Identification of Single-Atom Ni Site Active toward Electrochemical CO2Conversion to CO

Article abstract of DOI:10.1021/jacs.0c11008

Electrocatalytic conversion of CO2 into value-added products offers a new paradigm for a sustainable carbon economy. For active CO2 electrolysis, the single-atom Ni catalyst has been proposed as promising from experiments, but an idealized Ni-N4 site shows an unfavorable energetics from theory, leading to many debates on the chemical nature responsible for high activity. To resolve this conundrum, here we investigated CO2 electrolysis of Ni sites with well-defined coordination, tetraphenylporphyrin (N4-TPP) and 21-oxatetraphenylporphyrin (N3O-TPP). Advanced spectroscopic and computational studies revealed that the broken ligand-field symmetry is the key for active CO2 electrolysis, which subordinates an increase in the Ni redox potential yielding NiI. Along with their importance in activity, ligand-field symmetry and strength are directly related to the stability of the Ni center. This suggests the next quest for an activity-stability map in the domain of ligand-field strength, toward a rational ligand-field engineering of single-atom Ni catalysts for efficient CO2 electrolysis.

Full text of DOI:10.1021/jacs.0c11008

pubs.acs.org/JACS
Article
Identification of Single-Atom Ni Site Active toward Electrochemical
CO2 Conversion to CO
Haesol Kim,§ Dongyup Shin,§ Woojin Yang,§ Da Hye Won,§ Hyung-Suk Oh, Min Wook Chung,
Donghyuk Jeong, Sun Hee Kim, Keun Hwa Chae, Ji Yeon Ryu, Junseong Lee, Sung June Cho,
Jiwon Seo,* Hyungjun Kim,* and Chang Hyuck Choi*
Cite This: J. Am. Chem. Soc. 2021, 143, 925−933
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Electrocatalytic conversion of CO2 into value-added products offers a
new paradigm for a sustainable carbon economy. For active CO2 electrolysis, the
single-atom Ni catalyst has been proposed as promising from experiments, but an
idealized Ni−N4 site shows an unfavorable energetics from theory, leading to many
debates on the chemical nature responsible for high activity. To resolve this
conundrum, here we investigated CO2 electrolysis of Ni sites with well-defined
coordination, tetraphenylporphyrin (N4−TPP) and 21-oxatetraphenylporphyrin
(N3O−TPP). Advanced spectroscopic and computational studies revealed that the
broken ligand-field symmetry is the key for active CO2 electrolysis, which subordinates
an increase in the Ni redox potential yielding NiI. Along with their importance in
activity, ligand-field symmetry and strength are directly related to the stability of the Ni center. This suggests the next quest for an
activity−stability map in the domain of ligand-field strength, toward a rational ligand-field engineering of single-atom Ni catalysts for
efficient CO2 electrolysis.
electrode).27 In a gas diffusion electrode system, a similar
INTRODUCTION
■
Ni−N−C catalyst achieved a high CO partial current density
Electrochemical carbon dioxide reduction reaction (CO2RR) is
of >200 mA cm−2 as well as a stable CO Faradaic efficiency
a promising strategy for both balancing the global carbon cycle
(FE) of ca. 85%, verifying its immense potential for future
and production of fine chemicals.1,2 Among the various
industrial applications.21
CO2RR products, carbon monoxide is one of the simplest
Notwithstanding the high efficiency of Ni−N−C catalysts,
and most economically valuable substances by itself and serves
the molecular details of the catalytically active Ni moiety are
as a feedstock for bulk chemical manufacturing.3,4 To achieve
still puzzling. Density functional theory (DFT) calculations, as
selective CO2-to-CO conversion, noble metals such as gold5−8
a reliable supplement to experiments, have revealed a relatively
and silver9−13 have been widely investigated as irreplaceable
weak binding affinity of *COOH to the porphyrin-like Ni−N4
electrocatalysts due to their apposite binding affinities with the
site, predicting a substantially large overpotential (>1.5
CO2RR intermediate (*COOH) and the product (*CO) in
V).17,21,27 Considering the principles of organometallics, the
the scaling relationship.4,14,15 Nevertheless, noble metal
poor ability of NiII (i.e., d8 species) to form an axial bond is
catalysts have limitations such as high cost, vulnerability to
predictable, as it strongly prefers the D4h square-planar
impurities, and inadequate long-term durability.7,9,16 Hence,
geometry.32 Accordingly, different local structures around the
the development of new types of CO2RR catalysts is in great
Ni center with broken D4h symmetryfor instance, unsatu-
demand.
rated or heteroatom-coordinated Ni−N sites (i.e., Ni−NyX4−y
,
As an outlier that breaks the conventional properties of
metallic materials, single-atom metal catalysts, in which
atomically dispersed metal ions are ligated to nonmetallic
functional groups (e.g., Me−N−C catalysts, Me = Ni, Fe, Co,
Zn, Mn, etc.), have recently attracted considerable attention
for the selective CO production from CO2RR.17−31 In
particular, single-atom Ni catalysts have exhibited promising
CO2-to-CO conversion activities, outperforming other Me−
N−C catalysts and even comparable with noble metal
catalysts.19−31 For example, the Ni−N−C catalyst synthesized
by a topochemical transformation recorded near-unity CO
selectivity of 99% at −0.81 VRHE (reversible hydrogen
X = vacancy or heteroatom besides N such as S and C)have
been suggested as possible catalytic sites preferable for CO
production from CO2RR.22−24,33 Another issue of interest is
the oxidation state of Ni. Recently, Yang et al. reported the
Received: October 19, 2020
Published: January 7, 2021
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933
© 2021 American Chemical Society
925

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
formation of an uncommon oxidation state of NiI on an N(/
S)-doped graphene matrix, which was ascribed to the superior
CO2RR activity with 97% FE.31
species was NiII for both Ni-porphyrin compounds, as revealed
by the peaks at ca. 856.0 eV in X-ray photoelectron
spectroscopy (XPS)-Ni2p (Figure S15) and also by the Ni K-
edge position of 8345 eV in the X-ray absorption near edge
structure (XANES) spectra (Figure 1c and Figure S16).
The structure of Ni(−Cl)−N3O−TPP was further eluci-
dated by single-crystal X-ray diffraction (SC-XRD; Figure 1b,
Supporting Information Note 2, and Tables S1−S3). The
result shows a five-coordinate Ni complex with a vertical
distance of 0.303 Å between the Ni atom and O(1)N(2)-
N(3)N(4) plane and the axial chloride ion at a distance of
2.279 Å apart. The η1-furan ligand is coordinated to the Ni
atom with θ = 18.2° (θ is the angle between the furan plane
and Ni−O bond). This structural distortion, broken D4h
symmetry, is further verified by the silent 1s → 4pz transition
at ca. 8339 eV in the XANES spectrum of Ni(−Cl)−N3O−
TPP (Figure 1c), a fingerprint of square-planar Me−N4
moieties which can be clearly seen in that of Ni−N4−TPP.36
On the other hand, a pre-edge peak is observed at ca. 8335 eV
for Ni(−Cl)−N3O−TPP due to the increased dipole-allowed
1s → 4p transition, which is evidence of the distorted D4h
symmetry inducing hybridization of the 3d and 4p orbitals.37
Thus, NiII complexes with a four-coordinate dianionic ligand
(square-planar Ni−N4−TPP)38 and a monoanionic ligand with
an axial chloride ion (distorted Ni(−Cl)−N3O−TPP)34 were
identified.
For the rational design of Ni−N−C catalysts, we herein
investigate the CO2RR electrochemistry of Ni-porphyrin
model compounds using advanced in operando spectroscopies
and DFT calculations, with posing a fundamental question
which symmetry of the ligand field and which oxidation state
are responsible for CO2RR at the Ni center. The pyrolysis step
at >800 °C required for the usual synthesis of Ni−N−C
catalysts produces chemically inequivalent Ni species (e.g.,
metallic Ni, oxide, and various Ni−NyX4−y sites),21,22 which
renders the identification of the active Ni site highly vague.
