Achieving Selective and Efficient Electrocatalytic Activity for CO2 Reduction Using Immobilized Silver Nanoparticles

Article abstract of DOI:10.1021/jacs.5b06568

Selective electrochemical reduction of CO₂ is one of the most sought-after processes because of the potential to convert a harmful greenhouse gas to a useful chemical. We have discovered that immobilized Ag nanoparticles supported on carbon exhibit enhanced Faradaic efficiency and a lower overpotential for selective reduction of CO₂ to CO. These electrocatalysts were synthesized directly on the carbon support by a facile one-pot method using a cysteamine anchoring agent resulting in controlled monodispersed particle sizes. These synthesized Ag/C electrodes showed improved activities, specifically decrease of the overpotential by 300 mV at 1 mA/cm², and 4-fold enhanced CO Faradaic efficiency at 0.75 V vs RHE with the optimal particle size of 5 nm compared to polycrystalline Ag foil. DFT calculations enlightened that the specific interaction between Ag nanoparticle and the anchoring agents modified the catalyst surface to have a selectively higher affinity to the intermediate COOH over CO, which effectively lowers the overpotential.

Full text of DOI:10.1021/jacs.5b06568

Subscriber access provided by UNIV OF LETHBRIDGE
Article
Achieving Selective and Efficient Electrocatalytic Activity
for CO2 Reduction Using Immobilized Silver Nanoparticles
Cheonghee Kim, Hyo Sang Jeon, Taedaehyeong Eom, Michael Shincheon Jee,
Hyungjun Kim, Cynthia M. Friend, Byoung Koun Min, and Yun Jeong Hwang
J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b06568 • Publication Date (Web): 08 Oct 2015
Downloaded from http://pubs.acs.org on October 9, 2015
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
Journal of the American Chemical Society is published by the American Chemical
Society. 1155 Sixteenth Street N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
Achieving Selective and Efficient Electrocatalytic Activity for
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
CO Reduction Using Immobilized Silver Nanoparticles.
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
a
a,b
c
a,d
Cheonghee Kim, † Hyo Sang Jeon, † Taedaehyeong Eom, Michael Shincheon Jee, Hyungjun
c
e
a,b,f
a,b
Kim, Cynthia M. Friend, Byoung Koun Min, * and Yun Jeong Hwang *
a
Clean Energy Research Center, Korea Institute of Science and Technology, Seoul 02792, Republic of Korea
b
Korea University of Science and Technology, Daejeon 34113, Republic of Korea
c
Graduate School of Energy, Environment, Water, and Sustainability (EEWS), Korea Advanced Institute of Science
and Technology (KAIST), Daejeon, 34141, Republic of Korea
d
Department of Chemical and Biological Engineering, Korea University, Seoul 02841, Republic of Korea
e
Department of Chemistry and Chemical Biology, Harvard University, Cambridge, Massachusetts 02138, United
States
f
Green School, Korea University, Seoul 02841, Republic of Korea
KEYWORDS: Silver, Nanoparticle, Size control, CO RR, Activity, Selectivity
2
ABSTRACT: Selective electrochemical reduction of CO is one of the most sought-after processes because of the potential
2
to convert a harmful greenhouse gas to a useful chemical. We have discovered that immobilized Ag nanoparticles sup-
ported on carbon exhibit enhanced Faradaic efficiency and a lower overpotential for selective reduction of CO to CO.
2
These electrocatalysts were synthesized directly on the carbon support by a facile one-pot method using a cysteamine
anchoring agent resulting in controlled monodispersed particle sizes. These synthesized Ag/C electrodes showed im-
2
proved activities, specifically, decrease of the overpotential by 300 mV at 1 mA/cm , and four-fold enhanced CO Faradaic
efficiency at -0.75 V vs. RHE with the optimal particle size of 5 nm compared to polycrystalline Ag foil. DFT calculations
enlightened that the specific interaction between Ag nanoparticle and the anchoring agents modified the catalyst surface
to have a selectively higher affinity to the intermediate COOH over CO, which effectively lowers the overpotential.
INTRODUCTION
fore, it is necessary and important to design a new cata-
lyst with high efficiency and selectivity.
Atmospheric concentrations of carbon dioxide,
a
greenhouse gas, are ever increasing due to the continued
use of traditional fossil fuels, leading to adverse global
climate. In an effort to mitigate the world-wide depend-
Gold, silver, and copper have been identified as promis-
7
ing CO reduction reaction electrocatalysts because CO
2
1
,2
(*CO) relatively weakly binds on their surfaces compared
to hydrogen evolution electrocatalysts such as platinum.
Especially, Au and Ag had been reported to have highly
ence on fossil fuels and, thus, CO emission, there is in-
2
creasing emphasis on the development of renewable en-
ergy sources; and as a more active response, conversion of
carbon dioxide to a reusable carbon form, a sustainable
selective conversions from CO to CO in electrochemical
2
CO reduction reaction, while Cu produces a mixture of
2
3
,4
carbon recycling system, has been suggested. Among
more reduced form of hydrocarbon products such as CH4,
various methods for CO conversion, electrochemical CO2
C H , ethanol, etc. Although Ag shows as great selectivity
2
2
4
8
reduction in aqueous solution has attracted considerable
attention since it is environmentally clean and can be
combined with renewable energy resources such as solar
as Au but cheaper cost, Ag has not been widely studied
because the required overpotential is larger than Au.
