Modulating Single-Atom Palladium Sites with Copper for Enhanced Ambient Ammonia Electrosynthesis

Article abstract of DOI:10.1002/anie.202010159

The electrochemical reduction of N₂ to NH₃ is emerging as a promising alternative for sustainable and distributed production of NH₃. However, the development has been impeded by difficulties in N₂ adsorption, protonation of *NN, and inhibition of competing hydrogen evolution. To address the issues, we design a catalyst with diatomic Pd-Cu sites on N-doped carbon by modulation of single-atom Pd sites with Cu. The introduction of Cu not only shifts the partial density of states of Pd toward the Fermi level but also promotes the d-2π* coupling between Pd and adsorbed N₂, leading to enhanced chemisorption and activated protonation of N₂, and suppressed hydrogen evolution. As a result, the catalyst achieves a high Faradaic efficiency of 24.8±0.8 % and a desirable NH₃ yield rate of 69.2±2.5 μg h^?1 mg_cat.^?1, far outperforming the individual single-atom Pd catalyst. This work paves a pathway of engineering single-atom-based electrocatalysts for enhanced ammonia electrosynthesis.

Full text of DOI:10.1002/anie.202010159

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Modulating single-atom Pd sites with Cu for enhanced ambient
ammonia electrosynthesis
Authors: Lili Han, Zhouhong Ren, Pengfei Ou, Hao Cheng, Ning Rui,
Lili Lin, Xijun Liu, Longchao Zhuo, Jun Song, Jiaqiang Sun,
Jun Luo, and Huolin L.Xin
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202010159
Link to VoR: https://doi.org/10.1002/anie.202010159

10.1002/anie.202010159
Angewandte Chemie International Edition
RESEARCH ARTICLE
Modulating single-atom Pd sites with Cu for enhanced ambient
ammonia electrosynthesis
Lili Han,[a]† Zhouhong Ren,[b]† Pengfei Ou,[c]† Hao Cheng,[a] Ning Rui,[d] Lili Lin,[e] Xijun Liu*[b], Longchao
Zhuo[f], Jun Song*[c], Jiaqiang Sun[g], Jun Luo [b], and Huolin L. Xin*[a]
Abstract: The electrochemical reduction of N2 to NH3 is emerging as
a promising alternative for sustainable and distributed production of
NH3. However, the development has been impeded by difficulties in
N2 adsorption, protonation of *NN and inhibition of competing
hydrogen evolution. To address the issues, we design a catalyst with
diatomic Pd-Cu sites on N-doped carbon by modulation of single-
atom Pd sites with Cu. The introduction of Cu not only shifts the
partial density states of Pd toward the Fermi level but also promote
the d-2π* coupling between Pd and adsorbed N2, leading to
enhanced N2 chemisorption, activated protonation and suppressed
hydrogen evolution. As a result, the catalyst achieves a high
Faradaic efficiency of 24.8 ± 0.8 % and a desirable NH3 yield rate of
69.2 ± 2.5 μg h−1 mgcat.−1, far outperforming the individual single-
atom Pd catalyst. This work paves a pathway of engineering single-
Electrochemical N2 reduction reaction (NRR) to NH3 is a gentle
and ambient process with low energy consumption, considering
as a promising alternative to the Haber–Bosch process[6,7]
.
However, the development of NRR has been impeded by the
lack of efficient electrocatalysts. Single-atom catalysts (SACs)
are expected to have great potential to improve the NRR
performance due to the lowest site size, nearly 100% atomic
efficiency and unique low-coordinated features[8–11]. Many SACs
such as Ru[12,13], Mo[14], Fe[15-17], Au[18], Cu[19], Y[20], Ag[20], and
B[21] have been applied for the NRR in aqueous electrolytes.
However, two major challenges associated with the NRR in
aqueous media have to be faced[22]. For one thing, to efficiently
break N≡N triple bond (941 kJ mol−1) [23], an electrocatalyst
needs to possess strong N2 adsorption ability (* + N2 → *NN)
atom-based electrocatalysts for enhanced ammonia electrosynthesis. and low activation energy of the first proton-coupled electron
transfer step (H+ + e− + *NN → *NNH)[19,24]. For another, most of
the protons and electrons in the system go toward the hydrogen
evolution reaction (HER) rather than NRR, leading to an
extremely low Faradaic efficiency (FE) for NRR under ambient
conditions[25]. Therefore, addressing the challenges is critical to
Introduction
Ammonia (NH3) is one of the most highly produced inorganic
purposefully synthesizing active and durable single-atom-based
NRR electrocatalysts. Previous studies[22,26] have reported that
Pd during NRR can form Pd hydrides and promote surface
hydrogenation reactions of *NN, but Pd is subjected to poisoning
by molecular hydrogen due to more strong binding to hydrogen
adatoms than to nitrogen ones[27]. Cu can not only facilitate the
hydrogen dissociation at Pd sites but also enhance the electron
transfer rate by strong d-d coupling with Pd toward excellent
NRR[24,27,28]. Therefore, incorporation of Cu into single-atom Pd
sites is a promising way to modulate the electronic structure of
the Pd for NRR, which, however, has not been explored before.
