Co-deposition of Pt and ceria anode catalyst in supercritical carbon dioxide for direct methanol fuel cell applications

Article abstract of DOI:10.1016/j.electacta.2012.04.091

Pt and mixed Pt-ceria catalysts were deposited onto gas diffusion layers using supercritical fluid deposition (SFD) to fabricate thin layer electrodes for direct methanol fuel cells. Dimethyl (1,5-cyclooctadiene) platinum (II) (CODPtMe₂) and tetrakis (2,2,6,6-tetramethyl 3,5-heptanedionato) cerium (IV) (Ce(tmhd)₄) were used as precursors. Hydrogen-assisted Pt deposition was performed in compressed carbon dioxide at 60 °C and 17.2 MPa to yield high purity Pt on carbon-black based gas diffusion layers. During the preparation of the mixed Pt-ceria catalyst, hydrogen reduction of CODPtMe ₂ to yield Pt catalyzed the deposition of ceria from Ce(tmhd) ₄ enabling co-deposition at 150 °C. The catalyst layers were characterized using X-ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and scanning electron microscope-energy dispersive spectral (SEM-EDS) analyses. Their electrochemical performance toward methanol oxidation was examined in half cell mode using a three electrode assembly as well as in fuel cell mode. The thin layer electrodes formed via SFD exhibited higher performance in fuel cell operations compared to those prepared by the conventional brush-paint method. Furthermore, the Pt-ceria catalyst with an optimized composition exhibited greater methanol oxidation activity than pure platinum.

Full text of DOI:10.1016/j.electacta.2012.04.091

Electrochimica Acta 75 (2012) 191–200
Contents lists available at SciVerse ScienceDirect
Electrochimica Acta
journal homepage: www.elsevier.com/locate/electacta
Co-deposition of Pt and ceria anode catalyst in supercritical carbon dioxide
for direct methanol fuel cell applications
Eunyoung Youa, Rolando Guzmán-Blasb, Eduardo Nicolaub, M. Aulice Scibiohb,
Christos F. Karanikasa, James J. Watkinsc,∗, Carlos R. Cabrerab,∗
a Department of Chemical Engineering, University of Massachusetts, Amherst, MA 01003, USA
b Department of Chemistry and NASA Center for Advanced Nanoscale Materials, University of Puerto Rico, Rio Piedras Campus, PO Box 70377, San Juan, PR 00936-8377, USA
c Department of Polymer Science and Engineering, University of Massachusetts, Amherst, MA 01003, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Pt and mixed Pt-ceria catalysts were deposited onto gas diffusion layers using supercritical fluid deposi-
tion (SFD) to fabricate thin layer electrodes for direct methanol fuel cells. Dimethyl (1,5-cyclooctadiene)
platinum (II) (CODPtMe2) and tetrakis (2,2,6,6-tetramethyl 3,5-heptanedionato) cerium (IV) (Ce(tmhd)4)
were used as precursors. Hydrogen-assisted Pt deposition was performed in compressed carbon dioxide
at 60 ◦C and 17.2 MPa to yield high purity Pt on carbon-black based gas diffusion layers. During the prepa-
ration of the mixed Pt-ceria catalyst, hydrogen reduction of CODPtMe2 to yield Pt catalyzed the deposition
of ceria from Ce(tmhd)4 enabling co-deposition at 150 ◦C. The catalyst layers were characterized using X-
ray diffraction (XRD), X-ray photoelectron spectroscopy (XPS) and scanning electron microscope-energy
dispersive spectral (SEM-EDS) analyses. Their electrochemical performance toward methanol oxidation
was examined in half cell mode using a three electrode assembly as well as in fuel cell mode. The thin
layer electrodes formed via SFD exhibited higher performance in fuel cell operations compared to those
prepared by the conventional brush-paint method. Furthermore, the Pt-ceria catalyst with an optimized
composition exhibited greater methanol oxidation activity than pure platinum.
Received 29 December 2011
Received in revised form 28 February 2012
Accepted 25 April 2012
Available online 3 May 2012
Keywords:
Supercritical fluid deposition
Methanol oxidation
Platinum-ceria
Carbon dioxide
Thin-layer electrode
Direct methanol fuel cell
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
addition of transition metal oxides such as MoOx, VOx and WOx to
PtRu has also been evaluated for an enhanced methanol oxidation
Low-temperature fuel cells are attractive power sources for
portable and vehicular applications. Direct methanol fuel cells
(DMFCs) are at the forefront of commercialization due to their
favorable characteristics including high energy density, operation
at near ambient conditions and low emissions. Moreover, methanol
is a renewable fuel source and offers fast and convenient refueling,
as it is a liquid from −97.0 to 64.7 ◦C [1,2].
reaction [18–21]. These modified catalysts have been reported to
yield a higher methanol oxidation current than the conventional
PtRu catalysts. Ruthenium crossover [22], however, has triggered
serious concerns about anode stability.
Yet another impediment for early DMFC commercialization
is the high cost of noble metal catalysts employed in the elec-
trodes. Attempts have been made to reduce the catalyst loading
by increasing Pt utilization [22–26] using new fabrication methods
to enhance the three-phase boundary.
The use of non-noble catalysts is another approach for electrode
cost reduction. Yu et al. [27] incorporated an oxygen storage mate-
rial such as ceria into the cathode catalyst to increase the local
oxygen concentration and attained performance enhancement in
the DMFC. Xu and Shen [28,29] reported that the electrochemi-
cal oxidation of methanol, ethanol, glycerol and ethylene glycol on
Pt-CeO2/C catalyst constructed in alkaline media showed improved
performance compared to conventional Pt/C catalysts. Campos et al.
[30] prepared Pt/CeO2 catalyst layers on a glassy carbon electrode
by occlusion deposition method using a bath containing both ceria
and K2PtCl6. An increase in catalytic activity for methanol oxida-
tion with the ceria composite electrode was found when compared
to the Pt/C system. Yuan et al. [74] reported that carbon nano-
Despite the attractiveness of DMFCs, hurdles exist in the effec-
tive lifetime of the electrodes. Pt, considered to be one of the
best catalysts for the DMFCs, suffers from poisoning, especially at
the anode side during the methanol electro-oxidation by carbon
monoxide (CO) chemisorption, leading to a significant drop in the
overall performance [3,4]. Efforts have been made to mitigate CO
poisoning by the addition of other metals to platinum such as Ru
[5–10], Mo [11,12], W [11,13], Sn [11], Ni [14–16], and Nb [17].
