyzed by HPLC (Prominence HPLC, Shimadzu; Aminex HPX 87ꢀH column,
Biorad). Faradaic efficiencies were determined by analyzing 100 ꢂl aliquots of the electrolyte,
which were taken every 10 minutes during electrolysis, with HPLC.
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 and discussion
3.1 Characterization
The asꢀsynthesized nanoparticles were characterized with Xꢀray diffraction and ICPꢀOES (see
Fig. S1 and Table 1). The composition of the nanoparticles could be controlled very well with
the method that was used, since the maximum deviation in the composition from the targeted
PdxPt(100ꢀx) composition was 1.5%. In contrast, the loading of the nanoparticles was more
difficult to control. In the synthesis we aimed at a loading of 20 wt% PdxPt(100ꢀx) on Vulcan
XCꢀ72, the actual loading however varies between 11.8 wt% (for Pd90Pt10/C) and 33.7 wt%
(for Pd5Pt95/C). The loading of the nanoparticles does not influence the electrocatalytic
properties greatly, since no significant electrocatalytic differences were observed between
batches of the same PdxPt(100ꢀx)/C composition with different loading (see Fig. S2).
The Xꢀray diffraction spectra show a first peak around 25o which is ascribed to the
amorphous carbon support, and characteristic peaks for faceꢀcenteredꢀcubic (fcc) crystalline
Pd and Pt, corresponding to the (111), (200), (220) and (311) planes respectively. Especially
in the peak corresponding to the (111) plane a small shift to lower 2 theta values is seen with
increasing ratio of Pt in the alloy nanoparticles. This shift is caused by a lattice expansion due
to the incorporation of larger Pt atoms in the Pd lattice.18 The (111) peak in the Xꢀray
diffraction spectra was used to calculate the average size of the PdxPt(100ꢀx)/C nanoparticles
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 21
1
2
3
4
5
6
7
8
9
with the aid of the Scherrer equation. The calculated average size for all bimetallic
nanoparticles is around 4 nm, only the pure palladium nanoparticles deviate with a smaller
average size of 3.1 nm. TEM images of the nanoparticles (see Fig. S8 and S9), show that the
distribution of the nanoparticles over the Vulcan XCꢀ72 support is fairly even and that there is
no large deviation in the size of the particles. Particle size distribution histograms
corresponding to the TEM images show approximately the same average sizes for the
nanoparticle samples as the average sizes calculated with the Scherrer equation (see Fig. S10
and S11).
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 nanoparticles were further characterized electrochemically by performing cyclic
voltammetry in a 0.5 M H2SO4 electrolyte (Fig. 1). In Fig. 1b a clear shift in the cathodic peak
potential for the stripping of oxides is seen as a function of nanoparticle composition. A
negative shift of the oxide stripping peak potential is seen with increasing ratios of palladium,
in agreement with earlier studies.16 To evaluate the effect of the changing ratio of Pd and Pt in
the nanoparticles on the hydrogen adsorption and desorption region of the voltammograms, a
lower potential boundary was chosen in such a way that hydrogen evolution and oxidation
was excluded as much as possible. In the hydrogen adsorption and desorption region a shift
was seen in both the hydrogen adsorption and hydrogen desorption peak potential. The peak
potential shifted from 0.22 V vs. RHE for Pd50Pt50/C to 0.27 V vs. RHE for Pd/C for the
hydrogen adsorption, while for hydrogen desorption the peak potential shifted from 0.12 V vs.
RHE for Pd50Pt50/C to 0.23 V vs. RHE for Pd/C.
In Fig. 1a voltammograms are shown for nanoparticles ranging from Pd50Pt50/C to
Pd5Pt95/C. In these voltammograms no shift of the oxide stripping peak is observed. However,
it can be clearly seen that as the amount of Pt is increased in the nanoparticles the behaviour
of the nanoparticles in the hydrogen adsorption and desorption region shifts to more Ptꢀlike
behaviour, eventually evolving the two distinctive hydrogen adsorption and desorption peaks
for Pd5Pt95/C.