Inspired by the above breakthroughs and our previous
chemicals in the world because of its vast need in fertilizer
production, pharmaceutical production, and many other
industrial processes[1,2]
. The industrial production of NH3
commonly relies on the Haber−Bosch process, which requires
harsh conditions (350−550 °C and 150−350 atm) associated
with rather high energy consumption and CO2 emission[3–5]
.
[a]
L. L. Han, H Cheng, Prof. H. L. Xin
Department of Physics and Astronomy, University of California, Irvine,
CA 92697, USA.
E-mail: huolin.xin@uci.edu
[b]
[c]
Z. H. Ren, Dr. X. J. Liu, Prof. J. Luo
studies about NRR and SACs[14,15,29]
, we have rationally
Center for Electron Microscopy and Tianjin Key Lab of Advanced
Functional Porous Materials, Institute for New Energy Materials & Low-
Carbon Technologies, School of Materials Science and Engineering,
Tianjin University of Technology, Tianjin 300384, China.
E-mail: xjliu@tjut.edu.cn
modulated single-atom Pd sites and built diatomic Pd-Cu sites
on N-doped carbon (PdCu/NC) by the introduction of Cu. The
PdCu/NC achieves a high FE of 24.8 ± 0.8 % and a desirable
NH3 yield rate of 69.2 ± 2.5 μg h−1 mgcat. at −0.45 V vs
−1
P. F. Ou, J. Song
Department of Mining and Materials Engineering, McGill University,
Montreal H3A 0C5, Canada.
E-mail: jun.song2@mcgill.ca
reversible hydrogen electrode (RHE, all potentials in this work
are given vs RHE otherwise stated.) in 0.05 M H2SO4, which are
14.6 and 4.2 times of Pd/NC, respectively. Our density functional
theory (DFT) calculations reveal that introduction of Cu into
Pd/NC not only shifts the partial density of states (pDOS) of Pd
toward the Fermi level but also promote the d-2π* coupling
between Pd and adsorbed N2. It renders N2 chemisorption,
activates hydrogenation of the adsorbed N2 and suppresses
hydrogen evolution, eventually lowering the energy required by
the potential-determining step (PDS) to enhance the selectivity
of NRR.
[d]
[e]
[f]
N. Rui
Chemistry Division, Brookhaven National Laboratory, Upton, NY
11973, USA.
L. L. Lin
Institute of Industrial Catalysis, College of Chemical Engineering,
Zhejiang University of Technology, Hangzhou, China.
Prof. L. C. Zhuo
School of Materials Science and Engineering, Xi'an University of
Technology, Xi'an 710048, China
[g]
[†]
J. Q. Sun
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry,
Chinese Academy of Sciences, Taiyuan 030001, China
These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of
the document.
Results and Discussion
This article is protected by copyright. All rights reserved.

10.1002/anie.202010159
Angewandte Chemie International Edition
RESEARCH ARTICLE
A
B
PdCu/NC was synthesized via a dissolution-and-carbonization
method (see Experimental Section in Supporting Information for
details). The Cl2H14N4OPd, CuCl2 and (NH3OH)Cl aqueous
solutions were mixed with glucose ethanol solution followed by
being dried at 70 °C overnight. During this process, the chelated
[NH3] on Cl2H14N4OPd can chelate with Cu2+ to form a complex
of Pd and Cu[30]. Meanwhile, they were efficiently separated by
glucose backbones, which ensures a homogeneous distribution
of the Pd−Cu active sites in the subsequent calcination. After
carbonization at 600 °C, the carbon matrix, obtained from
carbonization of the glucose, was etched by NH3 produced via
the decomposition of hydroxylamine hydrochloride precursor
2 μm
5 nm
D
C
and formed a 3-dimensional (3D) porous structure[31]
.
The
pyrolyzed metal complexes were transformed into M−N−C sites
(M refers to metals)[32]. Therefore, Pd−Cu diatomic sites are
likely to be formed by the adjacent Pd and Cu atoms, which
coordinate with nitrogen and are uniformly dispersed on the 3D
carbon matrix. The obtained product was characterized by the
high-angle annular dark-field scanning transmission electron
2 nm
1 nm
Cu
microscopy
(HAADF-STEM).
