Vacancy Engineering of Iron-Doped W18O49 Nanoreactors for Low-Barrier Electrochemical Nitrogen Reduction

Article abstract of DOI:10.1002/anie.202002029

The electrochemical nitrogen reduction reaction (NRR) is a promising energy-efficient and low-emission alternative to the traditional Haber–Bosch process. Usually, the competing hydrogen evolution reaction (HER) and the reaction barrier of ambient electrochemical NRR are significant challenges, making a simultaneous high NH₃ formation rate and high Faradic efficiency (FE) difficult. To give effective NRR electrocatalysis and suppressed HER, the surface atomic structure of W₁₈O₄₉, which has exposed active W sites and weak binding for H₂, is doped with Fe. A high NH₃ formation rate of 24.7 μg h^?1 mg_cat^?1 and a high FE of 20.0 % are achieved at an overpotential of only ?0.15 V versus the reversible hydrogen electrode. Ab initio calculations reveal an intercalation-type doping of Fe atoms in the tunnels of the W₁₈O₄₉ crystal structure, which increases the oxygen vacancies and exposes more W active sites, optimizes the nitrogen adsorption energy, and facilitates the electrocatalytic NRR.

Full text of DOI:10.1002/anie.202002029

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Vacancy Engineering of Fe-doped W18O49 Nanoreactors for
Low-barrier Electrochemical Nitrogen Reduction
Authors: Yueyu Tong, Haipeng Guo, Daolan Liu, Xiao Yan, Ji Liang,
Panpan Su, Si Zhou, Jian Liu, Gao Qing (Max) Lu, and Shi
Xue Dou
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.202002029
Angew. Chem. 10.1002/ange.202002029
Link to VoR: http://dx.doi.org/10.1002/anie.202002029
http://dx.doi.org/10.1002/ange.202002029

10.1002/anie.202002029
Angewandte Chemie International Edition
RESEARCH ARTICLE
Vacancy Engineering of Fe-doped W18O49 Nanoreactors for Low-barrier
Electrochemical Nitrogen Reduction
Yueyu Tong, Haipeng Guo, Daolan Liu, Ji Liang*, Xiao Yan, Panpan Su, Si Zhou*, Jian Liu*,
Gao Qing (Max) Lu and Shi Xue Dou
Y.Y. Tong, Dr. H.P. Guo, D.L. Liu, Dr. J. Liang, Dr. X. Yan, Dr. S. Zhou, Prof. S. X. Dou
Institute for Superconducting & Electronic Materials, Australian Institute of Innovative Materials, University of Wollongong, North Wollongong, NSW 2500, Australia
E-mail: liangj@uow.edu.au
A/Prof. P.P. Su, Prof. J. Liu
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences, 457 Zhongshan Road, Dalian 116023, P. R. China
Prof. J. Liang
Key Laboratory for Advanced Ceramics and Machining Technology of Ministry of Education, School of Materials Science and Engineering, Tianjin University,
Tianjin 300350, China
Dr. X. Yan
Guangdong Key Laboratory of Membrane Materials and Membrane Separation, Guangzhou Institute of Advanced Technology, Chinese Academy of Sciences,
Guangzhou, 511458, China
Dr. S. Zhou
Key Laboratory of Materials Modification by Laser, Ion and Electron Beams (Dalian University of Technology), Ministry of Education, Dalian 116024, China
Email: siz@uow.edu.au
D.L. Liu, Prof. J. Liu, Prof. G. Q. Lu
DICP-Surrey Joint Centre for Future Materials, Department of Chemical and Process Engineering, and Advanced Technology Institute, University of Surrey,
Guilford, Surrey, GU2 7XH, UK
E-mail: jian.liu@surrey.ac.uk; jianliu@dicp.ac.cn
Supporting information and the ORCID identification number(s) for the author(s) of this article can be found under https://doi.org/.
Abstract: Electrochemical nitrogen reduction reaction (NRR) is a
promising energy-efficient and low-emission alternative to the
traditional Haber-Bosch method. Usually, the competing hydrogen
evolution reaction (HER) and reaction barrier of ambient
electrochemical NRR are most significant challenges, making the
simultaneous achievement of a high NH3 yielding rate and a high
Faradic efficiency (FE) extremely difficult for NRR. To address this
issue, W18O49, with exposed W sites and intrinsically weak binding for
H2, is doped by Fe to modify its surface atomic structure for effective
and large-scale Haber-Bosch plants is time, money, and space-
consuming. In contrast, direct electroreduction of nitrogen to
ammonia under ambient conditions, which was firstly
demonstrated by Stoukides and coworkers,[3a] presents
a
promising strategy to replace the traditional thermal catalysis
process with reduced emission, lower energy consumption, and
higher economic viability.[4] However, the major bottleneck of
electroreduction of nitrogen shifts to the competitive adsorption
between hydrogen ions and nitrogen molecules in the aqueous
environment and the activation of the ultra-stable N≡N bond with
a binding energy up to 940.95 kJ mol−1.[5]
NRR electrocatalysis and suppressed HER. On it, a high NH3 yielding
−1
cat.