Among these combinations, binary Pt Ru alloys are still considered
to be state of the art catalysts for methanol oxidation in DMFCs. The
∗
Corresponding authors.
E-mail address: carlos.cabrera2@upr.edu (C.R. Cabrera).
0013-4686/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.electacta.2012.04.091

192
E. You et al. / Electrochimica Acta 75 (2012) 191–200
tubes used as supports for platinum catalysts deposited with metal
oxides(CeO2, TiO2, and SnO2) were prepared for their application
as anode catalyst in DMFC. Catalyst with the addition of CeO2,
TiO2, and SnO2 presented higher catalytic activity than pure plat-
inum catalyst, and the catalysts with CeO2 were the best among.
Tang et al. [75] prepared PrxCeyOz(x/y = 3/1, 1/1, 1/3) modified Pt/C
catalysts by wet precipitation and reduction method. The results
showed that the Pt-(Pr1Ce1Oz)/C catalyst had the highest cata-
lyst activity, the best stability and CO-tolerante which could be
used as a suitable electrocatalyst for direct methanol fuel cell. Lee
and Manthiran [76] prepared Pt-CeO2/C catalysts prepared by one-
step syntheses of conventional borohydride reduction and reverse
microemulsion methods. The results suggested the possibility of
the Pt-CeO2/C-based catalysts, which are more stable than PtRu/C
and even Pt/C, as alterative anode catalysts for direct methanol
fuel cell. Guo et al. [31] employed PtRu-CeO2/C catalyst and found
that a particular composition, namely PtRu0.7(CeO2)0.3/C, yielded
a higher methanol oxidation current than an unmodified PtRu/C
catalyst. By wet-impregnation method, Scibioh et al. [23] prepared
Pt-CeO2/C catalyst and examined their methanol oxidation activity.
They found that 40 wt% Pt-9 wt% CeO2/C exhibited a better activity
and stability than the unmodified 40 wt% Pt/C catalyst.
aliquot of this catalyst containing solution was dropped onto glassy
carbon electrode surface. An and co-workers similarly pre-
pared PtRu carbon nanotube nanocomposites using different
metal precursors and dispersed the catalyst powder in N,N-
dimethylformamide to generate a black suspension that was coated
onto a glassy carbon electrode [57]. Jiang et al. impregnated Pd
directly onto a Nafion® electrolyte membrane in supercritical CO2
in order to reduce methanol crossover for DMFC [58]. It may be
desirable to directly deposit thin catalyst layers on top of the gas
diffusion layers (carbon paper) to replace the time- and labor-
intensive ink paint-brushing process currently in use.
Here we report the deposition of Pt and Pt-ceria anode catalysts
via SFD method directly onto gas diffusion layers and exam-
ined their activities for methanol oxidation in half cell and direct
methanol fuel cell systems. Further, these catalysts layers were
characterized using X-ray diffraction (XRD), X-ray photoelectron
spectroscopy (XPS) and scanning electron microscope-energy dis-
persive X-ray fluorescence spectroscopy (SEM-EDS) techniques.
2. Experimental
2.1. Fabrication of electrode layers
DMFC anode catalyst development requires not only the
exploration of new catalyst materials, which includes noble and
non-noble metals but also an innovative catalyst layer engineering
to fabricate thin, efficient electrodes.
2.1.1. Conventional brush-paint method
Gas diffusion layers (GDL) were made by coating Vulcan XC-
72R (1 mg-cm−2) onto teflonized carbon paper substrates (Toray
TGPH-060, 20% PTFE). 5 mg of Vulcan XC-72R was mixed with
200 ␮l of nanopure water and a mixture of 1:1 methanol and
100 ␮l of 2-propanol and sonicated for 1 h to create a slurry. Then,
5 wt% Nafion® solution (Sigma–Aldrich) was added to the mixture
to attain a 15 wt% polymer mixture. After sonication, the mix-
ture was stirred overnight. The carbon coating was applied by
spray-brushing the carbon ink over the oven-dried carbon paper
substrates until an approximate loading of 1 mg-cm−2 of ink was
achieved. After carbon coating, the electrodes are prepared for
catalyst ink deposition. Thus, catalyst ink dispersions were pre-
pared by mixing appropriate amounts of Pt-black (Johnson Matthey
PLC, London, UK), de-ionized water, isopropyl alcohol, and 5 wt%
Nafion® solution. The catalyst ink was brush-painted onto 5 cm2
of carbon paper (3 mg-cm−2). These brush-painted Pt-black elec-
trodes were used as the cathodes and anode control during the
course of this investigation in order to compare them with the
gas-diffusion electrodes made by SFD (Section 2.1.2).
Supercritical fluid deposition (SFD) yields pure and conformal
metal, metal oxide and alloy films on topographically complex sur-
faces with high deposition rates [32]. This combination of attributes
is enabled by the physico-chemical properties of supercritical fluids
(SCFs) including gas-like transport properties, liquid-like densities,
and zero surface tension [33]. Compared to conventional liquid
solvents, SCFs show enhanced mass transfer characteristics (low
viscosities and high diffusivities) while having equivalent solvat-
ing properties. The absence of surface tension enables the complete
wetting of surfaces. Carbon dioxide has been extensively used
as the process fluid for the supercritical deposition of metal and
metal oxide film depositions due to its easily accessible critical
point (Tc = 31 ◦C, Pc = 7.376 MPa), low-cost, non-flammability and
environmentally benign properties. Due to many organometallic
compounds being soluble in SCFs, such as carbon dioxide, SFD
provides orders of magnitude higher precursor concentrations
than traditional vapor based deposition methods such as chemical
vapor deposition (CVD). After the deposition, the reaction cham-
ber is simply depressurized to remove the reaction byproducts
to obtain high purity films [32]. Watkins, co-workers and others
have deposited various metal films, such as Ag [34], Au [35,36], Co
[37], Cu [33,38–42], Ni [33,37,43], Pd [36,44,45], Pt [36,46,47], Rh
[36], Ru [48,49] and oxides of Al [50,51], Bi [52], Ce [52], Ga [51],
Hf [52], Mn [50], Nb [52], Ru [50], Ta [52], Ti [52,53], Y [50,54],
Zr [50,52] using supercritical CO2. It has been demonstrated that
in supercritical carbon dioxide (scCO2), organometallic precursor,
such as dimethyl (1,5-cyclooctadiene) platinum (II), can be readily
reduced to yield high purity metal films on a Si wafer, organic sub-
strates, and within aluminum oxide membranes [47]. O’Neil et al.
carried out conformal deposition of various metal oxides, demon-
strating that the oxide deposition is owing to the hydrolysis of the
organometallic compounds in scCO2 [52]. Other work has focused
on the deposition of metal particles and films within porous inor-
ganic supports [45] and the deposition of metal particles within
polymers, including Teflon [47].