Figure
1a
shows
3D
2 nm
interconnected carbon frameworks with randomly opened
porous structure. The N2 adsorption−desorption isotherm and
pore distribution of PdCu/NC in Figure S1 demonstrates the
specific area of 214.34 m2 g−1 and the existence of many pores
with sizes of 2.3 ~ 7 nm. Such pores are beneficial to the
exposure of active sites and the mass transport of electrolytes
E
Pd
50 nm
during electrolysis[14]
. High-resolution transmission electron
Figure 1. Structural characterizations of PdCu/NC. (a) Low-magnification
HAADF-STEM image. (b) HRTEM image. (c) Atomic-resolution HAADF-STEM
image. (d) Magnified atomic-resolution HAADF-STEM image of the yellow-
frame area in (c). Some of atomic pairs are highlighted by yellow oval frames.
(e) HAADF-STEM image and corresponding EDS mapping images of Pd and
Cu.
microscopy (HRTEM) image and the selected area electron
diffraction (SAED) pattern demonstrate the amorphousness of
PdCu/NC (Figure 1b and Figure S2a). This finding coincides well
with the X-ray diffraction (XRD) results (Figure S2b). Atomic-
resolution HAADF-STEM images in Figure 1c show many
isolated bright dots distributed on the carbon matrix. When
zooming in to the magnified atomic-resolution HAADF-STEM
image (Figure 1d), it was found that many atomic pairs with two
different contracts were distributed on the carbon matrix. Further,
elemental mapping acquired by the energy-dispersive X-ray
spectroscopy (EDS) manifests homogeneous distribution of Pd,
Cu, N and C over the entire architectures (Figure 1e and Figure
S2c), suggesting that the atomic pair is composed of Pd and Cu.
The Pd and Cu loadings in PdCu/NC measured by the
inductively coupled plasma atomic emission spectroscopy (ICP-
AES) analyses are 2.23 wt% and 2.32 wt%, respectively.
X-ray absorption fine structure (XAFS) spectroscopy was
utilized to explore the local structures of Pd and Cu atoms in
PdCu/NC. Figure 2a shows that the Pd X-ray absorption near-
edge structure (XANES) edge of PdCu/NC is close to PdO,
indicating that the Pd atom valence state is +2. Figure 2b
presents the Cu K-edge XANES edge of PdCu/NC lies between
those of Cu2O and CuO, indicating that the Cu average valence
is between +1 and +2. The Pd K-edge Fourier transformed
extended XAFS (FT-EXAFS) spectrum of PdCu/NC (Figure 2c)
displays the main peak at 1.46 Å[33], which can be attributed to
the coordination of Pd to N. The Cu K-edge FT-EXAFS
spectrum of PdCu/NC (Figure 2d) displays a main peak at 1.47
Å[34], which can be assigned to the backscattering of Cu–N
coordination.
This article is protected by copyright. All rights reserved.

10.1002/anie.202010159
Angewandte Chemie International Edition
RESEARCH ARTICLE
consist of atomically dispersed Pd and Cu on N-doped carbon,
respectively (Figures S7 and S8). The Pd and Cu loadings in
Pd/NC and Cu/NC are determined to be 2.94 wt% and 5.53 wt%,
respectively, through ICP-AES measurements. Synchrotron
radiation XAFS measurements were performed on Cu/NC and
Pd/NC samples. Their XANES and EXAFS spectra are shown in
Figure S9. We fitted Cu K-edge EXAFS spectra of Cu/NC with
CuN3 and Cu-Cu (CuN2CuN2_556) configurations, and fitted Pd
K-edge EXAFS spectra of Pd/NC with PdN3 and Pd-Pd
(PdN2PdN2_556) configurations. The fitting results are shown in
Figure S9c,d and Table S2. The CuN3 and PdN3 configurations
show much better fitting qualities (lower R factor and less
deviation of fitting coordination numbers from the initial structure
models) than the PdN2PdN2 and CuN2CuN2 configurations,
suggesting that CuN3 and PdN3 are dominant sites in Cu/NC
and Pd/NC. Figure 2e,f displays that Pd 3d5/2 and Cu 2p3/2 XPS
peaks in PdCu/NC shift by −0.15 eV and −0.16 eV compared
with the Pd one in Pd/NC and the Cu one in Cu/NC, respectively,
which suggests that both Cu and Pd in PdCu/NC loss fewer
electrons than Pd in Pd/NC and Cu in Cu/NC, respectively. This
finding is consistent with the charge calculation result, where Pd
in PdCu/NC and Cu in PdCu/NC have 0.154 and 0.091 electrons
more than Pd in Pd/NC and Cu in Cu/NC, respectively (Figure
S10). On the contrary, C 1s peak of PdCu/NC has no shift
compared to those of Pd/NC and Cu/NC (Figure S11). These
comparisons demonstrate that the introduction of the SA Cu into
the SA Pd has a great effect on the electronic structures of both
the Cu and the Pd, which suggests bonding of Cu and Pd in
PdCu/NC. The finding aligns well with the XAFS result that
PdCu/NC is composed of the mixed PdN2CuN2_556 and
PdN3CuN3 structure.