rate of 24.7 μg h−1 mg
and a high FE of 20.0% have been
simultaneously gained at a very low overpotential of −0.15 V vs.
reversible hydrogen electrode. Ab initio reveals an intercalation-type
doping of Fe atoms in the tunnels of the W18O49 crystal structure,
which increases the oxygen vacancies for exposing more W active
sites, optimizes the nitrogen adsorption energy, and facilitates the
electrocatalytic NRR. Thus, this work will provide a rational design of
electrocatalysts for various electrochemical processes.
Addressing at this issue, various catalysts, especially the
noble-metal-based ones, have been designed for the
electrochemical nitrogen reduction reaction (NRR) under ambient
conditions.[6] However, their high costs and limited reserves
largely impede the broad application of those noble-metal-based
catalysts.[7] Apart from the noble-metal catalysts, other delicately
designed nanoreactors, in which the reactants could be
enriched/preassembled and the intermediate processes/products
could be stabilized, have also been studied.[8a-b] However, the
intrinsic hydrogen adsorption preference of these catalysts has
greatly restrained their nitrogen adsorption capability, let alone
the following electro-catalysis process.[8c-d] Thus, exploring
catalysts with a high selectivity for nitrogen adsorption is the
prerequisite for achieving highly efficient NRR electrocatalysis
with good economic feasibility for practical applications.
Introduction
Ammonia is an indispensable precursor for producing fertilizers,
synthetic fibers, and pharmacies to satisfy the needs from the
growing population and economy worldwide.[1] More importantly,
due to its high liquefied energy density and carbon-free nature,
ammonia has also been regarded as an ideal storage
intermediate for H2 and a promising carbon-free fuel.[2] Generally,
the production of ammonia is associated with 1.44% of global CO2
emission via the traditional Haber-Bosch process, which needs a
high pressure of over 200 atm and a high temperature of over
500 °C.[3] Apart from this, the construction of the highly centralized
With this regard, a number of pioneering works have been
conducted to overcome this difficulty, such as widening the
[9]
potential gap between NRR and HER
and designing active
centers with intrinsic preference for nitrogen adsorption[4a]
Although either superior Faradic Efficiency (FE) or an
.
a
outstanding ammonia yield rate has been achieved on these
catalysts, to simultaneously achieve these two merits is still
1
This article is protected by copyright. All rights reserved.

10.1002/anie.202002029
Angewandte Chemie International Edition
RESEARCH ARTICLE
significantly challenging. Furthermore, reaching this goal at a low Results and Discussion
overpotential for a more energy-saving electrocatalytic NRR is
even more challenging. This is, to some extent, resulted from the
limited nitrogen adsorption sites, the sluggish NRR kinetics, and
the dominating HER at high overpotentials.
The solvothermal fabrication of the material is illustrated in
Figure 1a, S1. By tuning the ratio of WCl6 and FeCl3, the doping
level in the obtained W18O49 can be tuned (denoted as W18O49-
xFe @ CFPs, x=4, 8, 16, 32, x was the designed mole percentage
of Fe). Fe-doped W18O49 without CPF was also prepared. The
color of W18O49 changes from navy blue to grey green and finally
to brown with Fe content increasing, clearly suggesting the
successful doping of Fe in W18O49 by this process (Figure S2).
