Complete quantitative online analysis of methanol electrooxidation products via electron impact and electrospray ionization mass spectrometry

Article abstract of DOI:10.1021/ac203276f

We report on a novel approach for complete quantitative online product analysis in electrocatalytic reactions, combining electron impact ionization mass spectrometry (EI-MS) and electrospray ionization mass spectrometry (ESI-MS) for simultaneous detection of both volatile and nonvolatile reaction products. The potential of this method is demonstrated using continuous methanol oxidation in a flow cell. The overall reaction rate was followed via the Faradaic current; CO₂ formation was monitored mass spectrometrically via a membrane inlet system, and formaldehyde and formic acid were detected by ESI-MS after a derivatization-extraction-separation procedure introduced recently (Zhao, W.; Jusys, Z.; Behm, R. J. Anal. Chem.2010, 82, 2472-2479) providing quantitative data on the product distribution. In a more general sense, this approach is applicable for a wide range of reactions at the solid-liquid interface or in liquid phase.

Full text of DOI:10.1021/ac203276f

Technical Note
pubs.acs.org/ac
Complete Quantitative Online Analysis of Methanol Electrooxidation
Products via Electron Impact and Electrospray Ionization Mass
Spectrometry
Wei Zhao,† Zenonas Jusys, and R. Jurgen Behm*
̈
Institute of Surface Chemistry and Catalysis, Ulm University, D-89069 Ulm, Germany
ABSTRACT: We report on a novel approach for complete quantitative online
product analysis in electrocatalytic reactions, combining electron impact
ionization mass spectrometry (EI-MS) and electrospray ionization mass
spectrometry (ESI-MS) for simultaneous detection of both volatile and
nonvolatile reaction products. The potential of this method is demonstrated
using continuous methanol oxidation in a flow cell. The overall reaction rate was
followed via the Faradaic current; CO2 formation was monitored mass
spectrometrically via a membrane inlet system, and formaldehyde and formic acid were detected by ESI-MS after a
derivatization−extraction−separation procedure introduced recently (Zhao, W.; Jusys, Z.; Behm, R. J. Anal. Chem. 2010, 82,
2472−2479) providing quantitative data on the product distribution. In a more general sense, this approach is applicable for a
wide range of reactions at the solid−liquid interface or in liquid phase.
he oxidative conversion of organic fuels into electric
energy in direct oxidation fuel cells such as direct
combination of electrospray ionization mass spectrometry (ESI-
MS) for the analysis of nonvolatile (ESI-MS) and electron
impact mass spectrometry (EI-MS) for volatile reaction
products.
T
methanol fuel cells (DMFCs) or direct ethanol fuel cells
(DEFCs) is a promising route in sustainable energy concepts,
with potential applications in various areas.1,2 A major problem
in the realization of these technologies is the fact that these
reactions may lead to a number of volatile and nonvolatile
incomplete oxidation reaction intermediates, in addition to the
stable final product CO2. Part of these products is highly toxic,
such as formaldehyde or acetaldehyde, and therefore, the
formation of these side products should be minimized. In
addition, their formation lowers the efficiency of the energy
conversion process significantly. Therefore, the complete,
quantitative analysis of the reaction products in these
electrocatalytic reactions, possibly online during the reaction,
has been a key issue in electrochemistry over the last decades.
A common method used for such kind of analysis, differential
electrochemical mass spectrometry (DEMS),3−5 has been
highly successful for the identification and quantitative product
analysis in “simple” reactions such as CO oxidation, involving
few and highly volatile products only,3 but is quickly limited in
its applicability for more complex reactions, due to the low
volatility of some reaction products and their fragmentation
upon electron impact ionization. Recently, DEMS has been
coupled with an additional detection electrode, allowing one to
detect also reactive nonvolatile species, though one cannot
distinguish between different species.6 Analysis by chromato-
graphic,7 spectroscopic,8−10 and analytical techniques,11−13
where these restrictions are absent, was either applicable only
for part of the reaction products or could not be used online,
for real time analysis.
