Au-supported Pt-Au mixed atomic monolayer electrocatalyst with ultrahigh specific activity for oxidation of formic acid in acidic solution

Article abstract of DOI:10.1039/c2cc37383a

Au-supported Pt-Au mixed atomic monolayer electrocatalyst was prepared by underpotential deposition of Cu on Au and then redox replacement with noble metal atoms, which shows an ultrahigh Pt-mass (or Pt-area) normalized specific electrocatalytic activity

Full text of DOI:10.1039/c2cc37383a

View Article Online / Journal Homepage / Table of Contents for this issue
Dynamic Article Links
ChemComm
Cite this: Chem. Commun., 2012, 48, 12106–12108
www.rsc.org/chemcomm
COMMUNICATION
Au-supported Pt–Au mixed atomic monolayer electrocatalyst with
ultrahigh specific activity for oxidation of formic acid in acidic solutionw
Zhao Huang, Yan Liu, Fangyun Xie, Yingchun Fu, Yong He, Ming Ma, Qingji Xie* and
Shouzhuo Yao*
Received 19th June 2012, Accepted 31st October 2012
DOI: 10.1039/c2cc37383a
Au-supported Pt–Au mixed atomic monolayer electrocatalyst was
prepared by underpotential deposition of Cu on Au and then redox
replacement with noble metal atoms, which shows an ultrahigh
Pt-mass (or Pt-area) normalized specific electrocatalytic activity of
of Pt, but such atomic-level attempts are not reported to date.
Mavrikakis et al. recently reported that the Pt-based MAMECs
on non-Pt metal electrodes exhibit facilitated oxygen reduction
reaction (ORR) kinetics and enhanced CO tolerance.8,9 Mixing Pt
overlayer atoms with more oxophilic metal atoms (e.g., (Pt80Ir20)/
Pd(111)) can notably accelerate the ORR, mainly by destabilizing
OH intermediates on the Pt sites of the catalytic surface.8
The enhanced CO tolerance results from an increased repulsive
interaction between adsorbed CO molecules, mediated by the
improved electronic structure of the Pt-based MAMEC surfaces
(e.g., Pt3Ir*/Ir(111)).9
À1
102 mA lgPt (124 mA cmPtÀ2) for oxidation of formic acid in
acidic aqueous solution.
In comparison with direct methanol fuel cells, direct formic acid
(FA) fuel cells (DFAFCs) of high current interest have the
advantages of higher theoretical open-circuit potential, higher
power density and lower fuel crossover through Nafion ion-
exchange membrane.1 Pt is the main anode catalyst in DFAFCs
at present, but its high cost and deactivation by COads formed
during FA oxidation (FAO) limit the wide commercialization of
DFAFCs. The effective management of the size, morphology and
composition of Pt-based catalysts can improve their catalytic
performance. Bimetallic and trimetallic catalysts of Pt,
e.g., Pt–Au,2,3 Pt–Pd,4 Pt–Pb5 and FePtAu,6 have been developed
for facilitated electrocatalytic oxidation of FA. This is due to the
synergetic effects between the components, including the ensemble
effect of Pt atoms (Pt atoms well isolated by additive elements are
preferable),3,7 and the electronic effect of Pt atoms derived from
their size and interaction with Pt-tethered/neighbored additive
elements.2,5 For instance, Zhang et al. constructed Pt-on-Au
nanostructures by chemically reducing PtCl62À to deposit Pt atoms
on a series of monodisperse Au nanoparticles of size 1.9–14 nm,
and found that fully dispersed Pt entities on the smallest Au
nanoparticles (1.9 nm) exhibit a very high mass-activity toward
FAO (7.5 mA mgPtÀ1), proving that smaller Au nanoparticle-
supported Pt atoms with a higher d band center possess a stronger
catalytic activity toward FAO.2
In addition, quantification of the mass of an electrode-supported
atomic monolayer electrocatalyst is obviously pivotal for evaluating
its mass-normalized specific electrocatalytic activity (SECAm). The
ex situ EDS, XPS and ICP-AES methods are difficult to use to
quantify the ultralow absolute mass of a small-area atomic mono-
layer, and the in situ electrical charge method is also interfered by
non-faradaic currents. The quartz crystal microbalance (QCM) and
its in situ combination with an electrochemical technique (electro-
chemical QCM, EQCM) can reliably weigh a change of ng-scale
mass (monolayer or sub-monolayer atoms) on the electrode surface
in either ex situ or in situ manner,10 which should be very promising
in quantitatively studying a Pt-based MAMEC and even its
components with discernible electrochemical features.
