Pd nanoparticles on carbon nitride-graphene for the selective electro-oxidation of glycerol in alkaline solution

Article abstract of DOI:10.1021/acscatal.5b00183

A hybrid support consisting of carbon nitride and graphene (CN_x/G) is prepared by annealing polypyrrole/GO at 800 °C. Pd nanoparticles (NPs) are then loaded onto the support by a microwave-polyol method. Pd-CN_x/G is used as a catalyst for glycerol electro-oxidation in 0.5 M NaOH aqueous solution. Electrochemical characterization and product analysis by high-performance liquid chromatography show that, in comparison to Pd NPs supported on carbon black, Pd NPs on a support containing nitrogen atoms has the ability to promote selectivity toward C3 products as well as increase activity. The main factors influencing the selectivity and activity are also discussed.

Full text of DOI:10.1021/acscatal.5b00183

Subscriber access provided by NEW YORK UNIV
Article
Pd Nanoparticles on Carbon Nitride-Graphene for the
Selective Electro-Oxidation of Glycerol in Alkaline Solution
Haibo Wang, Larissa Thia, Nan Li, Xiaoming Ge, Zhaolin Liu, and Xin Wang
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.5b00183 • Publication Date (Web): 17 Apr 2015
Downloaded from http://pubs.acs.org on April 19, 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 19
ACS Catalysis
1
2
3
4
5
Pd Nanoparticles on Carbon Nitride-Graphene for the Selective
Electro-Oxidation of Glycerol in Alkaline Solution
6
7
8
9
Haibo Wang,† Larissa Thia,‡§ Nan Li,† Xiaoming Ge,¶ Zhaolin Liu,¶
Xin Wang*†
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
†
School of Chemical and Biomedical Engineering, Nanyang Technological University, 62
Nanyang Drive, Singapore, 637459
‡
Interdisciplinary Graduate School, Nanyang Technological University, 50 Nanyang Avenue,
Block S2ꢀB3aꢀ01, Singapore, 639798
¶ Institute of Materials Research and Engineering, Agency for Science, Technology and Research
(A*STAR), 3 Research Link, Singapore, 117602
* Corresponding author. Email: WangXin@ntu.edu.sg. Phone: (65) 6316 8866. Fax: (65) 6794
7553.
ABSTRACT: A hybrid support consisting of carbon nitride and graphene (CNx/G) is prepared
by annealing polypyrrole/GO at 800 oC. Pd nanoparticles (NPs) are then loaded onto the support
by a microwaveꢀpolyol method. PdꢀCNx/G is used as a catalyst for glycerol electroꢀoxidation in
0.5 M NaOH aqueous solution. Electrochemical characterization and product analysis by high
performance liquid chromatography show that, compared to Pd NPs supported on carbon black,
Pd NPs on support containing nitrogen atoms has the ability to promote selectivity towards three
carbon products as well as increase activity. The main factors influencing the selectivity and
activity are also discussed.
KEYWORDS glycerol oxidation, selectivity, carbon nitride, Pd nanoparticle, graphene
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 19
1
2
3
4
1. INTRODUCTION
5
6
7
8
Recently, the growth of the biodiesel industry has greatly increased the production of
glycerol.1, 2 Employing glycerol as fuel for the coꢀgeneration of valuable chemicals and
electricity has been proposed as a promising way to utilize glycerol. Although the mechanistic
study of glycerol electroꢀoxidation in alkaline solution began twenty years ago, few
investigations have been devoted to this subject due to its complexity.3ꢀ6 Electroꢀoxidation of
glycerol gives rise to various products, such as glyceric acid, hydroxypyruvic acid, mesoxalic
acid, tartronic acid, and other two or one carbon products via CꢀC bond breaking.7ꢀ9 The diverse
range of products renders it challenging to control both activity and selectivity during the electroꢀ
oxidation process.