ESI-MS is an excellent technique for the direct detection of
complex liquid phase products of electrochemical reactions,14,15
since the soft ionization mode avoids fragmentation of the
product molecules. Its application for fuel cell related
electrocatalysis research, however, was not possible so far due
to the high acidity of the supporting electrolyte (ion
suppression and instrument corrosion) and due to the
adsorption/reaction of the nonaqueous solvents commonly
used for electrochemical ESI-MS with the noble metal
electrodes. In addition, aldehydes (one of the products in the
oxidation of alcohols) have a very low ionization probability in
ESI-MS. Recently, we developed a system for continuous
derivatization−extraction, which allowed us to extract organic
molecules into a mineral acid free organic phase and applied it
for the quantitative detection of formic acid and formaldehyde
in a highly acidic supporting electrolyte by ESI-MS.16 In that
case, the analyte was offline collected in separate methanol
electrochemical oxidation experiments and later injected into
the online derivatization−extraction−separation system to be
detected by ESI-MS. In the present study, we went a decisive
step forward, connecting this setup with an electrochemical cell
for online analysis of the methanol oxidation products, during
continuous reaction, and combining it with EI-MS via a
membrane inlet. In previous DEMS experiments, the formation
of formic acid was followed by the production of methyl-
formate, and formaldehyde formation was calculated assuming
In the present contribution, we describe a novel approach for
the complete quantitative real time online analysis of
continuous methanol oxidation in a flow cell, involving the
Received: December 10, 2011
Accepted: May 15, 2012
Published: May 15, 2012
© 2012 American Chemical Society
5479
dx.doi.org/10.1021/ac203276f | Anal. Chem. 2012, 84, 5479−5483

Analytical Chemistry
Technical Note
Figure 1. Schematic drawing of the combined setup for online analysis of gaseous and nonvolatile reaction products, using electron impact ionization
mass spectrometry via a membrane inlet in a thin-layer electrochemical (EC) flow cell and electrospray ionization mass spectrometry for the
simultaneous quantitative detection of the methanol electrooxidation products in a single experiment.
that formaldehyde, formic acid, and CO2 are the only reaction
products. Though plausible, the latter assumption has never
been proven quantitatively in an online experiment.18−20 In the
present study, we established a combination of the two
methods (DEMS and ESI-MS) for the quantitative online
analysis of both gas phase and liquid phase products in a
continuous electrocatalytic reaction,17 giving direct quantitative
access to the volatile and soluble reaction products of the
methanol oxidation reaction.
saturated calomel electrode (SCE) connected to the outlet of
the DEMS cell by a Teflon capillary served as reference
electrode. All potentials, however, are quoted against that of the
reversible hydrogen electrode (RHE).
Deaerated electrolyte (0.1 M methanol in 0.5 M sulfuric
acid) was introduced into the thin-layer electrochemical (EC,
Figure 1) flow cell using a gastight glass syringe (Harward
Apparatus) operated by a syringe pump (see Figure 1). The
electrolyte and products formed at the electrode were
transported to the second thin-layer compartment, which is
interfaced to the mass spectrometer via a porous membrane
(Scimat, 60 μm thick, 50% porosity, 0.2 μm pore diameter).21
Quantitative Detection of Volatile Reaction Products
via EI-MS. The online mass spectrometry setup for gaseous/
volatile product analysis was described in detail elsewhere.19,20
It consists of a differentially pumped two-chamber system with
a Balzers QMS 112 quadrupole mass spectrometer, a Pine
Instruments potentiostat, and a computerized data acquisition
system.