Many of the Ag-based electrocatalysts have been studied
9
and wind energies to store them in the form of chemical
in non-aqueous electrolyte such as ionic-liquids. Recent-
ly, density functional theory (DFT) calculation reported
that Ag-based catalysts doped by S or As can decrease the
5
energy. However, electrochemical CO reduction reac-
2
tion suffers from poor efficiency for several reasons, in-
cluding the requirement of a large overpotential, difficul-
ty in controlling selectivity, and competition with hydro-
1
0
overpotential for CO to CO electrochemical conversion,
2
which motivates the development of Ag-based electrocat-
6
gen evolution reaction in aqueous environment. There-
alysts for CO reduction reaction.
2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Recently, the surfaces of bulk polycrystalline metals
Page 2 of 9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
have been modified to form nanostructures in order to
improve catalytic activity for CO reduction reaction. For
2
example, oxide-derived Au nanoparticles or Ag nanopo-
rous demonstrated low overpotentials and high stability
for CO production, where high-index facets of the curved
1
1,12
surfaces were suggested to be related to high activity.
In addition, the size effect of nanoparticles associated
with changes in the ratio between the edge and corner
1
3,14
sites has been demonstrated with Au and Cu metals
in
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
the aqueous electrolyte. Zhu et al. proposed that edge
sites of Au nanoparticles prefer CO evolution; whereas,
1
3,15
corner sites are active for the competitive H evolution.
2
Therefore, the existence of the optimal particle size has
been suggested for the CO reduction reaction with high
2
activity and selectivity. Because the prevalence of high-
index facets or edge sites varies as the morphology and
the size of the nanoparticle changes, uniform synthesis is
important to understand nano-morphology influence on
the CO reduction reaction activity.
2
Wet chemistry synthesis allows metallic nanoparticles
with uniformly controlled sizes and shapes that can easily
16-19
be obtained in large quantities.
To use them for elec-
trocatalysts, they have been deposited on a carbon sup-
port of a high surface area after the nanoparticle synthesis
as a two-step process. An alternative and more ad-
vantages approach for electrocatalytic applications is the
direct growth of nanoparticles on the carbon supports
with anchoring agents. The anchoring agent acts a role of
nucleating sites that induce monodisperse particle syn-
20,21
thesis and immobilize on the support simultaneously.
In this study, we developed a direct one-pot synthesis
of Ag nanoparticles on carbon support using an anchoring
agent (cysteamine) and successfully controlled uniform
particle sizes. Three different sized silver particles (3, 5,
and 10 nm) supported on carbon (nano-sized Ag/C) were
Figure 1. TEM image of a; 3 nm, b; 5 nm and c; 10 nm sized
Ag nanoparticles directly synthesized on carbon support. d;
HR-TEM image of 5 nm Ag/C, and e; XPS spectra of 5 nm
Ag/C, Cysteamine-C and Carbon black for N 1s and S 2p.
evaluated for electrochemical CO reduction reaction in
an aqueous environment, and the influence of the an-
2
nm Ag/C) and 10.6 ± 2.8 nm at 200 °C for 1 hour (denoted
as 10 nm Ag/C) as shown in Figure 1a-c. A high resolution
transmission electron microscopy (HR-TEM) image con-
firmed the single crystalline Ag nanoparticles with a d-
spacing of 0.23 nm, which corresponds to the (111) crystal
plane of face-centered cubic silver (Figure 1d). Further,
the X-ray diffraction (XRD) data in Figure S1 show that
these nanoparticles are face-centered cubic Ag crystal
structure. Without cysteamine, we found that, from TEM
images (Figure 1 and Figure S2), Ag particles cannot be
directly grown on the carbon support and the sizes of the
nanoparticles were not controlled because cysteamine
molecules helps to anchor nanoparticles in the initial nu-
choring agent on the CO reduction activity was investi-
2
gated. Particles with a diameter of 5 nm showed a high
activity, enhanced durability, and maximum Faradaic
efficiency (84.4%) for the CO2 to CO electrochemical
conversion. Notably, the overpotential for CO production
2
had reduced about 300 mV at 1 mA/cm with 5 nm sized
Ag/C when compared with polycrystalline Ag foil. We
propose that the anchoring agent modifies the spatial
spin density of the Ag nanoparticles, which decreases the
overpotential for CO to CO reduction by DFT calcula-
2
tions.
20,21
cleation step.
