Insulating diamond particles as substrate for Pd electrocatalysts

Article abstract of DOI:10.1039/c1cc12387d

The suitability of insulating highly crystalline diamond particles as support for Pd based electrocatalysts is explored for the first time by evaluating the electrochemical stripping of CO and oxidation of formic acid in acid solutions.

Full text of DOI:10.1039/c1cc12387d

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2011, 47, 7656–7658
www.rsc.org/chemcomm
COMMUNICATION
Insulating diamond particles as substrate for Pd electrocatalystsw
a
b
a
a
´
´
Amy Moore, Veronica Celorrio, Marıa Montes de Oca, Daniela Plana,
a
Wiphada Hongthani,a Marı
a J. Lazaro and David J. Fermın*
´ ´
´
b
Received 24th April 2011, Accepted 20th May 2011
DOI: 10.1039/c1cc12387d
The suitability of insulating highly crystalline diamond particles
as support for Pd based electrocatalysts is explored for the first
time by evaluating the electrochemical stripping of CO and
oxidation of formic acid in acid solutions.
done to demonstrate that conducting boron doped diamond
(BDD) films have unique properties as electrode supports,
particularly in the context of dimensionally stable anodes.6a,10
Recent reports have described electrocatalysts supported on
BDD powders for fuel cell applications.10,11 Here, we shall
demonstrate that Pd nanostructures supported on insulating
crystalline diamond particles can be used effectively as electrode
material. Our findings not only raise fundamental questions
about charge transport on wide band gap materials, but also
demonstrate the possibility of employing material such as
High Pressure High Temperature (HPHT) diamond powders
as electrocatalyst support, a material more readily available
than BDD powders.
Carbon supports play a fundamental role on the efficiency of
electrocatalysts in low temperature fuel cells.1 A conventional
approach in the preparation of highly dispersed metal nano-
structures involves the impregnation of the metal precursor
in a highly porous carbon matrix, followed by chemical
reduction employing compounds such as hydrazine, boro-
hydride, formic acid or fomaldehyde.2 The presence of
functional groups at the carbon surface can have a strong
influence on the metal particle size, morphology, stability and
homogeneity of the dispersion.3 Vulcan XC-72R (Vulcan) is
the most widely used carbon support in this context, due to a
combination of high specific surface area and electrical
conductivity.4 On the other hand, novel materials such as
carbon xerogels, nanotubes and graphene may provide new
avenues for developing highly active electrocatalyts.4,5 In this
communication, we shall explore an entirely new approach,
based on highly crystalline insulating diamond particles (DP)
acting as supports for metal nanostructures.
HPHT type Ib diamond particles featuring a nominal size of
500 nm (Microdiamant AG-MSY005) were employed as the
base material. HPHT diamond exhibits a significant higher
degree of crystallinity and low density of sp2 carbon impurities
in comparison to particles obtained by detonation. These
surface groups are of paramount importance on the electro-
chemical properties of diamond powders.7c,8a The Pd electro-
catalyst supported on DP (Pd/DP) was prepared by impregnating
the carbon supports with a solution of sodium hexachloro-
palladate(IV) tetrahydrate followed by reduction with sodium
borohydride (details in the electronic supplementary information
ESI). In order to compare the properties of the Pd/DP catalyst,
the same procedure was followed using Vulcan as support
(Pd/Vulcan). The metal loading in both catalysts was between
35 and 40%, as estimated from EDX.
Diamond is characterised by high chemical, thermal and
mechanical stability and is currently being considered not only
in the electrocatalysis field,6 but also for a host of applications
ranging from thermal emitters to sensing devices and bio-
markers.7 Undoped diamond is essentially an insulator with a
band gap of 5.5 eV. However, hydrogen terminated surfaces
are characterised by an unusual surface conductivity in con-
tact with air or aqueous solution,8 which is not readily
observed in oxygen terminated diamond surfaces. Insulating
diamond powders have been used as supports for metals such
as Pd and Co in heterogeneous catalytic systems.9 However, it
would be counterintuitive to consider this material in fuel cell
electrocatalysis in view of its wide band gap and complex
surface chemistry. In recent years, significant work has been
Characteristic X-ray diffractograms of the DP and the
Pd/DP composite are illustrated in Fig. 1a. The XRD pattern
for the diamond shows peaks at 2y = 43.31, 74.71 and 911. The
Pd/DP composite showed additional peaks at 2y = 39.41,
45.61, 67.41 and 81.31 associated with the (111), (200), (220)
and (311) Pd planes, respectively. The sharp, well-defined
peaks of DP indicate the highly crystalline structure of HPHT
diamond. The crystallite sizes of Pd particles in the Pd/DP
composite were calculated through the Scherrer equation
using the (220) peak at 2y = 67.41, yielding an average crystal
size of 10 nm. This region was chosen in order to avoid the
influence of the signals due to the carbon substrate. Identical
crystal size was estimated from the equivalent XRD peak for
Pd particles supported on Vulcan (results not shown).
