Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial Photosynthesis

Article abstract of DOI:10.1016/j.chempr.2017.09.014

Utilizing solar energy to fix CO₂ with water into chemical fuels and oxygen, a mimic process of photosynthesis in nature, is becoming increasingly important but still challenged by low selectivity and activity, especially in CO₂ electrocatalytic reduction. Here, we report transition-metal atoms coordinated in a graphene shell as active centers for aqueous CO₂ reduction to CO with high faradic efficiencies over 90% under significant currents up to ～60 mA/mg. We employed three-dimensional atom probe tomography to directly identify the single Ni atomic sites in graphene vacancies. Theoretical simulations suggest that compared with metallic Ni, the Ni atomic sites present different electronic structures that facilitate CO₂-to-CO conversion and suppress the competing hydrogen evolution reaction dramatically. Coupled with Li⁺-tuned Co₃O₄ oxygen evolution catalyst and powered by a triple-junction solar cell, our artificial photosynthesis system achieves a peak solar-to-CO efficiency of 12.7% by using earth-abundant transition-metal electrocatalysts in a pH-equal system. Using clean electricity to reduce CO₂ to chemicals or fuels is becoming increasingly important to renewable energy applications and environmental protection. The challenge comes from the strong competition with the hydrogen evolution reaction in aqueous solutions, especially for those earth-abundant transition metals such as Ni, which dramatically lowers the CO₂ reduction selectivity. Isolating the transition-metal single atoms into a graphene matrix can significantly tune their catalytic behaviors to favor the CO₂-to-CO reduction pathway, reaching a high CO selectivity of more than 90%. This work creates an important platform in designing active and low-cost CO₂ reduction catalysts with high selectivity toward fuels, opening up great opportunities for both technological applications in renewable energies and fundamental mechanism studies in catalysis. State-of-the-art three-dimensional atom probe tomography provides direct evidence of Ni single atoms coordinated in graphene vacancies for highly selective CO₂ reduction to CO and suppressed hydrogen evolution in water.

Full text of DOI:10.1016/j.chempr.2017.09.014

Article
Transition-Metal Single Atoms in a Graphene
Shell as Active Centers for Highly Efficient
Artificial Photosynthesis
Kun Jiang, Samira Siahrostami,
Austin J. Akey, ..., Jens K.
Nørskov, Yi Cui, Haotian Wang
hwang@rowland.harvard.edu
HIGHLIGHTS
Single Ni atoms coordinated in
graphene exhibit superb catalytic
2
CO RR performance
Atomic-resolution 3D APT
provides direct identification of
single Ni atoms
DFT simulations suggest that the
tuning of electronic structure
2
favors CO RR over HER
State-of-the-art three-dimensional atom probe tomography provides direct
evidence of Ni single atoms coordinated in graphene vacancies for highly selective
2
CO reduction to CO and suppressed hydrogen evolution in water.
Jiang et al., Chem 3, 1–11
December 14, 2017 ª 2017 Elsevier Inc.
https://doi.org/10.1016/j.chempr.2017.09.014

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
Article
Transition-Metal Single Atoms in a Graphene
Shell as Active Centers for Highly
Efficient Artificial Photosynthesis
Kun Jiang,1,11 Samira Siahrostami,2,11 Austin J. Akey, Yanbin Li, Zhiyi Lu, Judith Lattimer,
3
4
4
5
6
1
1
4
7
1
Yongfeng Hu, Chris Stokes, Mahesh Gangishetty, Guangxu Chen, Yawei Zhou, Winfield Hill,
Wen-Bin Cai, David Bell,3,8 Karen Chan, Jens K. Nørskov, Yi Cui,3,10 and Haotian Wang1,12,
7
9
2,9
*
SUMMARY
The Bigger Picture
Using clean electricity to reduce
CO to chemicals or fuels is
2
Utilizing solar energy to fix CO with water into chemical fuels and oxygen, a
mimic process of photosynthesis in nature, is becoming increasingly important
but still challenged by low selectivity and activity, especially in CO electrocata-
lytic reduction. Here, we report transition-metal atoms coordinated in a
graphene shell as active centers for aqueous CO reduction to CO with high
2
becoming increasingly important
to renewable energy applications
and environmental protection.
The challenge comes from the
strong competition with the
hydrogen evolution reaction in
aqueous solutions, especially for
those earth-abundant transition
metals such as Ni, which
2
2
faradic efficiencies over 90% under significant currents up to ꢀ60 mA/mg. We
employed three-dimensional atom probe tomography to directly identify the
single Ni atomic sites in graphene vacancies. Theoretical simulations suggest
that compared with metallic Ni, the Ni atomic sites present different electronic
structures that facilitate CO
2
-to-CO conversion and suppress the competing
+
hydrogen evolution reaction dramatically. Coupled with Li -tuned Co
3
O
4
oxygen evolution catalyst and powered by a triple-junction solar cell, our
artificial photosynthesis system achieves a peak solar-to-CO efficiency of
dramatically lowers the CO
2
reduction selectivity. Isolating the
transition-metal single atoms into
a graphene matrix can
1
2.7% by using earth-abundant transition-metal electrocatalysts in a pH-equal
system.
significantly tune their catalytic
2
behaviors to favor the CO -to-CO
INTRODUCTION
reduction pathway, reaching a
high CO selectivity of more than
Effectively converting clean solar energy into carbon fuels via electrocatalytic carbon
dioxide (CO
2
) reduction and water oxidation, a mimic process of photosynthesis in
9
0%. This work creates an
important platform in designing
active and low-cost CO reduction
nature, can potentially play a critical role in sustaining the global energy demands
1
–6
and in preventing further CO
currently challenged by the low activity and selectivity of the CO
CO RR) because of the high kinetic barriers and competition with the hydrogen
evolution reaction (HER) in aqueous media. Strategies, including exploring novel
2
emissions.
However, this practical application is
2
2
reduction reaction
catalysts with high selectivity
toward fuels, opening up great
opportunities for both
(
2
7
–9
8
,9
10–12
catalysts or using non-aqueous electrolyte such as ionic liquid,
extensively studied to reduce reaction barriers or suppress HER. Highly selective
CO reduction requires catalysts to have specific electronic structures that could
facilitate the CO activation process and also properly bind reaction intermediates,
not too strong or too weak. A representative example is a transition-metal (TM)
catalyst such as Au, which has been demonstrated to convert CO to carbon
whereas Pt, with only one d-band electron
have been
technological applications in
renewable energies and
fundamental mechanism studies
in catalysis.
2
2
1
3
2
1
4–16
monoxide (CO) with high selectivity,
less, generates H exclusively and can be easily poisoned by CO.
abundant TMs such as Fe, Co, and Ni are rarely studied as CO
mainly because of their good HER activities as well as the strong bonding between
1
7,18
2
Other earth-
to CO catalysts,
2
1
7–21
CO and the metal surfaces.
Therefore, how to effectively tune the catalytic
RR catalysts.
electronic properties plays a critical role in searching for active CO
2
Chem 3, 1–11, December 14, 2017 ª 2017 Elsevier Inc.
1

