Catalysis of multi-walled carbon nanotubes supported pdxcoy nanoparticles prepared by a pyrolysis method using ionic liquids as the solvent toward ethanol oxidation reaction

Article abstract of DOI:10.1002/jccs.201300009

Multi-walled carbon nanotubes (MWCNTs) decorated with PdxCoy (the nominal atomic ratios of Pd to Co were 3:1, 3:1.5, 3:2, 3:3, respectively) nanoparticles (denoted as PdxCoy/MWCNTs ) were fabricated by a simple pyrolysis process, in which room temperature ionic liquids (RTILs) of butyl-3-methylimidazolium hexafluorophosphate (denoted as [BMIM]PF6) was used as the solvent. X-ray diffraction (XRD) and transmission electron microscopy (TEM) were all used to characterize the PdxCoy/MWCNTs catalysts, showing that the PdxCoy particles were dispersed on the surface of theMWCNTs with an average particle size of ～25.0 nm. The electro-catalytic activity of the PdxCoy/MWCNTs catalysts toward ethanol oxidation reaction (EOR) was examined by cyclic voltammetry (CV). It was revealed that the onset potential was ～90 mV lower and the peak current was about four times higher for ethanol oxidation for Pd3Co1.5/MWCNTs compared to those of Pd3Co1/MWCNTs. The possible catalysis mechanisms of the Pd3Co1.5/MWCNTs toward EOR were also discussed.

Full text of DOI:10.1002/jccs.201300009

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
Article
Catalysis of Multi-walled Carbon Nanotubes Supported Pd Co Nanoparticles
x
y
Prepared by a Pyrolysis Method Using Ionic Liquids as the Solvent toward Ethanol
Oxidation Reaction
a
a
a
a
a
a
a
Hongwei Yang, Yahui Wang, Yanli Cao, Chunbao Zheng, Lu Liu, Likun Liu, Keqiang Ding, *
b
b
b
Sowjanya B. Rapole, Yiran Wang and Zhanhu Guo *
College of Chemistry and Materials Science, Hebei Normal University, Shijiazhuang, 050024, P. R. China
a
b
Integrated Composites Laboratory (ICL), Dan F. Smith Department of Chemical Engineering, Lamar University,
Beaumont, TX 77710, USA
(Received: Jan. 4, 2013; Accepted: Apr. 1, 2013; Published Online: ??; DOI: 10.1002/jccs.201300009)
Multi-walled carbon nanotubes (MWCNTs) decorated with Pd
were 3:1, 3:1.5, 3:2, 3:3, respectively) nanoparticles (denoted as Pd
simple pyrolysis process, in which room temperature ionic liquids (RTILs) of butyl-3-methylimidazolium
hexafluorophosphate (denoted as [BMIM]PF ) was used as the solvent. X-ray diffraction (XRD) and
transmission electron microscopy (TEM) were all used to characterize the Pd Co /MWCNTs catalysts,
showing that the Pd Co particles were dispersed on the surface of the MWCNTs with an average particle
size of ~25.0 nm. The electro-catalytic activity of the Pd Co /MWCNTs catalysts toward ethanol oxida-
x
Co
y
(the nominal atomic ratios of Pd to Co
x
Co /MWCNTs ) were fabricated by a
y
6
x
y
x
y
x
y
tion reaction (EOR) was examined by cyclic voltammetry (CV). It was revealed that the onset potential
was ~90 mV lower and the peak current was about four times higher for ethanol oxidation for
Pd
Pd
3
Co1.5/MWCNTs compared to those of Pd
Co1.5/MWCNTs toward EOR were also discussed.
3 1
Co /MWCNTs. The possible catalysis mechanisms of the
3
Keywords: Pd
INTRODUCTION
x y
Co nanoparticles; Ionic liquids; Ethanol oxidation reaction (EOR).
8
reduction reaction. For example, Zhang et al. described
the preparation of Pd-Co alloys dispersed on Vulcan-XC-
72R carbon black powder by an ultrasonic-assisted chemi-
cal reduction approach, in which ammonium hexachloro-
palladate and cobalt nitrate were employed as the starting
materials, and sodium borohydride the reducing agent.
Recently, the ethanol oxidation reaction (EOR) has
gained much attention in the research field of direct liquid
fuel cells, mainly due to its low toxicity, abundant avail-
ability, low permeability (but not negligible) across proton
exchange membrane and higher energy density (8030 Wh
-
1
-1
1
9
kg ) compared to that (6100 Wh kg ) of methanol. Thus,
developing novel catalysts for EOR has become into an im-
portant research topic in electrochemistry. Although plati-
num (Pt) has been recognized as the most active catalyst for
ethanol oxidation, the high cost and limited supply of Pt
have constituted a major limitation to the development of
Wang and co-workers reported the preparation of Pd-Co
alloy catalysts, in which Pd(NO
3
)
2
and Co(NO
3
)
2
·6H
2
O
were used as the precursors, and the flowing H
2
/N
2
gas as
the reducing agent. The second method is the electrochemi-
1
0
cal reduction reaction. For instance, Osaka et al. electro-
deposited two types of Pd-Co films onto Au substrate by
2
,3
direct ethanol fuel cells (DEFC). Therefore, Pt-free cata-
applying different current density, where [Pd(NH
and CoCl ·6H O were utilized as the starting materials. To
the best of our knowledge, the preparation of Pd Co com-
3 4 2
) ]Cl
4
lysts such as Pd have been intensively studied. Among the
2
2
developed Pd-based catalysts, the binary composite parti-
cles of Pd-Co were rarely probed as a catalyst for EOR, es-
pecially in an alkaline solution, though the composite nano-
particles of Pd-Co have been widely used as catalysts for
x
y
posite NPs via a facile method of pyrolysis using the ionic
liquids as solvent, has been rarely reported yet, though
many papers concerning Pd
lished so far.
x y
Co composite NPs were pub-
5
,6
oxygen reduction reaction (ORR) and formic acid oxida-
7
tion (FAO).