a School of Chemistry, University of Bristol, Cantocks Close,
Bristol BS8 1TS, UK. E-mail: david.fermin@bristol.ac.uk;
Fax: +44 1179250612; Tel: +44 117928981
b
´
Instituto de Carboquımica (CSIC), Miguel Luesma Castan 4,
50018 Zaragoza, Spain
´
Fig. 1b and c illustrate a typical SEM image and the
corresponding Pd distribution employing X-ray elemental
mapping (recorded at 2.839 eV). The topographic features
w Electronic supplementary information (ESI) available: Preparation,
thermogravimetric analysis and background voltammetric responses in
acid solution of the Pd/DP composite. See DOI: 10.1039/c1cc12387d
c
7656 Chem. Commun., 2011, 47, 7656–7658
This journal is The Royal Society of Chemistry 2011

View Article Online
set of evidence of good electronic communication through the
diamond support.
The electrocatalytic properties of the Pd/DP composite
towards CO oxidation, in H2SO4 electrolyte, were contrasted
with Pd/Vulcan and the commercial Pd/C catalyst from E-Tek
(20 wt% Pd supported on Vulcan). High purity CO was
bubbled into the electrolyte solution for 10 min while keeping
the electrode potential at À0.166 V, in order to achieve the
maximum coverage of CO at the Pd centres. Subsequently,
dissolved CO was purged out of the electrolyte by bubbling Ar
for 20 min. Finally, two consecutive voltammograms were
recorded between À0.2 and 1.0 V, at 0.02 V sÀ1. Identical
procedure was implemented for all three electrocatalysts.
The voltammetric features associated with CO stripping at
Pd/DP, Pd/Vulcan and E-Tek are contrasted in Fig. 2 (left).
The hydrogen adsorption region appears blocked in the initial
forward scan due to adsorbed CO at the Pd surface. The key
feature in the first forward scan is the CO stripping peak
located at 0.68 V in the Pd/DP catalysts. The disappearance of
the CO stripping peak on subsequent scans, and the reappearance
of hydrogen peaks at negative potentials, indicate complete
removal of CO in the first scan. A somewhat larger H absorption
current was observed in the presence of the Pd/Vulcan composite,
which is surprising as the Pd crystalline size domains are similar.
This could be caused by the presence of large Pd domains
due to agglomeration of small crystallites. However, the main
feature at positive potentials of Fig. 2 (left), is associated with
the potential of CO stripping, which is comparable for all
three catalysts. This rather unexpected result clearly shows
Fig. 1 XRD diffractograms of the HPHT diamond particles and
the Pd/DP composite (a). The XRD data is characterised by the
sharp peaks of the crystalline diamond particles and the nanoscopic
Pd domains with an average particle size of 10 nm. SEM image of the
Pd/DP composite (b) and Pd elemental mapping shown as white
spots (c) with identical magnification. The scale bar in (b) corresponds
to 1 mm. A 35–40% metal loading was obtained from EDX analysis.
of the composite are determined by the morphology of the
500 nm DP with sharp edges characteristic of HPHT materials.8
Identical topographic features are observed in the absence of Pd
particles. The elemental mapping shown in Fig. 1c reveals the
areas rich in Pd as white dots scattered in a dark background
corresponding to the diamond support. This image, recorded in
the same scale as the SEM image in Fig. 1b, demonstrates that
Pd grows as discrete particles over the entire composite.
The thermal stability of the Pd/DP and Pd/Vulcan composites,
as monitored by thermogravimetric analysis, are contrasted in
the supplementary information. Fig. S1w shows that the onset
of Vulcan degradation shifts towards lower temperatures by
300 1C in the presence of Pd nanoparticles. On the other hand,
the DP support is completely stable up to 600 1C in the
presence and absence of Pd particles. A similar behaviour
has been previously reported for Pt-coated BDD.12
Electrodes were prepared by standard drop-casting of a
Nafion based ink of the Pd/DP composite on a glassy carbon
electrode. The ink was prepared by mixing 2 mg of the Pd/DP
composite with 50 ml of Nafion dispersion (5 wt%, Sigma-Aldrich)
in 500 ml of ultrapure water (Millipore Milli-Q system). A 20 ml
aliquot of this suspension was drop-casted onto a freshly
polished glassy carbon electrode. The working electrode was
placed in contact with the electrolyte solution in a meniscus
configuration. The electrochemical cell incorporated a platinum
wire as counter-electrode and a Ag/AgCl reference electrode
connected by a luggin capillary. The Pd active surface area of
the electrocatalyst was estimated from the CO stripping,
Fig. 2 CO stripping voltammograms (left) obtained for the Pd/DP,
Pd/Vulcan and E-Tek electrocatalysts, at 0.02 V sÀ1, in 0.5 mol dmÀ3
H2SO4. CO stripping is observed in the first scan (continuous black
lines), while hydrogen adsorption/absorption responses are observed
assuming a charge per unit area of 490 mC cmÀ2 13
A typical
.