2.1.2. Supercritical fluid deposition method
Supercritical CO2 deposition of Pt and ceria was performed on
the GDLs, which were made as described in Section 2.1.1. The GDL
was placed in a high pressure reactor with an internal volume of
25 ml. The reactor configuration is shown in Fig. 1a. The GDL was
secured to the stage using, a stainless steel gasket, which has a
square through hole resulting in an exposed area of ∼5 cm2. A
known amount of metal precursor was put in the high pressure
reactor and the reactor was sealed with a Viton® O-ring.
A schematic of the sample structures with the sample notations
is shown in Fig. 1b along with the mass uptake in each layer after
the serial depositions. The GDL was dried in an oven maintained at
110 ◦C for 10 min before and after the SFD and the weights were
recorded. A typical SFD of the Pt layer used 0.05 mmol of dimethyl
(1,5-cyclooctadiene) platinum (II) (Strem Chemicals, Newburyport,
MA). The concentration of the precursor in scCO2 at the reaction
condition is at least an order of magnitude lower than its solubil-
ity limit reported in the literature [59]. The reactor and ancillary
tubing were initially purged with N2 while the reactor wall tem-
perature was set and controlled at 60 ◦C for 1 h. Supercritical CO2
was delivered via a 60 ◦C jacketed syringe pump (Teledyne ISCO
Recently, Lin et al. demonstrated preparing catalyst layers using
SCF, particularly for use in DMFCs [55]. PtRu nanoparticles were
deposited on carbon nanotubes in scCO2 using a methanol co-
solvent (20 vol%) and H2 at 200 ◦C [56]. The catalyst was then
mixed into a Nafion® solution to obtain a black suspension and an

E. You et al. / Electrochimica Acta 75 (2012) 191–200
193
Fig. 1. (a) Schematic representation of the supercritical fluid deposition reactor set-up which was used for the preparation of the anode electrodes for the direct methanol
fuel cells. (b) Anode catalyst samples prepared using supercritical fluid deposition and the cathode catalyst sample employed for the direct methanol fuel cell study. The
active area of each electrode was 5 cm2.
Inc. Lincoln, NE) and the reactor was pressurized to 17.2 MPa. The
reactor was maintained at these conditions for 1 h to dissolve the
precursor in CO2. A 200% molar excess of H2 was then dosed into
the reactor to induce the reduction of Pt precursor to Pt. After 1 h
the reactor was depressurized and purged with fresh scCO2 (2–3
times the reactor volume) while the effluent was passed through
an activated carbon packed bed cylinder before it was vented to the
air. This typical deposition was repeated until the desired amount
of Pt uptake is achieved.
For platinum and ceria co-deposition, tetrakis (2,2,6,6-
tetramethyl 3,5-heptanedionato) cerium (IV) (Strem Chemicals,
Newburyport, MA) was purchased and used as-received. Both the Pt
and Ce precursors (0.05 mmol each) were placed inside the reactor
and purged with N2 as it was done for the platinum-only deposi-
tion. scCO2 was charged to the reactor at 10.34 MPa and the wall
temperature was then heated to 150 ◦C for 1 h, resulting in a pres-
sure rise up to 19.58 MPa. The wall temperature was then reduced
to 60 ◦C and scCO2 was charged to the reactor to achieve a pres-
sure of 17.24 MPa. A 200% molar excess of H2 was then dosed into
the reactor to induce the precursor reduction to metal. The reactor
was depressurized and purged with fresh CO2 following the similar
sequences used for the Pt-only deposition. The intermediate heat-
ing step was necessary to ensure that the Ce precursor was fully
dissolved in scCO2.
mated Al K␣ X-ray source (1486.6 eV), settings at 25 W, 15 kV and
100 ␮m beam size. The take-off angle was set at 45. For survey spec-
tral acquisition (0–1000 eV), 187 eV pass energy was used with a
step size of 1.6 eV, while for the multiplexes, 46 eV with 0.2 eV step
size was used. For the sputter depth profiles, the samples were
subject to 500 eV Ar+ bombardment for 1 min on 1 mm × 1 mm
area and the multiplex spectra were collected after every 1 min
of sputtering for 20 cycles. The raw XPS data were analyzed using
Multipak software (v. 6.1a Physical Electronics, USA), and the spec-
tra were smoothened and fitted with least square method for the
deconvolution procedure. All of the multiplex spectra were shifted
according to the reference peak position of C1s at 284.5 eV. Each
spectrum was fitted with an analytical function consisting of a sum
of Gauss–Lorentzian shape functions for O1s and Ce3d, asymmetric
peaks for Pt4f, with a Shirley base-line correction. X-ray diffrac-
tograms were recorded on a PANalytical X’Pert PRO X’Celerator
˚
using Ni filtered Cu K␣ radiation (ꢀ = 1.54187 A). The data were
collected with 2ꢁ varying from 10 to 120, with an increment of
0.0167113◦. X-ray generator voltage was 45 kV and the tube cur-
rent was 40 mA. Scanning electron microscope (SEM) images and
EDS mapping results were assessed by the use of a JEOL JSM 6480
LV equipped with an EDAX Genesis 2000 detector.
2.3. Electrochemical measurements
We note that for the purposes of this study, only the anode layers
are prepared via supercritical fluid deposition, whereas the cathode
employed in fuel cell study is of the same Pt/C composition for all
the experiments and was prepared via a conventional brush-paint
method. One set of conventional Pt/C anode catalyst layer was fab-
ricated via brush-paint method and tested in DMFC for comparison
to the anodes prepared by SFD.