Therefore, we have employed molecular catalysts based on Ni-
complexed porphyrins, tetraphenylporphyrin (N4−TPP) and
21-oxatetraphenylporphyrin (N3O−TPP). Their immobiliza-
tion on a carbon substrate provided a heterogeneous catalytic
platform with well-defined and equivalent organometallic
active species, enabling us to unravel the structure−activity
(and structure−stability) relationship in single-atom Ni
electrocatalysis in CO2RR.
RESULTS AND DISCUSSION
■
Synthesis of the Ni-Porphyrin Compounds. The meso-
substituted porphyrins, i.e., N4−TPP and N3O−TPP, were
synthesized by modified Lindsey’s methods (Figure 1a and
In Situ Characterizations of Porphyrin Compounds.
Electrocatalysis of CO2RR on Ni−N4−TPP and Ni(−Cl)−
N3O−TPP was investigated in a 0.5 M KHCO3 electrolyte
using a rotating disk electrode (RDE). Electrodes were
prepared by mixing the porphyrins with carbon black to
provide sufficient electrical conductivity. Linear sweep
voltammetry (LSV) measurements at a scan rate of 50 mV
s−1 reveal similar polarizations of Ni−N4−TPP in Ar- and
CO2-saturated electrolytes (Figure 2a; pH = ca. 8.9 and 7.2,
respectively). On the other hand, Ni(−Cl)−N3O−TPP shows
a substantial increase in the reductive current density in the
CO2-saturated electrolyte, which is two times higher at −0.8
VRHE than that in the Ar-saturated electrolyte. The reductive
currents are attributed to electrocatalysis on the Ni-porphyrin
compounds, as confirmed by the much lower electrocatalytic
activity on glassy carbon and carbon black without the model
Ni-porphyrin compounds (Figure S17). Hence, the difference
in the polarization curves, clearly observed on the Ni(−Cl)−
N3O−TPP electrode but not on the Ni−N4−TPP electrode,
implies better electrocatalytic CO2RR activity of Ni(−Cl)−
N3O−TPP than that of Ni−N4−TPP.
To confirm the above hypothesis, electrocatalytic CO2-to-
CO conversion and competitive H2 evolution, i.e., the two
main reactions on the single-atom Ni catalysts under similar
CO2RR conditions,19−31 were monitored online by differential
electrochemical mass spectroscopy (DEMS) coupled to a
scanning flow cell (SFC) system (Figure S18). Under
potentiodynamic polarizations at a scan rate of 5 mV s−1, the
CO2 level (m/z = 44) in the electrolyte is stable in the high-
potential region but starts to decrease at a potential below ca.
−0.8 and −0.6 VRHE for Ni−N4−TPP and Ni(−Cl)−N3O−
TPP, respectively (Figure 2b and c). Simultaneously, the CO
signal (m/z = 28) is newly evolved and intensified as the
overpotential increases. These changes are hardly detectable in
the Ar-saturated electrolyte, in which the gaseous product is
only H2 (Figure S19). Identical onset potentials for CO2
consumption and CO evolution indicate that CO is produced
Figure 1. Synthesis and physical characterizations of the prepared
porphyrins. (a) Synthetic scheme for Ni−N4−TPP and Ni(−Cl)−
N3O−TPP by modified Lindsey’s methods. (b) 3D molecular
structure of Ni(−Cl)−N3O−TPP determined by SC-XRD. (c) Ni
K-edge XANES spectra of Ni−N4−TPP, Ni(−Cl)−N3O−TPP, and
Ni foil.
Supporting Information Note 1), where 2,5-bis-
(phenylhydroxymethyl)furan was used as a precursor for the
preparation of N3O−TPP (Figure S1).34 After metalation,34,35
Ni−N4−TPP and Ni(−Cl)−N3O−TPP were recrystallized
and used for subsequent analyses. Successful synthesis of the
complexes was confirmed by matrix-assisted laser desorption
ionization time-of-flight mass (MALDI-ToF MS), liquid
chromatography (LC), ultraviolet−visible (UV−vis) absorp-
tion, and 1H-/13C-nuclear magnetic resonance (NMR) spectra
(Figures S2−S14). The oxidation state of the central Ni
926
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
Figure 2. Electrochemical and in situ spectroscopy studies. (a) Polarization curves of Ni−N4−TPP and Ni(−Cl)−N3O−TPP measured at a scan
rate of 50 mV s−1 in Ar-/CO2-saturated 0.5 M KHCO3 electrolytes. (b, c) Online SFC-DEMS results of Ni−N4−TPP (b) and Ni(−Cl)−N3O−
TPP (c). The CO2 (m/z = 44), CO (m/z = 28), and H2 (m/z = 2) signals were collected during two cycles of CV from 0 to −0.9 VRHE in a CO2-
saturated 0.5 M KHCO3 electrolyte. The potential profile during the DEMS measurements is indicated by a dotted line. (d) CV results of Ni−N4−
TPP and Ni(−Cl)−N3O−TPP measured at a scan rate of 50 mV s−1 in a potential range from −0.5 to 0.4 VRHE in a CO2-saturated electrolyte.
Redox couple A is indicated by an arrow. (e) In situ XANES spectra of Ni−N4−TPP and Ni(−Cl)−N3O−TPP measured at the OCP and −0.65
VRHE. The differences between the spectra obtained at the OCP and −0.65 VRHE are also shown (dotted line). (f) Linear combination fitting of the
Ni K-edge XANES spectrum of Ni(−Cl)−N3O−TPP collected at −0.65 VRHE
.
by the electrochemical reduction of CO2 dissolved in the
electrolyte. Along with the higher onset potential, more
substantial ionic current changes in CO2 consumption and CO
evolution are recorded for Ni(−Cl)−N3O−TPP than for Ni−
N4−TPP. Therefore, these results reveal a better CO2-to-CO
conversion on Ni(−Cl)−N3O−TPP than on Ni−N4−TPP.
To clarify the origin of the diametrically opposite CO2-to-
CO conversion activities between the Ni-porphyrins, variations
in the Ni oxidation state, which has been debated as a key
parameter determining CO2RR performance,21−23,30,31 were
investigated under the polarization conditions. Cyclic
voltammetry (CV) measurement for the Ni−N4−TPP shows
a double-layer capacitive characteristic without noticeable
redox signals of the central Ni ion in the CO2-saturated
electrolyte (Figure 2d), indicating stabilization of the initial
oxidation state of NiII in a potential range from 0.4 to −0.5
VRHE (even to −0.8 VRHE, Figure S20). In the CV result of
Ni(−Cl)−N3O−TPP, however, a clear redox couple is
observed at ca. −0.22 VRHE (peak A), a much lower value of
the open circuit potential (OCP) of ca. 0.98 VRHE, at which the
initial oxidation state of NiII is thermodynamically stable.
Because Ni(−Cl)−N3O−TPP is adsorbed on the electrode,
i.e., a surface-confined redox reaction, the peak current is
linearly proportional to the scan rate (Figure S21).39,40
Considering the amount of Ni(−Cl)−N3O−TPP on the
electrode and the charge passed during the reaction, the
number of electrons in redox couple A can be estimated to be
ca. 1.9. Similar results are obtained in the Ar-saturated
electrolyte (Figure S22), but the redox potential (E0) of
Ni(−Cl)−N3O−TPP is positively shifted by ca. 110 mV on
the RHE scale. However, on the standard hydrogen electrode
(SHE) scale without pH compensation, the redox couple A is
located at an identical E0 of −0.62 VSHE in both Ar- and CO2-
saturated electrolytes (Figure S23). Therefore, the voltammo-
gram results indicate that the redox reaction A accompanies
two-electron transfer without proton transfer.