Figure 2. Synchrotron radiation XAFS and XPS measurements. (a) Pd K-edge
XANES spectra of PdCu/NC and reference samples (Pd foil, PdO, PdCl2). (b)
Cu K-edge XANES spectra of PdCu/NC and reference samples (Cu foil, Cu2O,
CuO). (c) Pd K-edge and (d) Cu K-edge FT-EXAFS spectra of PdCu/NC and
their reference samples. Cu and Pd FT-EXAFS spectra of PdCu/NC were
fitted with the PdN2CuN2_556 and PdN3CuN3 mixed configuration. (e) Pd 3d
XPS spectra of PdCu/NC and Pd/NC. (f) Cu 2p XPS spectra of PdCu/NC and
Cu/NC.
The NRR performance of the PdCu/NC catalyst was
examined in N2-saturated 0.05
M H2SO4 under ambient
To determine the structure of PdCu/NC, we established
various theoretically optimized models based on DFT (see
Computational Section in Supporting Information for more
details). Their XANES results were further calculated and shown
in Figures S3 and S4, which suggests that the calculated Pd and
Cu XANES spectra of PdN2CuN2_556 and PdN3CuN3 structures
match well with the experimental ones. Based on the above
findings, a fit to the experimental Pd and Cu EXAFS data was
performed, and a joint PdN2CuN2_556 and PdN3CuN3 structures
in the respective percentage of 65 ± 5% and 35 ± 5% was
determined (Figure 2c,d, Figure S5 and Table S1). Such a result
is consistent with many Pd, Cu atomic pairs distributed on N-
doped carbon in Figure 1d. To explore the existence of the Pd-
Pd and Cu-Cu sites in PdCu/NC, we fitted the Pd and Cu K-
edge EXAFS spectra of PdCu/NC using the Pd-Pd
(PdN2PdN2_556) and Cu-Cu (CuN2CuN2_556) configuration
models. The fitting results in Figure S6 and Table S1 indicate
poor fitting qualities (higher R-factor) for both of the two
configurations. The finding demonstrates that the neglectable
amount of Pd-Pd and Cu-Cu sites formed in PdCu/NC.
conditions according to our previous work[14,15] (see
Experimental Section in Supporting Information for more details).
To exclude the possible contaminants of NH3 and NOx in the
feeding gases, N2 and Ar gases were purified by 2 M NaOH, 0.1
M FeSO4 and 5 mM H2SO4 solutions[35]. The NH3 yield rates
were estimated by the indophenol blue method[36]. As shown in
Figure 3a, the NH3 yields over the PdCu/NC electrocatalyst
increase with more negative potential until reaching −0.45 V,
where the maximum value of NH3 yield rate is calculated as 69.2
± 2.5 μg h−1 mgcat.−1, 4.2 and 3.0 times of Pd/NC and Cu/NC,
respectively. It achieved the maximum FE of 24.8 ± 0.8 % at
−0.45 V, 14.6 and 3.6 times of Pd/NC and Cu/NC, respectively
(Figure 3b). These results demonstrate that the PdCu/NC after
the modulation with Cu is far more selective and active than
Pd/NC for NRR to NH3.
To confirm the origin of the NH3 produced, the controlled
experiments and isotope measurements were performed[35]. As
shown in Figure 3c, no NH3 was detected after NRR
experiments performed over the PdCu/NC catalyst for 2 h in Ar-
saturated catholyte, solely on carbon paper, and in the N2-
saturated electrolyte without electrolysis (that is, open circuit).
Moreover, the 15N isotopic labeling experiment (Figure 3d)
demonstrated a triplet coupling for 14NH4+ and a double coupling
Comparison of Pd 3d and Cu 2p X-ray photoelectron
spectroscopy (XPS) spectra of PdCu/NC, with respective those
of control samples including Pd/NC and Cu/NC, was performed
to explore the electronic structure of PdCu/NC. The Pd/NC and
Cu/NC were synthesized via the same method of PdCu/NC (see
Experimental Section in Supporting Information for details). They
for 15NH4 . The isotopic 15N measurement result for the
+
quantification of ammonia shows that the concentration of
15NH4 product after 15N2 reduction was determined to be 51.2
+
This article is protected by copyright. All rights reserved.

10.1002/anie.202010159
Angewandte Chemie International Edition
RESEARCH ARTICLE
μM (Figure S12), close to the value (50.9 μM) measured by the
indophenol blue method. More importantly, the two valules are in
the range of (50.7 ± 1.8 μM) measured by performing the
electrocatalysis of 14N2 (Figure 3a). These confirm that the
generated ammonia is produced from the NRR. In addition,
stability is a critical parameter of NRR performance for practical
applications. Under sustained N2 gas flow, 30-h electrolysis on
the PdCu/NC catalyst at a potential of −0.45 V only leads to a
slight decrease in current density and NH3 yield rate (Figure 3e),
indicating that the high stability of PdCu/NC for NRR. Additional
essential prerequisite for an efficient NRR process[9,37]. Figure
S14a compares PdN3CuN3 with PdN4 and CuN4 in their
interaction with N2, showing that they all weakly interact with N2
and fail to chemically adsorb and activate the inert triple bond of
N2. This is consistent with the previous theoretical studies[19,37]
,
indicating that PdN4 or CuN4 structures are not suitable catalyst
candidates for NRR. This finding also suggests that the active
sites of the PdCu/NC catalyst for NRR are not from the
PdN3CuN3 structure.