Details about the fabrication process can be found in the
Considering this, defect-rich W18O49, the only nonstoichiometric
oxide compositions of tungsten that can be isolated in a pure
phase,[10] might be an ideal candidate due to its intrinsically weak
binding with hydrogen.[11] In this case, however, the main
challenge turns to be: (1) maximizing the exposure of possible
active sites (i.e., the W sites) and (2) the activation of N≡N to
accelerate the sluggish NRR kinetics. For the former, the inherent
oxygen vacancies (Ovac) in W18O49 have already laid a good
foundation for disclosing more tungsten and facilitating the
cleavage of nitrogen triple bond.[12] While for the latter, surface
chemical modification (e.g., heteroatoms doping and/or defect
engineering) has long been considered as a high-efficiency
method for modulating the surface electronic structures of the
electrochemical catalysts, which has been proven to be able to
significantly enhance catalysts’ activity for a range of applications,
including oxygen reduction, oxygen evolution, hydrogen evolution,
and nitrogen reduction.[13]
For the defect-rich W18O49, the abundant tunnels in its lattice
structure make it ideal for such modifications, which may generate
enhanced NRR performance in return. Enlightened by the
biological nitrogenase, iron (Fe) may enhance both the adsorption
and activation of nitrogen molecules due to their strong interaction
with nitrogen. In addition, as a transition metal element located in
the middle part of the periodic table, the unoccupied d orbitals of
Fe endow it both identities of electron-acceptor and electron-
donator.[14] More importantly, comparing to the d-spacing of
W18O49 (3.8 ꢀÅ), the relatively smaller radius of Fe atoms (1.2 Å)
would greatly facilitate its intercalation into W18O49.[15] Thus, the
Fe-doping of W18O49 should not only modify its electronic structure
but also tune the intrinsic covalency of its metal-oxygen bond,
resulting in the refinement of the defect statement, improvement
of the conductivity, and enhancement of N2 adsorption capability.
Based on these considerations, we herein for the first time
report a Fe-doped and defect-rich W18O49 nanowire, which is
grown on carbon fiber papers (CFPs), as a freestanding and high-
performance electrocatalyst for NRR. By simply changing the Fe-
doping level, the defect status of W18O49 can be finely tuned,
which results in the optimized nitrogen adsorption, suppressed
hydrogen evolution, and enhanced charge transfer for
electrocatalytic NRR. Due to these multi-synergetic effects, a
greatly improved electrocatalytic NRR performance has been
achieved. The W18O49-16Fe @ CFP simultaneously acquires a
supplementary information.
Figure 1. (a) Schematic illustration of the preparation of Fe-doped W18O49
nanowires @ CFP. SEM images of (b-c) CFP, (d-e) W18O19-16Fe nanowires @
CFP.
After the formation of W18O49, a large number of thorn-like
structures can be observed on the surface of CFP, which is
originally smooth (Figure 1b, c). High magnification images
further show those thorn-like structures are nanowires with a
length of several micrometers, homogeneously tangling on the
CFP (Figure 1d, e). With the variation of the designed Fe-doping
concentration from 0 to 32%, the thorn-like structure on CFP
gradually aggregates, but the homogeneous nanowires still
remain (Figure S3a-h). This is in good agreement with the
transmission electron microscopy (TEM) observation, which also
reveals the homogeneous nanowire architecture of Fe-doped
W18O49 (Figure 2a, S3i-l). Besides, the select-area electron
diffraction (SAED) patterns of the nanowires also confirm their
crystalline nature along the (010) direction, regardless of the Fe-
doping level (Figure 2b inset and S3i-l), indicating the Fe-doping
does not change the crystal structure of the W18O49 host, which is
beneficial for the stable NRR electrocatalysis.
The microstructure of these Fe-doped W18O49 nanowires was
then studied using high-angle annular dark-field scanning
transmission electron microscope (HAADF-STEM). These
nanowires, with diameters of less than 5 nm, clearly show lattice
fringes of ~0.38 nm, which belong to (010) planes of W18O49 and
further verifies their preference growth along the [010] direction
(Figure 2b, S3m-p). Notably, Fe-doping does not induce obvious
change of the lattice parameter of W18O49 according to both SAED
and HAADF-STEM measurements, indicating the robustness of
W18O49 structures upon Fe-doping. The XRD patterns of the Fe-
doped W18O49 nanowires, which were scraped off from CFP, show
monoclinic W18O49 phase, with characteristic peaks at around 23°
high ammonia yield rate (24.7 μg h−1 mg ) and a high FE of
-1
cat.
20.0% at −0.15 V, which are much superior to those un-modified
counterparts and most previous reported non-noble-metal
catalysts. Especially, the quite low overpotential (−0.15 V) further
improved the energy efficiency of electrocatalytic NRR, making it
more energy-saving. Ab initio calculations further disclose that it
is the Fe atoms, which are intercalated in the tunnels of W18O49,
that lead to the redistribution surface electrons of W18O49 to
substantially promote its N2 fixation ability and minimize the NRR
overpotentials. Thus, our work will blaze a new trial for both the
rational design of catalysts and the elaborate clarification of
mechanism for NRR at ambient conditions.
2
This article is protected by copyright. All rights reserved.