The variation of the oxidation state of the NiII center was
further studied by in situ XANES analysis. The XANES spectra
were recorded in a CO2-saturated electrolyte at the OCP and
−0.65 VRHE (Figure 2e and Figure S24); the latter potential is
lower than the redox potential of Ni(−Cl)−N3O−TPP (i.e., −
0.22 VRHE), and CO2RR is maximized at this potential
(discussed later). For Ni−N4−TPP, the XANES spectrum
measured at −0.65 VRHE is almost identical to that measured at
the OCP. However, the Ni K-edge position of Ni(−Cl)−
N3O−TPP decreases by 2 eV once the electrode is polarized at
−0.65 VRHE. The absorption edge position at 8.343 eV
corresponds to the average Ni oxidation state of ca. +1.6
(Figure S16), indicating a part of NiII reduced to Ni0 and/or
NiI. Therefore, it can be presumed that the electronic state of
the Ni site may play a decisive role in the electrochemical CO2-
to-CO conversion, and either metallic or monovalent Ni sites
are likely to be responsible for efficient CO evolution rather
than NiII moieties.
However, the XANES spectrum of Ni(−Cl)−N3O−TPP at
−0.65 VRHE shows a poor fitting value (R-factor = 0.075) in
linear combination with the spectra of Ni0 (Ni foil) and NiII
(Ni(−Cl)−N3O−TPP at the OCP) components (Figure 2f).
The downshift of the edge in the region is assigned to Ni1.6+
(negative residual at edge) but the overestimation of the NiII
contribution (positive residual at the white line). The failure of
the fitting thus points out that the redox couple A of
Ni(−Cl)−N3O−TPP at −0.22 VRHE is not fully attributed to
the electron transfer process between NiII and Ni0.
Accordingly, one-electron reduction of NiII to monovalent
NiI could be reasonably considered, which accompanies
another one-electron reduction of the N3O−TPP ligand
because the redox couple A follows the two-electron process
927
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
as verified by CV (Figure 2d). It is worth noting that the
stabilization of monovalent NiI as well as NiII has been
successfully achieved using monoheteroatom porphyrin ligands
similar to N3O−TPP,34,41,42 while this character is rarely found
in typical N4-porphyrins due to the formation of porphyrin π-
anion radicals.43
Electronic State of Active Ni Species in Operando
and CO2RR Chemistry. Using DFT (Supporting Information
Note 4), which was confirmed to predict reliable E0 for various
Me−N4−TPP systems (Me = Fe, Co, Ni, Cu, and Zn) (Table
S6), the E0 of Ni-porphyrin compounds was theoretically
investigated. DFT predicts the one-electron reduction
potential of Ni−N4−TPP as −1.04 VSHE and the additional
electron to be added to the π*-orbital of N4−TPP (Figure
S25), which are in agreement with the previous experimen-
tal44,45 and theoretical studies.46 Thus, DFT results support
that the NiII center of Ni−N4−TPP is redox-innocent in the
potential range for CO2RR.
Figure 3. Suggested CO2RR mechanism. (a) Catalytic cycles of Ni−
N4−TPP (left) and Ni(−Cl)−N3O−TPP (right) and (b) corre-
sponding DFT-calculated free energy profiles when a potential of
−0.6 VRHE is applied. CO2 adsorption to the Ni site is an uphill
process at a long distance (Figure S32), and thus the CO2-adsorbed
species is considered to be a transient configuration (denoted using
parentheses).
The two-electron redox of Ni(−Cl)−N3O−TPP is consid-
ered to occur via the following reaction.
Ni(−Cl)−N3O−TPP + 2e−
⌉−
F Ni−N3O−TPP + Cl−(aq)
(1)
to NiI of Ni−N3O−TPP⌉− (Figure S33). For an active CO2RR,
it is thus concluded that breaking the symmetry of the in-plane
ligand field is pivotal. Additional CV experiments and DFT
calculations further supported the in situ generation of the Ni
active center with a weakened D4h ligand field during CO2RR
of another high-performing single-atom Ni catalyst of Ni-
complexed phthalocyanine (NiPc; Supporting Information
Note 6). Thus, our conclusion is not a specific nature of
Ni(−Cl)−N3O−TPP but is general for single-atom Ni
catalysts for CO2RR.
DFT predicts the E0 of eq 1 as −0.54 VSHE. This corresponds
to our experimental observation of the redox couple A located
near −0.62 VSHE. During two-electron reduction, DFT further
reveals that one electron reduces the NiII center to NiI as
supported by the spin density plot of the Ni−N3O−TPP⌉0
species (Figure S26a) and the +0.98 Mulliken spin population
at the Ni center. The other electron is found to occupy the π*-
orbital of N3O−TPP with marginally changing the Mulliken
spin population at the Ni center from +1 but significantly
increasing the spin density in the porphyrin π-system (Figure
S26b). For a detailed understanding about the electronic
structural changes during the reduction of NiII−N4−TPP and
NiII(−Cl)−N3O−TPP, readers are referred to the Supporting
Information Note 5 (and Figures S25−S30).
As shown in Figure 4a and b, NiI−N3O−TPP⌉− can form a
stronger Ni−C bond than NiII−N4−TPP, which is the origin
Consequently, NiII of Ni−N4−TPP and NiI of Ni−N3O−
TPP⌉− are the actual chemical species responsible for the
CO2RR. Thus, the CO2RR chemistry at these two different Ni
centers is of interest. In order to decouple the effect of change
in oxidation state and ligand field, a hypothetical system of
Ni−N3O−TPP⌉+, which can be formed by removing the axial
chloride from Ni(−Cl)−N3O−TPP without redox, is addi-
tionally examined using DFT. Unlike Ni(−Cl)−N3O−TPP,
the lack of fifth coordination of Ni−N3O−TPP⌉+ stabilizes the
low-spin NiII state (S = 0) with virtually no out-of-plane
distortion (Figure S31). Thus, the metal oxidation state and
coordination geometry of both Ni−N3O−TPP⌉+ and Ni−N4−
TPP are nearly the same, but the D4h ligand field is broken only
for Ni−N3O−TPP⌉+ due to the substitution of one N with O.
In Figure 3, the CO2RR free energies are shown for Ni−N4−
TPP (NiII with D4h ligand field) and Ni−N3O−TPP⌉− (NiI
with broken D4h ligand field) when a potential of −0.6 VRHE is
applied. At the *COOH step, which shows the most
unfavorable energetics, NiII of Ni−N4−TPP exhibits the low
stabilization ability of *COOH, predicting a large over-
potential. On the contrary, the active species of Ni(−Cl)−
N3O−TPP in operando, NiI of Ni−N3O−TPP⌉−, can
substantially stabilize *COOH, yielding a much reduced
overpotential. Interestingly, the hypothetical active site of
NiII of Ni−N3O−TPP⌉+ shows comparable CO2RR energetics
Figure 4. Origin of different activities. DFT-optimized structures of
COOH-bound (a) Ni−N4−TPP and (b) Ni−N3O−TPP⌉−. Binding
energy of ·COOH to the Ni center is also displayed, where the more
negative value means the stronger binding. (c) A schematic MO
diagram is shown. To form a strong Ni−COOH bond, the electron-
rich Ni center needs to redistribute its electron density to the N3X
ligand (orange arrow); the energy cost for such an electron
redistribution becomes large when the ligand-field splitting, Δ,
increases.
928
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
of the larger electrochemical reactivity of NiI−N3O−TPP⌉−.