XRD,
HAADF-STEM
and
electrochemical
impedance
spectroscopy (EIS) characterizations further confirmed the
stability of PdCu/NC (Figure S13).
Figure 4. Computed electronic properties for Pd/NC, Cu/NC, and PdCu/NC,
represented by PdN3, CuN3 and PdN2CuN2_556 respectively. (a) Partial
density of states (pDOS) of N2 adsorption on Pd/NC, Cu/NC, and PdCu/NC.
Blue, orange, and cyan lines indicate the 2p orbital of N, 3d orbital of Cu, and
4d orbital of Pd, respectively. (b) Projected crystal orbital Hamilton population
(−pCOHP) and its integrated value (ICOHP) of N2 adsorption on Pd/NC,
Cu/NC, and PdCu/NC. Orange and cyan lines indicate the bondings between
N-Cu and N-Pd. The energy value is with reference to the Fermi level. Yellow
and blue color areas depict the bonding and antibonding states.
Meanwhile, Figure S14b shows the case of PdN2CuN2_556
in comparison with PdN3 and CuN3, demonstrating that N2 can
be only physisorbed on the sites of PdN3 and CuN3 via the end-
on configuration with weak binding strength, with the free-energy
changes of N2 adsorption (∆G*N2) being 0.46 eV and −0.05 eV
respectively. The weakly adsorbed N2 exhibits a bond length
similar to that of its gas-phase (1.11 Å, Figure S14a), signaling
failure to activate the inert triple bond of N2, hence preventing
the reaction from proceeding to the subsequent NRR steps. On
the other hand, PdN2CuN2_556 shows strong interaction with N2,
leading to considerably elongated N≡N triple bond lengths of
1.17 and 1.15 Å for the side-on and end-on configurations
(Figure S14b), respectively. The elongation significantly
weakens the N≡N triple bond to yield chemisorption with much
Figure 3. NRR performances. (a) NH3 yield rates and (b) FEs of PdCu/NC,
Pd/NC, and Cu/NC at each given potential in N2-saturated 0.05 M H2SO4. (c)
UV-vis absorption spectra of electrolytes stained with indophenol blue
indicator after NRR on PdCu/NC, carbon paper, Ar control experiment, and
open circuit condition at −0.45 V. (d) 1H NMR spectra of 0.05 M H2SO4
solutions after 2-h electrochemical reduction on PdCu/NC at −0.45 V using Ar
(in black), 14N2 (in blue) and 15N2 (in red) as the feeding gases. (e)
Chronoamperometric curve and FE stabilities of PdCu/NC during 30-h
electrolysis.
larger ∆G*N values of −0.71 eV and −0.58 eV. The result
2
suggests the PdN2CuN2_556 structure to be the responsible
active center on the PdCu/NC catalyst to facilitate NRR. In the
following, for simplicity of the discussion, PdCu/NC, Pd/NC and
Cu/NC are represented by the PdN2CuN2_556, PdN3 and CuN3
respectively.
The above results highlight that weakening of the triple
bond and chemical bonding formation of N2 on PdCu/NC
together contribute obvious activation of N2 that is beneficial for
the subsequent NRR steps. To further analyze the interactions
between the adsorbed N2 and metals and understand the
underlying mechanism of N2 activation on the PdCu/NC, we
examined the pDOS and projected crystal orbital Hamilton
population (pCOHP)[38]. From the pDOSs in Figure 4a, the
To better understand the observations and provide
theoretical insights, DFT calculations were performed on the
PdCu/NC, Pd/NC and Cu/NC systems to examine their
respective catalytic activity and selectivity for NRR (see
Computational Section in Supporting Information for more
details). In particular, based on the EXAFS fitting results in
Figure 2c,d, we built the PdN3CuN3 and the PdN2CuN2_556
models for the PdCu/NC system, and the CuN4, PdN4, CuN3 and
PdN3 models for the reference Pd/NC and Cu/NC systems
(Figure S14). Then, N2 adsorption on different models was
investigated, as adequate N adsorption and activation is the
This article is protected by copyright. All rights reserved.

10.1002/anie.202010159
Angewandte Chemie International Edition
RESEARCH ARTICLE
introduction of Cu into Pd/NC moves the pDOS of Pd 4d toward
more positive values, and consequently, more pDOSs of Pd 4d
are involved in the overlap region with those of N 2p near the
Fermi level. Additionally, from the computed −pCOHP as shown
in Figure 4b, the capability of PdCu/NC to adsorb and activate
N2 is ascribed to two aspects: the 2π and 3σ molecular orbitals
of N2 donate electrons to the unoccupied d orbitals of Pd and Cu
to form the bonding states to strengthen the N2 adsorption.