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
Several strategies have been developed to improve product selectivity during glycerol electroꢀ
oxidation. One approach is via the mediation of other molecules and adatoms. For example, with
the presence of TEMPO (2,2,6,6ꢀtetramethylpiperidineꢀ1ꢀoxyl) in buffer solution (pH = 9.1), 1,3ꢀ
dihydroxyacetone (DHA) can be selectively oxidized from glycerol.10 Kwon et al. applied
several kinds of adatoms (Bi, Sb, Pb, In and Sn) on the surface of Pt and found that glycerol was
fully converted to DHA when Bi was present in the solution.11, 12 Selectivity can also be
controlled by tuning the operating conditions. Kim et al. achieved the glyceric acid selectivity of
o
87% in the continuousꢀflow electrocatalytic reactor at 60 C after 10 h.13 The group of Li
obtained high selectivity toward tartronic acid, glycolic acid, mesoxalic acid in a fuel cell when
temperature, potential and flow rate of the cell were optimized.14ꢀ17 In other studies, Pd alloyed
with other metals was used to investigate the electroꢀoxidation of glycerol. Simões et al. showed
different Pd alloys produced different products.18 Zalineeva at al. studied the electrocatalytic
behaviour of PdBi alloys during glycerol oxidation.19, 20 On the other hand, Marchionni et al.
2
ACS Paragon Plus Environment

Page 3 of 19
ACS Catalysis
1
2
3
4
5
6
loaded Pd nanoparticles (NPs) on (NiꢀZn)/C support. However, limited improvements in
selectivity were observed.21
7
8
9
Herein, we demonstrated that selectivity of Pd NPs for glycerol electroꢀoxidation could be
improved by tuning metal support interaction. Our previous study showed that extended polymer
chains can largely change the electronic property of Au NPs and greatly enhance the selectivity
to glyceric acid.22 As such, we developed a catalyst containing Pd NPs which are supported on
the surface of carbon nitride/graphene (PdꢀCNx/G) support which was annealed from
polypyrrole/GO (PPy/G). Existence of nitrogen in carbon materials could enhance the interaction
between metal NPs and support, and thus increase its activity in reactions such as the electroꢀ
oxidation of methanol, the liquid phase oxidation of alcohols and oxygen reduction.23ꢀ26 However,
there has been no report on the effect of metalꢀnitrogen interaction in the glycerol electroꢀ
oxidation. In this work, we found that the interaction between Pd NPs and nitrogen containing
carbon materials not only increased activity but also enhanced selectivity of three carbon
products during glycerol electroꢀoxidation. Furthermore, due to the high surface area of
graphene,27 the synthesized CNx/G results in Pd NPs of smaller sizes. This also benefits the
production of three carbon products.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. EXPERIMENTAL SECTION
Materials. Chemicals used in this work were purchased from SigmaꢀAldrich.
Preparation of catalysts. GO was synthesized from graphite flakes by using the modified
Hummers method.28 It includes two steps, the preꢀoxidation step and oxidation step. In the preꢀ
oxidation step, 2 g of K2S2O8 and 2 g of P2O5 were first dissolved into 8 ml of 98% H2SO4 at 80
oC after being stirred for 10 min. 2 g of graphite flakes was then added and the suspension was
o
kept stirring for 4.5 h at 80 C. After being cooled down to room temperature, the ingredients
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 19
1
2
3
4
5
6
7
8
9
were diluted with 300 ml of DI water and filtered with 0.2 µm Nylon membrane. The black
powder was washed with DI water three times and dried in air overnight. In the oxidation step,
the preꢀoxidized graphite flakes was added into 80 ml of 98% H2SO4. While keeping the
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
temperature at 5 C by ice bath, KMnO4 was gradually added under vigorous stirring. After
KMnO4 was fully dissolved, the suspension was stirred at 35 oC for 2 h. 160 ml of DI water was
o
then added under stirring while the temperature was maintained at 35 C for 2 h. After that, 460
ml of DI water and 13 ml of 30% H2O2 was added in sequence under vigorous stirring. The
suspension was placed overnight, then centrifuged and washed with 300 ml of 10% HCl and 300
ml of DI water in sequence. After the pH of suspension was almost neutral, 150 ml of DI water
was added. The solution was ultrasonicated for 30 min. Upper solution was collected and diluted
into 200 ml to obtain GO suspension. The final concentration was estimated at about 2 wt.%.