10.1002/anie.202002029
Angewandte Chemie International Edition
RESEARCH ARTICLE
To further confirm the successful doping of Fe element into
W18O49, energy dispersion spectrum (EDS) -based elemental
mappings and Fe-L2,3 edge EELS spectra of the doped nanowires
were obtained (Figure 3a-c, S6), which show homogeneous
distributions of W, O, and Fe without any other elements detected.
With the increasing amount of FeCl3 in the precursor, the Fe
content in the doped W18O49 nanowires also increases, as
confirmed by both EDS and XPS results (Table S2). The Fe/W
ratio acquired from HAADF-STEM-EDS is quite close to that in
the FeCl3 and WCl6 precursors, indicating that the Fe-doping
concentration in W18O49 can be precisely controlled. On the other
hand, the Fe/W ratio derived from XPS spectra is much higher
than the EDS results, which implies that the Fe dopants mostly
exist in the near-surface region of W18O49 nanowires, more
efficient for the surface modification to participate in the NRR
electrocatalysis.
X-ray photoelectron spectroscopy (XPS) was then applied to
further identify the chemical states of the elements in the
synthesized samples. The XPS survey scan confirms the
presence of W, Fe, O, and C elements in all the Fe-doped W18O49
@ CFP analogues (Figure S7a). The deconvoluted high-
resolution W 4f spectra of pristine and W18O49-16Fe nanowires
suggests both W5+ and W6+ species present in the samples
(Figure 3d, Figure S7b), which is consistent with the distorted
edge-sharing crystal structure of W18O49 forming low-valence W
elements and metallic W-W interaction in the lattice.[16] Compared
with the pristine W18O49, the W18O49-16Fe showed an increased
W5+/W6+ ratio from 1.2 to 1.7 and a clear shift to the lower binding
energy (Table S3), indicating the partial reduction of W6+ into W5+
Figure 2. (a) TEM image, (b) HAADF-STEM images of W18O19-16Fe nanowires
@ CFP. Insert is the corresponding SAED pattern. (c) XRD patterns of W18O19
-
16Fe nanowires and pure W18O49 nanowires. (d) enlarged XRD patterns of
W18O19-16Fe nanowires and pure W18O49 nanowires in the 2θ=22°-26° region.
and 47° that can be ascribed to (010) and (020) facets (JCPDS
No. 05-4539, Figure 2c-d and S4), respectively. Interestingly,
with increasing Fe-doping level from 0 to 32%, a small shift of the
(010) peak towards a lower degree can be observed (Figure 2d,
S4b-c, and Table S1), indicating the very slight expansion of the
monoclinic structure of W18O49 due to the Fe-doping, which will be
discussed later (Figure S5).
associated with the increase of Ovac.
[11, 17] On the other hand, for
the Fe 2p spectra, it can be deconvoluted into four peaks at
around 710.5, 712.3, 723.8 and 725.6 eV (Figure 3e, Figure S7c),
corresponding to the Fe2+ 2p3/2, Fe3+ 2p3/2, Fe2+ 2p1/2, and Fe3+
2p1/2 states, respectively.[17-18] Besides, the additional pre-peak in
the low-binding-energy region (708.7 eV) accounts for the Fe ions
with a lower oxidation state than Fe2+, further revealing the
existence of Ovac in the neighboring sites of Fe dopants.[18]
Electron paramagnetic resonance (EPR) spectroscopy was
used to investigate the unpaired electrons in the synthesized
samples, the amount of which is proportional to the Ovac content
(Figure 3f). Compared with the pure W18O49, the increased EPR
intensity (i.e., a higher unpaired electron/ Ovac content) of the Fe-
doped counterparts can be attributed to Fe atoms located next to
the charge-compensating oxygen vacancies, resulting in the
lower-oxidation-state of Fe atoms.[17, 19] The highest EPR intensity
of W18O49 means the most intense charge-compensating effect,
indicating the highest concentration of Ovac in it, which is essential
to generate more active sites for NRR catalysis as will be
discussed later (Figure 3f). Moreover, the absent signal of free
electrons in the region of mT > 180 indicates that electrons tend
to be captured by Fe ions rather than delocalization within the
lattice of W18O49, suggesting an ‘electron reservoir’ role of the Fe
dopants.[20] This ‘electron reservoir’ is believed to be of great
benefit for the hydrogenation of nitrogen during the ammonia
production process.
The ambient electrocatalytic NRR performance of the as-
prepared samples was investigated in a N2-saturated 0.25 M
LiClO4 aqueous solution and the ammonia yield was determined
by the salicylic acid method (Figure S8, S9-S13). In general, the
ammonia yield rate and FE for all the Fe doped-W18O49 @ CFP
catalysts gradually increase with the negative shift of testing
Figure 3. (a-b) EDS spectrum and STEM-EDS images of W18O49-16Fe
nanowires. (c) Fe-L2, spectra of as-prepared Fe-doped W18O49 nanowires.