3.2 Formic acid oxidation on PdxPt(100-x)/C nanoparticles
The electroꢀoxidation of formic acid on the synthesized PdxPt(100ꢀx)/C nanoparticles was
studied in an electrolyte containing 0.5 M H2SO4 and 0.5 M formic acid, and cyclic
voltammograms with a scanrate of 20 mV/s from 0.05 V to 1.0 V vs. RHE were recorded (see
Fig. 2a and Fig 2b). For the nanoparticles with compositions from Pd5Pt95/C to Pd50Pt50/C
similar current densities are measured for the oxidation of formic acid (see Fig. 2a). The
6
ACS Paragon Plus Environment

Page 7 of 21
ACS Catalysis
1
2
3
4
5
6
7
8
9
measured current was always smaller in the forward going scan than in the back scan. A shift
in the peak potential is seen for these nanoparticles, with a peak potential of 0.9 V vs. RHE
for Pd5Pt95/C going to a peak potential between 0.5 V and 0.6 V vs. RHE for Pd50Pt50/C.
In Fig. 2b the cyclic voltammograms for nanoparticles with compositions from Pd50Ptꢀ
50/C to Pd/C are shown. With an increasing amount of Pd a sharp peak in the measured
current density starts to develop. Whereas Pd50Pt50/C gave a voltammogram with a broad peak
for the oxidation of formic acid, Pd90Pt10 gives a sharp peak with a peak potential of 0.5 V vs.
RHE. In accordance with earlier reports the highest current densities are achieved with a
composition of Pd90Pt10/C, both nanoparticles with higher and lower Pd ratios give lower
current responses than Pd90Pt10/C.15,16 Interestingly, in our experiments the peak current
density for the oxidation of formic acid shifts to lower potentials with an increasing ratio of
Pd in the nanoparticles. This observation does not agree with the results of Cai et al., since
they reported the opposite trend for these nanoparticles with Pd50Pt50/C giving a peak at the
lowest potential.16 Comparing our voltammograms with other literature studies, we find that
our voltammogram for Pd50Pt50/C is very comparable to the results of Hsing et al.,15 where a
similar broad oxidation peak is seen that peaks between 0.5 V and 0.6 V vs. RHE.
The electrochemical behaviour of formic acid oxidation on the PdxPt(100ꢀx)/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
nanoparticles is explained by the hypothesis that in the forward going scan the surface is
blocked by adsorbed CO that is formed from the dehydration of formic acid.19 Once the
potential increases, the adsorbed CO is stripped from the surface giving rise to a peak in the
forward scan. The peak potential of this stripping of CO from the surface is dependent on the
composition of the nanoparticles. Since the surface sites that were previously blocked by CO
are now available to participate in the oxidation of formic acid, the current starts to increase
reaching a peak around 0.8 ꢀ 0.9 V vs. RHE. At this potential the surface of the nanoparticles
starts to oxidize, thus diminishing the activity towards formic acid oxidation. In the back scan,
the surface oxide is stripped off the surface, leading to an increasing formic acid oxidation
current. The current peak in the back scan is a consequence of the potential dependent formic
acid oxidation and the renewed blocking of the surface by adsorbed CO, which cannot be
stripped off at these low potentials.