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
Different methods, such as metal oxidation and metal alloying, have been demon-
strated to be successful in engineering TM electronic states for improved CO
2
RR ac-
7
,8,16,22–24
tivities.
However, these engineering processes on TM catalysts usually
result in complicated atomic structures and coordination, making it difficult to study
and understand the possible catalytic active sites. Instead, introducing TM atoms
into a well-established material matrix could open up great opportunities to
(
1) tune the electronic properties of TMs as CO
2
RR active sites and (2) at the same
time maintain relatively simple atomic coordination for fundamental mechanism
9
,25
studies.
be easily moved around during catalysis, which prevents the nucleation or recon-
In addition, those TM atoms trapped in a confined environment cannot
2
6–29
structions of surface atoms observed in many cases.
particular interest as host for TM atoms because of their high electron
conductivity, chemical stability, and inertness to both CO reduction and HER
Graphene layers are of
2
3
0
catalytic reactions. TM atoms can be trapped in the naturally or chemically formed
defects of the graphene such as single vacancies (SVs) and double vacancies (DVs),
3
1–33
presenting distinctively different properties from bulk metal materials.
The
introduction of N dopants can generate graphene defects that could significantly in-
3
4
crease the concentration of TM atoms coordinated within the layers. Although
some previous studies suggested CO -to-CO catalysis on M-N-C sites, there was
no direct evidence to demonstrate the single-atom morphology or the coordination
2
3
5,36
environment of the active sites.
atoms embedded as a highly active electrocatalyst for CO
aqueous electrolyte. We employed three-dimensional (3D) atom probe tomography
APT) to directly identify the single Ni atomic sites in graphene vacancies. With den-
sity functional theory (DFT) calculations, we show that, compared with metallic Ni, Ni
atomic sites in graphene can dramatically lower the CO activation barrier, weaken
Here, we report graphene shells (GSs) with Ni
2
reduction to CO in an
(
2
the binding with CO for facile product release, and suppress the proton reduction
side reaction. As a result, the catalyst exhibits a high faradic efficiency (FE)
of ꢀ93.2% toward CO formation under a significant current of ꢀ20 mA/mg, which
1Rowland Institute, Harvard University,
Cambridge, MA 02142, USA
2SUNCAT Center for Interface Science and
Catalysis, Department of Chemical Engineering,
Stanford University, Stanford, CA 94305, USA
À1
represents a turnover frequency (TOF) of ꢀ8 s for the active Ni center, and can
be further improved with a gas diffusion layer electrode in a flow cell.
3Center for Nanoscale Systems, Harvard
University, Cambridge, MA, USA
4Department of Materials Science and
RESULTS
Engineering, Stanford University, Stanford, CA
94305, USA
The catalysts were synthesized by electrospinning of polymer nanofibers (NFs) with Ni
and N precursors homogeneously distributed (Supplemental Experimental Proced-
ures). Graphitized carbon NFs (CNFs, ꢀ200 nm in diameter) catalyzed by uniformly
dispersed Ni nanoparticles (NPs) (ꢀ20 nm in diameter) were obtained after the carbon-
5Department of Chemistry and Chemical Biology,
Harvard University, Cambridge, MA 02138, USA
6Canadian Light Source Inc., University of
Saskatchewan, 44 Innovation Boulevard,
Saskatoon, SK S7N 2V3, Canada
3
7–39
ization process of polymers (Figures 1A and 1B).
Carbon atoms alloyed with Ni un-
7Shanghai Key Laboratory of Molecular Catalysis
and Innovative Materials, Collaborative
Innovation Center of Chemistry for Energy
Materials, Department of Chemistry, Fudan
University, Shanghai 200433, China
der high temperature precipitated out and were catalyzed to form graphene layers on
the Ni metal surface during the cooling down process. A closer observation of the Ni
NPs by aberration-corrected scanning transmission electron microscopy (STEM) in Fig-
ure 1C reveals that the NP is tightly encapsulated by a few layers (ꢀ10 nm) of graphene,
as confirmed by the averaged layer spacing of ꢀ0.34 nm (NiN-GS; Figure S1). No Ni
clusters were observed within the GS. This shell prevents the Ni NP from direct contact
with the aqueous electrolyte and can thus dramatically suppress HER. The existence of
Ni atoms in the surface shell was confirmed by energy-dispersive X-ray spectroscopy
8School of Engineering and Applied Sciences,
Harvard University, Cambridge, MA 02142, USA
9
SUNCAT Center for Interface Science and
Catalysis, SLAC National Accelerator Laboratory,
575 Sand Hill Road, Menlo Park, CA 94025, USA
2
10Stanford Institute for Materials and Energy
Sciences, SLAC National Accelerator Laboratory,
2575 Sand Hill Road, Menlo Park, CA 94025, USA
(
EDS) mapping in Figure 1D (Figure S2 and Supplemental Experimental Procedures).
The Z-contrast STEM image on the left shows three bright areas representing three
Ni NPs, with one of them highlighted by the yellow circle. In the Ni mapping image
11These authors contributed equally
12Lead Contact
(
marked by green dots), in addition to the NP regions with concentrated signals, Ni
peaks were also detected in the neighboring carbon areas (Figure S2), demonstrating
the successful incorporation of Ni atoms in the GS. N doping here plays a critical role in
*Correspondence: hwang@rowland.harvard.edu
https://doi.org/10.1016/j.chempr.2017.09.014
2
Chem 3, 1–11, December 14, 2017