RESULTS AND DISCUSSION
To investigate the role of cysteamine, X-ray photoelec-
tron spectroscopy (XPS) was conducted on the carbon
black, cysteamine treated carbon (denoted as Cysteam-
ine-C) and prepared Ag nanoparticles/C (Figure 1e and
Figure S3a). Figure 1e shows a clear N 1s peak for the Cys-
teamine-C and 5 nm Ag/C while its absence for carbon
black indicates that cysteamine was immobilized on the
carbon support. On the other hand, Ag/C particles com-
Immobilized Ag nanoparticle synthesis with an an-
choring agent
Ag nanoparticles of three different sizes were easily
synthesized by using a Ag precursor, carbon black, and
cysteamine. The size of the Ag nanoparticles was 3.4 ± 0.6
nm when synthesized at 160 °C for 1 hour (denoted as 3
nm Ag/C), 5.0 ± 0.9 nm at 160 °C for 3 hours (denoted as 5
ACS Paragon Plus Environment

Page 3 of 9
Journal of the American Chemical Society
monly show additional S 2p peaks (see an arrow in Figure
e, and pink lines in Figure S3a) unlike those of carbon
measured at the steady-state current density from chron-
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
1
oamperometry. The CO partial current density of 5 nm
Ag/C showed highest values at the low cathodic potential
region (-0.3 ~ -0.9 V vs. RHE). Meanwhile, the 3 nm and
10 nm Ag/C started to show CO partial current density at
a higher biased potential region than 5 nm Ag/C, while Ag
foil had an even larger cathodic onset potential meaning a
black and Cysteamine-C. The additional S 2p3/2 peak pos-
sesses the binding energy at 162.1 eV, associated with an
22
Ag-S bond which results from the interaction of the cys-
teamine with the Ag surface. Such interaction affects the
electronic structure of Ag where the relative intensity of a
minor Ag 3d peak (orange lines in Figure S3b) increases as
that of the S 2p3/2 peak at 162.1 eV increases. Cysteamine
amount was estimated by comparing the Ag 3d and S 2p
signal in XPS (see supporting information). The Ag-S
bond ratio of 5 nm Ag/C (22.0%) was observed to be
higher than those of 3 nm Ag/C (18.6%) and 10 nm Ag/C
larger required overpotentials for CO reduction reaction.
2
Our Ag foil data is well matched with the previous re-
8
port.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
In order to compare the required overpotentials further,
we obtained overpotentials at a fixed current density and
overpotentials dependent CO partial current density
(
16.3%). The carbon black as purchased also showed S 2p
(
Figure 2b and Figure S5). The thermodynamic reduction
23
peaks remnant from its production process.
potential for CO to CO is -0.11 V vs. RHE. All of the pre-
2
pared Ag/C samples showed a noticeable decrease in
overpotential and 5 nm Ag/C showed the largest dimin-
ishment. To be specific, 5 nm Ag/C showed about 300 mV
Electrochemical CO reduction with nanoparticu-
2
lated Ag/C and Ag foil
2
anodic shift of overpotential at 1 mA/cm , when compared
The electrochemical CO reduction reaction was per-
2
with Ag foil. Furthermore, as seen in Figure 2c mass activ-
ity of three different nanoparticle Ag/C electrodes at vari-
ous potentials followed a similar trend with CO current
density normalized with an electrode area since the
amount of Ag metal content was controlled using induc-
formed using the prepared 3, 5, and 10 nm Ag/C samples,
and all showed enhanced catalytic activity for CO produc-
tion compared with the polycrystalline Ag foil (Figure 2a
and Figure S4). Figure 2a shows the iR-corrected potential
dependent CO partial current densities, which were
1
.0
.8
0
3 nm Ag/C
nm Ag/C
10 nm Ag/C
Ag foil
b
5
0
a
-
-
-
-
2
4
6
8
0.6
.4
Carbon black
Cysteamine-C
nm Ag/C
nm Ag/C
0 nm Ag/C
0
3
5
1
0.2
0.0
-
-
10
12
Ag foil
2
2
2
0.01 mA/cm 0.1 mA/cm 1 mA/cm
-
1.2 -1.0 -0.8 -0.6 -0.4 -0.2
Potential
(V, E vs RHE
)
CO partial current density
100
10
8
3
nm Ag/C
5 nm Ag/C
0 nm Ag/C
3 nm Ag/C
5 nm Ag/C
10 nm Ag/C
Ag foil
8
6
4
2
0
0
0
0
0
1
6
4
2
c
d
0
-
0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1.0 -1.1
-0.3 -0.4 -0.5 -0.6 -0.7 -0.8 -0.9 -1.0 -1.1
Potential (V, E vs RHE)
Potential (V, E vs RHE)
Figure 2. a; CO partial current density depending on applied potential, b; overpotentials at fixed current densities, c; mass activ-
ity depending on applied potential and d; Specific activity depending on applied potential. Specific activity was normalized by
electrochemically active surface area of silver measured from Pb UPD. CO electrochemical reduction was performed in CO -
2
2
saturated 0.5 M KHCO . The thermodynamic reduction potential for CO to CO is -0.11 V vs. RHE.
3
2
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 4 of 9
tively coupled plasma (ICP) results in all fabricated elec-
trodes. Additionally, the electrochemical surface area
ECSA) of Ag electrode was measured by using an
10 nm and 3 nm Ag/C, and Ag foil followed sequentially.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
In addition, the maximum CO Faradaic efficiencies were
76.8, 84.4, 72.6, and 70.5% for 3, 5, 10 nm Ag/C, and Ag
foil, respectively, with 5 nm Ag/C achieving the highest
CO Faradaic efficiency. When CO Faradaic efficiencies of
all samples were compared at a fixed potential of -0.75 V
vs. RHE, the maximum selectivity potential for 5 nm Ag/C
(Figure 3b), Faradic efficiency of the 5 nm Ag/C (average
F.E.CO = 79.2%) showed 4.8 times enhancement compared
to Ag foil (average F.E.CO = 16.5%). In our study, mass ac-
tivity, overpotentials, and CO Faradaic efficiency demon-
strated that 5 nm was the optimal Ag nanoparticle size for
(
underpotential deposition (UPD) method (Figure S6).