Since the dehydration of formic acid to form CO is believed to take place on at least
two neighboring Pt sites (called the ‘ensemble effect’),20 an increase in the amount of Pd can
effectively reduce the number of these Pt ensembles, thereby decreasing the effect of CO
poisoning. Furthermore, the mixing of Pd and Pt will lead to a shift in the dꢀband center
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 21
1
2
3
4
5
6
7
8
9
values for the nanoparticles.21 Since the dꢀband center values for Pt and Pd are ꢀ2.25 and ꢀ1.83
eV, respectively, according to DFT calculations,22 the addition of Pt will lower the dꢀband
center value. Thus the dꢀband center values for the PdxPt(100ꢀx)/C nanoparticles can be finely
tuned, which has been hypothesized to have an additional beneficial effect on the formic acid
oxidation.16
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
3.3 CO2 reduction on PdxPt(100-x)/C nanoparticles
Blank cyclic voltammograms of the PdxPt(100ꢀx)/C were recorded in a pH 6.7 phosphate buffer
electrolyte that was purged with argon and are compared with cyclic voltammograms in the
same electrolyte purged with CO2 in Fig. 3 and Fig. S9. For all compositions of the PdxPt(100ꢀx)
nanoparticles a decrease in current is seen in the CO2 scan in comparison with the blank scan,
indicating that the nanoparticles have an interaction with CO2. The particles with high Pd
ratios show a bigger decrease in current between the blank and CO2 voltammogram, with the
exception of Pd95Pt5/C, indicating that those particles have a stronger interaction with CO2
than the particles with high ratios of Pt. In the CO2 scans of Pd/C and Pd95Pt5/C a reduction
wave between 0 V and ꢀ0.7 V vs. RHE is observed, similar to what has been seen on the Pd
layers deposited on Pt (PdꢀPt)9 and Pd electrodes.23 This reduction current is ascribed to the
ꢀ
either direct or indirect reduction of HCO3 to formate/formic acid. Since the pH near the
electrode surface will increase as the potential is decreased, due to continuous hydrogen
ꢀ
evolution,24 HCO3 will be generated in situ from CO2 near the electrode surface due to the
ꢀ
equilibrium between HCO3 and CO2. For CO2 reduction on the PdꢀPt layer system,9 we
observed that the reduction peak disappeared after the first scan, indicating a fast poisoning by
CO. For the Pd/C and Pd95Pt5/C nanocatalysts studied here, the same observation was made,
suggesting that CO poisoning for these nanoparticle compositions is fast as well.
The production of formic acid from CO2 as a function of potential was followed with
online HPLC (Fig. 4). The production of formic acid is normalized to the total mass of
nanoparticles that was dropcasted on the electrode. Except for Pd/C, all other PdxPt(100ꢀx)/C
nanoparticles had very comparable average sizes, thus this normalization should give a fair
comparison for the production of formic acid between the different nanoparticles. For all
nanoparticle compositions a peak in the production of formic acid is observed around ꢀ0.5 V
vs. RHE. The observation of this peak appears to be related to the presence of bicarbonate,
and has been observed on PdꢀPt layer systems as well.9 However, on PdꢀPt layer electrodes
production of formic acid was also observed at more negative potentials, starting form ꢀ0.9 V
8
ACS Paragon Plus Environment

Page 9 of 21
ACS Catalysis
1
2
3
4
5
6
7
8
9
vs. RHE, whereas the PdxPt(100ꢀx)/C nanoparticles show no production of formic acid at these
more negative potentials. The onset potential for formic acid production is around 0 V vs.
RHE for all the nanoparticle compositions, which is the same for the PdꢀPt electrodes. This
onset potential approaches the theoretical equilibrium potential of 0.02 V vs. RHE for the
production of formic acid from CO2 in this electrolyte.
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
We note that nanoparticles with low ratios of Pd, up to Pd20Pt80/C, produce only small
amounts of formic acid. With an increasing ratio of Pd in the nanoparticles, the production of
formic acid grows, reaching a maximum at Pd70Pt30/C, while Pd60Pt40/C and Pd80Pt20/C also
show a high production of formic acid. For nanoparticles with higher amounts of Pd, the
production of formic acid decreases again.
Fig. 5a shows the faradaic efficiencies for the production of formic acid on PdxPt(100ꢀ
x)/C nanoparticles after a period of 40 minutes and 2 hours at a potential of ꢀ0.5 V vs. RHE.