For the quantitative detection of carbon dioxide, the mass
spectrometer setup was calibrated in separate experiments via
potential-step electro-oxidation of formic acid, using the same
flow rate to ensure an identical collection efficiency of the
membrane inlet in the calibration experiments. The sensitivity
factor K* was determined in these calibration experiments as
the ratio of mass spectrometric current (charge), multiplied by
two (number of electrons in oxidation of formic acid molecule
to CO2), and the Faradaic current (charge). Using the same
proportionality, the ion current (charge) was converted into
partial current for methanol oxidation to CO2 by accounting for
a release of six electrons per one molecule.
Quantitative Detection of Nonvolatile Reaction
Products via ESI-MS. The out-flowing electrolyte at the
outlet of the cell is mixed in a T connector (Upchurch
Scientific) with a saturated 2,4-DNPH solution in 0.5 M
sulfuric acid, which is delivered from a separate syringe pump,
for the derivatization of formaldehyde to form 2,4-dinitro-
phenyl hydrazone in a subsequent capillary.19 After derivatiza-
tion, the streams of the aqueous reaction mixture and an
organic extraction phase (isobutylacetate) are guided against
each other to meet in the next T connector. Due to the
immiscibility of the phases and their equal flow rates (0.05 mL
min−1), small and rather regular separate segments of the
EXPERIMENTAL SECTION
■
Continuous electrooxidation of methanol at different constant
potentials was performed over a commercial carbon-supported
Pt/C catalyst (E-Tek, Inc.) in methanol containing sulfuric acid
solution, which was fed into a dual thin-layer electrochemical
flow cell. A schematic drawing of the combined setup for the
online quantitative analysis of the methanol electrooxidation
reaction products, employing a combination of EI-MS and ESI-
MS for simultaneous detection of both volatile and nonvolatile
species, is depicted in Figure 1. The flow cell was interfaced to
an EI-MS setup via a membrane inlet system. The outlet of the
flow cell was linked to the analyte pretreatment system for
sequential derivatization−extraction−separation of nonvolatile
components, which in turn was connected to the ESI-MS. In
this configuration, the overall methanol oxidation reaction rate
was monitored via the Faradaic current, and CO2 formation was
followed by EI-MS via the membrane inlet system, while
formaldehyde and formic acid were both detected after online
extraction into an immiscible organic phase by ESI-MS.
Electrode Preparation and Electrochemical Measure-
ments. The thin-film Pt/Vulcan (E-TEK, Inc.) electrode was
prepared on a mirror-polished glassy carbon disk (Sigradur G,
Hochtemperatur Werkstoffe GmbH, Germany) by sequential
pipetting/drying of 20 μL of an aqueous Pt/Vulcan (20 wt %
Pt) suspension and aqueous Nafion solution,19,20 resulting in a
catalyst film of 6 mm in diameter with a platinum loading of 28
μgPt cm−2. The electrode was mounted in the thin-layer DEMS
flow cell,21 so that the catalyst film was exposed to the solution
within a centered circular tightener (inner diameter of 6 mm,
ca. 50 μm thick), forming a thin electrolyte layer of ca. 5 μL of
volume on the electrode. Two Pt counter electrodes
interconnected via the external resistance were used. A
5480
dx.doi.org/10.1021/ac203276f | Anal. Chem. 2012, 84, 5479−5483

Analytical Chemistry
Technical Note
individual phases are formed and pass through a Teflon
capillary (see magnified sketch in Figure 1). This enables an
efficient extraction (ca. 25%19) of the organic molecules from
the aqueous phase into the organic phase at the liquid−liquid
phase interface due to the high ratio of the contact area
between the phases and their volume. (Note that the
esterification reaction of methanol and formic acid occurs in
acid solution, with an equilibrium concentration of the resulting
methylformate of ∼10% of that of formic acid. During
extraction, the hydrolysis of the ester prevails, allowing the
ester to decompose and thus to extract formic acid into the
organic phase.)