Moreover, we note that the appearance of occupied 2π*
molecular orbitals near the Fermi level for PdCu/NC, that is, the
occupied d orbitals of Pd and Cu back-donate electrons to the
2π* molecular orbitals of N2 to activate N2 hydrogenation[10]. This
further distinguishes PdCu/NC from the Pd/NC and the Cu/NC
systems. For a more quantitative analysis, the integrated
pCOHP (ICOHP) (Figure 4b) was also computed to compare the
d-2π* coupling PdCu/NC with Pd/NC and Cu/NC. From the
figure, a more negative ICOHP was observed for PdCu/NC,
indicating a much stronger d-2π* coupling.
PdCu/NC favors the distal pathway energetically, different from
Pd/NC and Cu/NC which would prefer the alternating pathway.
Furthermore, the selectivity of Pd/NC, Cu/NC and PdCu/NC
is also quantified by computing the difference between
thermodynamic limiting potentials for NRR and HER (i.e.,
UL(NRR) – UL(HER), where UL is the minimum potential that
required for all reaction steps being downhill in free-energies)
[39,40]. The results are summarized in Figure 5d, along with
calculated ∆G*H for HER presented in Figure 5a–c. For
PdCu/NC, the UL(NRR) – UL(HER) value (−0.15 V) is more
positive than those for Pd/NC (−0.91 V) and Cu/NC (−0.94 V),
indicating higher selectivity for NRR and inhibition of the
competing HER. Collectively, the enhanced catalytic
performance of PdCu/NC for NRR results from enhanced N2
adsorption, a different reaction mechanism with reduced ∆GPDS
,
as well as the suppression of HER.
Conclusion
A
B
In summary, the incorporation of Cu into single-atom Pd sites to
form diatomic Pd-Cu sites on N-doped carbon is designed and
synthesized as an NRR electrocatalyst. Consequently, the FE
and the yield rate of NH3 are substantially higher than the
counterpart values of individual single-atom Pd catalyst. Our
DFT calculations reveal that the introduction of Cu into Pd/NC
not only moves the pDOS of Pd toward the Fermi level but also
promote the d-2π* coupling between Pd and adsorbed N2. This
renders N2 chemisorption and activates the adsorbed N2 to be
hydrogenation, which result in lowered energy required by the
potential-determining step and eventually enhanced selectivity of
NRR to NH3. These findings may further inspire the design of
efficient single-atom-based electrocatalysts toward NRR.
C
D
Figure 5. Reaction activity and selectivity from DFT calculations. Calculated
free-energy diagrams of NRR and HER on (a) Pd/NC, (b) Cu/NC, and (c)
PdCu/NC systems. The alternating pathway is favored on Pd/NC and Cu/NC,
whereas the distal pathway is the most energetically preferred on PdCu/NC.
The asterisk (*) represents a reactive intermediate or a vacant surface site. (d)
Difference between the limiting potentials for NRR and HER of Pd/NC, Cu/NC,
and PdCu/NC.
Acknowledgements
Work done at UCI was supported by the startup funding of H.L.X.
Work done at TJUT was supported by National Key R&D
Program of China (2017YFA0700104), National Natural Science
Foundation of China (51971157, 21601136, and 51761165012),
and Tianjin Science Fund for Distinguished Young Scholars
(19JCJQJC61800). This research used the 7-BM beamline of
the National Synchrotron Light Source II, a U.S. DOE Office of
Science User Facility operated for the DOE Office of Science by
Brookhaven National Laboratory under Contract No. DE-
SC0012704. P.O and J.S. acknowledge financial support by
Natural Sciences and Engineering Research Council of Canada
(NSERC) Discovery Grants Program (grant #: NSERC RGPIN-
2017-05187), the McGill William Dawson Scholar Award, the
McGill Engineering Doctoral Award, and computing resources
from Compute Canada.
In terms of the subsequent NRR steps, the protonation of
adsorbed *N2 (*N2 + (H+ + e−) → *NNH) is the PDS in both distal
and alternating pathways for Pd/NC and Cu/NC. The *NNH
formation exhibits huge ∆GPDS values for Pd/NC and Cu/NC,
being 1.31 eV and 1.85 eV respectively, effectively rendering
them NRR inactive (Figure 5a,b). Considering the intensive
energy demand of *N2H2 formation (0.72 and 0.74 eV) in the
distal pathway for Pd/NC and Cu/NC, the alternating pathway is
more energetically favorable on these single-atom sites. In
contrast, PdCu/NC shows a much lower the free-energy change
for the *NNH formation (∆G*NNH) , i.e., 0.67 eV, thus
demonstrating a substantial enhancement in catalytic activity
compared to Pd/NC and Cu/NC. As a result of a much reduced
∆G*NNH, the PDS on PdCu/NC is changed from *NNH formation Conflict of interest
to the protonation of *NH2 (*NH2 + (H+ + e−) → *NH3, 0.78 eV),
as shown in Figure 5c. Additionally, it is worth mentioning that
The authors declare no competing interests.