All the nanoparticle compositions show the production of formic acid over the course of 2
hours. Remarkably, the Pd95Pt5/C particles consistently gave faradaic efficiencies that were a
little higher than the trend observed in Fig. 5a, for reasons that are not well understood; the
faradaic efficiency data for this sample can be found in Table S3. We assume that besides
formic acid hydrogen is produced as the major side product, similar to PdꢀPt,9 which accounts
for the remainder of the faradaic efficiency. The amount of gaseous CO that is produced as a
side product at these potentials is assumed to be small, in correspondence to what has been
observed for Pd and Pt electrodes.25,26 We expect that the formation of strongly surfaceꢀbound
CO may still play a role in the slow deactivation of the catalyst. However, as was also shown
by Cai et al.16 for formic acid oxidation, these PdPt nanocatalysts have been designed to show
minimal CO poisoning, and we consider that this effect explains the considerably lower
poisoning rate of these PdPt nanocatalysts in comparison with the PdꢀPt layer system studied
in our previous paper.9 Similar to the online HPLC experiments, only small amounts of
formic acid are produced with nanoparticles with low amounts of Pd. The particles with a
composition of Pd20Pt80 or with lower amounts of Pd show efficiencies between 2ꢀ3% towards
formic acid. With increasing amount of Pd in the nanoparticles, an increase in the faradaic
efficiency towards formic acid is observed, reaching an optimum at Pd70Pt30/C, similar to the
results obtained with online HPLC. The Pd70Pt30/C nanoparticles are able to convert CO2 to
formic acid with a maximum faradaic efficiency of 67%. For nanoparticles with a higher
amount of Pd than Pd70Pt30/C the faradaic efficiency towards formic acid decreases again. It
should be noted however, that the faradaic efficiency of the Pd70Pt30/C nanoparticles, and also
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 21
1
2
3
4
5
6
7
of the other nanoparticle compositions, is not constant and gradually decreases over time. For
the Pd70Pt30/C nanoparticles this means that at the end of the electrolysis experiments, after 2
hours, the faradaic efficiency has dropped to 38%.
In an effort to avoid deactivation of the nanoparticles the potential for the reduction of
CO2 to formic acid on the Pd70Pt30/C nanoparticles was optimized (Fig. 5b). A clear optimum
in the applied potential was observed at ꢀ0.4 V vs. RHE. At this potential a maximum faradaic
efficiency of 88% towards formic acid was achieved after 1 hour of electrolysis. For
potentials higher than ꢀ0.4 V vs. RHE the maximum faradaic efficiency decreases to 73%,
22% and 21% for ꢀ0.3 V vs. RHE, ꢀ0.2 V vs. RHE and ꢀ0.1 V vs. RHE respectively. For all
potentials a deactivation was seen over the course of time. However, for ꢀ0.3, ꢀ0.2 and ꢀ0.1 V
vs. RHE the decrease in faradaic efficiency was less prominent, decreasing to 59%, 16% and
17% respectively, as compared to what was observed for ꢀ0.5 V vs. RHE and ꢀ0.4 V vs. RHE,
at which potentials a decrease to 38% and 58%, respectively.
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
Pd based catalysts have been previously reported to reduce CO2 or bicarbonate to
formic acid at low overpotentials.9,13,27 Electrodeposited Pd layers on Pt have been shown to
be able to produce formic acid almost selectively from CO2 at potentials close to 0 V vs.
RHE.9,13 However, since a fast poisoning effect diminishes the catalytic properties quickly,
this material does not satify the requirements for an applicable catalyst. With the alloying of
Pd with Pt we observe a drastic decrease in the poisoning rate, while the onset potential for
the production of formic acid and the potential at which the production of formic acid peaks
are not affected.
A PdCl42ꢀ impregnated polymer on a tungsten or platinum electrode has been shown to
be able to reduce bicarbonate to formic acid with an efficiency of 85% at approximately ꢀ0.1
V vs. RHE.27 This efficiency however was achieved at a current density of 100 ꢂA/cm2 which
decreased over the course of electrolysis to 50 ꢂA/cm2, while the authors state that “as the
electrolyses proceed the more negative potentials required to sustain current flow are a
consequence of the buildup of reduction product”, indicating that long term electrolysis at
high current densities with this catalyst will seriously increase the overpotential that is
necessary. Our catalyst is able to reduce CO2 to formic acid at current densities that are almost
two orders in magnitude higher. For the first 10 minutes of electrolysis with Pd70Pt30/C at ꢀ0.4
V vs. RHE we observe a current density of approximately 7 mA/cm2 which decreases to 3.5
mA/cm2 after 2 hours of electrolysis (see Fig. S6). Therefore, the significant advance of this
10
ACS Paragon Plus Environment

Page 11 of 21
ACS Catalysis
1
2
3
4
5
paper is that Pt is not only required to enhance the stability of the catalyst, but it is also
essential to achieve high faradaic efficiency at high current densities.