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
Figure 1. Characterizations of NiN-GS Catalysts
(
(
A) Scanning electron microscopy (SEM) image of carbonized electrospun polymer NFs. Scale bar, 5 mm.
B) TEM image of a ball-milled NiN-GS catalyst. The dark dots (highlighted by red circles as examples) uniformly distributed in the CNF are Ni NPs.
Scale bar, 200 nm.
C) Aberration-corrected STEM image of a Ni NP tightly wrapped by a few graphene layers. The Ni NP is ꢀ20 nm in diameter. The GS is ꢀ10 nm thick.
The layer spacing is measured to be 0.34 nm. Scale bar, 5 nm.
D) EDS mapping of the NiN-GS catalyst. Three Ni NPs were observed in the STEM image on the left, which is consistent with the Ni mapping image, with
(
(
one of the NPs indicated by the yellow circle. Ni signals were detected in areas away from the NPs, demonstrating the successful incorporation of
Ni atoms in graphene layers. The selected area spectra are shown in Figures S2 and S5. Scale bar, 20 nm.
creating defects in the graphene layers, which helps to trap and bond a significant num-
3
4
ber of Ni atoms in the GS. This is demonstrated by both Raman and transmission
electron microscopy (TEM) characterizations where the graphene layers in NiN-GS pre-
sent a defective nature compared with the sample without N incorporation (Ni-GS;
40
Figures S3 and S4 and Supplemental Experimental Procedures). In addition, no Ni
signals were detected in the GS outside the Ni NP in Ni-GS, because of the high quality
of graphene synthesized (Figure S5). Those coordinated Ni atoms within the graphene
layers in NiN-GS showed distinctively different oxidation states from Ni NPs covered
below by X-ray photoelectron spectroscopy (XPS; Figure S6), suggesting the successful
tuning of Ni electronic structures and thus the possible tuning of its catalytic activities.
The electrocatalytic CO
glassy carbon current collector, was performed in 0.1 M potassium bicarbonate
KHCO ) electrolyte in a customized H cell (Supplemental Experimental Procedures).
The different cyclic voltammograms (CVs) in CO and N saturated electrolyte
suggest that reactions other than HER occur when CO is present (Figures 2A and
2
RR performance of the NiN-GS catalyst, drop casted on a
(
3
2
2
2
S7). Gas products were analyzed by gas chromatography (GC) at potential
steps of 100 mV and further zoomed into 50 mV around the FE peak (Figures 2B
and S8 and Supplemental Experimental Procedures). Detectable CO signals start
at –0.35 V versus reversible hydrogen electrode (RHE), suggesting the onset overpo-
2
tential of CO to CO to be less than 230 mV (Figure S9). As the potential goes more
negative, the FE of CO increases as the HER decreases (Figure 2B). The overall FE
under different potentials ranges from 91.2% to 105.8% (Figures 2B, Table S1).
Chem 3, 1–11, December 14, 2017
3

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
100
0
0
0
Partial Current
0
8
0
-
20
-4
-8
-12
-
10
-2
-4
-6
6
0
H2
CO
-40
40
CO (pH = 6.8)
H2
2
-20
N (pH = 8.8)
CO
.1 M KHCO3
2
20
0
-60
0
-30
-0.9
-0.6
-0.3
0.0
0.3
-1.0
-0.8
-0.6
-0.4
-1.0
-0.8
-0.6
-0.4
E vs RHE (V)
E vs RHE (V)
E vs RHE (V)
100
10
80
0
6
0
-
10
20
30
40
40
-
-
20
0
=
- 0.7 V
-
0
5
10
Time (h)
15
20
1
00
0
0
100
80
60
40
20
0
0
0
H
2
CO
H2
CO
8
0
-20
-40
-4
-
10
-10
-20
-30
-40
6
0
-8
-
20
4
0
-60
-12
-16
-20
H2
CO
Overall
H2
-
30
CO
Overall
1 MKHCO3
2
0
-80
0
.5 M KHCO3
-
40
0
-100
-0.9
-0.8
-0.7
E vs RHE (V)
-0.6
-0.5
-0.9
-0.8
-0.7
-0.6
-0.5
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4
E vs RHE (V)
-1.0 -0.9 -0.8 -0.7 -0.6 -0.5 -0.4
E vs RHE (V)
E vs RHE (V)
Figure 2. Electrocatalytic CO
2
Reduction Performance of NiN-GS
and N saturated electrolyte, suggesting a different reaction pathway when CO
and CO under different applied potentials for NiN-GS. The error bars are based on three identical samples. The highest CO FE is 93.2%
(
(
A) CVs of NiN-GS in CO
B) FEs of H
2
2
2
is present.
2
under –0.82 V versus RHE, with an overpotential of À0.7 V. The error bars represent three identical samples.
(
(
C) Partial currents of H
2
and CO. The CO evolution current density is around 20 mA/mg under the highest CO FE.
D) Long-term electrolysis test under –0.7 V overpotential. The high FE of CO is maintained for more than 20 hr, suggesting good stability of the NiN-GS
catalyst. The error bars represent three identical samples.
(
E) The TEM image of post-catalysis NiN-GS. Scale bar, 5 nm.
F and G) CO RR catalytic performance (current densities in F and FEs in G) of 1 mg/cm NiN-GS catalyst on a high-surface-area CFP electrode in 0.5 M
electrolyte.
H and I) CO RR performance (current densities in H and FEs in I) of NiN-GS catalyst on a gas diffusion layer electrode in a flow cell configuration. The
catalyst mass loading is the same as for glassy carbon electrode in (A)–(D).
2
(
2
KHCO
3
(
2
The highest FE of CO
2
to CO reaches 93.2% at 0.7 V overpotential, with a CO
2
2
evolution current density of ꢀ20 mA/mg (4 mA/cm at 0.2 mg/cm mass loading;
1
3
Figures 2C and S10). An isotope CO
on GC-mass spectroscopy to confirm that the gas product of CO comes from
CO reduction (Figure S11 and Supplemental Experimental Procedures).
The Tafel slope plotted with electrolysis currents and overpotentials was
2
labeling experiment was performed
2
9
,40–42
1
38.5 mV/decade (Figure S12).
No other gas or liquid products were detected
1
by GC or H NMR (Figure S13). Whether these surrounded GSs will be stable under
long-term operations is a concern, because if the protection layer breaks, the Ni NPs
1
9
2
will be exposed to water and could produce H heavily. Around 80% FE of CO was
4
Chem 3, 1–11, December 14, 2017