PPy/GO was synthesized according to the literature procedure with modification.29 Firstly, 7.2
g of cetyltrimethylammonium bromide (CTAB) was dissolved into 80 ml of 1 M HNO3 solution,
then 0.1 g GO dispersion (10 mg mlꢀ1) was added under stirring. After the temperature reached to
o
10 C, 0.8 g of pyrrole was added and stirred for 60 min. 20 ml of ammonium persulfate (APS,
2.7g, molar ratio of APS to pyrrole was 1:1) was added to initiate the polymerization. After 180
min, the solid product was filtered and washed with DI water and methanol in sequence, then
o
was vacuum dried at 40 C for one day. Wire structured PPy was synthesizd at the same
condition without the addition of GO.
To prepare PdꢀCNx/G, PdꢀCNx and PdꢀCB, firstly PPy/GO and PPy were annealed under Ar
o
atmosphere at 800 C for 2 h to obtain CNx/G and CNx, then 4 mg of annealed support and 1.6
ml of K2PdCl4/EG (3 mg mlꢀ1) was added into 10 ml of EG. The solution was ultrasonicated for
0.5 h. After adding 8.4 ml of EG and 0.2 ml of 2 M NaOH/EG, the suspension was microwaved
4
ACS Paragon Plus Environment

Page 5 of 19
ACS Catalysis
1
2
3
4
5
6
7
8
9
at medium power (300 W) for 110 s using a household microwave oven. The solution was
filtered and washed with ethanol several times after it was cooled down. The final product was
dried at 60 oC for one day.
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
Physicochemical characterization. The morphology was observed by using scanning electron
microscopy (SEM) at JEOL JSM6701F and transmission electron microscopy (TEM) at JEOL
JEMꢀ3010. The Xꢀray diffraction patterns were collected by Bruker D2 Phaser. The Pd loadings
in different samples were determined by inductively coupled plasma atomic emission
spectroscopy (ICPꢀAES) on a Perkin Elmer ICP Optima 2000DV instrument after samples were
dissolved by 36% HCl/ 69% HNO3 (v/v=3/1), Pd loading in PdꢀCB, PdꢀCNx and PdꢀCNx/G is
24.55%, 26.35% and 23.93%, respectively. The Xꢀray photoelectron spectroscopy (XPS) spectra
were collected at VG ESCALAB 200iꢀXL and calibrated by the C1s peak (284.6 eV). The BET
surface area of catalyst was measured by nitrogen adsorption/desorption at 77 K on the
13
Quantachrome Autosorbꢀ6B. C Nuclear magnetic resonance (NMR) spectroscopy was carried
out at Bruker Avance 300 NMR. To concentrate the sample, 5 ml original solution was stirred
and evaporated at 80 oC for several hours till the final volume reached around 0.4 ml. Then 1 ml
13
D2O was added to dissolve most of the precipitates. 0.7 ml mixed solution was taken for the C
NMR test.
Electrochemical characterization. Electrochemical tests of glycerol oxidation were carried out
at room temperature in a conventional threeꢀelectrode cell using Princeton Applied Research
VMP2 multichannel potentiostat. Pt coil and Hg/HgO (6 M KOH) were used as the counter and
reference electrode, a glassy carbon electrode with 5 mm diameter or carbon paper (CP, 1
×
1 cm2,
Gashub) was used as the working electrode. Glassy carbon based working electrode was
prepared as follows: firstly 1 mg of the catalyst was dispersed in 1.45 ml of ethanol solution by
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 19
1
2
3
4
5
6
7
8
9
ultrasonicating for 5 min; then 0.05 ml of 0.5 wt.% Nafion solution was added and ultrasonicated
for 10 min to obtain the black dispersion. After that, 30 µl of dispersion was dropped on the
glassy carbon electrode and dried to reach a loading of 0.01 mg cmꢀ2. For the preparation of CP
based working electrode, 1 mg of the catalyst was dispersed in 0.3 ml of ethanol and 0.05 ml of
0.5 wt.% Nafion solution, then the dispersion was dropped onto CP to prepare sample with the
loading of 1 mg cmꢀ2. After being dried, the working electrode was firstly immersed in 0.5 M
NaOH + 0.5 M glycerol under stirring for 10 h to fully saturate the electrode, then washed
carefully with DI water. It was immersed inside 20 ml of 0.5 M NaOH + 0.5 M glycerol solution
during the test. Then 1 ml of solution was collected for the high performance liquid
chromatography (HPLC) analysis after applying the specific potential on the working electrode
for 2 h.