6
7
Compared to other catalysts reported in the recent literature, our best catalyst,
Pd70Pt30/C, shows considerably lower onset potentials for the reduction of CO2 to formic acid
and correspondingly higher faradaic efficiencies towards formic acid at these low
overpotentials. A Sn/SnOx tinꢀfilm catalyst reported by Chen et al.7 started producing formic
acid at ꢀ0.5 V vs. RHE and reached a maximum efficiency towards formic acid of 40% at ꢀ0.7
V vs. RHE, with CO accounting for the remaining 60%. Zhang et al.8 reported that the
production of formic acid from CO2 on nanostructured tin catalysts starts at ꢀ1.0 V vs. SCE (ꢀ
0.36 V vs. RHE) and reached a maximum faradaic efficiency of 86.2 % at ꢀ1.8 V vs. SCE (ꢀ
1.16 V vs. RHE). A homogeneous iridium pincer complex in a gas diffusion electrode
reported by Kang et al.28 started producing formic acid at an onset potential of ꢀ1.0 V vs. NHE
(ꢀ0.6 V vs. RHE) and achieved a faradaic efficiency of 96 % at ꢀ1.2 V vs. NHE (ꢀ0.8 V vs.
RHE) with a decreasing activity for higher potentials. Our catalyst thus achieves competing or
better efficiencies towards formic acid at substantially lower overpotentials. Moreover, our
catalyst also has a high activity for the reverse reaction, i.e. the oxidation of formic acid. This
catalytic reversibility has not been demonstrated for the catalysts reported in the literature,
other than for immobilized electroꢀactive enzymes.29
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
The comparison between the composition sensitivity of formic acid oxidation and CO2
reduction on the PdxPt(100ꢀx)/C nanoparticles shows a clear difference in optimal composition.
Whereas the optimum composition for formic acid oxidation is Pd90Pt10/C, the optimal
composition for CO2 reduction is Pd70Pt30/C, with Pd60Pt40/C and Pd80Pd20/C also showing
good efficiencies for the reduction of CO2 to formic acid. Therefore, the ensemble effect and
the breaking up of Pt ensembles must be less important for CO2 reduction than for formic acid
oxidation. We hypothesize that the fine tuning of the dꢀband center by mixing Pt and Pd may
also play a role in the enhancement of the catalytic activity. Futhermore, the addition of the
bigger Pt atoms can induce lattice strain effects in the nanoparticles, which can have an
additional beneficial effect.
Conclusions
In this paper, we have shown that PdxPt(100ꢀx)/C nanoparticles have a low onset potential for
the production of formic acid from carbon dioxide, with formic acid production starting from
0 V vs. RHE, very close to the theoretical equilibrium potential of 0.02 V vs. RHE, in
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 21
1
2
3
4
5
6
7
8
9
combination with very good selectivity. Depending on the composition of the nanoparticles
the faradaic efficiency towards formic acid can be very high, reaching a maximum of ca. 90%
for the Pd70Pt30/C nanoparticles at an applied potential of ꢀ0.4 V vs. RHE during one hour of
electrolysis. We show that this catalyst is able to avoid fast poisoning and is able to achieve
good faradaic efficiencies at high current densities, which was problematic with earlier
formulated Pd based catalysts. This catalyst is among the best catalysts to date with respect to
the onset potential for the formation of formic acid and it is able to achieve higher or
competing faradaic effiencies than other recently reported non Pdꢀbased catalysts at
substantially lower overpotential. Moreover, the catalyst obeys the reversibility principle in
the sense that it also catalyzes the reverse reaction with a low overpotential.
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
Acknowledgements
This work is supported by NanoNextNL, a micro and nanotechnology consortium of the
Government of the Netherlands and 130 partners. The authors gratefully acknowledge the
support of the Smart Mix Program of the Netherlands Ministry of Economic Affairs and the
Netherlands Ministry of Education, Culture and Science. Jos van Brussel is kindly
acknowledged for assisting with ICPꢀOES analysis and Gerda Lamers is kindly acknowledged
for assisting with TEM experiments.