2
The ECSAs of Ag electrodes were 1.8, 1.2, 1.2 and 0.17 cm
for 3 nm, 5 nm, 10 nm Ag/C, and Ag foil, respectively.
Figure 2d shows specific activities of the Ag electrodes
normalized with the measured ECSAs. 5 nm Ag/C has the
highest specific activity when compared with that of 3 nm
Ag/C, 10 nm Ag/C, and Ag foil before reaching mass
transfer limitation of CO2.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
To confirm that the activity for CO reduction of the
2
CO reduction reaction application in aqueous solution.
Ag/C electrodes is induced by the Ag nanoparticles, not
2
Figure S9 shows the Faradaic efficiency of the 5 nm Ag/C
for H and CO products in detail. Methane (CH ) is also
by the carbon support itself or Cysteamine-C, CO reduc-
2
tion reactions with both samples were performed under
the same condition. Small CO partial current densities
were detected for carbon black and Cysteamine-C (Figure
2
4
one of the gaseous products but less than 1% Faradaic
efficiency was detected at all potentials regardless the
type of working electrodes. Major products are H and CO,
2
a) although their total current densities sharply in-
creased as the applied potentials exceeded -0.8 V vs. RHE
Figure S7). Identification of the gas products by gas
2
and the sum of their Faradaic efficiencies was close to 100%
for each sample according to GC measurements indicat-
ing liquid products such as formic acid were negligible,
(
chromatography (GC) measurements confirmed that high
amount of hydrogen was generated as the major product
when Ag was not loaded on the carbon support. These
results indicated that both carbon support and Cysteam-
which are consistent with the previous study with Ag
8
foil. Additionally, the TEM and HR-TEM image after CO
2
reduction reaction showed that the 5 nm Ag/C has exhib-
ited no size change and had an average size of 5.0 ± 1.0
nm (Figure 3c).
ine-C were incapable of CO reduction reaction.
2
The selectivity of CO reduction reaction to CO was al-
2
The 5 nm Ag/C catalyst is the optimum size among the
three different nanoparticle sizes even though high sur-
face areas are expected to provide higher activity. Specifi-
cally, the 5 nm Ag particles yielded higher current density
so studied with the prepared Ag electrodes by varying the
applied potentials (Figure 3a and Figure S8). The en-
hanced CO production selectivity over hydrogen evolu-
tion reaction (HER) was commonly notable at less ca-
thodically biased region with the synthesized Ag/C of
three different sizes compared to that of Ag foil. Again,
and selectivity than the 3 nm Ag/C for CO to CO conver-
2
sion. This size dependency on the selectivity can be un-
derstood by the competition of catalytic active sites. For
the case of Au nanoparticles, previous study reported that
the edge sites can stabilize the intermediate species of the
where the 5 nm Ag/C showed the best CO reduction re-
2
action performance reaching its maximum CO Faradaic
efficiency with the smallest overpotential (635 mV), and
CO reduction reaction more efficiently and also more
2
13
readily lead to CO evolution compared to corner sites.
On the other hand corner sites, which represent a larger
fraction of smaller nanoparticles, are more active for
1
3,15
HER.
In our samples, we think that the optimum size
of Ag nanoparticle to maximize the selectivity of the CO2-
to-CO reduction reaction was achieved at 5 nm size. This
is also in consistent with the previous study performed in
9
an organic electrolyte, which showed that 5 nm Ag na-
noparticles had the highest CO partial current density at
a given bias potential.
Notably, the reduction of overpotentials with our Ag
nanoparticle/C samples relative to Ag foil is also a signifi-
cant improvement. Beyond the increase of the mass activ-
ity as well as the CO Faradaic efficiency with 5 nm Ag/C,
we found that the onset potentials for CO reduction re-
2
action (Figure 2) and the overpotentials at the maximum
CO Faradaic efficiency (Figure 3) shifted in the anodic
direction, both of which indicates the overpotentials for
CO production decreased with Ag/C samples.
Figure 3. CO Faradaic efficiency a; depending on applied
potential and b; fixed potential at -0.75 V (vs. RHE). CO2
The 5 nm Ag/C is also more durable than polycrystal-
line Ag foil (Figure S10). Durability tests were performed
with Ag foil and the prepared 5 nm Ag/C at a specific po-
tentials (-1.1 V vs. RHE for Ag foil, and -0.8 V vs. RHE for 5
electrochemical reduction was performed in CO -saturated
2
0
.5 M KHCO . c; TEM image and HR-TEM image of 5 nm
3
Ag/C after CO reduction.
2
ACS Paragon Plus Environment

Page 5 of 9
Journal of the American Chemical Society
nm Ag/C) for 5 hours. Although current densities were
stable for both Ag foil and the 5 nm Ag/C during 5 hour
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
durability test, the decline of the CO Faradaic efficiency
was dramatically different. The decline of the CO Faradaic
efficiency was 58.2% (from 70.1% to 29.3%) for Ag foil
after 5 hours. Meanwhile, the CO Faradaic efficiency of 5
nm Ag/C after 5 hours decreased only by 18.9% (from 78.3%
to 63.5%). When the durability test with Ag foil was per-
formed at -0.8 V vs. RHE, a same potential applied to 5
nm Ag/C, the CO Faradaic efficiency decreased from 19.2%
to 7.7% within 1 hour, and became a HER catalyst. For
both samples, the CO Faradaic efficiency loss was trans-
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
3
ferred to H Faradaic efficiency. Lastly, CO isotope ex-
2
2
periment confirmed that the source of C was derived
completely from the dissolved CO in aqueous solution
2
(
Figure S11), and C from carbon support or cysteamine
molecules were not involved in the reduction reaction.