2.3.1. Half cell studies
The linear sweep and cyclic voltammetry studies were carried
out in a conventional, airtight, three-electrode cell with Ag/AgCl
as the reference electrode, platinum wire as the counter electrode
and thin layers of Pt/C or Pt-Ceria/C catalyst system as a working
electrode. The working electrode was prepared by mounting it on
a Teflon platform such that it had an effective area of 0.196 cm2
and it was exposed to an aqueous solution of 0.5 M H2SO4 and
1.0 M methanol. The tests were conducted at 25 ◦C at a scan rate of
20 mV s−1. All the electrochemical measurements were carried out
using a Potentiostat/Galvanostat model 263A EG & G Instruments
(Princeton Applied Research). Prior to making measurements with
2.2. Physicochemical characterizations
The deposited films were analyzed by XPS using a Physical
Instruments Quantum 2000 Scanning ESCA Microprobe (Physical
Electronics, USA). The instrument was operated with monochro-

194
E. You et al. / Electrochimica Acta 75 (2012) 191–200
Fig. 2. X-ray diffractogram of representative Pt/C and Pt-Ceria/C samples prepared using supercritical fluid deposition.
cyclic voltammetry and chronoamperometry, the electrolyte was
purged with N2 gas for 30 min. Chronoamperometry data were col-
lected for 1 h at 0.3 V vs. Ag/AgCl for the catalysts in a mixture of
0.5 M H2SO4 as electrolyte and 1.0 M methanol solution at 25 ◦C.
Throughout the paper, the potential values are indicated in SHE
scale. Water used for the experiments was previously distilled and
pumped through a Nanopure system (Barnstead) to give 18 Mꢂ-cm
nanopure water.
3. Results and discussion
3.1. Supercritical fluid deposition of Pt and Pt-Ceria on GDL
Crystalline Pt and amorphous ceria were deposited on the
GDL simultaneously using supercritical fluid deposition method.
SFD of ceria using tetrakis (2,2,6,6-tetramethy 3,5-heptanedionato)
cerium (IV) (Ce(tmhd)4) typically occurs at temperatures above
250 ◦C [52,60]. However, co-deposition of Pt and ceria occurred
at a temperature of 150 ◦C. These results are consistent with the
observations of Zhang and Puddephatt who reported catalyst-
enhanced chemical vapor deposition of yttrium oxide using
palladium precursors as catalysts. In their work, the presence of
Pd enabled the yttria deposition at a significantly lower tem-
perature [61]. Although the mechanism of catalyst-enhanced
chemical vapor deposition is not well understood, it is clear from
our XRD and XPS results, discussed below, that the presence
of Pt catalyzed ceria deposition at significantly lower tempera-
tures.
2.3.2. Single-cell testing
The thin layers of Pt/C or Pt-Ceria/C anode catalysts fabricated
via SFD used for testing in fuel cell assembly are shown in Fig. 1b-I
and b-II. Throughout the study, the Pt/C catalyst layers fabricated
via conventional paint method has been employed as the cathodes
(Fig. 1b-III).
A membrane electrode assembly (MEA) was fabricated by plac-
ing a Nafion® 117 membrane (Sigma–Aldrich) between a Pt or
Pt-Ceria (3 mg-cm−2) anode catalyst layer and a Pt (3 mg-cm−2
)
cathode followed by hot-pressing at 160 ◦C, 7.5 MPa for 5 min. The
MEAs had an effective area of 5 cm2. All experiments, including
the electrochemical measurements, were conducted with cells that
consisted of an MEA sandwiched between two graphite flow-field
plates. For all studies conducted in a fuel cell mode, 1.0 M methanol
solution was pumped through the anode side at a flow rate of
2 ml/min and air to the cathode side at a flow rate of 200 sccm.
The temperature and pressure of single cells were held at 25 or
60 ◦C and 0.101 MPa, respectively. Current–voltage curves were
measured galvanostatically.
The linear sweep and cyclic voltammetry studies were con-
ducted for the anode in the fuel cell mode by supplying 1.0 M
methanol to the anode and the cathode Pt/C was supplied with
a continuous stream of hydrogen (200 sccm) to make a standard
hydrogen electrode (SHE).
The deposition of cubic phase Pt crystals during the preparation
of Pt/C and Pt-Ceria/C was confirmed by X-ray diffraction (Fig. 2).
Debye–Scherrer analysis on the Pt (1 1 1) peaks of the two sam-
ples yields an estimated Pt crystal size of 9 1 nm. The diffraction
patterns do not reveal the presence of CeO2 nor Ce2O3. Mesguich
et al. performed CeO2 deposition in supercritical CO2 at substrate
temperatures of 250 and 300 ◦C. Their as-deposited films exhibited
crystallinity, while XRD and XPS studies revealed that the deposited
films were CeO2 [60]. For this study, the deposition temperature is
lower by 100 ◦C; therefore it is probable that the deposited ceria is
amorphous.
XPS survey spectra (not shown) indicated the presence of Pt and
C for Pt/C sample and Pt, Ce, O, and C for Pt-Ceria/C samples. The
multiplex spectra of Pt4f, C1s, O1s and Ce3d are shown in Fig. 3
after 1 min sputtering with Ar+ along with the results of curve
fitting. Analysis of the Pt/C and Pt-Ceria/C samples both revealed
Pt4f7/2 (71.2 eV) and Pt4f5/2 (74.5 eV) peaks (Fig. 3a and d). These
binding energies for Pt are in good accordance with the literature
values [62] indicating that the deposited Pt species are predomi-
nantly metallic. Pt/C sample did not show the O1s peak, whereas
Pt-Ceria/C sample did show O1s spectra. This further supports that
The methanol cross-over studies were made in the fuel cell
mode, by supplying 1.0 M methanol to the anode and the cathode
Pt/C was supplied with nitrogen (200 sccm). Linear sweep voltam-
mograms were made for the cathode to examine the extent of
methanol permeated from anode to cathode side while employing
different anode catalysts in the study.