Here, we report a novel Au-supported Pt–Au MAMEC for
FAO, which exhibit the highest Pt-mass/area specific activity
(102 mA mgPtÀ1; 124 mA cmPtÀ2) up to date. As shown in
Scheme 1, an Au-supported Pt atomic monolayer electrocatalyst
or a Pt–Au MAMEC is prepared by underpotential deposition
(UPD) of Cu (CuUPD) on Au (Fig. S1w), and then redox
replacement with noble metal atoms (Fig. S2w). The former is
prepared by using K2PtCl4 (or H2PtCl6) as a Pt source, while the
latter is prepared by varying the concentration ratio of H2PtCl6 to
HAuCl4 in their mixed solutions (for experimental details and
abbreviations of the prepared electrodes see Table S1w).
The above discussion logically implies that a ‘‘Pt-in-Au’’ mixed
atomic monolayer electrocatalyst (MAMEC) where the Pt atoms
are well isolated by their neighboring Au atoms may have
an ultrahigh activity for FAO at an extremely high availability
The replacement efficiency (Z) between CuUPD and the noble
metal atoms should be an important parameter for monolayer
electrocatalysts, however, little information on such Z values can be
found to date. Herein, we use the electrochemistry + QCM
method to quantify Z values, as depicted in ESIw. We define the
molar ratio of replacement-deposited Pt (or Au) atoms to electro-
deposited CuUPD atoms as ZPt (or ZAu). The reaction stoichiometry
for CuUPD-to-Pt replacement gives Cu/Pt ratios of 1 : 1 for
Key Laboratory of Chemical Biology and Traditional Chinese
Medicine Research (Ministry of Education of China), College of
Chemistry and Chemical Engineering, Hunan Normal University,
Changsha, 410081, China. E-mail: xieqj@hunnu.edu.cn;
Fax: +86-731-88865515
w Electronic supplementary information (ESI) available: Detailed
description of the experimental section and additional figures. See
DOI: 10.1039/c2cc37383a
c
12106 Chem. Commun., 2012, 48, 12106–12108
This journal is The Royal Society of Chemistry 2012

View Article Online
(Fig. S4, ESIw with discussion details). Hence, the redox
replacement between PtCl6 and CuUPD may go through a
2À
2À
PtCl62À - PtCl42À - Pt process, and partial PtCl4 formed
2À
after reduction of PtCl6 by CuUPD can be reoxidized by AuCl4À
during preparation of a Pt–Au MAMEC, leading to a dramatic
decrease in yPt at higher concentrations of Au source.
Fig. S5w shows the XPS spectra of Pt 4f (A) and Au 4f (B) for the
modified Au electrodes, respectively, and the binding energies (BEs)
of 4f signals for Pt0 and Au0 are summarized in Table S2w. A
Pt-load decrease leads to negatively and positively shifted BEs of
Pt 4f and Au 4f, respectively (Fig. S5w), confirming the electronic
effect between Pt and Au. Both the interatomic charge transfer
(from Au to Pt here) and intra-atomic charge redistributions should
take place, owing to a larger fraction of empty states in the valence
band of Pt (5d9) compared to Au (5d10) and the lattice mismatch
between Pt (2.78 A) and Au (2.88 A).13 Moreover, a more filled d
band of Pt, along with the upward shift of Pt d band center derived
from the expansion of the Pt overlayer on Au (narrowing the Pt d
band width),14 should strengthen the adsorbate–surface bond,15 as
experimentally proven by the negative shift of Pc-Pt in Fig. S3w.