Although its excellent features, like low-volatility,
non-toxicity, inflammability, higher conductivity com-
pared to common organic solvents, and higher solubility
Till present, there are two typical methods for synthe-
sizing Pd-Co nanoparticles (NPs). The first is the chemical
*
Corresponding author. K. Ding, Tel: +86-311-80787400; Fax: +86-311-80787403; Email: dkeqiang@263.net;
Z. Guo, Tel: (409) 880-7654/7195; Fax: (409) 880-2197 E-mail: zhanhu.guo@lamar.edu
J. Chin. Chem. Soc. 2013, 60, 000-000
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1

Article
Yang et al.
for organic substances when compared with aqueous solu-
tions have been well addressed, the application of room
and (220) located at 2q values of about 39.9, 47 and 68º, re-
spectively, are the characteristic diffraction peaks of face
centered cubic (fcc) crystalline Pd (JCPDS, Card No.05-
11
temperature ionic liquids (RTILs) is still very limited. To
our knowledge, RTILs are mainly applied in the following
1
4,21
0681), being consistent with the previous report
very
1
2
fields of chemistry, (1) as solvents in organic synthesis,
well. Also, no diffraction peaks of Co or its oxides/hydrox-
ides are observed, suggesting that Co particles or its ox-
1
3
and (2) as electrolytes in electrochemistry. Unfortunately,
the application of RTILs on a large scale has not been
achieved. In other words, extending the application of
RTILs is still a main challenge for chemical researchers.
2
2
ides/hydroxides exist in amorphous phases. It can also be
seen that the intensities of all diffraction peaks changed
correspondingly when the atomic ratios of Pd to Co were
varied. For example, as shown by the green-circled part,
1
4
Meanwhile, to our knowledge, except our previous work,
the application of RTILs in pyrolysis process has been
rarely reported.
the intensity of the diffraction peak at 39.9º for Pd
MWCNTs is obviously higher than that of Pd Co
Pd Co , indicating that the crystallinity of Pd
MWCNTs is superior to the other catalysts. For the cata-
lyst of Pd Co , no diffraction peaks corresponding to Pd are
3
Co1.5/
3
1
and
Although many novel kinds of carbon have been de-
3
2
3
Co1.5/
1
5
23
veloped recently, carbon nanotubes (CNTs) still attracted
a great deal of attention due to their unique properties, such
as high specific surface area, electrical conductivity, and
good thermal and chemical stability, which make them a
3
3
observed at all, indicative of a poor crystallinity. Also, no
obvious diffraction peak shift is observed in the four XRD
patterns, indicating that Co is hard to alloy with Pd using
1
6-18
good catalyst support for fuel cells.
Thus, immobilizing
2
4
metal NPs on CNTs has turned into an interesting field
mainly due to the key roles of CNTs and metal NPs in the
the present method. That is to say, the as-prepared
Pd Co /MWCNTs catalysts were composite particles other
than alloy particles.
The average particle sizes of the catalysts were also
evaluated from Pd (111) peak by the Scherrer formula us-
x
y
1
9
field of electrocatalysis, biosensors and so on. To the best
of our knowledge, no paper reporting the immobilization of
x y
Pd Co composite NPs on CNTs via a method of pyrolysis
2
5
using RTILs as the solvent was published, though we have
anchored platium (Pt) NPs on the surface of muti-walled
carbon nanotubes (MWCNTs) by a pyrolysis process using
ing equation (1):
kl
d(Å) =
(1)
2
0
bcos q
distilled water as the solvent successfully.
In present work, four kinds of Pd
x
Co
y
/MWCNTs
where k is a coefficient (0.9), l the wavelength of X-ray
used (1.54056 Å), b the full-width half maximum and q is
the angle at the position of peak maximum. The calculated
based nanocomposite catalysts with various atomic ratios
of Pd to Co were fabricated by a facile method of pyrolysis,
in which RTILs of [BMIM]PF
6
was employed as the sol-
particle sizes based on the plane (111) for Pd
MWCNTs and Pd
nm, respectively. Thus, the particle size of Pd
MWCNTs is smaller than the catalyst of Pd Co
ing a larger surface area can be provided by the nanoparti-
cles of Pd Co1.5 when the loadings are the same. Due to the
3
Co
1
/
vent. The crystalline structures and morphology of the
MWCNT supported nanoparticles were studied by XRD
and TEM, respectively. The electrochemical activities of
the as-prepared NPs for ethanol oxidation reaction (EOR)
were investigated by cyclic voltammetry (CV) and chrono-
3
Co1.5/MWCNTs, are 28.15 and 22.71
3
Co1.5
/
3
1
, suggest-
3
3
amperometry, revealing that the catalyst of Pd Co1.5/
MWCNTs exhibited the highest catalytic activity among all
the samples. The possible catalytic mechanisms of the
3
Pd Co1.5/MWCNTs toward EOR were also well discussed.
RESULTS AND DISCUSSION
XRD analysis
Fig. 1A shows the XRD profiles of the as-prepared
samples. According to our previous report, the main dif-
fraction peak appearing at around 26º can only be assigned
Fig. 1. XRD patterns for Pd
3
Co
1
/MWCNTs, Pd
3
Co1.5
/
/
MWCNTs, Pd
MWCNTs
3
Co
2
/MWCNTs and Pd Co
3 3
2
0
to the facet (002) of MWCNTs. The planes (111), (200)
2
www.jccs.wiley-vch.de
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Chin. Chem. Soc. 2013, 60, 000-000

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
x y
EOR on Pd Co Nanoparticles
poor crystallinity, the particle sizes of Pd
were not calculated using this method.