Supporting Information Includes additional information on the synthesis and
characterization of the PdxPt(100ꢀx)/C nanoparticles, faradaic efficiency experiments and
voltammetric analysis of CO2 reduction on the nanoparticles. This material is available free of
charge via the Internet at http://pubs.acs.org.
References
1
2
3
4
Gattrell, M.; Gupta, N.; Co, A. Energy Convers. Manage., 2007, 48, 1255ꢀ1265.
Whipple, D.T.K.; Kenis, P.J.A. J. Phys. Chem. Lett. , 2010, 1, 3451ꢀ3458.
Yu, X.; Pickup, P.G. J. Power Sources, 2008, 182, 124ꢀ132.
Y. Hori in Modern Aspects of Electrochemistry; Vayenas, C.; White, R.E.; Gamboaꢀ
Aldeco M.E. Eds.; Springer: New York, 2008; Vol. 42, pp. 89ꢀ189.
Chaplin, R.P.S.; Wragg, A.A. J. Appl. Electrochem., 2003, 33, 1107ꢀ1123.
5
12
ACS Paragon Plus Environment

Page 13 of 21
ACS Catalysis
1
2
3
4
5
6
7
8
9
6
7
8
9
Kwon, Y.; Lee, J. Electrocatal., 2010, 1, 108ꢀ115.
Chen, Y.; Kanan, M.W. J. Am. Chem. Soc., 2012, 134, 1986ꢀ1989.
Zhang, S.; Kang, P.; Meyer, T.J. J. Am. Chem. Soc., 2014, 136, 1734ꢀ1737.
Kortlever, R.; Balemans, C.; Kwon, Y.; Koper, M.T.M. Catal. Today, 2015, 244, 58ꢀ
62.
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
10
11
12
13
Koper, M.T.M. J. Electroanal. Chem., 2011, 660, 254ꢀ260.
Yu, X.; Pickup, P.G. J. Appl. Electrochem., 2011, 41, 589ꢀ597.
Uhm, S.; Lee, H.J.; Lee, J. Phys. Chem. Chem. Phys., 2009, 11, 9326ꢀ9336.
Podlovchenko, B.I.; Kolyadko, E.A.; Lu, S. J. Electroanal. Chem., 1994, 373, 185ꢀ
187.
14
Liu, B.; Li, H.Y.; Die, L.; Zhang, X.H.; Fan, Z.; Chen, J.H. J. Power Sources, 2009,
186, 62ꢀ66.
15
16
Li, X.; Hsing, M. Electrochimica Acta, 2006, 51, 3477ꢀ3483.
Zhang, H.ꢀX.; Wang, C.; Wang, J.ꢀX.; Zhai, J.ꢀJ.; Cai, W.ꢀB. J. Phys. Chem. C, 2010,
114, 6446ꢀ6451.
17
18
Kwon, Y.; Koper, M.T.M. Anal. Chem., 2010, 82, 5420ꢀ5424.
Wang, W.; Huang, Q.; Liu, J.; Zou, Z.; Li, Z.; Yang, H. Electrochemistry Commun.,
2008, 10, 1396ꢀ1399.
19
Baranova, E.A.; Miles, N.; Mercier, P.H.J.; Le Page, Y.; Patarachao, B.
Electrochimica Acta, 2010, 55, 8182ꢀ8188.
20
21
Park, S.; Xie, Y.; Weaver, M.J.; Langmuir, 2002, 18, 5792ꢀ5798.
Greeley, J.; Nørskov, J.K.; Mavrikakis, M.; Annu. Rev. Phys. Chem., 2002, 53, 319ꢀ
348.
22
23
Ruban, A.; Hammer, B.; Stoltze, P.; Skriver, H.L.; Nørskov, J.K. J. Mol. Catal. A:
Chem., 1997, 115, 421ꢀ429.
SpichigerꢀUlmann, M.; Augustynski, J. J. Chem. Soc. Faraday Trans., 1985, 81, 713ꢀ
716.
24
25
26
Gupta, N.; Gattrell, M.; MacDougall, B. J. Appl. Electrochem., 2006, 36, 161ꢀ172.