Separation of the two phases is achieved in a third T-
connector. After extraction and phase separation, only a mineral
acid-free organic phase is guided via a capillary to the injection
needle of the ESI-MS. The potentiostat, syringe pumps, and
other peripheral devices are connected to the power lines via an
isolation transformer (Tufvassons).
suppression is not favorable since it further reduces the analyte
concentration, especially when the methanol oxidation reaction
is performed at low potential with low product concentrations.
The mass spectrometric m/z 45 signal was monitored for
formic acid detection, and 2,4-dinitrophenyl hydrazone was
detected at the m/z 208.7 signal (single ion mode). The
presence of sulfuric acid in the organic phase was routinely
checked at m/z 97. The signals were quantified by integrating
the area of the corresponding mass spectrometric peaks using
the software provided by Varian Inc..
RESULTS AND DISCUSSION
■
Figure 2a,b presents the Faradaic current (a) and the mass
spectrometric (b) m/z 44 ion current transients upon stepping
the working electrode potential from 0.06 to 0.65 VRHE (vs the
Reversible Hydrogen Electrode, RHE) for 10 min and then
stepping back to the initial potential. Figure 2c,d depicts the
corresponding ESI-MS time responses for the mass signals of
The flow rates of both inlet and derivatization syringes (0.05
mL min−1) are controlled by a common multisyringe pump
(Harward PHD 2000), whereas the organic phase for extraction
(isobutylacetate) and the mobile phase (pure water) are driven
by separate syringe pumps (Harward 11Plus) at flow rates of
0.1 and 0.2 mL min−1, respectively (for details, see ref 16).
Special care was taken to avoid possible back-diffusion of the
analyte from the derivatization loop back into the cell via the
outflow capillary, by optimizing the diameter and the length of
the outflow and derivatization capillaries. (Note that the
calibration was performed using the same flow cell with a bare
glassy carbon substrate at an open-circuit and the same capillary
network as for the measurements.) For the ESI-MS measure-
ments, the electrospray ionization mass spectrometer (Varian,
Inc., model 1200 L) was used in the negative ion mode
(detector voltage, 1 kV; needle voltage, −4.5 kV; needle
current, ca. −17 μA; the ESI chamber at 50 °C and the
nebulizing gas at 350 °C (20 psi)), since (i) both analyte
molecules, formic acid and the derivatized product of
formaldehyde (2,4-dinitrophenyl hydrazone), have functional
groups that readily lose a proton and since (ii) this allows one
to avoid oxidation of the analyte during the electrospray
process possible at the positive ion mode. These settings were
used both in the experiments and in calibration measurements,
in which the mixtures of methanol with formaldehyde and
formic acid at known concentrations were analyzed after
passing via the same derivatization−extraction−separation
network. This way, we can avoid that possible electrochemical
reaction in the ESI source, such as the reduction of
dinitrophenyl−hydrazone in the negative ion ESI mode,
affecting the quantification of the reaction products. The
analyte was periodically injected manually into the ESI-MS
employing a Rheodyne 6-port valve, first sampling an aliquot (5
μL) from the continuously flowing organic phase into the
sample loop and then injecting these 5 μL of the analyte from
the sample loop into the continuously flowing mobile phase
(pure water). The Rheodyne 6-port valve and the connection
to the PEEK capillary leading to the ESI source were electrically
grounded. The choice of water as mobile phase was based on
the low resulting background signal, significantly lower than,
e.g., that observed for acetonitrile or water−methanol mixtures,
and its immiscibility with the organic phase (isobutyl acetate).
Because of the considerable loss of reaction products during
extraction and derivatization processes, addition of acetonitrile
to the mobile phase to reduce the water induced ion
Figure 2. Simultaneously measured response upon stepping the
working electrode potential from 0.06 V (at 1 min) to 0.65 V (RHE)
for 10 min and back to the initial potential. Faradaic current (a); m/z
44 ion current (b) transients; time response for the mass spectrometric
signals of formic acid at m/z 45 (c); and derivatized formaldehyde at
m/z 208.7 (d). Pt/C catalyst loading: 28 μgPt cm−2; solution: 0.1 M
methanol in 0.5 M sulfuric acid; room temperature.