The electrochemistry + QCM characterization allows us to
quantify the Pt loading in a Pt–Au MAMEC. The procedures
are given briefly as follows and detailed in ESIw. (1) A series of
Scheme 1 Schematic illustration of the construction of Pt and Pt–Au
atomic monolayer modified Au electrodes for FAO.
K2PtCl4 source and 2 : 1 for H2PtCl6 source, thus ZPt should
theoretically be 100% for Pt(Cu )/Au (K2PtCl4) and 50% for
UPD–Pt2+
Pt(Cu )/Au (H2PtCl6). As listed in Table 1, the experimental
UPD–Pt4+
ZPt (ZPt-exp) values are (74 Æ 6)% for Pt(Cu
)/Au and
UPD–Pt2+
(30 Æ 3)% for Pt(Cu
)/Au, respectively. The CuUPD-to-Au
UPD–Pt4+
replacement should take place at a Cu/Au ratio of 3 : 2, giving a
theoretical ZAu of 66.7%. Experimentally, the ZAu (ZAu-exp, for
Au(Cu
)/Au electrode in Table 1) is (49 Æ 3)%. The
UPD–Au3+
experimental Z values being below their theoretical values probably
result from the spontaneous oxidized stripping of CuUPD both after
turning off the applied potential in the UPD bath and during
solution changes from CuSO4 solution for UPD to ultrapure water
for rinsing, and finally to noble metal salt solution for redox
replacement under our experimental conditions.
Pt(Cu )/Au (or Pt(Cu )/Au) electrodes were prepared by
UPD–Pt2+ UPD–Pt4+
adjusting the amount of CuUPD using different applied potentials
during UPD of Cu (0.39–0.56 mg cmÀ2), followed by replacement
of CuUPD with K2PtCl4 (or H2PtCl6). The Pt loading for
Fig. S3w shows the cyclic voltammograms of several modified Au
electrodes in 0.1 M aqueous H2SO4. The Pt–Au or Pt modification
decreases the reduction peak of Au oxides (Pc-Au) at ca. 0.85 V
and increases the reduction peak of Pt oxides plus dissolved oxygen
(Pc-Pt) at ca. 0.3 V. The H UPD on deposited Pt at negative
potentials does not display two well-separated pairs of redox peaks
commonly observed on bulk Pt, implying an epitaxial growth of Pt
atoms on Au.11 The Pc-Pt shifts to more negative potentials with a
decrease of Pt loading, as reported previously.12 However, change
Pt(Cu
)/Au (or Pt(Cu
)/Au) electrode is obtained from
UPD–Pt4+
UPD–Pt2+
the QCM frequency responses; (2) the prepared electrodes were
subjected to potential cycling in 0.1 M aqueous H2SO4 to obtain
data of Pt-load-dependent |DQPc-Au| and H adsorption charge on
Pt surface (QH,UPD); (3) plots of the Pt loading vs. |DQPc-Au| or
QH,UPD are drawn to obtain their linear regression equations
(LREs), as shown in Fig. S6w; (4) the LREs are used to quantify
Pt loading in Au-supported Pt–Au MAMECs, as listed in
Table S3w, after determining |DQPc-Au| and QH,UPD of a Pt–Au
MAMEC in cyclic voltammetric experiments as shown in Fig. S3w.