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
still maintained after more than 20 hr of continuous electrolysis under 0.7 V overpo-
tential (Figure 2D), suggesting the excellent stability of the catalytic sites. A post-
catalysis TEM examination also confirmed that the core-shell structure remains
intact to prevent contact between Ni NPs and water (Figures 2E and S14). The
CO
0 mA/cm by loading more NiN-GS catalysts onto high-surface-area carbon fiber
paper (CFP) substrate while maintaining high CO FEs in 0.5 M KHCO electrolyte
Figures 2F and 2G). In addition, the overpotential can be further improved by
2
RR geometrical current density can be significantly improved to more than
2
2
3
(
2
applying NiN-GS catalyst (0.2 mg/cm ) onto a gas diffusion layer electrode in a
4
3
flow cell (Figures 2H and 2I), where a CO partial current of 31.4 mA/mg was ob-
tained at 0.48 V overpotential (Supplemental Experimental Procedures, Figure S15,
and Tables S1 and S2).
Control experiments were performed to provide important clues on the possible active
2
sites in NiN-GS for this highly selective CO reduction. First, it is unlikely that the Ni NP
cores are in contact with the electrolyte to participate in the gas reduction reactions. This
is confirmed by the TEM images of NiN-GS after an acid leaching process (AL-NiN-GS;
44
Figure S16 and Supplemental Experimental Procedures), which show the Ni NPs to be
wellprotectedfromconcentratedprotonsbythetightlysurroundedGSs.Onlybyviolent
ball milling followed by acid leaching can we break off some core-shell structures and
allow acids to attack Ni NPs (VAL-NiN-GS; Figure S16 and Supplemental Experimental
Procedures). In addition, Ni-GS with Ni NPs embedded presents nearly no activity to-
ward CO formation (Figures S17 and S18). Second, the dramatically decreased CO evo-
lutionactivity per electrochemical surface area (Figure S19) after acid leaching processes
suggest that Ni atoms in the surface GS play a more important role than the N dopants
3
6,40,41
(
Figure S20).
Third, Co and Fe catalysts with the same core-shell structure and N
doping (Figure S21 and Supplemental Experimental Procedures), however, present
lower activity and selectivity toward CO formation than NiN-GS (Figure S22), indicating
that the high CO FE is related to the specific electronic structure of Ni sites in the GS.
More detailed characterizations to elucidate the nature of Ni atomic sites in graphene
layers become important for a clear understanding of the catalytic active sites. Here,
4
5,46
we used APT technology (Figure S23) to reveal
whether the Ni atoms are isolated
single atoms or small clusters and whether the Ni atomic sites are coordinated with N
or not. The 3D tomography of NiN-GS catalyst is presented in Movie S1, and the
projected 2D image is shown in Figure 3A. Each pixel represents one single atom.
As shown in Figure 3B and Movie S2, away from the areas with concentrated Ni
(
Ni NPs), there are also a significant number of Ni atoms dispersed in carbon, consis-
tent with our EDX mapping in Figure 1D. The contour map with an interval of 2 at % in
Figure 3C provides detailed distribution information on Ni atoms in the catalyst, with
decreased Ni atom concentrations away from the Ni sources. The local coordination
environment of the Ni atoms is shown by taking a closer look at the graphene layers
in Figure 3D. There are a few Ni single atoms coordinated in graphene vacancies,
providing direct evidence of the single Ni atomic site. No Ni clusters were observed.
In addition, we also noticed that there is one Ni atom coordinated with one N atom
in a graphene vacancy, suggesting a small ratio of Ni atomic sites coordinated with
N atoms. More detailed information about the surrounding coordination of Ni atomic
sites can be extracted from statistics and quantitative analysis (Figure 3F). The selected
area with dispersed Ni atoms is indicated by the yellow circle in Figure 3A and enlarged
in Figure 3E. Among all the Ni atoms in this area, 83% are single atoms, without neigh-
˚
boring Ni atoms closer than 2.2 A (Figure S24). In addition, in those Ni single atoms,
˚
only 0.2% of them are directly coordinated with one neighboring N (less than 2 A), sug-
gesting that most of the Ni atomic sites are coordinated with C atoms. Single-atom
Chem 3, 1–11, December 14, 2017
5

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
Figure 3. ATP of the NiN-GS Catalyst
(
(
A) The 2D atom map of NiN-GS. The yellow circle represents the selected area for statistical studies in (E) and (F). Scale bar, 10 nm.
B) The 2D projected view of Ni atoms. The green areas represent Ni-rich areas (>50 at %). Away from the Ni sources, there are still a significant number
of Ni atoms dispersed in the carbon area. Scale bar, 10 nm.
(
(
C) The contour map of Ni concentration with an interval of 2 at %.
D) Zoomed in side view (upper) and top view (lower) of graphene layers with Ni single atoms coordinated in vacancies. Only one Ni atom is directly coordinated
with one N atom. Scale bars, 1 nm.
(
(
E) Atom map of the selected area in (A) as indicated by the yellow circle. Scale bar, 5 nm.
F) The statistics of the selected area in (E). Most of the Ni atoms are in single-atom morphology, and 0.2% of them are coordinated with N atoms.
coordination in the graphene shell is also observed by X-ray absorption spectroscopy
˚
as the small peak at ꢀ1.4 A in R space beside the dominating Ni–Ni bonding signal
from those embedded Ni NPs (Supplemental Experimental Procedures and Fig-
ure S24). In addition, the in situ electrochemical attenuated total reflection infrared
spectroscopy (ATR-IR) result with adsorbed monolayer CO as a surface probe further
demonstrates the single Ni atomic sites on the surface of the NiN-GS catalyst
6
Chem 3, 1–11, December 14, 2017

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
Figure 4. Simulations of CO
2
-to-CO Reduction on Different Atomic Sites
A) Schematic of the reaction steps of electrocatalytic CO to CO reduction.
B) Different atomic configurations in graphene for the DFT calculations.
C) The free energy diagram of CO to CO conversion on different atomic sites under an equilibrium potential of –0.12 V versus RHE.
(
(
(
2
2
2
4,47
(
Figure S25).
Therefore, we believe that the Ni sites in GS should be the
active centers for electrocatalytic CO
2
to CO conversion, which exhibits a high TOF
À1
of ꢀ8 s and a cumulative turnover number (TON) of up to 454,000 under –0.7 V over-
potential during 20 hr of continuous electrolysis (Figure S26 and Supplemental Exper-
imental Procedures).
DISCUSSION
To further investigate the origin of the activity of the NiN-GS catalyst, we studied the
reaction steps involved in the CO
calculations (Figure 4A). We considered Ni atoms trapped in single or double vacancies
Ni@SV and Ni@DV, respectively) of the graphene sheet, with a variety of possible N
2
to CO electrocatalytic reduction by using DFT
(
coordination to Ni sites as well as N dopants (N@SV) (Figure S27 and Supplemental
Experimental Procedures), among which the most stable and active structures are
shown in Figure 4B (Table S3). A simple thermochemical analysis shows that the
pathway for CO
2
8–50
reduction to CO goes through *COOH and *CO intermediates
The solvation correction is included in Figure S28 (Supplemental
4
(
Figures 4A).
Experimental Procedures). The free energy diagram for each reaction step on different
atomic structures, including the Ni (111) surface, was calculated at the equilibrium
potential (À0.12 V versus RHE) in Figure 4C, where CO
2
activation (*COOH formation)
or CO desorption are suggested to be the rate-limiting steps on different sites. As an
example, on Ni (111) metal surface, even though the *COOH formation is facile, the
binding with CO is too strong as indicated by the deep thermodynamic sink
Chem 3, 1–11, December 14, 2017
7