5481
dx.doi.org/10.1021/ac203276f | Anal. Chem. 2012, 84, 5479−5483

Analytical Chemistry
Technical Note
formic acid at m/z 45 (c) and derivatized formaldehyde (d) at
m/z 208.7. The CO2 ion current exhibits a 12−13 s delay in the
response of the reaction products due to the time required for
transporting the electrolyte into the second compartment of the
thin-layer cell containing the membrane inlet.
For the ESI-MS signals, the time delay is significantly longer,
with around 7 min until the reaction products reach the ESI-
MS and an additional few minutes until the signals reach stable
values. This time delay (7 min) reflects the residence time in
the derivatization and extraction capillaries and the subsequent
time required for the analyte in the organic phase to reach the
ESI-MS ionization chamber. Steady-state values are reached
around 12 min after the initial potential step and remain
constant for ca. 5 min for a potential step duration of 10 min.
The product concentrations were determined from the
averaged peak intensities during these 5 min by standard
calibration (see ref 15). At the reverse potential step, back to
the initial value, the time required for the ESI-MS signals to
reach the background level is slightly longer (Figure 2b), which
we attribute to a combination of slow diffusive release of
trapped product molecules from small gaps in the interconnects
between the capillaries and release of residual product
molecules in the ESI chamber.16 The considerable time for
the signal to reach its steady-state value (∼5 min, see Figure
2c,d) is caused by a diffusive broadening of the initial step
profile in the product concentration during the transport from
the flow cell to the 6-port valve. To be compatible with the ESI-
MS intensities and to correct for the small decrease in the
Faradaic current and m/z 44 ion current signals during the 10
min at elevated potential, we used average values, averaged over
these 10 min.
Analogous experiments were performed for a number of
different constant electrode potentials between 0.45 and 1.0
Figure 3. Activity and selectivity of the methanol electrooxidation
reaction over a Pt/C catalyst at different constant electrode potentials.
(a) Measured (net) Faradaic current, partial currents for methanol
oxidation to the corresponding products (see figure), and the sum of
the partial reaction currents; b) formation rates (in nanomoles per
second) for the individual reaction products and the total methanol
oxidation rate (methanol consumption).
VRHE, holding the respective potential for 10 min each time.
After that, the potential was stepped back to 0.06 VRHE and held
there for at least 10 min to return to the background intensities
of the ESI-MS signals. For lower and higher potentials the
methanol electrooxidation rates are rather low, with con-
versions below the detection limits.16 Furthermore, at
potentials above 1.0 VRHE, electrooxidation of the carbon
support to CO2 contributes increasingly to the apparent
product distribution. The results obtained for the activity and
product distribution in the methanol oxidation reaction at
different electrode potentials, in the above potential window,
are summarized in Figure 3.
Figure 3a displays the measured (net) Faradaic current, the
partial currents for CO2, formic acid, and formaldehyde
formation, and the sum of the partial currents during
potentiostatic oxidation of methanol over the Pt/C catalyst at
the different potentials. The partial currents were determined
by converting the corresponding mass spectrometric signals
into Faradaic currents using appropriate calibration constants
(see Experimental Section) and the number of electrons for the
formation of CO2, formic acid, and formaldehyde molecule of
6, 4, or 2 electrons per molecule, respectively. Both the overall
and partial reaction currents show that measurable methanol
oxidation starts at about 0.5 VRHE, passes through a maximum
assumption in previous studies that the only products during
continuous methanol oxidation are CO2, formaldehyde, and
formic acid,20 it is important to note that the sum of partial
currents agrees perfectly with the measured Faradaic net
current (Figure 3a). This also confirms the validity of the
previous assumption that only these three products are formed
during methanol electrooxidation and also the correctness of
the quantitative evaluation procedure employed in this work.