The average Pt loading in the Au-supported Pt–Au MAMECs for
SECAm estimation below is listed in Table 1.
in Pc-Au is hardly seen on Au(Cu )/Au. Pt coverage (yPt) is
UPD–Au3+
equal to |DQPc-Au/QPc-Au-bare|, where DQPc-Au is the charge change
under Pc-Au after Pt deposition on bare Au and QPc-Au-bare is the
charge under Pc-Au on bare Au. As listed in Table 1, yPt is 78.5% for
We have compared our in situ QCM method with the prevalent
ICP-AES method in determining Pt loading. On the Au electrode
modified with electroplated Pt (Ptpla/Au), the loading of Pt deposits
from the two methods agrees well (Fig. S7w). The Pt loading on
Pt(Cu )/Au and 30.7% for Pt(Cu )/Au, respectively.
UPD–Pt2+ UPD–Pt4+
Intriguingly, increasing the concentration of HAuCl4 in the mixed
solution largely decreases yPt. Even at a Pt/Au concentration ratio
of 3 : 1 (Pt3Au1(Cu )/Au), yPt is only 7.1%, implying the
UPD–Pt4+
Pt(Cu
)/Aupla/Au or Pt(Cu
)/Aupla/Au from the two
UPD–Pt2+
UPD–Pt4+
higher sensitivity of the composition of final Pt–Au deposit to
the concentration of the Au source. Under our experimental
methods also agrees fairly well on Au electrodes roughened by
electroplated Au (Aupla/Au) (Fig. S8 and Table S4w). Comparable
results of loading of electroplated and replacement-deposited Pt
obtained from in situ QCM and ex situ ICP-AES experiments serve
to validate our QCM method for quantifying the Pt loading here.
Fig. 1 shows the electrocatalytic activities for FAO in aqueous
2À
À
conditions, PtCl4 can quickly reduce AuCl4 to atomic Au, as
supported by QCM and UV-vis spectrophotometric experiments
Table 1 Data of total noble metal load (M load), Pt load, experi-
mental Z, yPt and SECAm for the modified Au electrodesa
1 M HCOOH + 0.1 M H2SO4. Au(Cu
)/Au (Fig. S9w)
UPD–Au3+
M load/ Pt load/ ZPt-exp or yPt SECAm/À1
mg cmÀ2 mg cmÀ2 ZAu-exp (%) (%) mA mgPt
shows a low electrocatalytic activity toward FAO, as reported
previously.16 For Pt(Cu
)/Au (Fig. 1A, inset), the high
UPD–Pt2+
Pt(Cu
Pt(Cu
Pt9Au1(Cu
Pt4Au1(Cu
Pt3Au1(Cu
Au(Cu
2+)/Au
4+)/Au
1.21
0.52
4+)/Au 0.84
4+)/Au 0.90
4+)/Au 0.99
1.21
0.52
0.24
0.16
0.11
0
74 Æ 6
30 Æ 3
—
—
—
49 Æ 3
78.5 0.51
30.7 26
14.9 63
10.1 102
7.1 92
UPD–Pt
oxidation peak at 0.25 V (0.61 mA cmÀ2, background-corrected
here and below) refers to the direct oxidation of FA to CO2
(dehydrogenation path), and the weak oxidation peak at 0.74 V
(0.05 mA cmÀ2) to oxidation of COads during indirect oxidation of
FA (dehydration path),6 implying FAO mainly via a dehydrogena-
tion path here, in contrast to a dominant dehydration path on
UPD–Pt
UPD–Pt
UPD–Pt
UPD–Pt
3+)/Au
UPD–Au
1.05
0
—
a
The 3rd column shows average Pt load in Pt–Au MAMECs.
c
This journal is The Royal Society of Chemistry 2012
Chem. Commun., 2012, 48, 12106–12108 12107

View Article Online
Fig. 1 Cyclic voltammograms of Pt(Cu 2+)/Au, Pt(Cu 4+)/Au, Pt9Au1(Cu 4+)/Au, Pt4Au1(Cu 4+)/Au, and Pt3Au1(Cu 4+)/Au in
UPD–Pt UPD–Pt UPD–Pt UPD–Pt UPD–Pt
aqueous 1 M HCOOH + 0.1 M H2SO4. Currents in panels A, B and C are normalized to geometric area (A), electroactive Pt surface area (B) and mass
of Pt load (C), respectively. Curve labels shown in A apply also for B and C. Inset in A is an enlarged plot for Pt(Cu
2+)/Au. Scan rate: 50 mV sÀ1
.