E. You et al. / Electrochimica Acta 75 (2012) 191–200
195
Fig. 3. De-convoluted XPS multiplex after 1 min Ar+ sputtering (a) Pt4f of Pt/C sample; (b) O1s, (c) Ce3d, (d) Pt4f of Pt-Ceria/C sample, they were prepared using supercritical
fluid deposition.
the Pt layer deposited is pure metallic for Pt/C sample. The O1s peak
was fitted with two major peaks centered at binding energies of
529.6 eV and 531.7 eV (Fig. 3b). The literature reports different val-
ues for O1s binding energy for cerium oxides; 530.4–530.7 eV [63],
530.0 eV [64], 529.5 eV [65], 529.2 eV [66], but it can be inferred
that the deposited cerium is in oxide form. The XPS indicates that
the oxidation state of ceria is predominantly Ce3+, since it does
not show the Ce4+ characteristic signal shapes. Mesguich et al. [60]
observed as-deposited CeO2 films exhibit Ce4+ characteristic Ce3d
binding energy signal shapes on the surface and peaks attributed to
Ce3+ after sputtering into the bulk. However, analysis of the Pt and
ceria co-deposited sample suggests the presence of Ce3+ at the sur-
face. The reduction of Ce4+ to Ce3+ can be caused by various means
including Ar+ sputtering, X-ray irradiation, and H2 reduction [67].
The preparation of Pt-Ceria/C sample involved dosing H2 to initiate
the deposition of Pt. The presence of H2 at the deposition conditions
could result in the reduction of CeO2 to Ce2O3. It is also possible that
CeO2 was reduced during the Ar+ sputtering for XPS analysis. We
further note that ceria can be re-oxidized by treatments with CO2
[68] or methanol [69].
SEM characterization was carried out for the representative
Pt/C sample (Fig. 4b) and Pt-Ceria/C sample (Fig. 4c) and both are
compared to a blank electrode (carbon paper brush-painted with
Vulcan) in Fig. 4a. The lighter areas in the SEM images are indicative
of highly conductive material, presumably platinum. The electrodes
are porous and catalyst particles are well dispersed on the gas dif-
fusion layer. Recent investigations have shown that the porosity
of the deposited material is critical to the electro-activity of the
deposited materials in fuel cell applications [70]. Fig. 4c shows
SEM and EDS element mapping of Pt and ceria in a representative
Pt-Ceria/C sample. This result suggests that the SFD process main-
tains the desirable porous structure inherent to the GDL, although
platinum and ceria were not uniformly distributed throughout the
sample.

196
E. You et al. / Electrochimica Acta 75 (2012) 191–200
Fig. 5. Half-cell cyclic voltammetry of Pt/C or Pt-Ceria/C anode catalyst in 0.5 M
H2SO4, at a scan rate of 20 mV s−1
.
vs. SHE with the peak current density of 101.18 mA cm−2, indicating
that the addition of ceria to platinum in the anode catalyst system
enhances methanol oxidation activity.
Fig. 6b shows the cyclic voltammetry traces made for the thir-
tieth cycle. The Pt/C system exhibited a peak current density of
77.17 mA-cm−2 and the Pt-Ceria/C system showed a peak current
density of 101.14 mA-cm−2, indicating that the performance levels
are maintained even after multiple cycles. It can be further noted
that the reverse peak of Pt-Ceria/C is sharper and the peak current
density is found at 0.935 V vs. SHE, while the – reverse peak of Pt/C
is broader and the peak current density found at 0.855 V vs. SHE,
indicating the facile removal of CO with the Pt-Ceria/C compared
to Pt/C as anode catalyst system.
The chronoamperometric measurements are an effective
method to evaluate the electrocatalytic activity and stability of
the electrode material. Fig. 6c shows typical current density-time
responses for methanol oxidation measured at a fixed potential
of 0.5 V vs. SHE in 0.5 M H2SO4 aqueous solution containing 1.0 M
methanol [72]. The Pt-Ceria/C catalyst system exhibited higher cur-
rent densities throughout the testing time of 1 h, with comparable
rates of decay in activity between Pt/C and Pt-Ceria/C catalyst sys-
tems.
Fig. 4. (a) SEM image of a representative Pt/C sample, (b) SEM image of a gas
diffusion layer (carbon paper coated with Vulcan), and (c) SEM-EDS results for
a representative Pt-Ceria/C sample on which the supercritical fluid deposition is
performed.
3.3. DMFC fuel cell study with anodes prepared by SFD and paint
methods
In order to examine the influence of fabrication techniques
employed in catalyst layer formations on overall performance, sin-
gle cell tests were conducted using a direct methanol fuel cell
with Pt/C anode catalyst layers prepared using conventional paint
methods (Fig. 1b-III) and compared with results for the thin layers
prepared using SFD (Fig. 1b-I) under similar experimental condi-
tions. The results are shown in Fig. 7. It can be seen that the DMFC
single cell tests made with anode catalyst layers fabricated via SFD
technique exhibited better performance both at room temperature
and at 60 ◦C and compared to anode catalyst layers made by con-
ventional route. This performance enhancement might arise from
thinner anode catalyst layers, which would provide facile mass
transfer of reactants to the active sites and the removal of the
products, in addition to enhanced three phase boundary.
3.2. Half cell study of anodes prepared by SFD method
Thin layers of Pt/C or Pt-Ceria/C catalyst systems were subjected
to electrochemical characterization by using cyclic voltammetry in
0.5 M H2SO4. The voltammogram shown in Fig. 5 clearly indicates
the presence of platinum in both samples. The potential region
between 0.3 and 0 V is due to hydrogen adsorption and desorption
phenomenon on platinum, respectively, and the region between
0.6 and 0.85 V is due to PtO reduction. Adsorbed oxygen is reflected
in the increase in oxidation current at about 0.80 V, and its reduc-
tion by a cathodic current peak at about 0.65 V in the reverse sweep
[71,73].