To explain this, a schematic molecular orbital (MO) diagram is
shown in Figure 4c. As most of the d orbitals of the low-valent
Ni species, such as NiII and NiI, are filled, the Ni center has a
low Lewis acidity, prohibiting the formation of a strong Ni−C
bond with COOH. To form a strong Ni−C bond with COOH,
therefore, the Ni center needs to redistribute its electron
density to the ligand. When the strong ligand field produces a
large splitting of Δ, as in the case of the N4−TPP ligand, the
energy cost required for the charge redistribution becomes
significant (see Figure 4c), which substantially weakens the
Ni−C bond strength. Therefore, when the symmetry of the in-
plane ligand field is broken and/or weakened, which is what
happened to the N3O−TPP ligand, the Ni center can form a
stronger Ni−C bond to stabilize the *COOH, which is the key
to achieving a favorable energetics for CO2RR at the Ni center.
But, it also accompanies a substantial positive shift of the Ni
reduction potential as the empty d orbital of the Ni center is
being lowered, facilitating the generation of a +1 oxidation
state under polarization.
Electrochemical CO2-to-CO Conversion. Along with the
fundamental understandings based on the operando spectro-
scopic and computational studies, the gaseous products were
analyzed by online gas chromatography (GC) in a two-
compartment H-cell to determine the FE and partial current
density for CO2RR. Figure 5a shows the FEs of formation of
CO and H2 measured under potentiostatic polarization
conditions for 1 h CO2RR operation. In both Ni-porphyrin
electrodes, the sum of FECO and FEH2 reached almost 100%,
indicating that the formation of other products was almost
negligible. The 1H NMR spectrum also confirmed the absence
(or imperceptible amount) of liquid products (e.g., formate or
methanol) in the electrolyte after the reaction (Figure S34).
During electrolysis, Ni−N4−TPP exhibits predominant H2
formation over the whole potential range (Figure S35), but
minor CO formation is detected at a high overpotential region
below −0.75 VRHE with a maximum FECO of only ca. 2%
(Figure 5a and b). On the other hand, Ni(−Cl)−N3O−TPP
shows an onset potential of CO formation at approximately
−0.55 VRHE, and its maximum FECO reaches ca. 80% at −0.65
V
RHE. The FECO of Ni(−Cl)−N3O−TPP and the partial
current density of CO formation, however, are deteriorated in
the high-overpotential region, while H2 production is
significantly enhanced (Figure 5a). This is in contrary to the
result measured by online SFC-DEMS, which shows enhanced
CO evolution as the overpotential increases.
Additional CV analysis for Ni(−Cl)−N3O−TPP with a
different lower potential limit (LPL) provides a clue to the
understanding of the discrepancy in CO selectivity (Figure 5c
and Figure S36). When the LPL decreases below −0.7 VRHE
,
the redox couple A of Ni(−Cl)−N3O−TPP at ca. −0.22 VRHE
in the CO2-saturated electrolyte disappears. Meanwhile, a
broad redox couple newly grows up at ca. 0.07 VRHE (Figure
5c; peak B), which can be more clearly differentiated in the Ar-
saturated electrolyte into two different redox couples at ca.
0.13 and 0.07 VRHE (peaks B1 and B2, respectively; Figure S36
and Supporting Information Note 3). The peak B at ca. 0.07
VRHE in the CO2-saturated electrolyte (or peak B1 at ca. 0.13
VRHE in the Ar-saturated electrolyte) is responsible for the
redox couple of Ni-free N3O−TPP (Figure S37), indicating
the instability (i.e., demetalation) of the Ni-coordination
structure at a potential of <−0.6 VRHE. Considering the
Pourbaix diagram,47 the precipitation of dissolved Ni ions to
metallic Ni or oxidized NiO and Ni(OH)2 is an exergonic
process in our pH conditions (i.e., pH = 7.2−8.9), and their
standard E0 are ca. 0.13 (E0Ni/NiO) and 0.11 (E0
)
Ni/Ni(OH)2
V
RHE, respectively. More evidently, the formation of bulk NiO
phase was clearly confirmed by the Raman spectrum of the
Ni(−Cl)−N3O−TPP electrode after CVs from −0.8 to 0.4
V
RHE (Figure 5d).48 It is of note that these bulk Ni phases are
highly active toward H2 evolution reaction.49,50 Although the
working electrode was replaced with a new one at each
potential in the H-cell measurements, the 1 h operating time is
much longer than a few minutes (only 80 s in potential below
−0.7 VRHE) of the CV protocol in SFC-DEMS analysis,
consequently resulting in a much more severe degradation of
Ni(−Cl)−N3O−TPP in the H-cell study. Despite the quick
measurements in the SFC-DEMS, the instability of Ni(−Cl)−
N3O−TPP can also be seen by the considerably decreased
ionic current from CO2 consumption and CO evolution at the
second CV polarization (Figure 2c).
OUTLOOK AND CONCLUSIONS
■
Figure 5. CO2-to-CO conversion in an H-type electrochemical cell.
(a) FETotal and FECO and (b) jTotal and jCO, acquired for 1 h CO2RR
operation at various potentials in a two-compartment H-cell. FETotal
was estimated by the sum of FECO and FEH2 (not shown). (c)
Baseline-subtracted oxidation peaks for Ni(−Cl)−N3O−TPP meas-
ured by CV with different LPLs. The voltammograms were recorded
by changing the LPL sequentially from −0.5 to −0.8 VRHE at a scan
rate of 50 mV s−1 in the CO2-saturated 0.5 M KHCO3 electrolyte.
The peaks (A and B) are indicated by arrows. (d) Raman spectra of
Ni(−Cl)−N3O−TPP before and after sequential CV measurements
from 0.4 to −0.8 VRHE. NiO signals in the Raman spectrum are
indicated by dotted lines.
Our experimental and theoretical findings on the electro-
chemistry of the two Ni-porphyrins offer important insights
into the rational design strategy of highly efficient Ni catalysts
for CO2RR. Given the similar case of single-atom Fe catalysts
for O2 reduction in fuel cells, it has been reported that, despite
the high instability of Fe-porphyrins, incorporation of their
active Fe moieties into a carbon lattice (i.e., Fe−N−C) can
effectively minimize the undesirable Fe demetalation even
under highly corrosive acidic conditions.51 In that sense,
anchoring the active Ni moieties on the carbon lattice (i.e.,
Ni−N−C) can thus be one promising approach to stabilization
929
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
performed using a synchrotron radiation light source at the Pohang
Accelerator Laboratory (8C-Nano XAFS). The samples were loaded
in a sample holder (w × l × d = 3 × 5 × 1 mm3) sealed by Kapton
tape. Before the measurement, the energy was calibrated using a Ni
foil standard. The XANES data were analyzed by Athena
implemented in the Demeter program package (ver. 0.9.26).
of the Ni sites without their dissolution. In many recent works,
indeed, this synthetic strategy has been successfully imple-
mented by pyrolyzing a mixture of Ni, N, and C precursors,
demonstrating their durable CO2-to-CO conversions over tens
of hours of operation.21,25,31
The next challenge will thus become how one can suppress
the in-plane ligand field and at the same time prevent the Ni
leaching under the reaction conditions. To provide further
synthetic guidelines, additional DFT calculations were
performed to investigate the Ni−N3X sites, where X = N, O,
C, S, and vacancy (V) using various active site models
embedded in the carbon lattice (Figure S38). The Ni binding
energies to N3X sites and *COOH binding energies to the Ni
center demonstrate an inverse proportionality (Figure S39),
implying a trade-off relation between the stability of the metal
site and CO2RR activity. Accordingly, it is important to
engineer the ligand-field strength to be optimal for both metal-
site stability and *COOH binding strength. Therefore, the
present understandings suggest that the next challenge is the
development of new synthetic routes to control the ligand-field
strength of the pyrolyzed Ni−N−C catalysts. With varying
heteroatom species and contents during pyrolysis, for example,
the ligand-field strength can be gradually tuned. We then
expect to map out the activity and stability in the domain of
ligand-field strength, which will enable a rational design of Ni−
N−C catalysts with achieving high CO2RR performance.