cyclic voltammogram of the Pd/DP catalyst on the glassy
carbon electrode in Ar-saturated 0.5 mol dmÀ3 H2SO4 is
shown in Fig. S2w (supplementary information). Well defined
responses associated with hydrogen adsorption and palladium
oxide formation/reduction were obtained, providing the first
in the second scan (dashed red lines). Formic acid (2 mol dmÀ3
)
oxidation was studied, under similar conditions, on the three electro-
catalysts (right).
c
This journal is The Royal Society of Chemistry 2011
Chem. Commun., 2011, 47, 7656–7658 7657

View Article Online
that the inherent electrocatalytic activity of Pd particles can be
exploited when supported on type Ib crystalline diamond
particles.
that this surface conductivity in the Pd/DP composite is onset by
partial hydrogenation of the diamond particle during the growth
of the metal centres.
Cyclic voltammograms associated with the oxidation of
formic acid (FA) at Pd/DP, Pd/Vulcan and E-Tek are also
contrasted in Fig. 2 (right). These experiments provide a
preliminary evaluation of the performance of the Pd/DP
composite for fuel cell applications. All three catalysts show
a broad peak on the forward scan and a drop in current as the
Pd oxide is formed, inhibiting further FA oxidation. On
reversal of the scan, once the Pd surface is recuperated, FA
is once again oxidised; similar current densities are obtained
on the forward and back scans indicating a high tolerance
towards electrode poisoning.14 Somewhat lower currents are
obtained on Pd/DP than on the other two electrocatalysts,
suggesting a smaller affinity of FA to the diamond surface as
opposed to the more porous Vulcan substrate. However, this
factor can be counterbalanced by the far superior stability
provided by the DP. In future work, we shall investigate this
reaction further, as well as the oxidation of other fuel cell
relevant compounds such as methanol and ethanol.
VC and MJL gratefully acknowledge CSIC and MICINN
for the JAE grant and the financial support given through
the Project MAT2008-06631-C03-01, respectively. DP and
DJF acknowledge the financial support by the EPRSC
(EP/H046305/1). WH and MMO acknowledge the support
from the Royal Thai Government and the Mexican National
Council for Science and Technology (CONACyT), respectively.
The authors are indebted to Dr Neil A. Fox and Mr Jonathan
Jones from the University of Bristol for their valuable support
to this work.
Notes and references
1 (a) E. Antolini, Appl. Catal., B, 2009, 88, 1–24; (b) A. L. Dicks,
J. Power Sources, 2006, 156, 128–141.
2 (a) J. B. Joo, P. Kim, W. Kim, Y. Kim and J. Yi, J. Appl.
Electrochem., 2009, 39, 135–140; (b) J. H. Tian, F. B. Wang,
Z. Q. Shan, R. J. Wang and J. Y. Zhang, J. Appl. Electrochem.,
2004, 34, 461–467.
3 (a) F. Rodriguez-Reinoso, Carbon, 1998, 36, 159–175; (b) E. Antolini,
J. Mater. Sci., 2003, 38, 2995–3005.
An important aspect highlighted by these studies is the
origin of the conductivity of the insulating DP support. It could
be argued that the electrochemical responses are associated
only with Pd particles in ‘‘physical contact’’ with the underlying
GC electrode. This is very unlikely considering the homogeneous
distribution of Pd in the ink and the strikingly similar responses
observed on conducting carbons. Additionally, a progressive
increase of the current is observed with successive layers of the
ink on the GC electrode. High uncompensated resistance would
arise if electron transport occurred only through discrete
networks of Pd particles. Such resistance is incompatible with
the voltammograms shown in Fig. 2 and Fig. S2w (supple-
mentary information). We propose that the origin of the
electronic communication is linked to partial hydrogenation
of the DP during the reduction of the metal precursor
anchored to the diamond support; generation of C–H bonds
on exposure of diamond particles to sodium borohydride has
recently been demonstrated through FTIR.15 As shown recently,
hydrogenation of diamond surfaces generates a shift of the
surface dipole pushing the onset potential for surface hole
accumulation to 0.1 eV vs. Ag/AgCl.8a Preliminary studies to
be reported elsewhere revealed the transport properties are
strongly dependent of the metal catalyst employed, as well as
the type of diamond particle.