TEM Analysis
3
Co
2
and Pd
3
Co
3
sponding to Pd, Co and O were displayed, indicating that
the nanoparticle of Pd
cle.
3
Co1.5 was not a pure bimetallic parti-
As shown in Fig. 2, the morphologies of the three typ-
Interestingly, no obvious aggregation of the as-pre-
pared particles was found in the TEM images (Fig. 2). How
does one understand this phenomenon? It was thought that
low charges and asymmetrically distributed charges are the
ical catalysts are characterized by TEM. It is evident from
3 1
image a that for the catalyst of Pd Co /MWCNTs after the
pyrolysis process, some black particles were decorated on
the surface of MWCNTs, implying that some nanoparticles
were fabricated by this pyrolysis method. Also, for the cat-
2
6
key factors causing nanoparticle aggregation. Probably,
2
7
[BMIM]PF
pared to water, can decrease the surface charge of the
Pd Co NPs due to the change in polarity of the solvent.
Therefore, similar to the case described by Wang et al.,
the newly formed Pd or Co atoms can migrate from the
[BMIM]PF to the [BMIM]PF /Pd Co interface in se-
6
, which has a lower dielectric constant com-
3 3 2
alysts of Pd Co1.5/MWCNTs and Pd Co /MWCNTs, some
small particles on the surface of MWCNTs are observed
clearly. It seems that smaller particles with a uniform distri-
bution of particle size were found in the catalyst of
x
y
2
8
Pd
served for the catalyst of Pd
Based on the TEM images, the estimated particle sizes for
the catalysts of Pd Co and Pd Co1.5 are about 27 and 23
3
Co1.5/MWCNTs (image b), and larger particles are ob-
6
6
x
y
3
Co /MWCNTs (image c).
2
quence, which is also analogous to the nanoparticle self-as-
sembly at the toluene/water or hexane/water interface in
2
9,30
3
1
3
ethanol-mediated methanol.
As a result, the agglomera-
nm, respectively. This result accords with the calculated
values from XRD patterns very well. Thus, it can be con-
cluded that the content of Co in the precursors has greatly
affected the resulted particle size and the morphologies of
tion of Pd Co particles was significantly suppressed, lead-
x
y
ing to a uniform dispersion. In other words, it is reasonable
to believe that the pyrolysis proces in RTILs should be dif-
ferent from that occurring in aqueous medium, which may
produce particles with some specific properties.
3
the catalysts as well. SEM image of Pd Co1.5/MWCNTs is
presented in image d, in which some white dotts were
found on the surface of MWCNTs. To determine the com-
position of the as-prepared samples, a typical spectrum of
Electrochemical response
The cyclic voltammograms (CVs) of EOR in 1 M
KOH on the as-prepared catalysts are shown in Fig. 3. It
can be seen that no peaks were found on the catalyst of
EDX for the catalyst of Pd
3
Co1.5/MWCNTs is shown by the
inset in image d. Except for the C element, the peaks corre-
Pd
3
Co
3
/MWCNTs, suggesting that more amount of Co in
Co would inhibit the
the composite nanoparticles of Pd
x
y
electrochemical oxidation process of ethanol. While for the
other three catalysts, EOR curves can be mainly character-
ized by two well-defined peaks, designated as Peak 1, cen-
tered at ~-0.2 V in the anodic sweep, and Peak 2, centered
at ~-0.45 V in the cathodic sweep, demonstrating that EOR
can take place on the three catalysts. Generally, the oxida-
tion peak appearing in the forward scan corresponds to the
Fig. 2. TEM images of (a) Pd
Pd Co1.5/MWCNTs, and (c) Pd
Photo d is the SEM image of Pd
3
Co
1
/MWCNTs, (b)
3
3
Co /MWCNTs,
2
Fig. 3. CVs obtained on the as-prepared Pd
MWCNTs coated glassy carbon electrode in a
solution of 1 M KOH + 1 M C OH at a scan
x y
Co /
3
Co1.5/
MWCNTs, and the inset is the corresponding
EDX spectrum.
2 5
H
-
1
rate of 20 mVs .
J. Chin. Chem. Soc. 2013, 60, 000-000
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jccs.wiley-vch.de
3

Article
Yang et al.
oxidation of freshly chemisorbed species coming from eth-
anol adsorption, and the reverse scan peak is primarily as-
sociated with the removal of carbonaceous species not
Pd
3
Co1.5/MWCNTs modified GC electrode, implying that
the Pd
3
Co1.5/MWCNTs catalyst showed the best electro-ca-
talysis toward EOR, which may be resulted from its higher
crystallinity (Fig. 1) and smaller particle size (Fig. 2).