UPD–Pt
E-TEK Pt/C (20%) (Fig. S10w). Intriguingly, with the decrease of
)/Au (Fig. 1A)
notably increases to 13.7 mA cmÀ2, while the oxidation peak of
COads disappears, demonstrating FAO fully via a dehydrogenation
path here. The direct oxidation peak current of 16.3 mA cmÀ2 at
FAO (Fig. S14w). Only direct oxidation of FA is observed on
these Pt–M MAMECs (optimized), and their current densities
(mA cmPtÀ2) are all lower than that of Pt–Au MAMEC, but far
higher than that of E-TEK Pt/C (20%), indicating the best
performance of Pt–Au MAMEC toward FAO among the
examined Pt–M MAMECs.
yPt, the direct oxidation peak current on Pt(Cu
UPD–Pt4+
Pt4Au1(Cu )/Au is the maximum among the examined Pt–Au
UPD–Pt4+
MAMECs. Moreover, the electroactive Pt surface area (from H
UPD charge) normalized specific electrocatalytic activity (SECAa,
In conclusion, Au-electrode-supported Pt–Au MAMECs
with excellent performance for FAO have been prepared by
redox replacement of CuUPD atoms. Pt loading in Pt–Au
MAMECs and Z values are evaluated by electrochemistry +
QCM for the first time. As far as we are aware, both the
À2
Fig. 1B) reaches the highest value of 124 mA cmPt (0.33 V) for
Pt4Au1(Cu
)/Au, which is 238Â that of Pt(Cu
)/Au
UPD–Pt2+
UPD–Pt4+
(0.52 mA cmPtÀ2, 0.25 V) and 1240Â that of E-TEK Pt/C (20%,
0.10 mA cmPtÀ2, 0.26 V, Fig. S10Aw). A Pt–Au MAMEC similarly
prepared on a new Au atomic layer shows an equivalent catalytic
SECAm and SECAa obtained on Pt4Au1(Cu
)/Au are the
UPD–Pt4+
highest among Pt-based electrocatalysts reported to date. The
protocols for catalyst design and characterization here can be
conveniently extended to studies on other new and efficient
monolayer or nanostructured electrocatalyst systems.
activity (Fig. S11w). In addition, Pt4Au1(Cu
)/Au has the
UPD–Pt4+
À1
largest SECAm of 102 mA mgPt (Fig. 1C, 0.33 V), which is
200Â that on Pt(Cu
)/Au (0.51 mA mgPtÀ1, 0.25 V), 1133Â
UPD–Pt2+
that on E-TEK Pt/C (20%, 0.09 mA mgPtÀ1, 0.26 V, Fig. S10Bw)
and 13.6Â that (7.5 mA mgPtÀ1) reported by Zhang et al.2
Chronoamperometry tests have also proven the highest activity
This work was supported by the National Natural Science
Foundation of China (21075036, 21175042, 21105026,
20875029), Hunan Lotus Scholars Program, Program for
Science and Technology Innovative Research Team in Higher
Educational Institutions of Hunan Province and the Graduate
Innovation Foundation of Hunan Province (CX2011B189).
on Pt4Au1(Cu )/Au among the examined electrocatalysts
UPD–Pt4+
(Fig. S12w). To our knowledge, the optimized Pt–Au MAMEC
here shows the highest specific activity toward FAO reported to
date. Performance comparison of our electrocatalysts with some
reported Pt-based ones toward FAO are given in Table S5w.