4 Y. Y. Shao, J. Liu, Y. Wang and Y. H. Lin, J. Mater. Chem., 2009,
19, 46–59.
5 R. L. McCreery, Chem. Rev., 2008, 108, 2646–2687.
6 (a) F. Montilla, E. Morallon, I. Duo, C. Comninellis and J. L.
Vazquez, Electrochim. Acta, 2003, 48, 3891–3897; (b) J. Wang,
G. M. Swain, T. Tachibana and K. Kobashi, Electrochem.
Solid-State Lett., 2000, 3, 286–289; (c) J. Wang and
G. M. Swain, J. Electrochem. Soc., 2003, 150, E24–E32;
(d) Y. R. Zhang, S. Asahina, S. Yoshihara and T. Shirakashi,
Electrochim. Acta, 2003, 48, 741–747.
7 (a) C. E. Nebel, S. Dongchan, B. Rezek, N. Tokuda, H. Uetsuka
and H. Watanabe, J. R. Soc. Interface, 2007, 4, 439–461;
(b) A. Krueger, Adv. Mater., 2008, 20, 2445–2449;
(c) K. B. Holt, Phys. Chem. Chem. Phys., 2010, 12, 2048–2058.
8 (a) W. Hongthani, N. A. Fox and D. J. Fermin, Langmuir, 2011,
27, 5112–5118; (b) F. Maier, M. Riedel, B. Mantel, J. Ristein and
L. Ley, Phys. Rev. Lett., 2000, 85, 3472; (c) J. A. Garrido, A. Hartl,
¨
M. Dankerl, A. Reitinger, M. Eickhoff, A. Helwig, G. Muller and
¨
M. Stutzmann, J. Am. Chem. Soc., 2008, 130, 4177–4181.
9 (a) O. V. Turova, E. V. Starodubtseva, M. G. Vinogradov, V. I.
Sokolov, N. V. Abramova, A. Y. Vul and A. E. Alexenskiy, Catal.
Commun., 2011, 12, 577–579; (b) V. L. Kirillov, V. I. Zaikovskii
and Y. A. Ryndin, React. Kinet. Catal. Lett., 1998, 64, 169–175;
(c) K. Nakagawa, M. Kikuchi, M. Nishitani-Gamo, H. Oda,
H. Ganto, K. Ogawa and T. Ando, Energy Fuels, 2008, 22,
3566–3570.
10 J. Hu, X. Lu and J. S. Foord, Electrochem. Commun., 2010, 12,
676–679.
11 (a) N. Spataru, X. T. Zhang, T. Spataru, D. A. Tryk and
A. Fujishima, J. Electrochem. Soc., 2008, 155, B264–B269;
(b) G. R. Salazar-Banda, K. I. B. Eguiluz and L. A. Avaca,
Electrochem. Commun., 2007, 9, 59–64.
12 L. Guo, V. M. Swope, B. Merzougui, L. Protsailo, M. H. Shao,
Q. Yuan and G. M. Swain, J. Electrochem. Soc., 2010, 157, A19–A25.
13 A. M. El-Aziz and L. A. Kibler, J. Electroanal. Chem., 2002, 534,
107–114.
14 H. Miyake, T. Okada, G. Samjeske and M. Osawa, Phys. Chem.
Chem. Phys., 2008, 10, 3662–3669.
15 L. La-Torre-Riveros, E. Abel-Tatis, A. Mendez-Torres, D. Tryk,
´
In conclusion, we have demonstrated that Pd nanostructures
supported on highly crystalline, insulating diamond particles can
potentially be used as electrocatalysts in fuel cells. A conventional
impregnation method was used for growing Pd centres onto
HPHT diamond powders, a material of tremendous chemical,
thermal, mechanical and electrical stability. Despite the inherent
bulk resistivity of DP, Pd centres appear interconnected to the
electrode surface, possibly by hole mediated surface transport
characteristic of hydrogenated diamond surfaces. We suggest
M. Prelas and C. Cabrera, J. Nanopart. Res., 2011, DOI: 10.1007/
s11051-11010-10196-11058.
c
7658 Chem. Commun., 2011, 47, 7656–7658
This journal is The Royal Society of Chemistry 2011