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 measurement was carried out in 0.5 M HClO4 by using Ag/AgCl (saturated KCl)
as the reference electrode. After holding the potential at 0 V for 40 min while purging CO into
the solution, the glassy carbon electrode was removed out of the solution and immersed into
fresh 0.5 M HClO4 which was pretreated by purging N2 for 30 min. CV then was collected by
scanning from ꢀ0.166 V to 1.2 V at a scan rate of 50 mV sꢀ1. The oxidation charge of CO
monolayer on Pd surface is assumed as 420 µC cmꢀ2.
Chromatographic analysis of products. The obtained products were analyzed by using Alltech
OAꢀ1000 Organic Acid column for high performance liquid chromatography (HPLC, Agilent
1260 Infinity). The eluent was sulfuric acid (4 mM). During the test, 10 ꢁl of sample was
injected into the column and the temperature of the column was kept at 80 °C. The flow rate was
0.2 ml minꢀ1. The separated compounds were detected with both the refractive index (RID)
6
ACS Paragon Plus Environment

Page 7 of 19
ACS Catalysis
1
2
3
4
5
6
7
8
9
detector and UVꢀVis detector. The standard solutions of the various products from glycerol
oxidation were prepared and analysed at the same condition to produce the calibration curves.
3. RESULTS AND DISCUSSION
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. N1s XPS spectra of PPy/GO and CNx/G which is obtained by annealing PPy/GO at 800 oC.
CNx/G and CNx were synthesized from PPy/GO and wire structured PPy respectively by
annealing them at high temperature (Figure S1). In Figure S1a, it was observed that wire
structured CNx experienced some degree of aggregation. While in the case of CNx/G, CNx is
loaded on the surface of graphene, which possesses a larger surface area. This is confirmed by
the calculated BET surface area (Figure S2). In XPS spectra, the N1s peak of CNx/G can be
deconvoluted into four peaks (Figure 1), with pyridinic N, pyrrolic N, quaternary N and pyridine
oxide at 398.2, 399.7, 400.8 and 402.1 eV, respectively.30 Compared to the N 1s peak of PPy/GO,
CNx/G shows the obvious increment in the pyridinic N and quaternary N peaks while
simultaneously showing a sharp decrement in the pyrrolic N. This demonstrates that most of
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 19
1
2
3
4
5
6
pyrrolic N in polypyrrole has been converted into pyridinic N and quaternary N after high
temperature annealing, resulting in the formation of carbon nitride materials.
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. TEM images of (a) PdꢀCB, (b) PdꢀCNx, (c) PdꢀCNx/G, insets are the corresponding particle size
histograms. (d) Xꢀray diffraction (XRD) patterns of PdꢀCB, Pdꢀ CNx and Pdꢀ CNx/G.
Pd NPs were then loaded onto the respective supports via the microwaveꢀpolyol method. XRD
patterns of PdꢀCB, PdꢀCNx and PdꢀCNx/G show the typical faceꢀcentered cubic (fcc) structure of
Pd (Figure 2d). Peaks present at 40.2o, 46.8o, 68.2o, 82.1o can be assigned to the (111), (200),
(220) and (311) facets, respectively. Pd NPs on the CNx/G support show lower peak intensity
compared with Pd NPs on CNx and CB. This indicates smaller particle size of Pd NPs on CNx/G.
From the Scherrer formula, we estimate that mean particle size of PdꢀCB, PdꢀCNx and PdꢀCNx/G
obtained from XRD patterns is 7.5 nm, 7.6 nm and 6.9 nm respectively.
8
ACS Paragon Plus Environment

Page 9 of 19
ACS Catalysis
1
2
3
4
5
6
7
8
9
TEM analysis was carried out to further confirm the different size distributions (Figure 2a, 2b,
2c). It is observed that the size of Pd NPs on CNx/G support is smaller than those present on the
CB and CNx support. The mean diameter of Pd NPs on CNx/G is 4.40 nm while it is 5.79 nm on
CB and 6.30 nm on CNx. From nitrogen adsorptionꢀdesorption isotherms (Figure S2), the BET
surface area of CNx, CNx/G and CB is calculated as 151, 168, 194 m2/g, respectively. Since CNx
has the smallest surface area, Pd NPs formed on its surface shows the largest size. Compared to
CNx, CNx/G shows larger surface area. Although its surface area is still lower than CB, the
existence of nitrogen atoms might faciliate the Pd NPs on CNx/G to exhibit the smallest size
distribution.