This article is protected by copyright. All rights reserved.

10.1002/anie.202010159
Angewandte Chemie International Edition
RESEARCH ARTICLE
[31] H. Liang, X. Zhuang, S. Brüller, X. Feng, K. Müllen, Nat. Commun.
2014, 5, 4973.
Keywords: ammonia electrosynthesis, active site modulation,
single-atom catalysis, diatomic Pd-Cu site, ambient condition
[32] P. Li, Z. Jin, Y. Qian, Z. Fang, D. Xiao, G. Yu, ACS Energy Lett. 2019, 4
(8), 1793-1802.
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
[9]
Y. Zhao, Y. Zhao, R. Shi, B. Wang, G. I. N. Waterhouse, L.-Z. Wu, C.-H.
Tung, T. Zhang, Adv. Mater. 2019, 31, 1806482.
[33] S. Zhou, L. Shang, Y. Zhao, R. Shi, G. I. N. Waterhouse, Y.-C. Huang,
L. Zheng, T. Zhang, Adv. Mater. 2019, 31, 1900509.
H. Cheng, L.-X. Ding, G.-F. Chen, L. Zhang, J. Xue, H. Wang, Adv.
Mater. 2018, 30, 1803694.
[34] H. Yang, Y. Wu, G. Li, Q. Lin, Q. Hu, Q. Zhang, J. Liu, C. He, J. Am.
Chem. Soc. 2019, 141, 12717-12723.
G. F. Chen, S. Ren, L. Zhang, H. Cheng, Y. Luo, K. Zhu, L. X. Ding, H.
Wang, Small Methods 2018, 3, 1800337.
[35] S. Z. Andersen, V. Čolić, S. Yang, J. A. Schwalbe, A. C. Nielander, J. M.
McEnaney, K. Enemark-Rasmussen, J. G. Baker, A. R. Singh, B. A.
Rohr, M. J. Statt, S. J. Blair, S. Mezzavilla, J. Kibsgaard, P. C. K.
Vesborg, M. Cargnello, S. F. Bent, T. F. Jaramillo, I. E. L. Stephens, J.
K. Nørskov, I. Chorkendorff, Nature 2019, 570, 504-508.
[36] D. Zhu, L. Zhang, R. E. Ruther, R. J. Hamers, Nat. Mater. 2013, 12,
836-841.
X. Ren, J. Zhao, Q. Wei, Y. Ma, H. Guo, Q. Liu, Y. Wang, G. Cui, A. M.
Asiri, B. Li, B. Tang, X. Sun, ACS Central Sci. 2019, 5, 116-121.
Z. Wang, F. Gong, L. Zhang, R. Wang, L. Ji, Q. Liu, Y. Luo, H. Guo, Y.
Li, P. Gao, X. Shi, B. Li, B. Tang, X. Sun, Adv. Sci. 2019, 6, 1801182.
C. Zhao, S. Zhang, M. Han, X. Zhang, Y. Liu, W. Li, C. Chen, G. Wang,
H. Zhang, H. Zhao, ACS Energy Lett. 2019, 4, 377-383.
L. Zhang, L. X. Ding, G. F. Chen, X. Yang, H. Wang, Angew. Chem. Int.
Ed. 2019, 58, 2612-2616.
[37] C. Ling, Y. Ouyang, Q. Li, X. Bai, X. Mao, A. Du, J. Wang, Small
Methods 2018, 3, 1800376.
[38] S. Maintz, V. L. Deringer, A. L. Tchougréeff, R. Dronskowski, J. Comput.
Chem. 2016, 37, 1030-1035.
X. Liu, Y. Jiao, Y. Zheng, M. Jaroniec, S. Z. Qiao, J. Am. Chem. Soc.
2019, 141, 9664-9672.
[39] D. Kim, C. Xie, N. Becknell, Y. Yu, M. Karamad, K. Chan, E. J. Crumlin,
J. K. Nørskov, P. Yang, J. Am. Chem. Soc. 2017, 139, 8329-8336.
[40] C. Shi, H. A. Hansen, A. C. Lausche, J. K. Nørskov, Phys. Chem.
Chem. Phys. 2014, 16, 4720-4727.
J. Zhao, Z. Chen, J. Am. Chem. Soc. 2017, 139, 12480-12487.
[10] X. Guo, J. Gu, S. Lin, S. Zhang, Z. Chen, S. Huang, J. Am. Chem. Soc.
2020, 142, 5709-5721.
[11] X. Guo, S. Huang, Electrochimica Acta 2018, 284, 392-399.