Nyquist plot is a typical curve in electrochemical im-
pedance spectroscopy (EIS), which can be used for evalu-
ating the electrochemical performance of a working elec-
3
1
completely oxidized in the forward scan. The incom-
pletely oxidized carbonaceous species, such as CH COads,
3
could accumulate on the electrode and poison the elec-
trode. Thus, the ratio of the forward anodic peak current
3
4
(
I
peak 1) to the reverse anodic peak current (peak 2), i.e.,
/I , could be used to evaluate the poisoning tolerance of
trode. Based on our previous report, the semicircle ap-
pearing at the high frequency region corresponds to a cir-
cuit having a resistance element parallel to a capacitance el-
ement, and a semicircle with a larger diameter corresponds
to a larger charge transfer resistance. Thus, approximately,
the diameter of the semicircle stands for the value of charge
transfer resistance. Unfortunately, it can be evidenced from
Fig. 6(A) that for all the catalysts in the whole frequency
region, no evident semicircle was displayed. However, for
1
2
3
1
1 2
catalyst. A larger I /I ratio indicates a better oxidation
ability of ethanol during the anodic scan and less accumula-
tion of carbonaceous residues on the electrode surface. The
ratios of I
to be 0.68, 0.76 and 0.64, respectively. That is to say,
Pd Co1.5 showed a better electrochemical oxidation behav-
1 2 3 1 3 3 2
/I for Pd Co , Pd Co1.5 andPd Co are estimated
3
ior among the as-prepared samples. Actually, no report on
the catalysis of Pd-Co nanoparticles toward EOR was pub-
lished, though the alloy or composite particles of Pd-Co
have been employed as catalysts for oxygen reduction
3
the Pd Co1.5/MWCNTs coated electrode, as shown by the
red-dotted line, a bended curve was presented, indicative of
a circuit containing charge transfer resistance and a capaci-
3
2,33
35
reaction (ORR) in the previously published works.
tance (i.e., a RC circuit).
To get a clear comparison, linear sweep voltammetry
LSV) was conducted on three typical catalysts as shown in
To acquire more useful information on the electrical
properties of the as-prepared catalysts, Bode plots for all
the samples are illustrated in Fig. 6(B). It is evident that all
the coated GC electrodes showed one symmetric peak,
which may correspond to the relaxation process of the elec-
(
Fig. 4. One can see that the peak current of EOR on
Pd
Pd
3
Co1.5 is about 4 times larger than that on the catalyst of
Co . While for Pd Co , the peak current of EOR was at-
3
1
3
2
3
6
tenuated greatly. Also, the onset potential for EOR in
trode/solution interface. Close inspection revealed that
for Pd Co modified GC electrode, a flat plateau is ob-
Pd
Pd
3
Co1.5 is about 90 mV negative compared to that on
Co . Thus, the catalysis of Pd Co1.5 for EOR is superior
3
1
3
1
3
served in the frequency region from 4 to 20 Hz, indicating
the formation of a novel surface structure different from
to the other catalysts used in this experiment.
The electrochemical stability of the as-prepared cata-
lysts for EOR was also investigated by chronoamperome-
try. Fig. 5 shows the current-time plots of all the catalysts in
Pd
the Pd
spectively. This result strongly demonstrated that all the
Pd Co /MWCNTs particles exhibited less capacitive be-
havior since ideal capacitive systems should give phase an-
3
Co1.5 and Pd
3
Co
2
. The lower phase angles at 0.01 Hz for
3
Co
1
, Pd
3
Co1.5 and Pd
3
Co are -36º, -45º and -60º, re-
2
1
M KOH and 1 M ethanol at -0.2 V. It can be seen that the
x
y
largest steady current value at 400s was exhibited in the
3
7,38
gles of ca. -90º.
Fig. 4. Linear sweep voltammetry (LSV) obtained on
the (a) Pd Co (b) Pd Co1.5 and (c) Pd Co
MWCNTs coated glassy carbon electrode in so-
3
1
3
3
2
/
Fig. 5. Chronoamperometry curves of as-prepared
samples-coated GC electrode in 1 M KOH + 1
M C H OH. The applied potential is -0.20V vs.
lution of 1 M KOH + 1 M C
2
H
5
OH at a scan rate
2
5
-
1
of 5 mVs .
SCE.
4
www.jccs.wiley-vch.de
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Chin. Chem. Soc. 2013, 60, 000-000

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
x y
EOR on Pd Co Nanoparticles
Chronopotentiometric (CP) experiments were per-
formed to provide further insight into the stability of each
electrode since CP is a powerful method to value the activ-
ity and stability of a catalyst especially when operated in a
fuel cell. Fig. 7 shows the chronopotentiometric curves of
ethanol oxidation for the three electrodes in which a con-
previously reported one concerning the bimetallic PdIr cat-
3
9
alysts. However, the electrode potentials for Pd
Pd Co reached the largest values of 2.12 V and 2.03 V at
around 2554s and 1817s, respectively. And then the elec-
trode potential of Pd Co drops gradually and stabilizes at
around 2.05 V, and the electrode potential of Pd Co drops
3 1
Co and
3
2
3
1
3
2
-
2
stant current density of 0.20 mAcm was applied for
6000s. It can be seen for the Pd Co1.5/MWCNTs-modified
evidently to be about 1.83 V at 35460s. The appearance of
the maximum electrode potential may correspond to the
formation of metal oxides, and the decrese of electrode po-
tential is thought to be related to the dissolution of these ox-
ides during the immersion based on the point of corrosion
3
3
GC electrode that the polarization potential gradually in-
creases with time, whose shape is very consistent with the
4
0
science. That is to say, for the catalysts of Pd
3 1
Co and
Pd Co , some oxides were formed on the surface of them
3
2
which may inhibit the process of EOR. This result can par-
tially account for the lowered peak current of EOR on the
catalysts of Pd
played on Pd Co1.5, as presented in Fig. 3. Thus, the cata-
lyst of Pd Co1.5 exhibited the best stability in this experi-
3 1 3 2
Co and Pd Co when compared to that dis-
3
3
ment, though with the increase in time an increased polar-
ization potential was observed. This also indicated that a
proper amount addition of Co to Pd can facilitate the re-
moval of the adsorbed ethoxi intermediates, and release
4
1
more active Pd sites for the EOR, and making the catalyst
more resistant to poisoning. This phenomenon observed in
Fig. 7 strongly suggested that the corrosion behavior of the
binary or ternary metal catalysts should be well considered
when evaluating their catalysis towards the electrochemi-
cal oxidation of small organic molecules especially in alka-
line medium.