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. CO stripping voltammetry of (a) PdꢀCB, (b) PdꢀCNx and (c) PdꢀCNx/G in 0.5 M HClO4 solution at a
scan rate of 50 mV sꢀ1. (d) CO stripping peaks plotted for different samples, the current has been derived by
subtracting the first cycle form the second one.
In the TEM images (Figure 2), the aggregation of Pd NPs is observed in all these samples due
to the high loading. In order to examine the extent of aggregation, we studied the Pd utilization
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 19
1
2
3
4
5
6
7
8
9
of these catalysts. CO stripping was carried out to obtain the electrochemical active surface area
(EASA) (Figure 3a, 3b, 3c, 3d). CO stripping is a common way to calculate the EASA of Pd
through oxidation of absorbed CO. As shown in Figure 3a, 3b and 3c, an obvious CO stripping
peak at around 0.75 V is observed. The calculated ECSA of PdꢀCB, PdꢀCNx and PdꢀCNx/G is
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
274, 258, 318 cm2 mgꢀ1, respectively. Pd NPs on CNx/G support showing the largest EASA.
3
ꢀ
Chemical surface area (CSA) is derived from the equation CSA=6/(ρPd*D) (where ρPd is 12 g/cm
ꢀ
and D is the mean diameter) (Table 1) and Pd utilization is derived from the equation Pd
utilization=EASA/CSA. PdꢀCNx/G shows the smallest Pd utilization while PdꢀCNx and PdꢀCB
show similar higher Pd utilization. This means that Pd NPs on CNx/G experienced the largest
degree of aggregation, although the differences among these catalysts are small.
Table 1 The particle size, ECSA, CSA and utilization of catalysts
Sample
Particle
ECSA /
CSA / cm2
ηPd
/ %
size / nm cm2 mgꢀ1
mgꢀ1
Pd
Pd
PdꢀCB
PdꢀCNx
PdꢀCNx/G
5.79
6.3
4.4
283
258
318
863
793
1136
32.8
32.5
28.0
10
ACS Paragon Plus Environment

Page 11 of 19
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 4. (a) Cyclic voltammograms (CV) and (b) accelerated durability test of PdꢀCB, PdꢀCNx, PdꢀCNx/G in
0.5 M glycerol + 0.5 M NaOH with a scan rate of 50 mV/s. In the accelerated durability test, the potential scan
was performed from ꢀ0.8 V to 0.5 V and data were collected after 20 cycles.
Electrocatalytic activity of the catalysts towards glycerol oxidation in the alkaline solution was
studied first using a glassy carbon working eletrode. It was observed that PdꢀCNx and PdꢀCNx/G
show higher massꢀspecific peak current density than PdꢀCB (Figure 4a). They also exhibited
higher peak current densities normalized by ECSA (Figure S3). Taking into account that Pd NPs
on CNx had the largest mean diameter whilst Pd supported on CNx/G had the smallest diameters,
the high peak current density of both these catalysts indicates that the interaction between Pd
NPs and nitrogen atoms is the dominant contributing factor for high electrocatalytic activity
instead of particle size. The accelerated durability test of these catalysts in the same setup
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 19
1
2
3
4
5
6
7
8
9
indicates Pd NPs on CNx/G show slightly better activity (in terms of massꢀspecific peak current
density) througout the test period. Improved activity and stability of PdꢀCNx/G makes it a
potential catalyst for glycerol oxidation.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5. Chronoamperometry curves of PdꢀCB, PdꢀCNx and PdꢀCNx/G at (a) 0 V and (b) 0.6 V with the
loading of 1 mg cmꢀ2 at 1 cm2 carbon paper. Inset is the enlarged curves of PdꢀCNx and PdꢀCNx/G.
Catalytic activity was also examined at the loading of 1 mg cmꢀ2 on carbon paper based
working electrode. It is observed that the current density of PdꢀCNx/G is much higher than that of
PdꢀCB (~ 1.5 times) at the potential of 0 V (Figure 5a). From this we can conclude that PdꢀCNx/G
demonstrates higher activity than PdꢀCB. Tests at other potentials also support this point (Figure
S4). However, when the potential increases to 0.6 V, current density of catalysts drops sharply
(Figure 5b). This was also observed in a previous study where Pd was used as a catalyst in a
direct formate fuel cell.31 This is due to the formation of Pd oxide on surface,32 which drastically
decreases the activity of catalysts. Interestingly, the concentration of formic acid reaches its
highest value while other components decrease drastically when the deactivation occurs (Figure
S5). This means that glycerol is mainly oxidized to formic acid (~70 %) on the surface of Pd
oxide at high potentials.