The rate determining step (RDS) of the CO reduction
2
reaction is generally thought to be formation of an ad-
-
-
sorbed CO2 (*CO ) intermediate on surfaces since the
2
anionic radical is highly unstable. The prepared Ag/C
samples and Ag foil had similar Tafel slopes around 140
-1
mVdec (Figure S12 and Table S1), which indicates that
-
the CO to *CO intermediate conversion step requires a
2
2
large overpotential, and the RDS is not changed for the
nanoparticulate Ag/C. However, exchange current densi-
ties commonly increased for Ag/C samples and more than
6
0 times enhancement was obtained with 5 nm Ag/C
compared to Ag foil (Table S1). A higher exchange current
density implies a higher intrinsic rate of electron transfer
between an electrolyte and the electrode and higher cata-
lytic kinetics. Moreover, the CO reduction reaction activ-
2
ities of the previously reported silver-based electrocata-
lysts are summarized in Table S2. Our cysteamine an-
chored Ag catalysts showed significantly high mass activi-
ty as well.
Figure 4. a; DFT results on the binding energies of the
COOH
COOH intermediate (ꢀE
) as a function of Ag nanopar-
B
ticle size. Ag13 (1.0 nm), Ag55 (1.6 nm), Ag147 (2.2 nm), and
Ag309 (2.7 nm) models with cuboctahedral geometry were
examined. b; Effect of anchoring agents on COOH and CO
binding energies examined using Ag(147-n)Cysn (n=0,1,2,4)
Understanding decrease of the overpotential: the
role of cysteamine anchoring agent
models. Colored map shows the theoretical CO reduction
2
0
potential (E ) as a function of COOH and CO binding free
Using DFT calculations, we examined the origin of en-
hanced catalytic activity of Ag/C. The previous theoretical
COOH
CO
and ꢀGB ; taken from the reference 10).
energies (ꢀGB
COOH
CO
Free energy values of ꢀGB
and ꢀGB
for Ag(147-n)Cysn
study on the CO reduction reaction mechanism suggest-
2
models were estimated from the DFT binding energies of
ed that the binding energy of the COOH intermediate
COOH
CO
COOH
ꢀEB
and ꢀEB using the fitted relationship between ꢀGB
(ꢀEB
) primarily determines the overpotential; higher
1
3,24
and ꢀE empirically derived from the previous study of the
B
COOH binding yields lower overpotential.
We thus
COOH CO
COOH
reference 10. For comparison, ꢀG
B
and ꢀG
B
calculated
calculated ꢀEB
to particles of various sizes: 1.0 nm, 1.6
for (111) slab surface of Cu, Au, Ag, and Zn are also displayed,
nm, 2.2 nm, and 2.7 nm cuboctahedral Ag nanoparticles
which well explains the experimental CO reduction poten-
2
(i.e., Wulff shape of Ag), which are Ag , Ag , Ag , and
1
3
55
147
tial of metal foil.
Ag309 respectively (Figure 4a). Unsurprisingly, low coordi-
nation number (CN) sites exposed at the corners (CN=5)
modulation observed from experiments varying the na-
noparticle size from 3 nm to 10 nm. Notably, the previous
Ag or Au nanoparticle reports only showed the changes of
or edges (CN=7) strongly bind to COOH. However, the
COOH
tunability of EB
(that is directly related to the tunabil-
ity of the overpotential) decreases as the nanoparticle size
the CO
particle size, and the onset overpotentials for CO
tion reaction were almost similar regardless of the particle
2
reduction current densities depending on the
increases, and becomes only within 126 meV (difference
2
reduc-
COOH
of ꢀEB
for CN=5 and CN=8) even when the nanopar-
9,13
size.
ticle size is 2.7 nm (that is the largest size available within
our limited computational cost). This suggests that size
effects alone cannot explain the strong overpotential
Consequently, the low overpotentials with our Ag/C
samples are expected to be associated with the anchoring
ACS Paragon Plus Environment

Journal of the American Chemical Society
agent, end group of –SH in cysteamine, that can change CONCLUSION
Page 6 of 9
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
the electronic structure in Ag surface as confirmed XPS.
To investigate this possibility, we performed DFT calcula-
tions of 2.2 nm Ag nanoparticle anchored with n number
In summary, silver nanoparticles with three different
sizes were directly synthesized on a carbon support by a
facile one-pot method using cysteamine as an anchoring
agent, which initiated the nucleation on the carbon sup-
port. These novel nanoparticulate Ag/C samples were
of cysteamine (Cys); Ag(147-n)Cys (n=0, 1, 2, 4) through Ag-
n
COOH
S bonds. As shown in Figure 4b, we found EB
shows a
monotonic increase as the cysteamine coverage increases
up to 261 meV as the cysteamine/Ag ratio increases until
effective as electrocatalysts for the CO reduction reaction
in an aqueous system, and CO isotope experiment con-
2
13
2
2.7%. Although it is difficult to experimentally quantify
firmed that CO was the carbon source to produce CO.