Fig. 6. (A) Nyquist plots for the catalysts coated GC
electrode in 1 M KOH + 1 M C
in which the catalysts are varied. (a) Pd
MWCNTs, (b) Pd Co1.5/MWCNTs, (c) Pd
2 5
H OH solution,
3
Co
Co
1
/
/
x y
Why did the Pd Co catalysts with various molar ra-
3
3
2
tios of Pd to Co show such different electrocatalytic activ-
ity toward EOR? To disclose the possible reasons, CVs of
MWCNTs, respectively; and (B): Bode plots,
obtained for the catalysts coated GC electrode
three typical catalysts, i.e., Pd
MWCNTs and Pd Co /MWCNTs, in 1 M KOH are plotted
in Fig. 8. It is revealed that there are no evident difference
of CVs between Pd Co /MWCNTs and Pd Co /MWCNTs,
3 1 3
Co /MWCNTs, Pd Co1.5/
in 1 M KOH + 1 M C
2
H
5
OH aqueous solution.
/MWCNTs, (b) Pd Co1.5/MWCNTs,
2
/MWCNTs, respectively.
(
a) Pd
c) Pd
3
Co
Co
1
3
3
2
(
3
3
1
3
2
Fig. 7. Chronopotentiometric curves of EOR obtained
on the as-prepared catalysts in 1 M KOH + 1
Fig. 8. CVs obtained on three catalysts in 1 M KOH at
-
2
-1
MCH
3
CH
2
OH at 0.20 mA cm .
a scan rate of 20 mVs .
J. Chin. Chem. Soc. 2013, 60, 000-000
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jccs.wiley-vch.de
5

Article
Yang et al.
excepting for two reduction peaks (one at -0.32 V and the
Mechanism of the pyrolysis process
other at -0.46 V) appearing in the catalyst of Pd
3
Co
2
. It is
To discuss the mechanism of this pyrolysis process,
generally regarded that except for the catalytic properties
owned by a catalyst, the electrochemically active surface
area (EASA) of the electrodes is a key parameter influenc-
ing the activity of a catalyst, thus the value of EASA was
usually estimated first for ananlyzing the catalysis owned
by a catalyst. However, in this case, there is no necessary to
photos of the filtered solutions are shown in Fig. 9(A). For
6
the pure RTILs of [BMIM]PF , a colorless solution is ob-
2
+
2+
served. After adding Pd and Co , a light blue solution
was observed instead. Interestingly, after the pyrolysis pro-
cess, the filtered solution showed a brown color. This im-
plies that there must be some interactions between
2
+
2+
evaluate the EASA of the catalyst of Pd
3
Co
2
since no en-
6
[BMIM]PF and Pd , Co or MWCNTs in the pyrolysis
hanced peak current for EOR was displayed on it. In con-
trast to the CVs obtained on the above catalysts, the catalyst
process, though the exact reaction mechanism cannot be
clarified by these photos only.
of Pd
sorption/desorption peaks (in the potential region from
0.85 to -1.25 V) as well as the prominent palladium oxide
3
Co1.5 exhibited the typical signatures of hydrogen ad-
Fig. 9(B) shows the ultraviolet-visible (UV-vis) ab-
sorption spectra for the [BMIM]PF
6
before and after pyrol-
, an
-
ysis process. As shown by curve a, for pure [BMIM]PF
6
reduction process at ~-0.45 V. More interestingly, a weak
oxidation peak is displayed at about 0.16 V, correspond-
ingly, and a reduction peak located at around -0.32 V is ob-
served on the negative-direction potential scan. The shape
of CVs curves in Fig. 8 is rather different from formerly re-
absorption peak located at around 230 nm was observed
clearly. It was reported that the spectra in the far-UV region
(200-250 nm) corresponds to the peptide n®p* electronic
4
5
2+
2+
transition. For the [BMIM]PF
6
containing Pd and Co
before pyrolysis, the intensities of all the absorption peaks
4
2
ported work of Pd electrode in alkaline solution. From the
obvious hydrogen adsorption and desorption peaks ob-
served in Fig. 8, it can be deduced that more hydrogen gas
were significantly decreased compared to pure [BMIM]PF
suggesting that the amount of some groups in [BMIM]PF
6
,
6
was lowered, probably due to the formation of some com-
or hydrogen atoms were fabricated in the vicinity of the
-
electrode of Pd
generated in the region close to the catalyst of Pd
comparison with that of Pd Co and Pd Co . Owing to the
fact that the process of EOR invloves many intermediate
products (such as linearly adsorbed CO and CO
3
Co1.5. Subsequently, more ions of OH were
3
Co1.5 in
3
1
3
2
2
, CHx,ads
4
3
2
species ) and final products (such as CO , acetaldehyde
and acetic acid), its mechanism still remains suspended.