12
ACS Paragon Plus Environment

Page 13 of 19
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 6. (a) The product selectivity of Pd NPs on different supports (CB, CNx, CNx/G) at various potentials.
The catalyst loading is 0.1 mg cmꢀ2 and working electrode is glassy carbon. (b) The mechanistic scheme of
glycerol oxidation.
The selectivity of different catalysts was studied on the glassy carbon based electrode (Figure
6a). When using PdꢀCB as catalyst, the concentration of formic acid is much higher than the total
concentration of glycolic acid and oxalic acid (~10%ꢀ20% higher). However, the amount of
formic acid produced is far more than the total amount of glycolic acid and oxalic acid,
espeically considering that some of the formic acid produced can be converted to
carbonate.21,33,34 We have also conducted carbon balance analysis on the glycerol electroꢀ
oxidation process and the results are listed in Table S1. It can be seen that there is around 20%
carbon deficiency. Since HPLC was used for the determination of various products and the
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 19
1
2
3
4
5
6
7
8
9
relevant errors introduced during the sampling and measurement are within 3%, the observed
carbon deficiency is mainly due to the formation of carbonate ion. The formation of carbonate
ion is confirmed by the peak at a chemical shift of 168 ppm in 13C NMR spectrum (Figure S6).21
Unfortunately, HPLC cannot detect the formation of carbonate or CO2. Nevertheless, the relative
ratio among all the products (i.e. selectivity) obtained will remain valid (Table S2). In the
oxidation process, it is likely that some glycolic acid was oxidized into formic acid through path
a (Figure 6b) on PdꢀCB. While in the case of PdꢀCNx/G, the amount of formic acid is slightly
lower than the total amount of glycolic acid and oxalic acid. This means that path a is
suppressed with PdꢀCNx/G. Moreover, the concentration of oxalic acid is slightly higher than
that on PdꢀCB, suggesting that the oxidation of glycolic acid on the PdꢀCNx/G favors the path of
oxalic acid.
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
From the product analysis, PdꢀCNx/G shows enhanced selectivity towards glyceric acid (~10ꢀ
15% higher) and reduced selectivity towards formic acid (~15% lower) as compared to PdꢀCB.
Although the concentration of oxalic acid also increases, the ratio between three carbon products
and other products in PdꢀCNx/G is still much higher than that in PdꢀCB (1.8 times at 0 V and 0.2
V and 3.4 times at 0.4 V compared to PdꢀCB). The ratio of three carbon products in PdꢀCNx/G is
also higher than PdꢀCNx (Figure 6a). This is likely due to the smaller size of Pd NPs on CNx/G
compared to the latter. Moreover, despite of its larger particle size, higher selectivity towards
three carbon products on PdꢀCNx than that on PdꢀCB was observed. This indicates that the
interaction between Pd NPs and nitrogen atoms plays a more important role in the improvement
of selectivity. As shown in previous study, the interaction between metal NPs and adatomꢀdoped
carbon support as well as the metal particle size can tailor the electronic property of metal
nanoparticles.35, 36 This may weak the adsorption ability of metal NPs towards three carbon
14
ACS Paragon Plus Environment

Page 15 of 19
ACS Catalysis
1
2
3
4
5
6
7
8
9
products, especially glyceric acid. Considering the competitive adsorption among three carbon
products and other species, weaker adsorption means the oxidation of three carbon products are
not preferred compared with other species and more three carbon products are obtained.
Therefore, these two factors enable PdꢀCNx/G to be more selective towads three carbon products
compared to the other two catalysts.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
4. CONCLUSIONS
Pd NPs supported on CNx/G is developed as an electrocatalyst for glycerol oxidation. The
larger surface area of CNx/G compared to CNx results in smaller sized Pd nanoparticles being
loaded on its surface. Due to the smaller particle size and nitrogen interaction with Pd NPs, Pdꢀ
CNx/G shows better activity and higher selectivity towards three carbon product. Product
analysis shows that part of glycolic acid is oxidized into formic acid when PdꢀCB is used as
catalyst. However, this path is likely to be suppressed when PdꢀCNx/G was used as the catalyst.