The linear sweep voltammetry and cyclic voltammetric traces
of Pt/C or Pt-Ceria/C catalyst systems made in aqueous solution
of 0.5 M H2SO4 and 1.0 M methanol are shown in Figs. 6a and 6b
respectively. From Fig. 6a, it can be seen clearly that the onset
potential of Pt/C is 0.51 V vs. SHE with the peak current density
of 77.28 mA cm−2, while the onset potential of Pt-Ceria/C is 0.418 V
3.4. Anodes performance and fuel cross over tests in DMFC
The linear sweep and cyclic voltammetric traces of Pt/C or Pt-
Ceria/C anode catalyst systems made in a direct methanol fuel cell

E. You et al. / Electrochimica Acta 75 (2012) 191–200
197
Fig. 7. Direct methanol fuel cell comparison of single cell test made with Pt/C elec-
trodes fabricated through conventional paint method (solid line) and supercritical
fluid deposition (dashes). Anode feed: 1.0 M methanol, 2.0 ml/min; cathode feed:
air, 200 sccm; cell temperatures: 25 and 60 ◦C.
voltammetry traces made for these systems. These results are in
agreement with the results from the half-cell studies (Fig. 6b). Fur-
ther, the reverse peak of Pt-Ceria/C is sharper and the peak current
density found at 0.853 V vs. SHE, while the slope of backward peak
Fig. 6. Half-cell (a) linear sweep voltammetry, (b) cyclic voltammetry, and (c)
chronoamperometric response of Pt/C or Pt-Ceria/C anode catalyst in 0.5 M H2SO4
and 1.0 M methanol at a scan rate of 20 mV s−1. Chronoamperometry was done at
0.30 V vs. Ag/AgCl for 1 h.
mode by feeding 1.0 M methanol in the anode and 200 sccm hydro-
gen gas in the cathode side are shown in Fig. 8a and b, respectively.
From Fig. 8a, it can be seen that the onset potential of Pt/C is 0.555 V
vs. SHE with the peak current density of 37.6 mA cm−2, while the
onset potential of Pt-Ceria/C is 0.508 V vs. SHE with the peak cur-
rent density of 74.6 mA cm−2, indicating that the addition of ceria
to platinum in anode catalyst system enhances the methanol oxi-
dation activity. Fig. 8b shows the cyclic voltammetry traces made
after 30 cycles. The Pt/C exhibited the peak current density of
33.26 mA cm−2, and the Pt-Ceria/C shows a peak current density of
77.06 mA cm−2, indicating that the performance levels are main-
tained after several cycling compared to the fresh linear sweep
Fig. 8. Direct methanol fuel cell (a) linear sweep voltammetry and (b) cyclic voltam-
metry of Pt/C or Pt-Ceria/C anode catalyst in single cell mode with a H2 cathode
feed. Anode feed: 1.0 M methanol, 2.0 ml/min; cathode feed: H2 gas, 200 sccm; cell
temperature: 25 ◦C.

198
E. You et al. / Electrochimica Acta 75 (2012) 191–200
Fig. 10. Direct methanol fuel cell (a) current polarization and (b) power curves
of Pt/C and Pt-Ceria/C as anode catalyst in single cell mode. Anode feed: 1.0 M
Fig. 9. Direct methanol fuel cell anode polarization curves of (a) Pt/C and (b) Pt-
Ceria/C as anode catalyst in single cell mode with a H2 cathode feed. Anode feed:
1.0 M methanol, 2.0 ml/min; cathode feed: H2 gas, 200 sccm; cell temperatures: 25,
60, 70 and 80 ◦C.
methanol, 2.0 ml/min; cathode feed: air, 200 sccm; cell temperatures: 25 and 60 ◦C.
temperatures, 25 and 60 ◦C as given in Fig. 10b, showing that the
Pt-Ceria/C anode catalyst system exhibited higher power at a given
current at both cell temperatures.
The performance stability of the anodes Pt/C or Pt-Ceria/C in the
direct methanol fuel cell-single cell assembly was tested and the
of Pt/C is steady and the peak current density found at 0.804 V vs.
SHE, indicating the facile removal of CO with the Pt-Ceria/C anode
catalyst system.
Anode polarization traces of Pt/C or Pt-Ceria/C anode cata-
lyst systems were made in single cell assembly by feeding 1.0 M
methanol in the anode and 200 sccm hydrogen gas in the cathode
side at various cell temperatures such as 25, 60, 70 and 80 ◦C, Fig. 9a
and b. It is found that at all temperatures of single cell operation
studied; the Pt-Ceria/C catalyst system exhibited higher current
density at any given potential.
The current-polarization and power curves for the direct
methanol fuel cell were determined by supplying 1.0 M methanol
to the anode and 200 sccm air at the cathode at two different oper-
ating temperatures 25 and 60 ◦C and the results are shown in Fig. 10.
The tested anodes were Pt/C or Pt-Ceria/C anode catalyst systems
prepared via SFD, while the cathode was Pt/C prepared via con-
ventional paint method. As shown in Fig. 10a, the Pt/C system
exhibited an open circuit voltage (OCV) of 0.403 V and 0.429 V at
25 and 60 ◦C, respectively. The Pt-Ceria/C anode catalyst systems
exhibited a 0.410 V and 0.443 V at 25 and 60 ◦C, respectively. The
higher OCV values shown by Pt-Ceria/C anode catalyst systems at
both temperatures indicate its higher methanol oxidation activ-
ity, consequently leading to higher fuel utilization at the anode and
lower methanol crossover. In addition, the power vs. current curves
were made for the anodes Pt/C or Pt-Ceria/C at two different cell
Fig. 11. Direct methanol fuel cell chronoamperometric traces of Pt/C or Pt-Ceria/C
anode catalyst in direct methanol fuel cell under a constant applied cell voltage of
0.1 V for 1 h. Anode feed: 1.0 M methanol, 2.0 ml/min; cathode feed: air, 200 sccm;
cell temperature: 25 ◦C. The cathode was used as a reference electrode, as well.

E. You et al. / Electrochimica Acta 75 (2012) 191–200
199
mesoporous carbon nanoparticles prior to fabrication of the GDL
using conventional techniques.
Acknowledgments
This work was supported by the NSF Center for Hierarchical
Manufacturing at the University of Massachusetts (CMMI-
0531171) and by NASA URC Center for Advanced Nanoscale
Materials (NNX10AQ17A). This material is based upon work sup-
ported in part by, the U.S. Army Research Office under grant number
54635CH. Eunyoung You and Rolando Guzmán-Blas equally con-
tributed to this work.
References
[1] B. Viswanathan, M.A. Scibioh, Fuel Cells: Principles and Applications, CRC Press,
Taylor and Francis Group, New York, 2007.
[2] M.A. Scibioh, B. Rajesh, B. Viswanathan, Proceedings of the Indian National
Science Academy Part A 68 (2002) 99.
[3] G.T. Burstein, C.J. Barnett, A.R. Kucernak, K.R. Williams, Catalysis Today 38
(1997) 425.
Fig. 12. Direct methanol fuel cell methanol cross over linear sweep voltammetry
traces for Pt/C cathode, while employing Pt/C or Pt-Ceria/C anode catalyst. The anode
was used as a reference electrode, as well. Anode feed: 1.0 M methanol, 2.0 ml/min;
cathode feed: air, 200 sccm; cell temperature: 25 ◦C.