However, a simplified mechanism has been provided re-
4
4
cently as follows
CH
CH
3
CH
CH
CH
2
OH « (CH
3
CH
2
OH)ads
(1)
-
(
3
2
OH)ads + 3OH ®
-
3
COads + 3H
2
O + 3e
(2)
(3)
(4)
(5)
-
-
OH « OHads + e
CH
CH
3
3
COads + OHads ®CH
3
COOH
-
-
3 2
COOH + OH ®CH COO + H O
Fig. 9. (A) Photos of the RTILs before and after pyrol-
Thus, according to above steps, due to the easier pro-
ysis. Photo a: pure [BMIM]PF ; Photo b:
6
2
+
2+
cess of hydrogen adsorption and desorption on the catalyst
[BMIM]PF
6
containing Pd and Co before
2
+
-
pyrolysis; Photo c: [BMIM]PF containing Pd
6
of Pd
3
Co1.5, more ions of OH were created yielding more
2
+
and Co after pyrolysis. (B) UV-vis absorption
spectra of (a) pure [BMIM]PF , (b) [BMIM]PF
containing Pd and Co before pyrolysis and
amounts of OHads. This can greatly accelerate the step (2),
3), (4) and (5). As a result, the process of EOR was signifi-
cantly facilitated, leading to an enhanced peak current and
lowered onset potential on the catalyst of Pd
6
6
(
2+
2+
2
+
2+
(c) [BMIM]PF containing Pd and Co after
6
Co1.5
.
pyrolysis.
3
6
www.jccs.wiley-vch.de
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Chin. Chem. Soc. 2013, 60, 000-000

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
x y
EOR on Pd Co Nanoparticles
2
+
plex-like molecules via the reaction of the introduced Pd
from Shenzhen nanotech port Co., Ltd. (China). All the electrodes
were purchased from Tianjin Aida Co., Ltd (China). All the chem-
icals were of analytical grade and used as-received without any
further treatment. Deionized water was used to prepare the aque-
ous solutions.
2
+
6
or Co with [BMIM]PF . Generally, the empty orbits, such
2
+
2+
as d-orbit, in Pd or Co can react with the electron pairs
in the imidazolium ring of [BMIM]PF , which can greatly
6
lower the intensity of the absorption peak. This kind of in-
teraction is very similar to that occurring in the formation
process of the self-assembled monolayer between the gold
Preparation of Pd
5.96 mg PdCl and an appropriate amount of Co(NO
were dissolved in 4 mL [BMIM]PF , in which the atomic ratios of
x
Co
y
/MWCNTs nanoparticles. Firstly,
2
3
)
2
·6H O
2
3
4
substrate and calmodulin. For the filtered [BMIM]PF
6
af-
6
ter pyrolysis, curve c, the intensity of the absorption peak
Pd to Co were different, namely, 3:1, 3:1.5, 3:2 and 3:3, respec-
tively. Then, 20 mg MWCNTs were added to the above solution,
leading to a suspension solution. This resultant solution was then
ultrasonicated for 30 min. Secondly, the resultant suspension so-
lution was placed in a home-made autoclave at room temperature,
and then the well-sealed autoclave was transferred to a box-type
furnace. Lastly, the temperature of the box-type furnace was in-
creased to 200 ºC within 20 min and was maintained for 3 h to ful-
fill the pyrolysis process, which was implemented in an SRJX-8-
13 box-type furnace equipped with a KSY12-16 furnace tempera-
ture controller. After cooling down to room temperature, the fil-
tered samples were thoroughly washed with ethylene chloride,
ethanol and distilled water successively, and dried in an ambient
was further attenuated, suggesting that more imidazolium
6
rings of [BMIM]PF were consumed or destroyed during
the pyrolysis process. It should be noticed that in all the
curves the absorption peak position at around 230 nm were
not much shifted before and after pyrolysis. However, the
absorption peak at about 210 nm was varied greatly. For in-
stance, for the pure RTILs, a broad peak at around 210 nm
2
+
2+
is displayed in curve a, after addition of Pd and Co , a
peak at about 213 nm is presented. It indicates that there
6
must be an interaction between [BMIM]PF and the added
ions though the position of the main peak at 230 nm was not
shifted. Interestingly, after the pyrolysis process, the peak
at 213 nm was reset to be around 210 nm, probably suggest-
condition to generate the MWCNTs supported Pd
(denoted as Pd Co /MWCNTs).
Preparation of Pd Co
x y
Co catalysts
2
+
2+
ing that the interaction between the Pd , Co and the
imidazolium ring of [BMIM]PF was demolished due to
the formation of Pd Co catalyst particles from the
BMIM]PF solution.
Obviously, the reduction potential of Pd and Co is
x
y
6
x
y
/MWCNTs modified electrode.
2
x
y
Glassy carbon (GC) electrode (geometric area of 0.07 cm ) was
polished to a mirror finish with 50 nm alumina nanopowder sus-
pensions before each experiment and served as a substrate for the
working electrode. The working electrodes were fabricated by
coating catalyst ink onto a glassy carbon electrode. The catalysts
ink was prepared by dispersing 1 mg catalyst in 1 mL Nafion etha-
nol solution (0.1 wt%). And after ultrasonication for 20 min,
about 15 mL ink was added to the surface of the GC electrode and
[
6
2
+
2+
4
6
2+
different from each other, and Pd has higher reduction
2
+
potential than Co . Thus, in the pyrolysis process, Pd at-
oms will precipitate first and enrich in the core and Co at-
oms then reduce mainly on the outer shell of the Pd atoms.
As a result, Pd-shell/alloy-core may form for the resultant
4
7
particles. Suo et al. reported that the key factor for design-
ing a Pd-based alloy catalyst displaying remarkable oxygen
reduction reaction (ORR) activity is to produce a “Pd-shell/
alloy-core” structure from the calculation of density func-
tion theory (DFT). Probably, in this experiment, compared
to the other particles, a well-defined “Pd-shell/alloy-core”
slowly dried in air, yielding a Pd
trode.
x y
Co /MWCNTs-coated GC elec-
Characterizations. The particle morphology was ob-
served by scanning electron microscopy (SEM, HITACHI, S-
570) and transmission electron microscopy (TEM, HITACHI, H-
7650). Energy Dispersive X-Ray Spectroscopy (EDX) spectrum
analysis was carried out on X-ray energy instrument (EDAX,
PV-9900, USA). XRD analysis of the catalyst was carried out on a
Bruker D8 ADVANCE X-ray diffractometer equipped with a Cu
Ka source (l = 0.154 nm) at 40 kV and 30 mA. The 2q angular re-
gion between 10 and 90° was explored at a scan rate of 1°/step.