This causes a decrease in the formation of formic acid in PdꢀCNx/G. Besides, compared to Pdꢀ
CNx which having larger particles size, smaller size of Pd NPs on CNx/G shows improved
selectivity towards three carbon products.
ACKNOWLEDGMENTS
This project is funded by the National Research Foundation (NRF), Prime Minister’s Office,
Singapore under its Campus for Research Excellence and Technological Enterprise (CREATE)
programme. We also acknowledge financial support from the academic research fund AcRF tier
1 (M4011020 RG8/12 and M4011253 RG 7/14) Ministry of Education, Singapore.
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 19
1
2
3
4
Supporting Information Available
5
6
7
8
SEM images of CNx and CNx/G, nitrogen adsorptionꢀdesorption isotherms of different
supports, cyclic voltammograms of catalysts normalized by ECSA, chronoamperometry curves
at 0.2 and 0.4 V, concentration of products of catalysts produced at different potentials, carbon
deficiency of the glycerol oxidation over different catalysts at the selected potentials, the ratio of
different products if assuming all other carbons are oxidized into carbonate ion and catalysts with
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
13
different loading have same carbon deficiency, C NMR spectra of solution using PdꢀCB as
catalyst at the potential of 0.2 V and loading of 1 mg cmꢀ2. This material is available free of
charge via the Internet at http://pubs.acs.org.
16
ACS Paragon Plus Environment

Page 17 of 19
ACS Catalysis
1
2
3
4
REFERENCES
5
6
7
8
(1) Bianchini, C.; Shen, P. K. Chem. Rev. 2009, 109, 4183ꢀ4206.
(2) Yu, E. H.; Krewer, U.; Scott, K. Energies 2010, 3, 1499ꢀ1528.
(3) AvramovꢀIvić, M. L.; Leger, J. M.; Lamy, C.; Jović, V. D.; Petrović, S. D. J. Electroanal.
Chem. 1991, 308, 309ꢀ317.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(4) AvramovꢀIvic, M.; Léger, J. M.; Beden, B.; Hahn, F.; Lamy, C. J. Electroanal. Chem. 1993,
351, 285ꢀ297.
(5) Roquet, L.; Belgsir, E. M.; Lbger, J. M.; Lamy, C. Electrochim. Acta 1994, 39, 2387ꢀ2394.
(6) Schell, M.; Xu, Y.; Zdraveski, Z. J. Phys. Chem. 1996, 100, 18962ꢀ18969.
(7) Bambagioni, V.; Bianchini, C.; Marchionni, A.; Filippi, J.; Vizza, F.; Teddy, J.; Serp, P.;
Zhiani, M. J. Power Sources 2009, 190, 241ꢀ251.
(8) Kwon, Y.; Schouten, K. J. P.; Koper, M. T. M. ChemCatChem 2011, 3, 1176ꢀ1185.
(9) Qi, J.; Xin, L.; Chadderdon, D. J.; Qiu, Y.; Jiang, Y.; Benipal, N.; Liang, C.; Li, W. Appl.
Catal. B: Environ. 2014, 154ꢀ155, 360ꢀ368.
(10) Ciriminna, R.; Palmisano, G.; Pina, C. D.; Rossi, M.; Pagliaro, M. Tetrahedron Lett. 2006,
47, 6993ꢀ6995.
(11) Kwon, Y.; Birdja, Y.; Spanos, I.; Rodriguez, P.; Koper, M. T. M. ACS Catal. 2012, 2, 759ꢀ
764.
(12) Kwon, Y.; Hersbach, T. J. P.; Koper, M. T. M. Top. Catal. 2014, 57, 1272ꢀ1276.
(13) Kim, H. J.; Lee, J.; Green, S. K.; Huber, G. W.; Kim, W. B. ChemSusChem 2014, 7, 1051ꢀ
1054.
(14) Xin, L.; Zhang, Z.; Wang, Z.; Li, W. ChemCatChem 2012, 4, 1105ꢀ1114.
(15) Zhang, Z.; Xin, L.; Li, W. Appl. Catal. B: Environ. 2012, 119ꢀ120, 40ꢀ48.