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
(
À1.14 eV) in Figure 4C, which dramatically limits the CO desorption process and thus
the overall catalytic performance. This is consistent with the observation that Ni metal
18
2 2
evolves primarily H under CO reduction conditions. However, when Ni atoms are
coordinated in graphene layers, the electronic structure is drastically changed, exhibit-
ing a much higher projected density of states around Fermi energy than that of Ni (111)
(
Figures S6 and S29). As a result, the binding with CO is significantly weakened for a
facile CO desorption (Figure 4C and Table S4). In addition, compared with N dopants
in graphene, the Ni atomic sites such as Ni@SV and NiN@SV present decreased
21
*
COOH formation energies (Figure 4C and Table S4). Direct coordination of N to
Ni atoms in graphene layers helps to pull the electrons away from Ni, which as a result
makes the *COOH formation thermodynamically downhill on NiN@SV. However, it
binds CO slightly more strongly than Ni@SV (Table S4). Combined with APT character-
izations, this analysis therefore suggests that the Ni atoms coordinated in the graphene
vacancies, such as Ni@SV and Ni@DV, are the major active sites for electrocatalytic CO
2
reduction to CO, where N atoms play a critical role in generating defects in graphene to
trap Ni atoms. The minority atomic sites in this catalyst system, such as NiN@SV or
2
NiN@DV, could also be highly active for CO RR as suggested by the
DFT simulations. The calculated H-binding energy, a good descriptor for HER
5
1–53
activity,
Ni@DV, and NiN@SV sites (DG
with Ni (111) metal surface (DG
mental results (Table S5). The CO
evaluated by the difference between the thermodynamic limiting potentials for CO
reduction and HER, U (CO (H ), which confirms the high selectivity of both
suggests significantly suppressed hydrogen generation on Ni@SV,
H
= 0.50, 0.56, and 0.52 eV, respectively) in comparison
5
2,53
H
= À0.26 eV),
which is consistent with our experi-
2
RR selectivity of different active sites can be further
2
L
2
) À U
L
2
Ni@SV and Ni@DV (Figure S30). Because of the different electronic structures, Co
atomic sites present stronger CO binding and higher HER catalytic activity than Ni sites
(
Table S6), which explains its lower CO selectivity (Figure S22).
For a practical electrosynthetic cell of CO
2
reduction to fuels, an active and stable
oxygen evolution reaction (OER) catalyst needs to be paired to efficiently oxidize
1
0,54
water and free the protons.
2
Different from previous literature where CO RR and
OER catalysts were usually separated by a bipolar membrane and operated in different
5
5,56
+
pH solutions,
3 4
here we developed the Li -tuned Co O NPs as an active and stable
OER catalyst in neutral pH to be coupled with NiN-GS for a pH-equal (pH 7.5) electrol-
20
ysis system (Figures S31–S35 and Supplemental Experimental Procedures). Powered
by a single cell of commercialized GaInP /GaAs/Ge TJ photovoltaic, our artificial
2
photosynthesis system delivers a peak solar-to-CO efficiency of 12.7% (Figures S36
and S37). Incorporating TM atoms into a well-defined 2D matrix can effectively tune
their electronic structures and thus favor the desired reaction pathways. Given the
wide variety of 2D layered materials, which provide different types of chemical environ-
ments for tuning different TM atoms, this approach creates an important platform for
2
designing active CO reduction catalysts with high selectivity toward CO and higher-
value hydrocarbons, which is of great importance to both technological applications in
renewable energies and fundamental mechanism studies in catalysis.
EXPERIMENTAL PROCEDURES
The NiN-GS catalyst was first prepared by electrospinning a precursor solution
containing polyacrylonitrile (M
w w
= 150,000), polypyrrolidone (M = 1,300,000),
Ni(NO 6H O, dicyandiamide, and dimethylformamide, followed by a polymer
3
)
2
2
nanofiber oxidation and a carbonization procedure (Supplemental Experimental
Procedures). The as-synthesized NiN-GS catalyst was ball milled for 5 min to
nanopowder for catalysis and characterization.
8
Chem 3, 1–11, December 14, 2017