3
High-resolution (d) W 4f and (e) Fe 2p XPS spectra of W18O49-16Fe nanowires
@ CFP. (f) EPR spectra of all Fe-doped W18O49 nanowires and pure W18O49
nanowires at −150 °C.
3
This article is protected by copyright. All rights reserved.

10.1002/anie.202002029
Angewandte Chemie International Edition
RESEARCH ARTICLE
potential till the peak value during the NRR tests. After that,
decrease in FE and NH3 yield rate occur due to the hydrogen
adsorption surpassing the nitrogen adsorption on the surface of
catalysts.[9] Among them, the 16% Fe-doped one showed both the
largest NH3 yield rate of 24.7 μg h−1 mgc−a1t. and the highest FE of
20.0% at −0.15 V vs. RHE, which are 12 times and 2 times higher
than the pristine W18O49 @ CFP catalyst, respectively (Figure 4a-
b). Remarkably, overpotential of −0.15 V is much lower than the
values of most reported electrocatalysts for NRR by far;[4a, 7a, 7b, 21]
and at this overpotential, the W18O49-16Fe @ CFP simultaneously
achieves the highest ammonia yield rate and FE with neither
feature sacrificed (Table S4). As the Fe-doping concentration
Apart from the NRR activity, the durability is another critical
quality for electrocatalysts. For the W18O49-16Fe @ CFP material,
a high ammonia yield rate retention up to 91% was obtained after
five continuous electrolysis cycles without any obvious decay in
its efficiency, clearly demonstrating its excellent stability for NRR
electrocatalysis (Figure 4d, S17a, b). Furthermore, the very
stable NRR current over 20 h for the catalyst also verifies its
superior electrocatalytic durability (Figure S17c). This stable
electrochemical behavior also agrees well with intact
microstructure of the catalyst after the long-term test (Figure S18).
Little changes were observed in the XPS spectrum of W18O49-
16Fe @ CFP catalyst after the durability test, further verifying its
excellent stability for NRR electrocatalysis in aqueous solution.
(Figure S19).
further increased to 32%, the ammonia yield rate and FE
−1
significantly dropped to 1.72 μg h−1 mg and 7.6%, which
cat.
indicates the optimized electrocatalytic performance of W18O49 at
an appropriate Fe doping level. The selective NH3 production
against hydrazine of all the Fe-doped W18O49 @ CFP samples
were also evaluated by the Watt and Chrisp methods (Figure S14,
S15), and no hydrazine was detected (Figure S15), which further
demonstrates the excellent selectivity of the as-prepared
catalysts.
Figure 4. (a) NH3 yield rate at various potentials vs. RHE. (b) Faradic efficiency
of as-prepared Fe-doped W18O49 nanowires @ CFP electrodes. (c) UV-vis
absorption spectra of various electrolytes stained with salicylic acid indicator.
OCV stands for open-circuit voltage. (d) cycling stability results at −0.15 V (vs.
RHE) of W18O49-16Fe nanowires @ CFP electrode.
Figure 5. (a) N2 adsorption energy (ΔEN2*) as a function of Ovac : W ratio for
W18O49 with and without Fe doping by DFT calculations. (b) The differential
charge density between Fe and W18O49 with Ovac : W ratio of 2 : 36 without N2
chemisorption. Purple and orange colors represent electron accumulation and
depletion regions, respectively, with isosurface value of 5×10−3 eÅ−3. (c, d) Free
energy diagrams of NRR on Fe-doped W18O49 with Ovac : W ratio of 1 : 36 and
2 : 36, respectively. The structures of reaction intermediates are shown next to
their energy segments. Left bottom of (c, d) inserted the differential charge
density distributions of N2 chemisorbed on Fe-doped W18O49 with Ovac : W ratio
of 1 : 36 and 2 : 36, respectively. The Fe/W ratio is fixed at 11%.
The origin of the produced ammonia was then verified to
eliminate any possible interference from contaminants (Figure
4c). Specifically, when the catalyst was placed in the electrolyte
at open-circuit voltages (OCVs) for 10 h, no ammonia could be
detected in either N2- or Ar-saturated electrolytes, confirming no
ammonia contamination from the gases or the catalysts used in
this work. When N2 was replaced by Ar for NRR electrocatalysis
over the samples, no apparent ammonia production was detected,
proving that the ammonia is produced from the feeding N2 (Figure
S16). Moreover, when a bare CFP was used as the catalytic
electrode, no apparent ammonia production was detected either,
further confirming that the ammonia was produced from the NRR
electrocatalysis over the Fe-doped W18O49, rather than from the
CFP electrocatalysis. On this basis, it is evidenced that the
excellent ammonia yield rate and FE are due to the NRR
electrocatalysis over the W18O49-16Fe @ CFP catalyst.