2
and/or control the cysteamine coverage to the Ag nano-
particles, we conceive that our 5 nm sized Ag/C sample
have an optimal activity due to suitable coverage of an-
choring agent, which was responsible for the remarkable
reduction of overpotential. More interestingly, we found
that the binding energy of the CO shows marginal chang-
es with respect to the effect of anchoring agent. We note
that modifying the catalyst surface to preferentially stabi-
lize the intermediate (COOH) over the final product (CO)
has its own importance in enhancing the catalytic activi-
Catalysts with 5 nm Ag/C sample exhibited the highest
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
CO reduction reaction activity in terms of low overpo-
2
tential, high mass activity, high CO Faradaic efficiency,
and high exchange current density with the enhanced
durability. Significant reduction of the overpotential with
directly grown Ag nanoparticle on carbon support was
associated with the cysteamine anchoring agent, which
developed Ag-S interaction. Our DFT studies elucidated
that the Ag-S interaction induces surface localization of
the unpaired electron, resulting in an enhanced interme-
diate stabilization and thus improved catalytic activity.
These studies demonstrate the potential for using sup-
ported Ag nanoparticles synthesized by simple wet chem-
1
0,25
ty.
To find the mechanistic origin of the selective modula-
tion of the surface binding energies for different adsorb-
ates, we analyzed the electronic structure of cysteamine
covered Ag nanoparticles. Figure 5 shows the change of
spatial spin density of the Ag nanoparticles as the number
istry methods as CO reduction catalysts with high mass
2
activities and low overpotential for CO production and
also suggest that anchoring groups such as cysteamine
can be used to control their materials properties.
of cysteamine increases (isosurface value is set as 0.001).
26,27
In the context of superatom complex model,
Ag(147-
Cys system can be understood as a superatom consist-
n)
n
EXPERIMENTAL SECTION
ing of single unpaired electron. When no anchoring agent
exists, this unpaired electron is delocalized over the entire
superatom, which is too stable to be utilized for the
chemical bonding with COOH radical. However, when
anchoring agents are introduced, symmetry breaking
leads to spatial localization of the unpaired electron par-
ticularly at the surface of Ag superatom (Figure 5). This
can provide more covalent character to the Ag-COOH
bond, which effectively stabilizes the intermediate
Synthesis and characterization of Ag nanoparticles
on carbon supports
For Ag nanoparticle/C, 20 mg of silver nitrate (AgNO ;
3
Aldrich, 99.9999%) was dissolved in 10 mL of ethylene
glycol (EG; Aldrich, 99.8%) with vigorous stirring and this
solution was slowly heated to 50 °C for 20 min. 20 mg of
carbon black (Ketjen black) was dispersed in 10 mL EG
and 1 mg of cysteamine solution (Aldrich) by ultrasoni-
cation for 30 min. Cysteamine was used as an anchoring
agent. The prepared carbon black solution was added to
the silver precursor solution at 50 °C, and the mixed solu-
tion was kept at 50 °C for 10 min prior to heating to 160 °C
with a ramping rate of 3-4 °C/min. 3 nm Ag/C and 5 nm
Ag/C were synthesized by controlling the maintaining
time at 160 °C for 1 hour and 3 hours, respectively. For 10
nm Ag/C, 20 mg of silver acetylacetonate (Ag(acac); Al-
drich, 98%) was dissolved in 10 mL 1-octadecene (ODE,
Aldrich) under a nitrogen flow with vigorous stirring. The
procedure was the same with the method above. The cys-
teamine-treatment carbon (denoted as Cysteamine-C)
was prepared without Ag precursor following the same
procedure.
(
COOH radical). However, the localized unpaired elec-
tron is less effective in stabilizing the donor-acceptor
bond between Ag and CO, resulting in a marginal change
of the binding energy of CO. It is thus of noted that the
effect of cysteamine can be understood using the previ-
ously suggested concept of covalency-aided electrochemi-
1
0
cal reaction (CAER) mechanism.
The product solution was cooled down to room tem-
perature, washed with isopropanol, filtered and dried.
The prepared catalysts were analyzed by X-ray photoelec-
tron spectroscopy (XPS; Thermo Scientific K-Alpha), X-
ray diffraction (XRD; Shimadzu XRD-6000), inductively
coupled plasma optical emission spectrometry (ICP-OES;
Thermo Scientific iCAP 6300 Duo), transmission electron
Figure 5. Electron spin density computed from DFT wave-
function is shown with the isosurface value of 0.001. Specific
interaction developed in between Ag nanoparticle and the
anchoring agents (cysteamine) leads the unpaired electron to
be localized at the surface state of Ag nanoparticle, which
helps to selectively stabilize COOH over CO.
2
microscopy (TEM; TECNAI F20 G at 200 kV) and high
ACS Paragon Plus Environment

Page 7 of 9
Journal of the American Chemical Society
at a fixed potential that produced the highest Faradaic
resolution transmission electron microscopy (HR-TEM;
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
JEM-ARM200F at 200 kV).
efficiency.