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
The STEM characterization in Figure 1C and other EDS analysis were carried out with
a JEOL ARM200F aberration-corrected scanning transmission electron microscope.
All other TEM images were obtained with a JEOL 2100 TEM. Specimens for 3D APT
were prepared with an FEI Helios 660 Nanolab Dual-Beam FIB/SEM. APT was then
performed with a Cameca LEAP 4000 HR operated in laser mode. Approximately
2
.3 million atoms were collected in the dataset presented. The data were
reconstructed and analyzed with Cameca’s IVAS software according to standard
reconstruction protocols.
In situ electrochemical ATR-IR measurements were run on a catalyst layer covered
hemicylindrical Si prism with a Nicolet IS50 FTIR spectrometer. Raman spectroscopy
was carried out on a WITEC CRM200 confocal Raman spectrometer with a 532 nm laser
source. XPS was obtained with a Thermo Scientific K-Alpha ESCA spectrometer, and
the surface componential content and peak fitting was analyzed with the Thermo
Avantage V5 program for selected elemental scans. The Ni K-edge XAS was acquired
in fluorescence mode with the SXRMB beamline of Canadian Light Source. Powder
XRD data were collected with a Bruker D2 Phaser diffractometer.
ꢁ
The electrochemical measurements were run at 25 C in a customized gas-tight H cell
(Figure S8) or flow cell (Figure S15) or a 3D-printed electrosynthetic cell (Figure S35)
separated by Nafion 117 membrane. A BioLogic VMP3 workstation was used to
record the electrochemical response. All measured potentials in this work were
converted to the RHE scale with manual iR compensation. During electrolysis, CO
2
gas
was delivered into the cathodic compartment containing CO -saturated KHCO electro-
2
3
lyte and vented into a Shimadzu GC-2014 gas chromatograph equipped with a thermal
conductivity detector and a flame ionization detector coupled with a methanizer.
All other experimental and setup details, as well the DFT calculations, are provided
in the Supplemental Experimental Procedures.
SUPPLEMENTAL INFORMATION
Supplemental Information includes Supplemental Experimental Procedures,
3
7 figures, 6 tables, and 2 movies and can be found with this article online at
https://doi.org/10.1016/j.chempr.2017.09.014.
AUTHOR CONTRIBUTIONS
H.W. conceived the project. H.W. and K.J. synthesized materials and measured
catalysts. H.W., K.J., A.J.A., Y.L., Z.L., J.L., Y.H., and Y.Z. characterized materials.
S.S., K.C., and J.K.N. performed simulations. All authors analyzed the results.
H.W., K.J., and S.S. wrote the manuscript. H.W. supervised the project.
ACKNOWLEDGMENTS
This work was supported by the Rowland Fellows Program at the Rowland Institute of
Harvard University. H.W. and K.J. acknowledge great support from the Friend group
at Harvard University. S.S., Y.L., Z.L., J.K.N., and Y.C. acknowledge support from
the Global Climate and Energy Project at Stanford University. K.C. and J.K.N.
acknowledge support from the Joint Center for Artificial Photosynthesis, a DOE
Energy Innovation Hub, supported through the Office of Science of the US Department
of Energy under award number DESC0004993. W.B.C. and Y.Z. acknowledge support
from the 973 Program of the Chinese Ministry of Science and Technology
(
2015CB932303) and Natural Science Foundation of China (21733004 and
1327901). This work was performed in part at the Center for Nanoscale Systems
2
Chem 3, 1–11, December 14, 2017
9