[4] A. Hamnett, Catalysis Today 38 (1997) 445.
[5] E. Antolini, Materials Chemistry and Physics 78 (2003) 563.
[6] C. Bock, C. Paquet, M. Couillard, G.A. Botton, B.R. MacDougall, Journal of the
American Chemical Society 126 (2004) 8028.
[7] G.A. Camara, M.J. Giz, V.A. Paganin, E.A. Ticianelli, Journal of Electroanalytical
Chemistry 537 (2002) 21.
[8] P. Costamagna, S. Srinivasan, Journal of Power Sources 102 (2001) 242.
[9] H.A. Gasteiger, N. Markovic, P.N. Ross, E.J. Cairns, Journal of Physical Chemistry
98 (1994) 617.
[10] V. Radmilovic, H.A. Gasteiger, P.N. Ross, Journal of Catalysis 154 (1995) 98.
[11] M. Götz, H. Wendt, Electrochimica Acta 43 (1998) 3637.
[12] G. Samjeske, H. Wang, T. Löffler, H. Baltruschat, Electrochimica Acta 47 (2002)
3681.
[13] P.K. Shen, K.Y. Chen, A.C.C. Tseung, Journal of the Electrochemical Society 142
(1995) 85.
[14] H.R. Colón-Mercado, H. Kim, B.N. Popov, Electrochemistry Communications 6
(2004) 795.
[15] T.C. Deivaraj, W. Chen, J.Y. Lee, Journal of Materials Chemistry 13 (2003) 2555.
[16] K.W. Park, J.H. Choi, B.K. Kwon, S.A. Lee, Y.E. Sung, H.Y. Ha, S.A. Hong, H. Kim,
A. Wieckowski, Journal of Physical Chemistry B 106 (2002) 1869.
[17] D.C. Papageorgopoulos, M. Keijzer, F.A. De Bruijn, Electrochimica Acta 48 (2002)
197.
chronoamperometry traces were made under a constant applied
cell voltage of 0.1 V for 1 h as shown in Fig. 11. The Pt-Ceria/C
exhibited higher current density values that were nearly constant
throughout the test period. A methanol cross over linear sweep
voltammetry study was carried out at the cathode side by supplying
1.0 M methanol in the anode and the cathode was fed with stream
of nitrogen gas with a flow rate of 200 sccm. These results are
compared with the Pt/C anode catalyst under similar experimental
conditions as shown in Fig. 12. The peak current density values are
indicative of the extent of permeated methanol oxidation in the
cathode side. It can be seen that while employing Pt-Ceria/C as the
anode catalyst, the permeated methanol is decreased and hence,
lower current density values were found, whereas with Pt/C as the
anode catalyst there is a higher extent of permeated methanol. This
could be indicative of better catalytic activity of Pt-Ceria/C toward
methanol oxidation at the anode side, thus enhancing the fuel uti-
lization, thereby restricting the methanol permeation.
[18] K.-W. Park, J.-H. Choi, K.-S. Ahn, Y.-E. Sung, Journal of Physical Chemistry B 108
(2004) 5989.
[19] Z. Jusys, T.J. Schmidt, L. Dubau, K. Lasch, L. Jörissen, J. Garche, R.J. Behm, Journal
of Power Sources 105 (2002) 297.
4. Conclusion
[20] K. Lasch, L. Jörissen, J. Garche, Journal of Power Sources 84 (1999) 225.
[21] L.X. Yang, C. Bock, B. MacDougall, J. Park, Journal of Applied Electrochemistry
34 (2004) 427.
[22] C.H. Park, M.A. Scibioh, H.J. Kim, I.H. Oh, S.A. Hong, H.Y. Ha, Journal of Power
Sources 162 (2006) 1023.
[23] M.A. Scibioh, S.-K. Kim, E.A. Cho, T.-H. Lim, S.-A. Hong, H.Y. Ha, Applied Catalysis
B: Environmental 84 (2008) 773.
[24] A. Bauer, C.W. Oloman, E.L. Gyenge, Journal of Power Sources 193 (2009) 754.
[25] S. Litster, G. McLean, Journal of Power Sources 130 (2004) 61.
[26] A.M. Zainoodin, S.K. Kamarudin, W.R.W. Daud, International Journal of Hydro-
gen Energy 35 (2010) 4606.
[27] H.B. Yu, J.H. Kim, H.I. Lee, M.A. Scibioh, J. Lee, J. Han, S.P. Yoon, H.Y. Ha, Journal
of Power Sources 140 (2005) 59.
[28] C. Xu, P.K. Shen, Chemical Communications 2004 (2004) 2238.
[29] C. Xu, P.K. Shen, Journal of Power Sources 142 (2005) 27.
[30] C.L. Campos, C. Roldán, M. Aponte, Y. Ishikawa, C.R. Cabrera, Journal of Electro-
analytical Chemistry 581 (2005) 206.
[31] J.W. Guo, T.S. Zhao, J. Prabhuram, R. Chen, C.W. Wong, Journal of Power Sources
156 (2006) 345.
[32] A.H. Romang, J.J. Watkins, Chemical Reviews 110 (2010) 459.
[33] J.M. Blackburn, D.P. Long, A. Cabanas, J.J. Watkins, Science 294 (2001) 141.
[34] B. Zhao, T. Momose, T. Ohkubo, Y. Shimogaki, Microelectronic Engineering 85
(2008) 675.
Supercritical fluid deposition (SFD) was utilized to deposit Pt
and ceria simultaneously to construct the fuel cell anode for a direct
methanol fuel cell system and for the study of the methanol oxida-
tion process. It was found that Pt catalyzes the deposition of ceria
and allows for co-deposition at a reduced temperature as low as
60 ◦C. The deposited Pt was predominantly metallic and crystalline
with estimated grain size of 9 nm. The thin layer electrode formed
via supercritical fluid deposition exhibited higher performance in
fuel cell operations compared to electrodes prepared with a con-
ventional way of coating the gas diffusion layers by the brush and
paint method when Pt/C was employed as anode catalyst. Further,
the Pt-ceria catalyst deposited by SFD exhibited higher methanol
oxidation activity compared to the platinum catalyst alone. Thus,
both the method of deposition as well as control of composition
are important factors. The linear sweep traces of the cathode made
for the methanol cross over study indicate that Pt-Ceria/C as the
anode catalyst, due to its better activity for methanol, improves the
fuel utilization, minimizing the methanol permeation from anode
to cathode compartment. Studies are underway, to improve the
three phase boundary by the further optimization of the deposi-
tion technique and fabrication of membrane electrode assembly to
realize improved DMFC performance. The current study has pri-
marily focused on directly depositing Pt and ceria onto the carbon
paper (GDL). However, with the current SFD approach it will be
feasible to deposit these catalysts to the carbon nanotube or other
[35] A. Caban˜as, D.P. Long, J.J. Watkins, Chemistry of Materials 16 (2004) 2028.