Electrochemical Analysis. Electrochemical measurements were
conducted with a RDE apparatus (RRDE-3A, ALS) and a potentiostat
(VMP3, Bio-Logic Science Ins.) in a three-electrode system. A
graphite rod and saturated Ag/AgCl reference electrode (RE-16, EC-
Frontier) were used as the counter and reference electrodes,
respectively. The electrochemical cell was made of Teflon, and the
reference electrode was separated by a double junction configuration
to avoid interference from glassware dissolution and halogen.52
A
glassy carbon RDE (diameter = 3 mm) was used as a working
electrode after mirror polishing with alumina slurry (1.0 and 0.05 μm,
R&B Inc.). For a sufficient electrical conductivity and homogeneous
dispersion of Ni-porphyrins, the Ni-porphyrins and Ketjenblack EC-
600JD were mixed together in dichloromethane (≥99.9%, Sigma-
Aldrich) with a target Ni loading of 1 wt %. It is of note that the
absence of the carbon black renders the electrochemical responses
rather irreversible in our case as compared to that with the carbon
black (data not shown). In the case of Ni-free porphyrins, an identical
number of porphyrins were introduced into the solution. After
evaporation of the solvent, catalyst inks were prepared by dispersing 5
mg of the Ni-porphyrin/carbon mixture and 23 μL of the Nafion
ionomer (5 wt %) in 4.3 mL of deionized (DI) water (>18.2 MΩ,
Arium mini, Sartorius). The thin film electrode was fabricated by
pipetting the inks onto the glassy carbon electrode (0.071 cm2) with a
targeted Ni loading of 1 μg cm−2. For comparison, a NiPc (Sigma-
Aldrich) electrode was fabricated identically. The electrolyte was 0.5
M KHCO3 solution, prepared using DI water and KHCO3 (99.7%,
Sigma-Aldrich). The reference electrode was calibrated in a H2-
saturated 0.5 M KHCO3 electrolyte. All potentials reported in this
work are given with respect to the RHE scale. Prior to the
electrochemical measurements, 30 cycles of CV were conducted in
the potential range of 0.05−1.00 VRHE at a scan rate of 200 mV s−1 to
stabilize the voltammetric response. The LSV measurements were
conducted at a scan rate of 50 mV s−1 at 1600 rpm rotation speed in
an Ar- or CO2-saturated electrolyte (pH = ca. 8.9 and 7.2,
respectively). The electrolyte pH was measured using an HM-30P
pH meter (DKK-TOA). The redox characteristics were measured at a
scan rate of 50 mV s−1 by CV from 0.4 VRHE to different LPLs,
decreasing from −0.5 to −0.8 VRHE (by 0.1 V). Before and after the
CV treatment, the precipitation of dissolved Ni ions into bulk phases
was investigated by Raman spectroscopy (micro-Raman spectroscopy
system, Renishaw) using a 514.5 nm laser (50 mW). The correlation
between the peak current of the redox couple and the scan rate was
analyzed at different scan rates (10, 20, 50, 100, 150, and 200 mV
s−1).
In Situ XANES Measurements. In situ XANES analysis was
performed with a synchrotron radiation light source at the Pohang
Accelerator Laboratory (1D, XRS KIST-PAL). A flow-type in situ
XAS cell, equipped with an electrolyte flow channel and a window for
X-ray radiation, was employed. The window was made of a carbon-
coated Kapton film (200RS100, 0.05 mm, DuPont), and it was
directly used as a working electrode after coating with the Ni-
porphyrin/carbon mixture (Ni loading = 110 μg cm−2). Due to the
low Ni content in the porphyrin/carbon mixture used for the
electrochemical measurements (ca. 1 wt % Ni), a mixture with a
higher Ni-porphyrin content on the carbon (7.5 wt % Ni) was
separately prepared for the in situ XANES measurements. Pt wire and
Ag/AgCl electrodes were used as counter and reference electrodes,
respectively, which made an electrochemical contact at the outlet of
the electrolyte stream. The electrolyte flow rate was set to 1 mL
min−1. The in situ XANES spectra were collected in the fluorescence
mode under the OCP and −0.65 VRHE in a CO2-saturated 0.5 M
KHCO3.
EXPERIMENTAL SECTION
■
Experimental details for the synthesis of Ni-porphyrins and computa-
tional predictions are given in Supporting Information Notes 1 and 4,
respectively.
Physical Characterizations. Liquid chromatography electrospray
ionization mass spectrometry (LC-ESI-MS) was performed using an
Agilent 1260 infinity liquid chromatography system with an Agilent
6120 single-quadrupole mass spectrometer. A binary mobile phase
system (solvent A: H2O + 0.1% trifluoroacetic acid (99.5%, Acors
Organics); solvent B: CH3CN) was used. The sample for LC-ESI-MS
was prepared as 100 μM solution in methanol (≥99.9%, Fisher
Scientific). MALDI-ToF MS was carried out with a Bruker
AutoflexTM speed MALDI-ToF spectrometer. The samples for
MALDI-ToF MS were prepared by mixing the porphyrin samples
with a CHCA matrix (saturated α-cyano-4-hydroxycinnamic acid
(Sigma-Aldrich) in 30% CH3CN (≥99.9%, Fisher Scientific) and 70%
H2O containing 0.1% trifluoroacetic acid (≥99.0%, Sigma-Aldrich)).
The spectra were recorded in a reflector positive ion mode using a
pulsed nitrogen laser as the ionization source. Data are reported as m/
z for both ESI-MS and MALDI-ToF MS. The UV−vis absorption
spectra were obtained on a GE Healthcare Ultrospec 2100 Pro in a
quartz cell with a 1 mm path length. The measurements were
performed in open air at room temperature. Each spectrum was
expressed after proper baseline correction with dichloromethane
(≥99.9%, Sigma-Aldrich). 1H NMR and 13C NMR spectra were
1
recorded with a JEOL ECS400 (400 MHz for H and 100 MHz for
13C) spectrometer in CDCl3 (D, 99.96%, Cambridge Isotope
Laboratories). The spectra were referenced to residual CHCl3 (7.26
ppm for 1H, 77.23 ppm for 13C). Chemical shifts are reported in ppm,
and signal multiplicities are indicated as s (singlet), d (doublet), and
m (multiplet). Elemental analysis was performed using an Elementar
vario MICRO cube. The chemical structure and atomic composition
of Ni−N4−TPP and Ni(−Cl)−N3O−TPP were analyzed by XPS
(PHI 5000 VersaProbe) with an Al Kα monochromator X-ray source.
The binding energies were calibrated against the C−C signal at 284.5
eV. The SC-XRD measurement was conducted on a Bruker APEX-II
CCD-based diffractometer with graphite-monochromated Mo Kα
radiation (λ = 0.7107 Å), and further details are given in Supporting
Information Note 2. Ex situ X-ray absorption spectroscopy (XAS) was
SFC-DEMS Analysis. The SFC connected to a mass spectrometer
(Max 300 LG, Extrel) was constructed for DEMS analysis. The SFC
930
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
was equipped with a 10 mm opening at the bottom of the cell for
electrical contact with the 3 mm diameter glassy carbon working
electrode. During the reactions, volatile and gaseous products were
directly collected by a tip equipped with a porous Teflon membrane
positioned ca. 100 μm above the working electrode. The graphite tube
(inner diameter = 3 mm) and reference electrode were electrically
connected by the U-shaped electrolyte channel. The working
electrode was fabricated in an identical manner to the setup for the
electrochemical measurements (Ni loading of 1 μg cm−2). The SFC-
DEMS measurements were conducted using two consecutive slow
CVs at a scan rate of 5 mV s−1 in Ar- or CO2-saturated electrolytes.