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
(
CNS), a member of the National Nanotechnology Infrastructure Network, which is sup-
ported by the National Science Foundation under award no. ECS-0335765. The CNS is
part of Harvard University. We acknowledge great support from Spectrolab for the tri-
ple-junction solar cell used in this work. We acknowledge the Nocera group at Harvard
University for helping with the solar simulator. We acknowledge the helpful discussion
on flow cells with Prof. Kenis and Mr. Sumit at the University of Illinois at Urbana-
Champaign.
Received: August 1, 2017
Revised: September 15, 2017
Accepted: September 26, 2017
Published: October 19, 2017
REFERENCES AND NOTES
1
. Cook, T.R., Dogutan, D.K., Reece, S.Y.,
Surendranath, Y., Teets, T.S., and Nocera, D.G.
Masel, R.I. (2011). Ionic liquid–mediated
selective conversion of CO2 to CO at low
overpotentials. Science 334, 643–644.
21. Shi, C., Hansen, H.A., Lausche, A.C., and
Norskov, J.K. (2014). Trends in electrochemical
CO2 reduction activity for open and close-
packed metal surfaces. Phys. Chem. Chem.
Phys. 16, 4720–4727.
(
2010). Solar energy supply and storage for the
legacy and nonlegacy worlds. Chem. Rev. 110,
474–6502.
1
1
2. Costentin, C., Robert, M., and Sav e´ ant, J.-M.
6
(2013). Catalysis of the electrochemical
2
. Meyer, T.J. (1989). Chemical approaches to
reduction of carbon dioxide. Chem. Soc. Rev.
42, 2423–2436.
22. Chen, Y., and Kanan, M.W. (2012). Tin oxide
dependence of the CO2 reduction efficiency
on tin electrodes and enhanced activity for tin/
tin oxide thin-film catalysts. J. Am. Chem. Soc.
134, 1986–1989.
artificial photosynthesis. Acc. Chem. Res. 22,
163–170.
3. Bligaard, T., Nørskov, J.K., Dahl, S.,
Matthiesen, J., Christensen, C.H., and
Sehested, J. (2004). The Brønsted–Evans–
Polanyi relation and the volcano curve in
heterogeneous catalysis. J. Catal. 224,
3
. Sakimoto, K.K., Wong, A.B., and Yang, P.
2016). Self-photosensitization of
(
nonphotosynthetic bacteria for solar-to-
chemical production. Science 351, 74–77.
23. Kim, D., Resasco, J., Yu, Y., Asiri, A.M., and
Yang, P. (2014). Synergistic geometric and
electronic effects for electrochemical reduction
of carbon dioxide using gold–copper
bimetallic nanoparticles. Nat. Commun. 5,
4948.
206–217.
4
. Liu, C., Col o´ n, B.C., Ziesack, M., Silver, P.A.,
and Nocera, D.G. (2016). Water splitting–
biosynthetic system with CO2 reduction
efficiencies exceeding photosynthesis. Science
14. Zhu, W., Michalsky, R., Metin, O
¨
., Lv, H., Guo,
S., Wright, C.J., Sun, X., Peterson, A.A., and
Sun, S. (2013). Monodisperse Au nanoparticles
for selective electrocatalytic reduction of CO2
to CO. J. Am. Chem. Soc. 135, 16833–16836.
352, 1210–1213.
24. Jiang, K., Wang, H., Cai, W.-B., and Wang, H.
(2017). Li electrochemical tuning of metal oxide
for highly selective CO2 reduction. ACS Nano
11, 6451–6458.
5
. Lewis, N.S., and Nocera, D.G. (2006). Powering
the planet: chemical challenges in solar energy
utilization. Proc. Natl. Acad. Sci. USA 103,
15. Liu, M., Pang, Y., Zhang, B., De Luna, P., Voznyy,
O., Xu, J., Zheng, X., Dinh, C.T., Fan, F., Cao, C.,
et al. (2016). Enhanced electrocatalytic CO2
reduction via field-induced reagent
15729–15735.
25. Nishihara, H., Hirota, T., Matsuura, K., Ohwada,
M., Hoshino, N., Akutagawa, T., Higuchi, T.,
Jinnai, H., Koseki, Y., Kasai, H., et al. (2017).
Synthesis of ordered carbonaceous
frameworks from organic crystals. Nat.
Commun. 8, 109.
6
. Jhong, H.-R.M., Ma, S., and Kenis, P.J.A. (2013).
Electrochemical conversion of CO2 to useful
chemicals: current status, remaining
challenges, and future opportunities. Curr.
Opin. Chem. Eng. 2, 191–199.
concentration. Nature 537, 382–386.
16. Chen, Y., Li, C.W., and Kanan, M.W. (2012).
Aqueous CO2 reduction at very Low
overpotential on oxide-derived Au
nanoparticles. J. Am. Chem. Soc. 134, 19969–
19972.
26. Manthiram, K., Beberwyck, B.J., and Alivisatos,
A.P. (2014). Enhanced electrochemical
methanation of carbon dioxide with a
dispersible nanoscale copper catalyst. J. Am.
Chem. Soc. 136, 13319–13325.
7
. Li, C.W., Ciston, J., and Kanan, M.W. (2014).
Electroreduction of carbon monoxide to liquid
fuel on oxide-derived nanocrystalline copper.
Nature 508, 504–507.
17. Kuhl, K.P., Hatsukade, T., Cave, E.R., Abram,
D.N., Kibsgaard, J., and Jaramillo, T.F. (2014).
Electrocatalytic conversion of carbon dioxide
to methane and methanol on transition metal
surfaces. J. Am. Chem. Soc. 136, 14107–14113.
8
. Gao, S., Lin, Y., Jiao, X., Sun, Y., Luo, Q., Zhang,
W., Li, D., Yang, J., and Xie, Y. (2016). Partially
oxidized atomic cobalt layers for carbon
27. Lei, F., Liu, W., Sun, Y., Xu, J., Liu, K., Liang, L.,
Yao, T., Pan, B., Wei, S., and Xie, Y. (2016).
Metallic tin quantum sheets confined in
graphene toward high-efficiency carbon
dioxide electroreduction. Nat. Commun. 7,
dioxide electroreduction to liquid fuel. Nature
18. Hori, Y., Wakebe, H., Tsukamoto, T., and Koga,
O. (1994). Electrocatalytic process of CO
selectivity in electrochemical reduction of CO2
at metal electrodes in aqueous media.
Electrochim. Acta 39, 1833–1839.
529, 68–71.
12697.
9
. Lin, S., Diercks, C.S., Zhang, Y.-B., Kornienko,
N., Nichols, E.M., Zhao, Y., Paris, A.R., Kim, D.,
Yang, P., Yaghi, O.M., et al. (2015). Covalent
organic frameworks comprising cobalt
porphyrins for catalytic CO2 reduction in water.
Science 349, 1208–1213.
28. Yang, M., Li, S., Wang, Y., Herron, J.A., Xu, Y.,
Allard, L.F., Lee, S., Huang, J., Mavrikakis, M.,
and Flytzani-Stephanopoulos, M. (2014).
Catalytically active Au-O(OH)x- species
19. Gong, M., Zhou, W., Tsai, M.-C., Zhou, J.,
Guan, M., Lin, M.-C., Zhang, B., Hu, Y., Wang,
D.-Y., Yang, J., et al. (2014). Nanoscale nickel
oxide/nickel heterostructures for active
hydrogen evolution electrocatalysis. Nat.
Commun. 5, 4695.
stabilized by alkali ions on zeolites and
mesoporous oxides. Science 346, 1498–1501.
1
0. Asadi, M., Kim, K., Liu, C., Addepalli, A.V.,
Abbasi, P., Yasaei, P., Phillips, P., Behranginia,
A., Cerrato, J.M., Haasch, R., et al. (2016).
Nanostructured transition metal
29. Manthiram, K., Surendranath, Y., and
Alivisatos, A.P. (2014). Dendritic assembly of
gold nanoparticles during fuel-forming
electrocatalysis. J. Am. Chem. Soc. 136, 7237–
7240.
2
0. Wang, H., Lee, H.-W., Deng, Y., Lu, Z., Hsu,
P.-C., Liu, Y., Lin, D., and Cui, Y. (2015).
Bifunctional non-noble metal oxide
dichalcogenide electrocatalysts for CO2
reduction in ionic liquid. Science 353, 467–470.
nanoparticle electrocatalysts through lithium-
induced conversion for overall water splitting.
Nat. Commun. 6, 7261.
1
1. Rosen, B.A., Salehi-Khojin, A., Thorson, M.R.,
Zhu, W., Whipple, D.T., Kenis, P.J.A., and
30. Novoselov, K.S., Geim, A.K., Morozov, S.V.,
Jiang, D., Zhang, Y., Dubonos, S.V., Grigorieva,
10
Chem 3, 1–11, December 14, 2017