[12] H. Tao, C. Choi, L.-X. Ding, Z. Jiang, Z. Han, M. Jia, Q. Fan, Y. Gao, H.
Wang, A. W. Robertson, S. Hong, Y. Jung, S. Liu, Z. Sun, Chem 2019,
5, 204-214.
[13] Z. Geng, Y. Liu, X. Kong, P. Li, K. Li, Z. Liu, J. Du, M. Shu, R. Si, J.
Zeng, Adv. Mater. 2018, 30, 1803498.
[14] L. Han, X. Liu, J. Chen, R. Lin, H. Liu, F. Lü, S. Bak, Z. Liang, S. Zhao,
E. Stavitski, J. Luo, R. R. Adzic, H. L. Xin, Angew. Chem. Int. Ed. 2019,
58, 2321-2325.
[15] F. Lü, S. Zhao, R. Guo, J. He, X. Peng, H. Bao, J. Fu, L. Han, G. Qi, J.
Luo, X. Tang, X. Liu, Nano Energy 2019, 61, 420-427.
[16] M. Wang, S. Liu, T. Qian, J. Liu, J. Zhou, H. Ji, J. Xiong, J. Zhong, C.
Yan, Nat. Commun. 2019, 10, 341.
[17] J. Li, S. Chen, F. Quan, G. Zhan, F. Jia, Z. Ai, L. Zhang, Chem 2020, 6,
1-17.
[18] Q. Qin, T. Heil, M. Antonietti, M. Oschatz, Small Methods 2018, 2,
1800202.
[19] W. Zang, T. Yang, H. Zou, S. Xi, H. Zhang, X. Liu, Z. Kou, Y. Du, Y. P.
Feng, L. Shen, L. Duan, J. Wang, S. J. Pennycook, ACS Catal. 2019, 9,
10166-10173.
[20] Y. Chen, R. Guo, X. Peng, X. Wang, X. Liu, J. Ren, J. He, L. Zhuo, J.
Sun, Y. Liu, Y. Wu, J. Luo, ACS Nano 2020, 14, 6938-6946.
[21] C. Ling, X. Niu, Q. Li, A. Du, J. Wang, J. Am. Chem. Soc. 2018, 140,
14161-14168.
[22] J. Wang, L. Yu, L. Hu, G. Chen, H. Xin, X. Feng, Nat. Commun. 2018, 9,
1795.
[23] H. Huang, F. Gong, Y. Wang, H. Wang, X. Wu, W. Lu, R. Zhao, H.
Chen, X. Shi, A. M. Asiri, T. Li, Q. Liu, X. Sun, Nano Res. 2019, 12,
1093-1098.
[24] W. Tong, B. Huang, P. Wang, L. Li, Q. Shao, X. Huang, Angew. Chem.
Int. Ed. 2020, 59, 2649-2653.
[25] X. Cui, C. Tang, Q. Zhang, Adv. Energy Mater. 2018, 8, 1800369.
[26] G. Deng, T. Wang, A. A. Alshehri, K. A. Alzahrani, Y. Wang, H. Ye, Y.
Luo, X. Sun, J. Mater. Chem. A 2019, 7, 21674-21677.
[27] M.-M. Shi, D. Bao, S.-J. Li, B.-R. Wulan, J.-M. Yan, Q. Jiang, Adv.
Energy Mater. 2018, 8, 1800124.
[28] F. Pang, Z. Wang, K. Zhang, J. He, W. Zhang, C. Guo, Y. Ding, Nano
Energy 2019, 58, 834-841.
[29] L. Han, S. Song, M. Liu, S. Yao, Z. Liang, H. Cheng, Z. Ren, W. Liu, R.
Lin, G. Qi, X. Liu, Q. Wu, J. Luo, H. Xin, J. Am. Chem. Soc. 2020, 142,
12563-12567.
[30] K. Eskandari, F. Ebadinejad, Mol. Phys. 2018, 116 (10), 1369-1376.
This article is protected by copyright. All rights reserved.

10.1002/anie.202010159
Angewandte Chemie International Edition
RESEARCH ARTICLE
RESEARCH ARTICLE
We have designed an efficent NRR
catalyst with diatomic Pd-Cu sites on
N-doped porous carbon by modulation
of single-atom Pd sites with Cu. The
catalyst achieves high Faradaic
Lili Han,† Zhouhong Ren,† Pengfei Ou,†
Hao Cheng, Ning Rui, Lili Lin, Xijun Liu*,
Longchao Zhuo, Jun Song*, Jiaqiang
Sun, Jun Luo, and Huolin L. Xin*
Page No. – Page No.
efficiency and a desirable yield rate for
NH3 electrosynthesis under ambient
conditions, outperforming the
Modulating single-atom Pd sites with
Cu for enhanced ambient ammonia
electrosynthesis.
individual single-atom counterparts.
This article is protected by copyright. All rights reserved.