To better understand the roles of Fe-doping and Ovac on the
electrocatalytic NRR activity of W18O49, DFT calculations were
carried out. A slab model of W18O49 (100) surface was adopted
based on the HAADF-STEM results, with different numbers of
Ovac being considered. As W18O49 possesses pore-like channels,
Fe atoms might either intercalate into the channel or substitute
the W atoms. Our calculations show that intercalation of Fe atom
into the W18O49 channel is both energetically and kinetically
favorable, with a low kinetic barrier of 0.37 eV and heat of
4
This article is protected by copyright. All rights reserved.

10.1002/anie.202002029
Angewandte Chemie International Edition
RESEARCH ARTICLE
formation of 1.43 eV (Figure S20). Intercalation at the Fe/W ratio Conclusion
of 11% (close to the EDS tested experimental doping
concentration 12%) leads to slight expansion of lattice parameters
of W18O49 below 1% (Figure S21a), which is consistent with the
TEM and XRD observations (Figure 2b, c). Therefore, it is
reasonable to deduce that the Fe dopants are most probably
intercalated in the channels of W18O49, which has also been
reported in the previous works.[15a, 22]
In summary, Fe-doped W18O49 nanowires have been
successfully fabricated on carbon fiber papers, through a facile
hydrothermal method, for high-performance electrocatalysis of
NRR. The W18O49-16Fe @ CFP nanoreactor achieved the highest
electrocatalytic NRR performance, in terms of a high ammonia
−1
cat.
yield rate of 24.7 μg h−1 mg
and high Faradic efficiency of
Such Fe intercalation can remarkably enhance the binding
capability of the W atoms in W18O49 to N2 molecules (Figure 5a).
20.0% at a very low over potential of −0.15 V vs. RHE. Our
experimental characterization combined with DFT calculations
reveal that it is the Fe intercalation strategy and the modified Ovac
state that simultaneously optimizes the electronic states and
surface properties of W18O49, leading to the enhanced binding
capability for nitrogen fixation-more active sites, the higher
electrical conductivity, the better nitrogen adsorption strength,
and the low-barrier NRR pathway. Thus, this work not only
provides an outstanding electrocatalyst for NRR, but also, more
importantly, opens a new avenue for modulating the electronic
and catalytic properties of nanoreactors for various
electrochemical processes.
For the undoped W18O49 with different designed levels of Ovac
,
only very small adsorption energies (ΔEN2*, relative to free N2) of
−0.10 to 0.20 eV can be achieved, which means N2 molecule can
only be very weakly adsorbed on their surface. In contrast, when
Fe atoms are doped on them, the N2 adsorption strength is
remarkably enhanced with ΔEN2* of −0.84 to −0.99 eV. Moreover,
when absorbed on the Fe-doped W18O49, the N≡N bond length of
the N2 molecules can also be significantly elongated from 1.10 Å
for free N2 molecule to 1.25 Å, suggesting the successful
activation of N2 molecule and high surface activity of the material
(Figure S21b-c). Note that we also considered substitutional
doping of Fe into W18O49, which, however, greatly weakens the
binding capability of W18O49 with N2 with regard to the pristine
system (Figure S22), and contradict the observed enhanced
activity of Fe-doped W18O49 in experiment.
The full reaction pathways of NRR over the Fe-doped W18O49
with different Ovac levels were then explored (Figure 5c, d). At a
low Ovac level (i.e., Ovac/W = 1/36) with fewer W sites exposed, on
which N2 can only be adsorbed at the W sites with the end-on
configuration, NRR occurs via the alternating pathway, involving
a small potential step of 0.20 eV during the formation of HN2*
species. At a higher Ovac level, N2 molecules favor the side-on
adsorption and NRR proceeds through the enzymatic pathway. In
this case, the whole reaction is almost barrierless except for the
desorption of NH3, in good accord with the extremely low
overpotential measured in experiment. Meanwhile, HER is also
largely inhibited due to the weaker binding of H* species on the
W sites in comparison with N2 (the adsorption strength of H* is
0.60 eV less than that for N2). Therefore, the Fe-doping and
presence of oxygen vacancies in W18O49 both contribute to the
enhanced N2 fixation and favorable reduction.