The electrolyte flow rate was set to 0.07 mL min−1. The mass signals
of H2 (m/z = 2), CO (m/z = 28), and CO2 (m/z = 44) were
collected simultaneously with the electrode polarizations. Due to the
fragmentation of CO2 to CO during its ionization, the contribution of
CO2 was subtracted (11%) from the ionic current collected at m/z =
28.
Online GC Measurement. Electrochemical CO2RR of the Ni-
porphyrins was conducted with a potentiostat (Ivium Stat-
potentiostat, Ivium Tech) in a two-compartment customized
electrochemical cell, where the cathode and anode regions were
separated using an anion exchange membrane (Selemion AMV). The
saturated calomel electrode (SCE) and Pt foil (0.05 mm, 99.99%, Alfa
Aesar) were used as the reference and counter electrodes, respectively.
The catalyst ink solution was prepared with 5 mg of the Ni-
porphyrin/carbon mixture and 40 μL of the Nafion solution (5 wt %)
in 3.78 mL of DI water. To prevent catalyst detachment, a higher
Nafion-to-catalyst ratio in the ink solution was employed for the CO2
electrolysis than that for the half-cell/SFC-DEMS studies. A working
electrode was fabricated by spray coating of the catalyst ink solution
on a glassy carbon substrate (active area = 0.28 cm2, Alfa Aesar). The
Ni loading on the glassy carbon was targeted at 100 μg cm−2. The
catalytic performance was examined through step-potential electrol-
ysis with periodic quantification of the gaseous products by online GC
(6500 GC, Young Lin Instrument Co. Ltd.). The products from CO2
electrolysis were CO and H2, and the net total FE was nearly 100%.
The electrolyte was a CO2-bubbled 0.5 M KHCO3 aqueous solution.
All potentials were compensated for the iR loss. The GC was
equipped with a pulsed discharge ionization detector (PDD) and a
capillary molecular sieve column (Restek, RT-Msieve 5A), using
ultrahigh purity He (99.9999%) as the carrier gas. During the
measurement, CO2 gas was continuously flowed through the
electrochemical cell at a velocity of 20 mL min−1. This helps to
sustain the steady reaction condition with almost unchanged
electrolyte pH during the electrolysis.53 After 1 h CO2RR using
Ni(−Cl)−N3O−TPP at −0.65 VRHE, 540 μL of the catholyte was
collected and mixed with 30 μL of standard solutionD2O solvent
(99.9%, Cambridge Isotope Laboratories) containing 2 mM dimethyl
sulfoxide (DMSO; ≥99.9%, Sigma-Aldrich) and 10 mM phenol
(≥99.0%, Sigma-Aldrich)to confirm the imperceptible formation of
the liquid products using 700 MHz 1D 1H NMR spectroscopy
(Bruker Avance).
AUTHOR INFORMATION
Corresponding Authors
■
Jiwon Seo − Department of Chemistry, Gwangju Institute of
Science and Technology, Gwangju 61005, Republic of Korea;
orcid.org/0000-0002-5433-5071; Email: jseo@gist.ac.kr
Hyungjun Kim − Department of Chemistry, Korea Advanced
Institute of Science and Technology, Daejeon 34141, Republic
of Korea; orcid.org/0000-0001-8261-9381;
Email: linus16@kaist.ac.kr
Chang Hyuck Choi − School of Materials Science and
Engineering, Gwangju Institute of Science and Technology,
Gwangju 61005, Republic of Korea; orcid.org/0000-
0002-2231-6116; Email: chchoi@gist.ac.kr
Authors
Haesol Kim − School of Materials Science and Engineering,
Gwangju Institute of Science and Technology, Gwangju
61005, Republic of Korea
Dongyup Shin − Department of Chemistry, Korea Advanced
Institute of Science and Technology, Daejeon 34141, Republic
of Korea
Woojin Yang − Department of Chemistry, Gwangju Institute
of Science and Technology, Gwangju 61005, Republic of
Korea
Da Hye Won − Clean Energy Research Center, Korea Institute
of Science and Technology, Seoul 02792, Republic of Korea
Hyung-Suk Oh − Clean Energy Research Center, Korea
Institute of Science and Technology, Seoul 02792, Republic of
Korea; orcid.org/0000-0002-0310-6666
Min Wook Chung − School of Materials Science and
Engineering, Gwangju Institute of Science and Technology,
Gwangju 61005, Republic of Korea
Donghyuk Jeong − Western Seoul Center, Korea Basic Science
Institute, Seoul 03759, Republic of Korea
Sun Hee Kim − Western Seoul Center, Korea Basic Science
Institute, Seoul 03759, Republic of Korea; orcid.org/
0000-0001-6557-1996
Keun Hwa Chae − Advanced Analysis Center, Korea Institute
of Science and Technology, Seoul 02792, Republic of Korea;
orcid.org/0000-0003-3894-670X
Ji Yeon Ryu − Department of Chemistry, Chonnam National
University, Gwangju 61186, Republic of Korea; orcid.org/
0000-0001-6321-5576
Junseong Lee − Department of Chemistry, Chonnam National
University, Gwangju 61186, Republic of Korea; orcid.org/
0000-0002-5004-7865
Sung June Cho − Department of Chemical Engineering,
Chonnam National University, Gwangju 61186, Republic of
Korea
ASSOCIATED CONTENT
■
Complete contact information is available at:
https://pubs.acs.org/10.1021/jacs.0c11008
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/jacs.0c11008.
Author Contributions
§H.K., D.S., W.Y., and D.H.W. contributed equally to this
work.
General methods; experimental procedure for the
preparation of porphyrins; X-ray crystallographic data
for Ni(−Cl)−N3O−TPP; assignment of redox couples
evolved after Ni dissolution from Ni(−Cl)−N3O−TPP;
computational details; detailed analyses of electronic
structure changes; catalytic nature of Ni-complexed
phthalocyanine (NiPc) in CO2RR (PDF)
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by the National Research
Foundation of Korea (NRF) grant funded by the Korea
government (MSIT) (No. NRF-2017M3D1A1039378,
Crystallographic data in CIF format (CIF)
931
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(18) Varela, A. S.; Ranjbar Sahraie, N.; Steinberg, J.; Ju, W.; Oh, H.
S.; Strasser, P. Metal-doped nitrogenated carbon as an efficient
catalyst for direct CO2 electroreduction to CO and hydrocarbons.
Angew. Chem., Int. Ed. 2015, 54, 10758−10762.
2018M3D1A1052659, and 2020R1A2C4002233) and sup-
ported by the Korea Institute of Science and Technology
(KIST) institutional program.
(19) Zheng, T.; Jiang, K.; Ta, N.; Hu, Y.; Zeng, J.; Liu, J.; Wang, H.
Large-scale and highly selective CO2 electrocatalytic reduction on
nickel single-atom catalyst. Joule 2019, 3, 265−278.
REFERENCES
■
(1) Olah, G. A.; Prakash, G. K. S.; Goeppert, A. Anthropogenic
chemical carbon cycle for a sustainable future. J. Am. Chem. Soc. 2011,
133, 12881−12898.
(2) Costentin, C.; Robert, M.; Saveant, J. M. Catalysis of the
electrochemical reduction of carbon dioxide. Chem. Soc. Rev. 2013, 42,
2423−2436.
(20) Liu, S.; Yang, H. B.; Hung, S.-F.; Ding, J.; Cai, W.; Liu, L.; Gao,
J.; Li, X.; Ren, X.; Kuang, Z.; Huang, Y.; Zhang, T.; Liu, B. Elucidating
the electrocatalytic CO2 reduction reaction over a model single-atom
nickel catalyst. Angew. Chem., Int. Ed. 2020, 59, 798−803.