The current efficiencies for methanol oxidation products in
the potential region from 0.65 to 0.80 VRHE are 70 5% for
CO2, 10 1% for formic acid, and 20 2% for formaldehyde,
respectively. The data of Figure 3a agree well with previous data
obtained on a similar catalyst for methanol oxidation in
potentiodynamic measurements20 and in constant potential
oxidation at 0.6 VRHE at different catalyst loadings,19 employing
online DEMS with a membrane inlet. In those measurements,
the production of formaldehyde was not measured directly but
calculated from the difference of the measured Faradaic current
and the sum of the partial currents for CO2 and formic acid
formation. In that case, we had to assume that only these three
components are formed during reaction, which can be proven
in the present measurements. The absolute steady-state rates
for the formation of the different methanol oxidation products
at ca. 0.7 VRHE, and decays to negligible rates at ca. 1.0 VRHE
.
The low methanol oxidation rates at lower potentials are due to
surface poisoning of the catalyst by COad, which results from
the dissociative adsorption of methanol, whereas at high
potentials oxygen/OH adsorption and Pt oxide formation
increasingly inhibit the reaction.18 Considering the tacit
5482
dx.doi.org/10.1021/ac203276f | Anal. Chem. 2012, 84, 5479−5483

Analytical Chemistry
Technical Note
at different constant potentials, at present catalyst loading and
flow rate, are shown in Figure 3b. In the same potential region,
from 0.65 to 0.80 VRHE, CO2 formation accounts for about half
of the overall reaction products, while methanol conversion to
formic acid and formaldehyde is only a quarter of the overall
amount (or half of that for CO2 formation), with product yields
of 53 5% for CO2, 23 2% for formic acid, and 24 2% for
formaldehyde formation.
Ongoing work on the product analysis of a much more
complex reaction, on the continuous oxidation of ethylene
glycol (ethylene glycol fuel cells22) with a large number of
incomplete oxidation products, underlines the potential of this
analysis scheme for studies of fuel cell relevant organic
molecule oxidation reactions. In addition to applications in
energy related electrocatalysis, this approach is applicable for
product analysis of various other reactions at the solid−liquid
interface and equally for analysis of homogeneous reactions in a
liquid phase, as well as for the detection and quantitative
analysis of gaseous/volatile and liquid phase species for the
environmental issues.
REFERENCES
■
(1) Lamy, C.; Leg
́
er, J.-M.; Srinivasan, S. In Modern Aspects of
Electrochemistry; Bockris, J. O’M., Conway, B., White, R. E., Eds.;
Kluwer Academic/Plenum Publishers: New York, 2001; pp 53−118.
(2) Lamy, C.; Lima, A.; Le Rhun, V.; Delime, F.; Coutanceau, C.;
́
Leger, J.-M. J. Power Sources 2002, 105, 283−296.
(3) Wolter, O.; Heitbaum, J. Ber. Bunsen-Ges. Phys. Chem. 1984, 88,
2−6.
(4) Iwasita, T. Electrochim. Acta 2002, 47, 3663−3674.
(5) Baltruschat, H. J. Am. Soc. Mass Spectrom. 2004, 15, 1693−1706.
(6) Wang, H.; Rus, E.; Abruna, H. D. Anal. Chem. 2010, 82, 4319−
4324.
(7) Batista, E. A.; Iwasita, T.; Vielstich, W. J. Phys. Chem. B 2004, 108,
14216−14222.
(8) Lin, W.-F.; Wang, J.-T.; Savinell, R. F. J. Electrochem. Soc. 1997,
144, 1917−1922.
(9) Sanicharane, S.; Bo, A.; Sompalli, B.; Gurau, B.; Smotkin, E. S. J.
Electrochem. Soc. 2002, 149, A554−A557.
(10) Gao, L.; Huang, H.; Korzeniewski, C. Electrochim. Acta 2004, 49,
1281−1287.