Deeper insights can be obtained from the differential charge
density distributions and electronic density of states (DOS)
presented in Figure 5b, left bottom illustrations in Figure 5c, d,
and Figure S23, respectively. Prominent electron transfer occurs
from Fe atoms to W18O49 (Figure 5b, left bottom illustrations in
Figure 5c, d). According to the Bader charge analysis, each W
atom gains 0.04 e from Fe dopant, and thus exhibits enhanced
binding capability with N2 compared to that of the pristine W18O49.
Moreover, the electronic states induced by Fe-doping fill the
original gap of W18O49 at around −1.0 eV in the DOS (Figure S23),
which may be beneficial to improve the electronic conductivity
during NRR. On the other aspect, the presence of Ovac not only
helps expose more W active centers for NRR, but also affects the
work function of W18O49. A higher Ovac level leads to the reduction
of work function and the approaching of W18O49’s Fermi level
towards the π* state of N2 molecule (−1.89 eV relative to vacuum),
which will facilitate the backdonation of electrons from W18O49 to
N2 for NRR. Consequently, these two factors synergistically
contribute to the enhanced NRR activity for the Fe-doped W18O49
nanowires.
Acknowledgements
Dr. X. Yan thanks China Scholarship Council for supporting his
visit to the University of Wollongong (No. 201804910018). This
work was supported by NCI supercomputer resources, National
Natural Science Foundation of China (No. 51502045 and
21905202), Dalian National Laboratory for Clean Energy (DNL),
CAS, DNL Cooperation Fund, CAS (DNL180402), and Australian
Research Council (ARC) through Discovery Early Career
Researcher Award (DECRA, No. DE170100871).
Keywords: Vacancy engineering • nitrogen reduction reaction •
electrocatalysts • tungsten oxide • nanoreactor
[1] a) I. Čorić, B. Q. Mercado, E. Bill, D. J. Vinyard, P. L. Holland, Nature 2015,
526, 96; b) T. Spatzal, K. A. Perez, O. Einsle, J. B. Howard, D. C. Rees, Science
2014, 345, 1620-1623; c) G.-F. Chen, S. Ren, L. Zhang, H. Cheng, Y. Luo, K.
Zhu, L.-X. Ding, H. Wang, Small Methods 2019, 3, 1800337.
[2] a) V. Rosca, M. Duca, M. T. de Groot, M. T. M. Koper, Chem. Rev. 2009,
109, 2209-2244; b) S. Licht, B. Cui, B. Wang, F.-F. Li, J. Lau, S. Liu, Science
2014, 345, 637-640; c) A. Klerke, C. H. Christensen, J. K. Nørskov, T. Vegge,
J. Mater. Chem. 2008, 18, 2304-2310.
[3] a) G. Soloveichik, Nat. Catal. 2019, 2, 377-380; b) B. H. R. Suryanto, H.-L.
Du, D. Wang, J. Chen, A. N. Simonov, D. R. MacFarlane, Nat. Catal. 2019, 2,
290-296.
[4] a) Y. Wang, M.-M. Shi, D. Bao, F.-L. Meng, Q. Zhang, Y.-T. Zhou, K.-H. Liu,
Y. Zhang, J.-Z. Wang, Z.-W. Chen, D.-P. Liu, Z. Jiang, M. Luo, L. Gu, Q.-H.
Zhang, X.-Z. Cao, Y. Yao, M.-H. Shao, Y. Zhang, X.-B. Zhang, J. G. Chen, J.-
M. Yan, Q. Jiang, Angew. Chem., Int. Ed. 2019, 58, 9464-9469; b) J. G. Chen,
R. M. Crooks, L. C. Seefeldt, K. L. Bren, R. M. Bullock, M. Y. Darensbourg, P.
L. Holland, B. Hoffman, M. J. Janik, A. K. Jones, M. G. Kanatzidis, P. King, K.
M. Lancaster, S. V. Lymar, P. Pfromm, W. F. Schneider, R. R. Schrock, Science
2018, 360, eaar6611.
[5] D. Bao, Q. Zhang, F.-L. Meng, H.-X. Zhong, M.-M. Shi, Y. Zhang, J.-M. Yan,
Q. Jiang, X.-B. Zhang, Adv. Mater. 2017, 29, 1604799.
[6]
X. Yan, D. Liu, H. Cao, F. Hou, J. Liang, S. X. Dou, Small Methods 2019,
3, 1800501.
[7] a) M.-M. Shi, D. Bao, S.-J. Li, B.-R. Wulan, J.-M. Yan, Q. Jiang, Adv. Energy
Mater. 2018, 8, 1800124; b) 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; c) M.-M. Shi, D. Bao, B.-R. Wulan, Y.-H. Li, Y.-F.