Please cite this article in press as: Jiang et al., Transition-Metal Single Atoms in a Graphene Shell as Active Centers for Highly Efficient Artificial
Photosynthesis, Chem (2017), https://doi.org/10.1016/j.chempr.2017.09.014
I.V., and Firsov, A.A. (2004). Electric field effect
carbon dioxide reduction. Nat. Commun. 4,
enzymes for CO2 reduction to CO. J. Phys.
in atomically thin carbon films. Science 306,
2819.
Chem. Lett. 4, 388–392.
6
66–669.
4
0. Sharma, P.P., Wu, J., Yadav, R.M., Liu, M.,
Wright, C.J., Tiwary, C.S., Yakobson, B.I., Lou,
J., Ajayan, P.M., and Zhou, X.-D. (2015).
Nitrogen-doped carbon nanotube arrays for
high-efficiency electrochemical reduction of
CO2: on the understanding of defects, defect
density, and selectivity. Angew. Chem. Int. Ed.
54, 13701–13705.
49. Hansen, H.A., Shi, C., Lausche, A.C., Peterson,
A.A., and Norskov, J.K. (2016). Bifunctional
alloys for the electroreduction of CO2 and CO.
Phys. Chem. Chem. Phys. 18, 9194–9201.
3
1. Rodr ı´ guez-Manzo, J.A., Cretu, O., and Banhart,
F. (2010). Trapping of metal atoms in vacancies
of carbon nanotubes and graphene. ACS Nano
4
, 3422–3428.
50. Peterson, A.A., and Nørskov, J.K. (2012).
Activity descriptors for CO2 electroreduction
to methane on transition-metal catalysts.
J. Phys. Chem. Lett. 3, 251–258.
3
2. Giovanni, M., Poh, H.L., Ambrosi, A., Zhao, G.,
Sofer, Z., Sanek, F., Khezri, B., Webster, R.D.,
and Pumera, M. (2012). Noble metal (Pd, Ru,
Rh, Pt, Au, Ag) doped graphene hybrids for
electrocatalysis. Nanoscale 4, 5002–5008.
41. Wu, J., Yadav, R.M., Liu, M., Sharma, P.P.,
Tiwary, C.S., Ma, L., Zou, X., Zhou, X.-D.,
Yakobson, B.I., Lou, J., et al. (2015). Achieving
highly efficient, selective, and stable CO2
reduction on nitrogen-doped carbon
51. Hinnemann, B., Moses, P.G., Bonde, J.,
Jørgensen, K.P., Nielsen, J.H., Horch, S.,
Chorkendorff, I., and Nørskov, J.K. (2005).
Biomimetic hydrogen evolution: MoS2
nanoparticles as catalyst for hydrogen
evolution. J. Am. Chem. Soc. 127, 5308–
5309.
3
3
3
3
3. Krasheninnikov, A.V., Lehtinen, P.O., Foster,
A.S., Pyykk o¨ , P., and Nieminen, R.M. (2009).
Embedding transition-metal atoms in
graphene: structure, bonding, and magnetism.
Phys. Rev. Lett. 102, 126807.
nanotubes. ACS Nano 9, 5364–5371.
42. Kornienko, N., Zhao, Y., Kley, C.S., Zhu, C., Kim,
D., Lin, S., Chang, C.J., Yaghi, O.M., and Yang,
P. (2015). Metal–organic frameworks for
4. Li, Y., Zhou, W., Wang, H., Xie, L., Liang, Y., Wei,
F., Idrobo, J.-C., Pennycook, S.J., and Dai, H.
52. Greeley, J., Jaramillo, T.F., Bonde, J.,
Chorkendorff, I., and Nørskov, J.K. (2006).
Computational high-throughput screening of
electrocatalytic materials for hydrogen
evolution. Nat. Mater. 5, 909–913.
electrocatalytic reduction of carbon dioxide.
J. Am. Chem. Soc. 137, 14129–14135.
(2012). An oxygen reduction electrocatalyst
based on carbon nanotube-graphene
complexes. Nat. Nanotech. 7, 394–400.
43. Jhong, H.-R.M., Brushett, F.R., and Kenis, P.J.A.
(
2013). The effects of catalyst layer deposition
5. Su, P., Iwase, K., Nakanishi, S., Hashimoto, K.,
and Kamiya, K. (2016). Nickel-nitrogen-
modified graphene: an efficient electrocatalyst
for the reduction of carbon dioxide to carbon
monoxide. Small 12, 6083–6089.
methodology on electrode performance. Adv.
Energy Mater. 3, 589–599.
53. Greeley, J., Nørskov, J.K., Kibler, L.A., El-Aziz,
A.M., and Kolb, D.M. (2006). Hydrogen
evolution over bimetallic systems:
understanding the trends. ChemPhysChem 7,
1032–1035.
44. Zhao, Y., Nakamura, R., Kamiya, K., Nakanishi,
S., and Hashimoto, K. (2013). Nitrogen-doped
carbon nanomaterials as non-metal
electrocatalysts for water oxidation. Nat.
Commun. 4, 2390.
6. Varela, A.S., Ranjbar Sahraie, N., Steinberg, J.,
Ju, W., Oh, H.-S., and Strasser, P. (2015). Metal-
doped nitrogenated carbon as an efficient
catalyst for direct CO2 electroreduction to CO
and hydrocarbons. Angew. Chem. Int. Ed. 54,
54. Schreier, M., Curvat, L., Giordano, F., Steier, L.,
Abate, A., Zakeeruddin, S.M., Luo, J., Mayer,
M.T., and Gr a¨ tzel, M. (2015). Efficient
photosynthesis of carbon monoxide from CO2
using perovskite photovoltaics. Nat. Commun.
6, 7326.
45. Tedsree, K., Li, T., Jones, S., Chan, C.W.A., Yu,
K.M.K., Bagot, P.A.J., Marquis, E.A., Smith,
G.D.W., and Tsang, S.C.E. (2011). Hydrogen
production from formic acid decomposition at
room temperature using a Ag-Pd core-shell
nanocatalyst. Nat. Nanotech. 6, 302–307.
10758–10762.
3
7. Chen, Z., Ren, W., Gao, L., Liu, B., Pei, S., and
Cheng, H.-M. (2011). Three-dimensional
flexible and conductive interconnected
55. Schreier, M., H e´ roguel, F., Steier, L., Ahmad, S.,
Luterbacher, J.S., Mayer, M.T., Luo, J., and
Gr a¨ tzel, M. (2017). Solar conversion of CO2 to
CO using earth-abundant electrocatalysts
prepared by atomic layer modification of CuO.
Nat. Energy 2, 17087.
graphene networks grown by chemical vapour 46. Kelly, T.F., and Miller, M.K. (2007). Atom probe
deposition. Nat. Mater. 10, 424–428.
tomography. Rev. Sci. Instrum. 78, 031101.
3
8. Deng, D., Novoselov, K.S., Fu, Q., Zheng, N.,
Tian, Z., and Bao, X. (2016). Catalysis with two-
dimensional materials and their
47. Huo, S.-J., Wang, J.-Y., Yao, J.-L., and Cai,
W.-B. (2010). Exploring electrosorption at iron
electrode with in situ surface-enhanced
56. Zhou, X., Liu, R., Sun, K., Chen, Y., Verlage, E.,
Francis, S.A., Lewis, N.S., and Xiang, C. (2016).
Solar-driven reduction of 1 atm of CO2 to
formate at 10% energy-conversion efficiency by
use of a TiO2-protected III–V tandem
photoanode in conjunction with a bipolar
membrane and a Pd/C Cathode. ACS Energy
Lett. 1, 764–770.
heterostructures. Nat. Nanotech. 11, 218–230.
infrared absorption spectroscopy. Anal. Chem.
82, 5117–5124.
3
9. Kumar, B., Asadi, M., Pisasale, D., Sinha-Ray, S.,
Rosen, B.A., Haasch, R., Abiade, J., Yarin, A.L.,
and Salehi-Khojin, A. (2013). Renewable and
metal-free carbon nanofibre catalysts for
48. Hansen, H.A., Varley, J.B., Peterson, A.A., and
Nørskov, J.K. (2013). Understanding trends in
the electrocatalytic activity of metals and
Chem 3, 1–11, December 14, 2017
11