Solution resistance was obtained by measuring the elec-
trochemical impedance spectroscopy (EIS) at various po-
tential. The measured potentials for CO reduction reac-
tion were compensated for iR loss, and were reported ver-
sus the reversible hydrogen electrode (RHE) by using the
following equation.
Preparation of working electrode
2
The catalysts were dispersed into a mixture of isopro-
panol and Nafion solution (5 wt%, DuPont) to prepare a
catalyst ink by sonication for 30 min. The actual metal
amount in the solution was measured by inductively cou-
pled plasma (ICP) elemental analysis. Glassy carbon plate
(Alfa Aesar) was used as the electrode substrate after me-
chanically polishing and complete cleaning with deion-
ized water. The catalyst ink solution was sprayed onto the
glassy carbon plate. The active area of the electrode was
E (vs. RHE) = E (vs. Ag/AgCl) + 0.197 V + 0.0591 V × pH
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Density functional theory calculations
Density functional theory (DFT) calculations were per-
formed using Spanish Initiative for Electronic Simulations
with Thousands of Atoms (SIESTA) program. We chose
exchange-correlation functional of Perdew-Burke-
Ernzerhof (PBE) method (ref), and double zeta level of
basis sets were employed. For the sampling of reciprocal
space, only the gamma point is sampled because we are
considering the non-periodic nanoparticles, and Mesh
cutoff energy and force tolerance were set as 250 Rydberg
and 0.1 eV/Å, respectively, for the geometry optimization.
2
0.5 cm and the sprayed amount of the metal catalyst was
2
0.09 ± 0.01 mg /cm .
Ag
The carbon and Cysteamine-C ink solutions were pre-
pared by the same procedure and sprayed on the glassy
carbon plate. For the comparison, Ag foil was used as a
working electrode after mechanically polishing the sur-
2
face, which electrode active area was also 0.5 cm .
Electrochemical measurements
Platinum and Ag/AgCl (saturated KCl) were used as the
counter electrode and the reference electrode, respective-
ly. Electrochemical measurements were performed by
using Ivium potentiostat (Iviumtechnology) in a two-
compartment electrochemical cell, and a Nafion 117, a
proton exchange membrane, was used to separate the
catholyte and the anolyte. The electrolyte solution (0.5 M
ASSOCIATED CONTENT
Supporting Information. Additional TEM image, XRD, and
XPS spectra. Further electrocatalytic CO reduction data with
error bars. This material is available free of charge via the
Internet at http://pubs.acs.org.
2
KHCO ) was purged with high purity CO gas for at least 1
AUTHOR INFORMATION
Corresponding Author
3
2
hour, where the pH was 7.0 after saturation. For obtaining
stable electrochemical signals, chronoamperometry was
*B. K. Min: bkmin@kist.re.kr
performed at -2.0 V (vs. Ag/AgCl) for 10 min prior to CO2
*Y. J. Hwang: yjhwang@kist.re.kr
reduction reaction measurements. The CO2 reduction
reaction were measured by using chronoamperometry at
each fixed potential and gaseous products (i.e. H and CO)
were quantified by a gas chromatography (GC; Younglin
Author Contributions
These authors contributed equally to this work.
2
†
6
500 GC) equipped with a pulsed discharge detector
Notes
(
PDD). Ultra high purity helium (He; 99.9999%) was used
The authors declare no competing financial interests.
as the carrier gas. CO average flow rate (Q) was 120
2
cc/min measured by a universal flow meter (Agilent tech-
nologies ADM 2000) at the exit of the electrochemical cell.
The partial current and Faradaic efficiency of H or CO
ACKNOWLEDGMENT
This work was supported by the program of the Korea Insti-
tute of Science and Technology (KIST) and University-
Institute cooperation program of the National Research
Foundation of Korea, funded by the Korean Government
(MSIP). We also thank Young Scientists Fellowship program
of National Research council of Science & Technology and
the Rowland Institute at Harvard University for support
through a Rowland Visiting Scholar Award.
2
production (iH₂ or CO and F.E.H₂ or CO) were calculated from
GC chromatogram peak areas where VH₂or CO is volume
concentration of H or CO based on calibration of the GC,
2
itotal is measured current, F is Faradaic constant, p is pres-
0
sure, T is temperature, and R is ideal gas constant, 8.314
-
1 11,28
J·mol·K .
2
Fꢊꢋ
REFERENCES
ꢀꢁꢂꢃꢄꢅꢃꢆꢇ
= ꢈꢁꢂꢃꢄꢅꢃꢆꢇ × ꢉꢃ ×
ꢌꢍ
(1)
Turner, J. A. Science 1999, 285, 687-689.
ꢀꢁꢂ
ꢄꢅꢃꢆꢇ
(2)
Goeppert, A.; Czaun, M.; May, R. B.; Prakash, G. K. S.;
ꢎ. ꢏ.ꢁꢂꢃꢄꢅꢃꢆꢇ =ꢃ
ꢃ× 100
ꢀꢐꢄꢐꢑꢒ
Olah, G. A.; Narayanan, S. R. J. Am. Chem. Soc. 2011, 133, 20164-
0167.
3)
S. Renew. Sustainable Energy Rev. 2011, 15, 1-23.
2
(
Graves, C.; Ebbesen, S. D.; Mogensen, M.; Lackner, K.