[36] D.P. Long, J.M. Blackburn, J.J. Watkins, Advanced Materials 12 (2000) 913.
[37] E.T. Hunde, J.J. Watkins, Chemistry of Materials 16 (2004) 498.
[38] A. Caban˜as, X. Shan, J.J. Watkins, Chemistry of Materials 15 (2003) 2910.
[39] E. Kondoh, Japanese Journal of Applied Physics 43 (2004) 3928.
[40] E. Kondoh, K. Shigama, Thin Solid Films 491 (2005) 228.
[41] M. Matsubara, M. Hirose, K. Tamai, Y. Shimogaki, E. Kondoh, Journal of the
Electrochemical Society 156 (2009) H443.
[42] Y. Zong, J.J. Watkins, Chemistry of Materials 17 (2005) 560.
[43] Q. Peng, J.C. Spagnola, G.N. Parsons, Journal of the Electrochemical Society 155
(2008) D580.

200
E. You et al. / Electrochimica Acta 75 (2012) 191–200
[44] J.M. Blackburn, D.P. Long, J.J. Watkins, Chemistry of Materials 12 (2000) 2625.
[45] N.E. Fernandes, S.M. Fisher, J.C. Poshusta, D.G. Vlachos, M. Tsapatsis, J.J. Watkins,
Chemistry of Materials 13 (2001) 2023.
[61] Y. Zhang, R.J. Puddephatt, Chemistry of Materials 11 (1999) 148.
[62] J.F. Moulder, J. Chastain, Handbook of X-ray Photoelectron Spectroscopy,
Perkin-Elmer, Eden Prairie, MN, 1992.
[46] C.S. Lin, F.L.Y. Lam, X. Hu, W.Y. Tam, K.M. Ng, Journal of Physical Chemistry C
112 (2008) 10068.
[47] J.J. Watkins, J.M. Blackburn, T.J. McCarthy, Chemistry of Materials 11 (1999)
213.
[48] C.F. Karanikas, J.J. Watkins, Microelectronic Engineering 87 (2010) 566.
[49] A. O’Neil, J.J. Watkins, Chemistry of Materials 18 (2006) 5652.
[50] T. Gougousi, D. Barua, E.D. Young, G.N. Parsons, Chemistry of Materials 17
(2005) 5093.
[51] Q. Peng, D. Hojo, K.J. Park, G.N. Parsons, Thin Solid Films 516 (2008) 4997.
[52] A. O’Neil, J.J. Watkins, Chemistry of Materials 19 (2007) 5460.
[53] H. Uchida, A. Otsubo, K. Itatani, S. Koda, Japanese Journal of Applied Physics 44
(2005) 1901.
[54] T. Gougousi, Z. Chen, Thin Solid Films 516 (2008) 6197.
[55] Y. Lin, X. Cui, C. Yen, C.M. Wai, Journal of Physical Chemistry B 109 (2005) 14410.
[56] Y. Lin, X. Cui, C.H. Yen, C.M. Wai, Langmuir 21 (2005) 11474.
[57] G. An, P. Yu, L. Mao, Z. Sun, Z. Liu, S. Miao, Z. Miao, K. Ding, Carbon 45 (2007)
536.
[58] R. Jiang, Y. Zhang, S. Swier, X. Wei, C. Erkey, H.R. Kunz, J.M. Fenton, Electro-
chemical and Solid-State Letters 8 (2005) A611.
[59] O. Aschenbrenner, S. Kemper, N. Dahmen, K. Schaber, E. Dinjus, Journal of
Supercritical Fluids 41 (2007) 179.
[63] D.R. Mullins, S.H. Overbury, D.R. Huntley, Surface Science 409 (1998) 307.
[64] P. Concepción, A. Corma, J. Silvestre-Albero, V. Franco, J.Y. Chane-Ching, Journal
of the American Chemical Society 126 (2004) 5523.
[65] J.P. Holgado, G. Munuera, J.P. Espinós, A.R. González-Elipe, Applied Surface
Science 158 (2000) 164.
[66] D. Barreca, G.A. Battiston, R. Gerbasi, E. Tondello, Surface Science Spectra 7
(2000) 297.
[67] L. Qiu, F. Liu, L. Zhao, Y. Ma, J. Yao, Applied Surface Science 252 (2006) 4931.
[68] T. Staudt, Y. Lykhach, N. Tsud, T. Skála, K. Prince, V. Matolín, J. Libuda, Journal
of Catalysis 275 (2010) 181.
[69] C. Binet, A. Badri, J.C. Lavalley, Journal of Physical Chemistry 98 (1994) 6392.
[70] S.K. Natarajan, J. Hamelin, Journal of Power Sources 195 (2010) 7574.
[71] Z. Liu, L. Ming Gan, L. Hong, W. Chen, J. Yang Lee, Journal of Power Sources 139
(2005) 73.
[72] G.Q. Lu, W. Chrzanowski, A. Wieckowski, Journal of Physical Chemistry B 104
(2000) 5566.
[73] S. Patra, N. Munichandraiah, Langmuir 25 (2009) 1732.
[74] H. Yuan, D. Guo, X. Qiu, W. Zhu, L. Chen, Journal of Power Sources 188 (2009) 8.
[75] Z.C. Tang, G.X. Lu, Applied Catalysis B: Environmental 79 (2008) 1.
[76] E. Lee, A. Manthiram, Journal of Physical Chemistry C 114 (2010) 21833.
[60] D. Mesguich, C. Aymonier, J.M. Bassat, F. Mauvy, E. You, J.J. Watkins, Chemistry
of Materials 23 (2011) 5323.