Kolbe, D.; Vielstich, W. Electrochimica Acta , 1996, 41, 2457ꢀ2460.
Hori, Y.; Wakebe, H.; Tsukamoto, T.; Koga, O. Electrochimica Acta, 1994, 39, 1833ꢀ
1839.
27
Stalder, C.J.; Chao, S.; Wrighton, M.S. J. Am. Chem. Soc., 1984, 106, 3673ꢀ3675.
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 21
1
2
3
4
5
6
7
28
29
Kang, P.; Zhang, S.; Meyer, T.J.; Brookhart, M. Angew. Chem. Int. Ed., 2014, 53,
8709ꢀ8713.
Armstrong, F.A.; Hirst, J. PNAS, 2011, 108, 14049ꢀ14054.
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
14
ACS Paragon Plus Environment

Page 15 of 21
ACS Catalysis
1
2
3
4
Figures and tables
5
6
7
8
9
Table 1. Actual compositions of the synthesized PtxPd(100ꢀx)/C nanoparticles and their corresponding loading on
Vulcan XCꢀ72 as measured with ICPꢀOES and their average size as measured with XRD and calculated with the
Scherrer equation.
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
Calculated
composition
Pd5Pt95/C
Measured
composition
Pd4.5Pt95.5/C
Pd11.5Pt88.5/C
Loading
(wt%)
33.7
Particle
size (nm)
4.4
Pd10Pt90/C
24.4
3.8
Pd20Pt80/C
Pd30Pt70/C
Pd40Pt60/C
Pd50Pt50/C
Pd60Pt40/C
Pd70Pt30/C
Pd80Pt20/C
Pd90Pt10/C
Pd95Pt5/C
Pd/C
Pd20.9Pt79.1/C
Pd29.5Pt70.5/C
Pd39.3Pt60.7/C
Pd49.6Pt50.4/C
Pd59.7Pt40.3/C
Pd70.9Pt29.1/C
Pd78.8Pt21.2/C
Pd88.6Pt11.4/C
Pd94.1Pt5.9/C
Pd/C
21.4
17.2
32.3
16.5
15.6
20.1
16.4
11.8
14.8
12.0
4.1
4.1
3.8
4.0
3.7
4.3
4.2
4.2
4.1
3.1
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 21
1
2
3
4
5
6
7
8
a)
b)
400
0
-400
-800
-1200
-1600
400
0
9
-400
-800
Pd/C
Pd50Pt50/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
400
0
400
0
-400
Pd40Pt60/C
Pd95Pt5/C
-400
400
0
400
0
-400
-800
-1200
Pd30Pt70/C
Pd90Pt10/C
-400
400
0
400
0
-400
-800
Pd20Pt80/C
Pd80Pt20/C
-400
400
0
400
0
Pd10Pt90/C
Pd70Pt30/C
-400
-400
400
400
0
0
-400
-400
Pd5Pt95/C
Pd60Pt40/C
0.2 0.4 0.6 0.8 1.0 1.2
0.2 0.4 0.6 0.8 1.0 1.2
E (V) vs. RHE
E (V) vs. RHE
Figure 1. Cyclic voltammograms of the synthesized PdxPt(100ꢀx)/C nanoparticles in a 0.5 M H2SO4 solution at a
scan rate of 20 mV/s; a) Cyclic voltammograms of Pd5Pt95/C, Pd10Pt90/C, Pd20Pt80/C, Pd30Pt70/C, Pd40Pt60/C and
Pd50Pt50/C; b) Cyclic voltammograms of Pd60Pt40/C, Pd70Pt30/C, Pd80Pt20/C, Pd90Pt10/C, Pd95Pt5/C and Pd/C.