Electrochemical measurements including cyclic voltam-
metry (CV) and electrochemical impedance spectroscopy (EIS)
were carried out on a CHI 660B electrochemical workstation
(Shanghai Chenhua Apparatus, China) connected to a personal
3
structure has formed in the catalyst of Pd Co1.5, which can
lead to an enhanced electrochemical response of EOR.
EXPERIMENTAL
Reagents and materials. Room temperature ionic liquid of
butyl-3-methylimidazolium hexafluorophosphate (denoted as
[
BMIM]PF
Hangzhou Chemer Chemical Co., Ltd. (China). MWCNTs (purity
95%) with an average diameter of 10-20 nm were purchased
6
) a purity of more than 99% was obtained from
>
J. Chin. Chem. Soc. 2013, 60, 000-000
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jccs.wiley-vch.de
7

Article
Yang et al.
computer. EIS was performed in the frequency range from 0. 01 to
REFERENCES
5
1
2
. Demirci, U. B. J. Power Sources 2007, 169, 239.
. Shen, S. Y.; Zhao, T. S.; Xu, J. B.Electrochim. Acta 2010, 55,
179.
1
0 Hz with an amplitude of 5 mV. Aconventional three-electrode
x y
system was employed, in which a Pd Co /MWCNTs modified GC
9
electrode and a platinum wire were used as the working electrode
and counter electrode, respectively. It should be noted that the ref-
erence electrode was a saturated calomel electrode (SCE). All po-
tentials in this paper were reported with respect to SCE. A solu-
tion of 1 M KOH containing 1 M ethanol was used to study etha-
nol oxidation activity. Prior to each electrochemical test, the elec-
trolyte was bubbled with high purity nitrogen for 30 min to avoid
the influence of oxygen dissolved in the electrolyte. All the exper-
iments were carried out at room temperature.
3
. Tian, N.; Zhou, Z. Y.; Sun, S. G.; Ding, Y.; Wang, Z. L. Sci-
ence 2007, 316, 732
. Tian, N.; Zhou, Z.-Y.; Yu, N.-F.; W, L.-Y.; Sun, S.-G. J. Am.
Chem. Soc. 2010, 132, 7580.
4
5. Wang, W.; Zheng, D.; Du, C.; Zou, Z.; Zhang, X.; Xia, B.;
Yang, H.; Akins, D. L. J. Power Sources 2007, 167, 243.
. Shao, M.; Sasaki, K.; Marinkovic, N. S.; Zhang, L.; Adzic,
R. R. Electrochem. Commun. 2007, 9, 2848.
. Morales-Acosta, D.; Ledesma-Garcia, J.; Godinez, L. A.;
Rodríguez, H. G.; Álvarez-Contreras, L.; Arriaga, L. G. J.
Power Sources 2010, 195, 461.
6
7
8
9
. Zhang, L.; Lee, K.; Zhang, J. Electrochim. Acta 2007, 52,
CONCLUSIONS
3
088.
x y
For the first time, Pd Co composite nanoparticles
. Wei, Y.-C.; Liu, C.-W.; Chang, Y.-W.; Lai, C.-M.; Lim, P.-Y.;
Tsai, L.-D.; Wang, K.-W. Int. J. Hydrogen Energy 2010, 35,
with an average diameter close to 25 nm, generated by a
facile method of pyrolysis in the presence of MWCNTs us-
1
864.
ing RTILs of [BMIM]PF
alysts for EOR. Analysis revealed that the smallest particle
size and the better crystallinity of Pd Co1.5 may be respon-
6
as solvent, were employed as cat-
1
0. Xiao, Y.; Yu, G.; Yuan, J.; Wang, J.; Chen, Z. Electrochim.
Acta 2006, 51, 4218.
11. Welton, T. Chem. Rev. 1999, 99, 2071.
3
1
1
1
1
2. Wilkes, J. S. J. Mol. Catal. A: Chem. 2004, 214, 11.
3. Zhang, Y.; Zheng, J. Electrochem. Commun. 2008, 10, 1400.
4. Ding, K.; Yang, G. Electrochim. Acta 2010, 55, 2319.
5. Xue, M.; Chen, G.; Yang, H.; Zhu, Y.; Wang, D.; He, J.; Cao,
T. J. Am. Chem. Soc. 2012, 134, 6536.
sible for its better electrocatalytic activity toward EOR
when compared to the other samples. Results from LSV
showed a 90 mV decrease in the onset oxidation potential
and a four times enhancement in the peak current of EOR
for Pd
Pd Co
samples in alkaline medium indicated that hydrogen evolu-
tion became easier on the catalyst of Pd Co1.5 as compared
to that on Pd Co and Pd Co , which may be partially con-
3
Co1.5/MWCNTs composite catalyst compared to
16. Iijima, S. Nature 1991, 354, 56.
1
1
1
7. Sathiya, M.; Prakash, A. S.; Ramesha, K.; Tarascon, J. M.;
Shukla, A. K. J. Am. Chem. Soc. 2011, 133, 16291.