In our opinion, the exceptional atomic structure of Au-supported
Pt–Au MAMEC leads to its enhanced performance toward FAO,
namely, (1) the Pt atoms in an Au-supported Pt–Au MAMEC are
well separated/isolated by the coplanar and underlying Au atoms,
and such a favourable ensemble effect of Pt atoms with high Pt
availability greatly boosts their catalytic activity and results in the
disappearance of the COads-related current peak,3,7 as supported by
the stripping tests of the COads formed by dissociating FA at open
circuit potential (Fig. S13, ESIw with discussion details); (2) the
electronic effect of Au on Pt (upward shift of d band center and
more filled d band for Pt) markedly enhances the activity of Pt–Au
MAMEC;2,15 and (3) the Pt atoms residing on the corners and steps
of Au electrode surface (active sites) should be more active than
those on terraces.17 During the redox replacement, Au atoms may
show a priority to deposit on these active sites due to the
Notes and references
1 X. L. Ji, K. T. Lee, R. Holden, L. Zhang, J. J. Zhang, G. A. Botton,
M. Couillard and L. F. Nazar, Nat. Chem., 2010, 2, 286.
2 G. R. Zhang, D. Zhao, Y. Y. Feng, B. S. Zhang, D. S. Su, G. Liu
and B. Q. Xu, ACS Nano, 2012, 6, 2226.
3 R. Y. Wang, C. Wang, W. B. Cai and Y. Ding, Adv. Mater., 2010,
22, 1845.
4 B. Liu, H. Y. Li, L. Die, X. H. Zhang, Z. Fan and J. H. Chen,
J. Power Sources, 2009, 186, 62.
5 D. J. Chen, Z. Y. Zhou, Q. Wang, D. M. Xiang, N. Tian and
S. G. Sun, Chem. Commun., 2010, 46, 4252.
6 S. Zhang, S. J. Guo, H. Y. Zhu, D. Su and S. H. Sun, J. Am. Chem.
Soc., 2012, 134, 5060.
7 M. Neurock, M. Janik and A. Wieckowski, Faraday Discuss.,
2009, 140, 363.
8 A. U. Nilekar, K. Sasaki, C. A. Farberow, R. R. Adzic and
M. Mavrikakis, J. Am. Chem. Soc., 2011, 133, 18574.
9 J. L. Zhang, M. B. Vukmirovic, K. Sasaki, A. U. Nilekar,
M. Mavrikakis and R. R. Adzic, J. Am. Chem. Soc., 2005, 127, 12480.
10 D. A. Buttry and M. D. Ward, Chem. Rev., 1992, 92, 1355.
11 S. H. Yoo and S. Park, Electrochim. Acta, 2008, 53, 3656.
12 B. C. Du and Y. Y. Tong, J. Phys. Chem. B, 2005, 109, 17775.
13 E. Bus and J. A. V. Bokhoven, J. Phys. Chem. C, 2007, 111, 9761.
14 S. Kumar and S. Z. Zou, Langmuir, 2007, 23, 7365.
15 M. P. Hyman and J. W. Medlin, J. Phys. Chem. C, 2007, 111, 17052.
16 N. Kristian, Y. S. Yan and X. Wang, Chem. Commun., 2008, 353.
17 S. W. Lee, S. Chen, W. C. Sheng, N. Yabuuchi, Y. T. Kim, T. Mitani,
E. Vescovo and Y. Shao-Horn, J. Am. Chem. Soc., 2009, 131, 15669.
À
2À
rapid redox reaction of AuCl4 with the PtCl4 intermediate
as discussed above, which decreases the number of Pt atoms
residing on these active sites, thus the Pt3Au1(Cu )/Au of
UPD–Pt4+
an increased Au source concentration shows a lower activity
than Pt4Au1(Cu )/Au.
UPD–Pt4+
We also examined the catalytic performance of other similarly
prepared Pt–M (M = Ir, Rh, Pd, Ru and Os) MAMECs for
c
12108 Chem. Commun., 2012, 48, 12106–12108
This journal is The Royal Society of Chemistry 2012