̈
(21) Moller, T.; Ju, W.; Bagger, A.; Wang, X.; Luo, F.; Ngo Thanh,
(3) Kumar, B.; Asadi, M.; Pisasale, D.; Sinha-Ray, S.; Rosen, B. A.;
Haasch, R.; Abiade, J.; Yarin, A. L.; Salehi-Khojin, A. Renewable and
metal-free carbon nanofibre catalysts for carbon dioxide reduction.
Nat. Commun. 2013, 4, 2819.
(4) Lim, H.-K.; Shin, H.; Goddard, W. A.; Hwang, Y. J.; Min, B. K.;
Kim, H. Embedding covalency into metal catalysts for efficient
electrochemical conversion of CO2. J. Am. Chem. Soc. 2014, 136,
11355−11361.
T.; Varela, A. S.; Rossmeisl, J.; Strasser, P. Efficient CO2 to CO
electrolysis on solid Ni−N−C catalysts at industrial current densities.
Energy Environ. Sci. 2019, 12, 640−647.
(22) Cheng, Y.; Zhao, S.; Li, H.; He, S.; Veder, J.-P.; Johannessen,
B.; Xiao, J.; Lu, S.; Pan, J.; Chisholm, M. F.; Yang, S.-Z.; Liu, C.;
Chen, J. G.; Jiang, S. P. Unsaturated edge-anchored Ni single atoms
on porous microwave exfoliated graphene oxide for electrochemical
CO2. Appl. Catal., B 2019, 243, 294−303.
(5) Liu, M.; Pang, Y.; Zhang, B.; De Luna, P.; Voznyy, O.; Xu, J.;
Zheng, X.; Dinh, C. T.; Fan, F.; Cao, C.; de Arquer, F. P. G.; Safaei, T.
S.; Mepham, A.; Klinkova, A.; Kumacheva, E.; Filleter, T.; Sinton, D.;
Kelley, S. O.; Sargent, E. H. Enhanced electrocatalytic CO2 reduction
via field-induced reagent concentration. Nature 2016, 537, 382−386.
(6) Feng, X.; Jiang, K.; Fan, S.; Kanan, M. W. Grain boundary
dependent CO2 electroreduction activity. J. Am. Chem. Soc. 2015, 137,
4606−4609.
(7) Zhu, W.; Zhang, Y. J.; Zhang, H.; Lv, H.; Li, Q.; Michalsky, R.;
Peterson, A. A.; Sun, S. Active and selective conversion of CO2 to CO
on ultrathin Au nanowires. J. Am. Chem. Soc. 2014, 136, 16132−
16135.
(23) Yan, C.; Li, H.; Ye, Y.; Wu, H.; Cai, F.; Si, R.; Xiao, J.; Miao, S.;
Xie, S.; Yang, F.; Li, Y.; Wang, G.; Bao, X. Coordinatively unsaturated
nickel−nitrogen sites towards selective and high-rate CO2 electro-
reduction. Energy Environ. Sci. 2018, 11, 1204−1210.
(24) Jeong, H.-Y.; Balamurugan, M.; Choutipalli, V. S. K.; Jeong, E.;
Subramanian, V.; Sim, U.; Nam, K. T. Achieving highly efficient CO2
to CO electroreduction exceeding 300 mA cm−2 with single-atom
nickel electrocatalysts. J. Mater. Chem. A 2019, 7, 10651−10661.
(25) Jiang, K.; Siahrostami, S.; Zheng, T.; Hu, Y.; Hwang, S.;
Stavitski, E.; Peng, Y.; Dynes, J.; Gangisetty, M.; Su, D.; Attenkofer,
K.; Wang, H. Isolated Ni single atoms in graphene nanosheets for
high-performance CO2 reduction. Energy Environ. Sci. 2018, 11, 893−
903.
(26) Zhao, C.; Dai, X.; Yao, T.; Chen, W.; Wang, X.; Wang, J.; Yang,
J.; Wei, S.; Wu, Y.; Li, Y. Ionic exchange of metal−organic frameworks
to access single nickel sites for efficient electroreduction of CO2. J.
Am. Chem. Soc. 2017, 139, 8078−8081.
(27) Li, X.; Bi, W.; Chen, M.; Sun, Y.; Ju, H.; Yan, W.; Zhu, J.; Wu,
X.; Chu, W.; Wu, C.; Xie, Y. Exclusive Ni−N4 sites realize near-unity
CO selectivity for electrochemical CO2 reduction. J. Am. Chem. Soc.
2017, 139, 14889−14892.
(28) Ju, W.; Bagger, A.; Hao, G.-P.; Varela, A. S.; Sinev, I.; Bon, V.;
Roldan Cuenya, B.; Kaskel, S.; Rossmeisl, J.; Strasser, P. Under-
standing activity and selectivity of metal-nitrogen-doped carbon
catalysts for electrochemical reduction of CO2. Nat. Commun. 2017,
8, 944.
(29) Su, P.; Iwase, K.; Nakanishi, S.; Hashimoto, K.; Kamiya, K.
Nickel nitrogen-modified graphene: an efficient electrocatalyst for the
reduction of carbon dioxide to carbon monoxide. Small 2016, 12,
6083−6089.
(30) Su, P.; Iwase, K.; Harada, T.; Kamiya, K.; Nakanishi, S.
Covalent triazine framework modified with coordinatively-unsatu-
rated Co or Ni atoms for CO2 electrochemical reduction. Chem. Sci.
2018, 9, 3941−3947.
(8) Kauffman, D. R.; Alfonso, D.; Matranga, C.; Qian, H.; Jin, R.
Experimental and computational investigation of Au25 clusters and
CO2: a unique interaction and enhanced electrocatalytic activity. J.
Am. Chem. Soc. 2012, 134, 10237−10243.
(9) Liu, S.; Tao, H.; Zeng, L.; Liu, Q.; Xu, Z.; Liu, Q.; Luo, J.-L.
Shape-dependent electrocatalytic reduction of CO2 to CO on
triangular silver nanoplates. J. Am. Chem. Soc. 2017, 139, 2160−2163.
́
(10) Ma, M.; Trzesniewski, B. J.; Xie, J.; Smith, W. A. Selective and
efficient reduction of carbon dioxide to carbon monoxide on oxide-
derived nanostructured silver electrocatalysts. Angew. Chem., Int. Ed.
2016, 55, 9748−9752.
(11) Hsieh, Y.-C.; Senanayake, S. D.; Zhang, Y.; Xu, W.; Polyansky,
D. E. Effect of chloride anions on the synthesis and enhanced catalytic
activity of silver nanocoral electrodes for CO2 electroreduction. ACS
Catal. 2015, 5, 5349−5356.
(12) Lu, Q.; Rosen, J.; Zhou, Y.; Hutchings, G. S.; Kimmel, Y. C.;
Chen, J. G.; Jiao, F. A selective and efficient electrocatalyst for carbon
dioxide reduction. Nat. Commun. 2014, 5, 3242.
(13) Rosen, B. A.; Salehi-Khojin, A.; Thorson, M. l. R.; Zhu, W.;
Whipple, D. T.; Kenis, P. J. A.; Masel, R. I. Ionic liquid−mediated
selective conversion of CO2 to CO at low overpotentials. Science
2011, 334, 643−644.
(14) Feaster, J. T.; Shi, C.; Cave, E. R.; Hatsukade, T.; Abram, D. N.;
Kuhl, K. P.; Hahn, C.; Nørskov, J. K.; Jaramillo, T. F. Understanding
selectivity for the electrochemical reduction of carbon dioxide to
formic acid and carbon monoxide on metal electrodes. ACS Catal.