8. Wang, S.; Yang, F.; Jiang, S. P.; Chen, S.; Wang, X. Electro-
chem. Commun. 2010, 12, 1646.
3
1
/MWCNTs catalyst. More importantly, CVs of the
3
3
1
3
2
9. Davies, T. J.; Hyde, M. E.; Compton, R. G. Angew. Chem.
Int. Edit. 2005, 44, 5121.
tributed to the enhanced peak current of EOR. Apart from
discussing the electro-catalysis of the composite nanopar-
ticles toward EOR, presenting a novel method of pyrolysis
using RTILs as the solvent to synthesize nanoparticles is
the other main contribution of this work, which is expected
to be helpful for both extending the application of RTILs
and preparing nanoparticle supported CNTs on a large
scale. The present work may open a new path for the syn-
thesis of promising electrocatalysts in the fuel cells.
20. Ding, K.; Cao, M. Russ. J. Electrochem. 2008, 44, 977.
21. Zhang, Z.; Xin, L.; Sun, K.; Li, W. Int. J. Hydrogen Energy
2
011, 36, 12686.
2
2
2
2. Wang, Z. B.; Zuo, P. J.; Wang, G. J.; Du, C. Y.; Yin, G. P. J.
Phys Chem C 2008, 112, 6582.
3. Yang, M. R.; Teng, T. H.; Wu, S.-H. J. Power Sources 2006,
1
59, 307.
4. Li, B. J. Prakash. Electrochem. Commun. 2009, 11, 1162.
25. Radmilovic, V.; Gasteiger, H. A.; Ross, P. N. J. Catal. 1995,
54, 98.
6. Shipway, A. N.; Lahav, M.; Gabai, R.; Willner, I. Langmuir
000, 16, 8789.
1
2
2
2
2
ACKNOWLEDGEMENTS
2
This work was financially supported by the National
Natural Science Foundation of China (No. 21173066), Nat-
ural Science Foundation of Hebei Province of China (No.
B2011205014). Z. Guo acknowledges the support from US
National Science Foundation (EAGER:CBET11-37441)
managed by Dr Rosemarie D. Wesson.
7. Zhang, D.; Okajima, T.; Matsumoto, F.; Ohsaka, T. J.
Electrochem. Soc. 2004, 151, D31.
8. Wang, M.; Li, Y.-J.; Xie, Z.-X.; Liu, C.; Yeung, E. S. Mater.
Chem. Phys. 2010, 119, 153
9. Li, Y. J.; Huang, W. J.; Sun, S. G. Angew. Chem. Int. Ed.
2006, 45, 2537.
8
www.jccs.wiley-vch.de
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
J. Chin. Chem. Soc. 2013, 60, 000-000

JOURNAL OF THE CHINESE
CHEMICAL SOCIETY
x y
EOR on Pd Co Nanoparticles
3
3
3
3
0. Yun, S.; Park, Y.-K.; Kim, S. K.; Park, S. Anal. Chem. 2007,
9, 8584.
1. Zhao, G. Y.; Xu, C. L.; Guo, D. J.; Li, H.; Li, H- L. J. Power
Sources 2006, 162, 492.
39. Shen, S. Y.; Zhao, T. S.; Xu, J. B. Electrochim. Acta 2010,
55, 9179.
40. Li, M. C.; Zeng, C. L.; Luo, S. Z.; Shen, J. N.; Lin, H. C.;
Cao, C. N. Electrochim. Acta 2003, 48, 1735.
41. Ding, K.; Yang, G.; Wei, S.; Mavinakuli, P.; Guo, Z. Ind.
Eng. Chem. Res. 2010, 49, 11415.
42. Ramulifho, T.; Ozoemena, K. I.; Modibedi, R. M.; Jafta, C.
J.; Mathe, M. K. Electrochim. Acta 2012, 59, 310.
43. Wang, H.; Jusys, Z.; Behm, R. J. J. Power Sources 2006, 154,
351.
44. Liang, Z. X.; Zhao, T. S.; Xu, J. B.; Zhu, L. D. Electrochim.
Acta 2009, 54(8), 2203.
7
2. Zhang, L.; Lee, K.; Zhang, J. Electrochim. Acta 2007, 52,
7
964.
3. Wei, Y.-C.; Liu, C.-W.; Chang, Y.-W.; Lai, C.-M.; Lim, P.-Y.;
Tsai, L.-D.; Wang, K.-W. Int. J. Hydrogen Energy 2010, 35,
1
864.
3
3
4. Ding, K.; Jia, Z.; Wang, Q.; He, X.; Tian, N.; Tong, R.;
Wang, X. J. Electroanal. Chem. 2001, 513, 67.
5. Adekunle, A. S.; Ozoemena, K. I. Electrochim. Acta 2008,
5
3, 5774.
45. Wu, X.; Du, P.; Wu, P.; Cai, C. Electrochim. Acta 2008, 54,
738.
46. Bard, A. J.; Parsons, R.; Jordan, J. Standard Potentials in
Aqueoussolutions; Marcel Dekker; New York, 1985.
47. Suo, Y.; Zhuang, L.; Lu, J. Angew Chem. Int. Ed. 2007, 46,
2862.
3
3
6. Hu, C.; Yuan, S.; Hu, S. Electrochim. Acta 2006, 51, 3013.
7. Adekunle, A. S.; Pillay, J.; Ozoemena, K. I. Electrochim.
Acta 2010, 55, 4319.
8. Ding, K.; Wang, Q. Portugaliae Electrochim. Acta 2007, 25,
4
3
01.
J. Chin. Chem. Soc. 2013, 60, 000-000
© 2013 The Chemical Society Located in Taipei & Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.jccs.wiley-vch.de
9