A durability test was performed for 5 hours in CO2-
saturated 0.5 M KHCO solution by chronoamperometry
3
ACS Paragon Plus Environment

Journal of the American Chemical Society
Page 8 of 9
(4)
Centi, G.; Quadrelli, E. A.; Perathoner, S. Energy Envi-
(17)
Wang, C.; Daimon, H.; Onodera, T.; Koda, T.; Sun, S.
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ron. Sci. 2013, 6, 1711-1731.
Angew. Chem. Int. Ed. 2008, 47, 3588-3591.
(5)
Mikkelsen, M.; Jorgensen, M.; Krebs, F. C. Energy Envi-
(18)
Grass, M. E.; Joo, S. H.; Zhang, Y.; Somorjai, G. A. J.
ron. Sci. 2010, 3, 43-81.
Phys. Chem. C 2009, 113, 8616-8623.
(
6)
, 3451-3458.
7) Hori, Y. In Modern Aspects of Electrochemistry;
Whipple, D. T.; Kenis, P. J. A. J. Phys. Chem. Lett. 2010,
(19)
Kim, C.; Oh, J.-G.; Kim, Y.-T.; Kim, H.; Lee, H. Electro-
1
chem. Commun. 2010, 12, 1596-1599.
(20) Kim, C.; Kim, S. S.; Yang, S.; Han, J. W.; Lee, H. Chem.
Commun. 2012, 48, 6396-6398.
(
Vayenas, C., White, R., Gamboa-Aldeco, M., Eds.; Springer New
York: 2008; Vol. 42, p 89-189.
(21)
Kim, C.; Kim, J.; Yang, S.; Lee, H. RSC Advances 2014, 4,
(
8)
Jaramillo, T. F. Phys. Chem. Chem. Phys. 2014, 16, 13814-13819.
(9) Salehi-Khojin, A.; Jhong, H.-R. M.; Rosen, B. A.; Zhu,
W.; Ma, S.; Kenis, P. J. A.; Masel, R. I. J. Phys. Chem. C 2012, 117,
627-1632.
10)
Min, B. K.; Kim, H. J. Am. Chem. Soc. 2014, 136, 11355-11361.
11) Chen, Y.; Li, C. W.; Kanan, M. W. J. Am. Chem. Soc.
012, 134, 19969-19972.
12) Lu, Q.; Rosen, J.; Zhou, Y.; Hutchings, G. S.; Kimmel, Y.
C.; Chen, J. G.; Jiao, F. Nat. Commun. 2014, 5, 6.
(13) Zhu, W.; Michalsky, R.; Metin, Ö.; Lv, H.; Guo, S.;
Wright, C. J.; Sun, X.; Peterson, A. A.; Sun, S. J. Am. Chem. Soc.
013, 135, 16833-16836.
14) Reske, R.; Mistry, H.; Behafarid, F.; Roldan Cuenya, B.;
Strasser, P. J. Am. Chem. Soc. 2014, 136, 6978-6986.
15) Zhu, W.; Zhang, Y.-J.; Zhang, H.; Lv, H.; Li, Q.;
Michalsky, R.; Peterson, A. A.; Sun, S. J. Am. Chem. Soc. 2014, 136,
Hatsukade, T.; Kuhl, K. P.; Cave, E. R.; Abram, D. N.;
63677-63680.
(22) Battocchio, C.; Fratoddi, I.; Fontana, L.; Bodo, E.; Por-
caro, F.; Meneghini, C.; Pis, I.; Nappini, S.; Mobilio, S.; Russo, M.
V.; Polzonetti, G. Phys. Chem. Chem. Phys. 2014, 16, 11719-11728.
(23) Lee, J.-S.; Park, G. S.; Kim, S. T.; Liu, M.; Cho, J. Angew.
Chem. Int. Ed. 2013, 52, 1026-1030.
(24) Gao, D.; Zhou, H.; Wang, J.; Miao, S.; Yang, F.; Wang,
G.; Wang, J.; Bao, X. J. Am. Chem. Soc. 2015, 137, 4288-4291.
(25) Hansen, H. A.; Varley, J. B.; Peterson, A. A.; Nørskov, J.
K. J. Phys. Chem. Lett. 2013, 4, 388-392.
(26) Hakkinen, H. Chem. Soc. Rev. 2008, 37, 1847-1859.
(27) Walter, M.; Akola, J.; Lopez-Acevedo, O.; Jadzinsky, P.
D.; Calero, G.; Ackerson, C. J.; Whetten, R. L.; Grönbeck, H.;
Häkkinen, H. Proc. Natl. Acad. Sci. 2008, 105, 9157-9162.
(28) Jeon, H. S.; Koh, J. H.; Park, S. J.; Jee, M. S.; Ko, D.-H.;
Hwang, Y. J.; Min, B. K. J. Mater. Chem. A 2015, 3, 5835-5842.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
(
Lim, H.-K.; Shin, H.; Goddard, W. A.; Hwang, Y. J.;
(
2
(
2
(
(
16132-16135.
(16) Zhang, J.; Liu, H.; Wang, Z.; Ming, N. Adv. Funct. Mater.
2007, 17, 3295-3303.
ACS Paragon Plus Environment

Page 9 of 9
Journal of the American Chemical Society
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
9
ACS Paragon Plus Environment