(16) Zhang, Z.; Xin, L.; Qi, J.; Wang, Z.; Li, W. Green Chem. 2012, 14, 2150ꢀ2152.
(17) Zhang, Z.; Xin, L.; Qi, J.; Chadderdon, D. J.; Sun, K.; Warsko, K. M.; Li, W. Appl. Catal.
B: Environ. 2014, 147, 871ꢀ878.
(18) Simões, M.; Baranton, S.; Coutanceau, C. Appl. Catal. B: Environ. 2010, 93, 354ꢀ362.
(19) Coutanceau, C.; Zalineeva, A.; Baranton, S.; Simoes, M. Int. J. Hydrogen Energy 2014, 39,
15877ꢀ15886.
(20) Zalineeva, A.; Serov, A.; Padilla, M.; Martinez, U.; Artyushkova, K.; Baranton, S.;
Coutanceau, C.; Atanassov, P. B. J. Am. Chem. Soc. 2014, 136, 3937ꢀ3945.
(21) Marchionni, A.; Bevilacqua, M.; Bianchini, C.; Chen, Y. X.; Filippi, J.; Fornasiero, P.;
Lavacchi, A.; Miller, H.; Wang, L.; Vizza, F. ChemSusChem 2013, 6, 518ꢀ528.
(22) Wang H.; Thia L.; Li N.; Ge X.; Liu Z.; Wang, X. Appl. Catal. B: Environ. 2015, 166–167,
25–31.
(23) Ma, Y.; Jiang, S.; Jian, G.; Tao, H.; Yu, L.; Wang, X.; Wang, X.; Zhu, J.; Hu, Z.; Chen, Y.
Energy Environ. Sci. 2009, 2, 224ꢀ229.
(24) Imran Jafri, R.; Rajalakshmi, N.; Ramaprabhu, S. J. Mater. Chem. 2010, 20, 7114ꢀ7117.
(25) Wang H., Maiyalagan T., Wang X. ACS Catal. 2012, 2, 781ꢀ794.
(26) ChanꢀThaw, C. E.; Villa, A.; Veith, G. M.; Kailasam, K.; Adamczyk, L. A.; Unocic, R. R.;
Prati, L.; Thomas, A. Chem. Asian J. 2012, 7, 387ꢀ393.
(27) Stoller, M. D.; Park, S.; Zhu, Y.; An, J.; Ruoff, R. S. Nano Lett. 2008, 8, 3498ꢀ3502.
(28) Kovtyukhova, N. I.; Ollivier, P. J.; Martin, B. R.; Mallouk, T. E.; Chizhik, S. A.; Buzaneva,
E. V.; Gorchinskiy, A. D. Chem. Mater. 1999, 11, 771ꢀ778.
(29) Zhang, L. L.; Zhao, S.; Tian, X. N.; Zhao, X. S. Langmuir 2010, 26, 17624ꢀ17628.
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 19
1
2
3
4
5
6
7
8
9
(30) Xin, L.; Zhang, Z.; Wang, Z.; Li, W. ChemCatChem 2012, 4, 1105ꢀ1114.
(31) Bartrom, A. M.; Haan, J. L. J. Power Sources 2012, 214, 68ꢀ74.
(32) Wang, Y.; Wang, X.; Li, C. Appl. Catal. B: Environ. 2010, 99, 229ꢀ234.
(33) Bambagioni, V.; Bevilacqua, M.; Bianchini, C.; Filippi, J.; Marchionni, A.; Vizza, F.; Wang,
L. Q.; Shen, P. K. Fuel Cells 2010, 10, 582ꢀ590.
(34) Lin, J.; Ren, J.; Tian, N.; Zhou, Z.; Sun, S. J. Electroanal. Chem. 2013, 688, 165ꢀ171.
(35) Holme, T.; Zhou, Y.; Pasquarelli, R.; O’Hayre, R. Phys. Chem. Chem. Phys. 2010, 12,
9461ꢀ9468.
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
(36) van Bokhoven, J. A.; Miller, J. T. J. Phys. Chem. C 2007, 111, 9245ꢀ9249.
18
ACS Paragon Plus Environment

Page 19 of 19
ACS Catalysis
1
2
3
4
TOC
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
19
ACS Paragon Plus Environment