2017, 7, 4822−4827.
(15) Hansen, H. A.; Varley, J. B.; Peterson, A. A.; Nørskov, J. K.
Understanding trends in the electrocatalytic activity of metals and
enzymes for CO2 reduction to CO. J. Phys. Chem. Lett. 2013, 4, 388−
392.
(31) Yang, H. B.; Hung, S.-F.; Liu, S.; Yuan, K.; Miao, S.; Zhang, L.;
Huang, X.; Wang, H.-Y.; Cai, W.; Chen, R.; Gao, J.; Yang, X.; Chen,
W.; Huang, Y.; Chen, H. M.; Li, C. M.; Zhang, T.; Liu, B. Atomically
dispersed Ni(i) as the active site for electrochemical CO2 reduction.
Nat. Energy 2018, 3, 140−147.
(32) Ballhausen, C. J. Introduction to Ligand Field Theory; McGraw-
Hill: New York, 1962.
(33) Hossain, M. D.; Huang, Y.; Yu, T. H.; Goddard, W. A.; Luo, Z.
Reaction mechanism and kinetics for CO2 reduction on nickel single
atom catalysts from quantum mechanics. Nat. Commun. 2020, 11,
2256.
(16) Wuttig, A.; Surendranath, Y. Impurity ion complexation
enhances carbon dioxide reduction catalysis. ACS Catal. 2015, 5,
4479−4484.
́
(34) Chmielewski, P. J.; Latos-Grazynski, L.; Olmstead, M. M.;
̇
Balch, A. L. Nickel complexes of 21-oxaporphyrin and 21, 23-
́
(17) Costentin, C.; Drouet, S.; Robert, M.; Saveant, J.-M. A local
proton source enhances CO2 electroreduction to CO by a molecular
Fe catalyst. Science 2012, 338, 90−94.
dioxaporphyrin. Chem. - Eur. J. 1997, 3, 268−278.
932
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933

Journal of the American Chemical Society
pubs.acs.org/JACS
Article
(35) Adler, A. D.; Longo, F. R.; Kampas, F.; Kim, J. On the
preparation of metalloporphyrins. J. Inorg. Nucl. Chem. 1970, 32,
2443−2445.
on a hexagonal Zn catalyst. Angew. Chem., Int. Ed. 2016, 55, 9297−
9300.
(36) Colpas, G. J.; Maroney, M. J.; Bagyinka, C.; Kumar, M.; Willis,
W. S.; Suib, S. L.; Mascharak, P. K.; Baidya, N. X-ray spectroscopic
studies of nickel complexes, with application to the structure of nickel
sites in hydrogenases. Inorg. Chem. 1991, 30, 920−928.
(37) Yamamoto, T. Assignment of pre-edge peaks in K-edge x-ray
absorption spectra of 3d transition metal compounds: electric dipole
or quadrupole? X-Ray Spectrom. 2008, 37, 572−584.
(38) Maclean, A. L.; Foran, G. J.; Kennedy, B. J.; Turner, P.;
Hambley, T. W. Structural characterization of nickel(II) tetraphe-
nylporphyrin. Aust. J. Chem. 1996, 49, 1273−1278.
(39) Bard, A. J.; Faulkner, L. R. Electrochemical Methods:
Fundamentals and Applications, 2nd ed.; Wiley: New York, 2001.
(40) Elgrishi, N.; Rountree, K. J.; McCarthy, B. D.; Rountree, E. S.;
Eisenhart, T. T.; Dempsey, J. L. A practical beginner’s guide to cyclic
voltammetry. J. Chem. Educ. 2018, 95, 197−206.
́
(41) Latos-Grazynski, L.; Olmstead, M. M.; Balch, A. L. The first
̇
structural characterization of a nickel(I) macrocyclic system: structure
of nickel(I) diphenyldi-p-toyl-21-thiaporphyrin. Inorg. Chem. 1989,
28, 4065−4066.
́
(42) Pacholska, E.; Chmielewski, P. J.; Latos-Grazynski, L.
̇
Formation and reactivity of paramagnetic organometallic nickel
1
complexes of 21-oxa- and 21-selenaporphyrins H NMR and EPR
investigations. Inorg. Chim. Acta 1998, 273, 184−190.
(43) Lovecchio, F. V.; Gore, E. S.; Busch, D. H. Oxidation and
reduction behavior of macrocyclic complexes of nickel. Electro-
chemical and electron spin resonance studies. J. Am. Chem. Soc. 1974,
96, 3109−3118.
(44) Fang, Y.; Senge, M. O.; Van Caemelbecke, E.; Smith, K. M.;
Medforth, C. J.; Zhang, M.; Kadish, K. M. Impact of substituents and
nonplanarity on nickel and copper porphyrin electrochemistry: first
observation of a CuII/CuIII reaction in nonaqueous media. Inorg.
Chem. 2014, 53, 10772−10778.
(45) Kumar, R.; Yadav, P.; Rathi, P.; Sankar, M. Synthesis,
spectroscopic, electrochemical redox, solvatochromism and anion
binding properties of β-tetra- and -octaphenylethynyl substituted
meso-tetraphenylporphyrins. RSC Adv. 2015, 5, 82237−82246.
(46) Liao, M.-S.; Scheiner, S. Electronic structure and bonding in
metal porphyrins, metal = Fe, Co, Ni, Cu, Zn. J. Chem. Phys. 2002,
117, 205−219.
(47) Huang, L. F.; Hutchison, M. J.; Santucci, R. J.; Scully, J. R.;
Rondinelli, J. M. Improved electrochemical phase diagrams from
theory and experiment: the Ni−water system and its complex
compounds. J. Phys. Chem. C 2017, 121, 9782−9789.
(48) Cazzanelli, E.; Kuzmin, A.; Mariotto, G.; Mironova-Ulmane, N.
Study of vibrational and magnetic excitations in NicMg1‑cO solid
solutions by Raman spectroscopy. J. Phys.: Condens. Matter 2003, 15,
2045−2052.
(49) Gong, M.; Zhou, W.; Tsai, M.-C.; Zhou, J.; Guan, M.; Lin, M.-
C.; Zhang, B.; Hu, Y.; Wang, D.-Y.; Yang, J.; Pennycook, S. J.; Hwang,
B.-J.; Dai, H. Nanoscale nickel oxide/nickel heterostructures for active
hydrogen evolution electrocatalysis. Nat. Commun. 2014, 5, 4695.
(50) Chhetri, M.; Sultan, S.; Rao, C. N. R. Electrocatalytic hydrogen
evolution reaction activity comparable to platinum exhibited by the
Ni/Ni(OH)2/graphite electrode. Proc. Natl. Acad. Sci. U. S. A. 2017,
114, 8986.
(51) Santori, P. G.; Speck, F. D.; Li, J.; Zitolo, A.; Jia, Q.; Mukerjee,
S.; Cherevko, S.; Jaouen, F. Effect of pyrolysis atmosphere and
electrolyte pH on the oxygen reduction activity, stability and
spectroscopic signature of FeNx moieties in Fe-N-C catalysts. J.
Electrochem. Soc. 2019, 166, F3311−F3320.
(52) Mayrhofer, K. J. J.; Crampton, A. S.; Wiberg, G. K. H.; Arenz,
M. Analysis of the impact of individual glass constituents on
electrocatalysis on Pt electrodes in alkaline solution. J. Electrochem.
Soc. 2008, 155, P78−P81.
(53) Won, D. H.; Shin, H.; Koh, J.; Chung, J.; Lee, H. S.; Kim, H.;
Woo, S. I. Highly efficient, selective, and stable CO2 electroreduction
933
https://dx.doi.org/10.1021/jacs.0c11008
J. Am. Chem. Soc. 2021, 143, 925−933