(11) Ota, K.-I.; Nakagawa, Y.; Takahashi, M. J. Electroanal. Chem.
1984, 179, 179−186.
(12) Korzeniewski, C.; Childers, C. L. J. Phys. Chem. B 1998, 102,
489−492.
CONCLUSIONS
■
(13) Childers, C. L.; Huang, H.; Korzeniewski, C. Langmuir 1999, 15,
786−789.
In conclusion, we have presented a novel approach for online
analysis of the reaction products in continuous electrocatalytic
reactions such as organic molecule oxidation, which allows for
the first time to quantitatively detect all reaction products, in
this case of the methanol oxidation reaction, simultaneously. It
combines electron impact ionization mass spectrometry, via a
membrane inlet system, for detection of gaseous reaction
products, and electrospray ionization mass spectrometry for
simultaneous quantitative detection of nonvolatile reaction
products. The potential of that method was demonstrated for
continuous methanol oxidation over a carbon supported Pt/C
catalyst. The sum of the partial currents for CO2, formic acid,
and formaldehyde formation was found to agree perfectly with
the measured Faradaic current, confirming the previous
assumption that these are the only products formed during
continuous methanol electrooxidation. The current efficiencies
and product yields were determined at different potentials
between 0.45 and 1.0 VRHE, revealing that in the range of
highest activities (0.65−0.80 VRHE) CO2 formation prevails
with about 50% product yield, while formaldehyde and formic
acid formation each contribute about 25%. The potential of this
analysis approach for energy related electrocatalysis and
product detection in liquid phase reactions, both at the
solid−liquid interface and in the homogeneous liquids, was
outlined.
(14) Zettersten, C.; Sjoberg, P. J. R.; Nyholm, L. Anal. Chem. 2009,
̈
81, 5180−5187.
(15) Gun, J.; Bharathi, S.; Gutkin, V.; Rizkov, D.; Voloshenko, A.;
Shelkov, R.; Sladkevich, S.; Kyi, N.; Rona, M.; Wolanov, Y.; Rizkov, D.;
Koch, M.; Mizrahi, S.; Pridkhochenko, P. V.; Modestov, A.; Lev, O. Isr.
J. Chem. 2010, 50, 360−373.
(16) Zhao, W.; Jusys, Z.; Behm, R. J. Anal. Chem. 2010, 82, 2472−
2479.
(17) Zhao, W.; Jusys, Z.; Behm, R. J. 219th ECS Meeting, The
Electrochemical Society, Pennington, NJ, USA, Montreal, 2011,
Abstract 357.
(18) Jusys, Z.; Behm, R. J. J. Phys. Chem. B 2001, 105, 10874−10883.
(19) Jusys, Z.; Kaiser, J.; Behm, R. J. Langmuir 2003, 19, 6759−6769.
(20) Jusys, Z.; Behm, R. J. In Fuel Cell Catalysis: A Surface Science
Approach; Koper, M. T. M., Ed.; Wiley & Sons: Chichester, 2009; pp
411−464.
(21) Jusys, Z.; Massong, H.; Baltruschat, H. J. Electrochem. Soc. 1999,
146, A1093−A1098.
(22) Livshits, V.; Philosoph, M.; Peled, E. J. Power Sources 2008, 178,
687−691.
AUTHOR INFORMATION
Corresponding Author
*E-mail: juergen.behm@uni-ulm.de.
■
Present Address
†Agilent Technologies (Shanghai) Co. Ltd., Yinglun Road 412,
Waigaoqiao Free Trade Zone, Shanghai, 200131, P.R. China.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the Deutsche Forschungsgemein-
schaft (JU 2781/2-1) and by the German Federal Ministry of
Education and Research (03SF0311C). W.Z. is greatly indebted
for a fellowship by the Alexander von Humboldt Foundation.
5483
dx.doi.org/10.1021/ac203276f | Anal. Chem. 2012, 84, 5479−5483