16
ACS Paragon Plus Environment

Page 17 of 21
ACS Catalysis
1
2
3
4
5
6
7
8
9
25
20
15
10
5
a)
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
0
Pd5Pt95/C
Pd10Pt90/C
Pd20Pt80/C
Pd30Pt70/C
Pd40Pt60/C
Pd50Pt50/C
0,0
0,2
0,4
0,6
0,8
1,0
E (V) vs. RHE
60
50
40
30
20
10
b)
0
Pd60Pt40/C
Pd70Pt30/C
Pd80Pt20/C
Pd90Pt10/C
Pd95Pt5/C
Pd/C
0,0
0,2
0,4
0,6
0,8
1,0
E (V) vs. RHE
Figure 2. Cyclic voltammograms recorded of PdxPt(100ꢀx)/C nanoparticles in a 0.5 M H2SO4 electrolyte containing
0.5 M formic acid at a scan rate of 20 mV/s; a) Cyclic voltammograms of Pd5Pt95/C (grey), Pd10Pt90/C (orange),
Pd20Pt80/C (purple), Pd30Pt70/C (blue), Pd40Pt60/C (green) and Pd50Pt50/C (red); b) Cyclic voltammograms of
Pd60Pt40/C (grey), Pd70Pt30/C (orange), Pd80Pt20/C (purple), Pd90Pt10/C (blue), Pd95Pt5/C (green) and Pd/C (red).
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 21
1
2
3
4
a)
0
-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
-8
-12
-16
Pd/C
-1.5
-1.0
-0.5
0.0
E (V) vs. RHE
b)
0
-4
-8
-12
-16
-20
Pd60Pt40/C
-1.5
-1.0
-0.5
0.0
E (V) vs. RHE
Figure 3. a) Cyclic voltammograms of Pd/C nanoparticles recorded in a 0.1 M KH2PO4 / 0.1 M K2HPO4
electrolyte (pH 6.7) at a scan rate of 50 mV/s purged with either argon (black) or CO2 (red); b) Cyclic
voltammograms of Pd60Pt40/C nanoparticles recorded in a 0.1 M KH2PO4 / 0.1 M K2HPO4 electrolyte (pH 6.7) at
a scan rate of 50 mV/s purged with either argon (black) or CO2 (red).
18
ACS Paragon Plus Environment

Page 19 of 21
ACS Catalysis
1
2
3
4
5
a)
6
7
8
9
200
150
100
50
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
0
20
40
60
80
100
b)
Pd %
Pd5Pt95/C
Pd10Pt90/C
Pd20Pt80/C
Pd30Pt70/C
Pd40Pt60/C
Pd50Pt50/C
Pd60Pt40/C
Pd70Pt30/C
Pd80Pt20/C
Pd90Pt10/C
Pd95Pt5/C
Pd/C
200
150
100
50
0
-1.5
-1.0
-0.5
E (V) vs. RHE
Figure 4. a) The maximum concentration of formic acid that was observed with online HPLC, normalized by the
mass of nanoparticles that was dropcasted on the electrode, as a function of the percentage of Pd in the
nanoparticles; b) Formation of formic acid as detected by online HPLC on PdxPt(100ꢀx)/C nanoparticles in a 0.1 M
K2HPO4 / 0.1 M KH2PO4 electrolyte.
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 21
1
2
3
4
5
6
7
8
9
a)
70
60
50
40
30
20
10
0
40 min.
2 hours
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
0
20
40
60
80
100
Pd %
b)
100
80
60
40
20
0
1 hour
2 hours
-0.5
-0.4
-0.3
-0.2
-0.1
E (V) vs. RHE
Figure 5. a) Faradaic efficiencies towards the production of formic acid on PdxPt(100ꢀx)/C nanoparticles in a pH
6.7 phospate buffer (0.1 M KH2PO4/ 0.1 M K2HPO4) at ꢀ0.5 V vs. RHE after 40 minutes (■) and after 2 hours
(●); b) Faradaic efficiencies towards the production of formic acid on Pd70Pt30/C nanoparticles in a pH 6.7
phospate buffer (0.1 M KH2PO4/ 0.1 M K2HPO4) at different potentials after 1 hour (■) and after 2 hours (●).
20
ACS Paragon Plus Environment

Page 21 of 21
ACS Catalysis
1
2
3
4
TOC Graphic
5
6
7
8
9
70
60
50
40
30
20
10
CO2
HCOOH
40 min.
2 hours
2(H+ + e-)
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
PdxPt(100-x)
0
0
20
40
60
80
100
Vulcan XC-72
Pd %
21
ACS Paragon Plus Environment