5
This article is protected by copyright. All rights reserved.

10.1002/anie.202002029
Angewandte Chemie International Edition
RESEARCH ARTICLE
Zhang, J.-M. Yan, Q. Jiang, Adv. Mater. 2017, 29, 1606550; d) J. Wang, L. Yu,
L. Hu, G. Chen, H. Xin, X. Feng, Nat. Commun. 2018, 9, 1795.
[8] a) H. Tian, J. Liang, J. Liu, Adv. Mater. 2019, 31, 1903886; b) Y.-Y. Tong, X.
Yan, J. Liang, S. X. Dou, Small 2019, 1904126; c) C. Guo, J. Ran, A. Vasileff,
S.-Z. Qiao, Energy Environ. Sci. 2018, 11, 45-56; d) J. K. Nørskov, T. Bligaard,
A. Logadottir, J. R. Kitchin, J. G. Chen, S. Pandelov, U. Stimming, J.
Electrochem. Soc. 2005, 152, J23-J26.
[9] M. Wang, S. Liu, T. Qian, J. Liu, J. Zhou, H. Ji, J. Xiong, J. Zhong, C. Yan,
Nat. Commun. 2019, 10, 341.
[10] C. Guo, S. Yin, M. Yan, M. Kobayashi, M. Kakihana, T. Sato, Inorganic
Chemistry 2012, 51, 4763-4771.
[11] X. Zhong, Y. Sun, X. Chen, G. Zhuang, X. Li, J.-G. Wang, Adv. Funct. Mater.
2016, 26, 5778-5786.
[12] a) Y. Zhao, Y. Zhao, G. I. N. Waterhouse, L. Zheng, X. Cao, F. Teng, L.-Z.
Wu, C.-H. Tung, D. O'Hare, T. Zhang, Adv. Mater. 2017, 29, 1703828; b) H. Li,
J. Shang, Z. Ai, L. Zhang, J. Am. Chem. Soc. 2015, 137, 6393-6399.
[13] a) Q. Wang, Y. Lei, D. Wang, Y. Li, Energy Environ. Sci. 2019, 12, 1730-
1750; b) S. Wang, F. Ichihara, H. Pang, H. Chen, J. Ye, Adv. Funct. Mater. 2018,
28, 1803309; c) D. Yan, H. Li, C. Chen, Y. Zou, S. Wang, Small Methods 2019,
3, 1800331.
[14] C. Ling, Y. Ouyang, Q. Li, X. Bai, X. Mao, A. Du, J. Wang, Small Methods
2019, 3, 1800376.
[15] a) M. R. Thalji, G. A. M. Ali, H. Algarni, K. F. Chong, J. Power Sources 2019,
438, 227028; b) L.-J. Guo, G.-F. Zhao, Y.-Z. Gu, X. Liu, Z. Zeng, Phys. Rev. B
2008, 77, 195417.
[16] N. Zhang, A. Jalil, D. Wu, S. Chen, Y. Liu, C. Gao, W. Ye, Z. Qi, H. Ju, C.
Wang, X. Wu, L. Song, J. Zhu, Y. Xiong, J. Am. Chem. Soc. 2018, 140, 9434-
9443.
[17] T. Wu, Z. Xing, S. Mou, C. Li, Y. Qiao, Q. Liu, X. Zhu, Y. Luo, X. Shi, Y.
Zhang, X. Sun, Angew. Chem., Int. Ed. 2019, 58, 18449.
[18] A. P. Grosvenor, B. A. Kobe, M. C. Biesinger, N. S. McIntyre, Surf. Interface
Anal. 2004, 36, 1564-1574.
[19] W. Zhao, J. Zhang, X. Zhu, M. Zhang, J. Tang, M. Tan, Y. Wang, Appl.
Catal., B 2014, 144, 468-477.
[20] Z. Ying, S. Chen, S. Zhang, T. Peng, R. Li, Appl. Catal., B 2019, 254, 351-
359.
[21] a) L. Li, C. Tang, B. Xia, H. Jin, Y. Zheng, S.-Z. Qiao, ACS Catal. 2019, 9,
2902-2908; b) Y. Zhang, W. Qiu, Y. Ma, Y. Luo, Z. Tian, G. Cui, F. Xie, L. Chen,
T. Li, X. Sun, ACS Catal. 2018, 8, 8540-8544; c) Y. Fang, Z. Liu, J. Han, Z. Jin,
Y. Han, F. Wang, Y. Niu, Y. Wu, Y. Xu, Adv. Energy Mater. 2019, 9, 1803406;
d) X. Li, T. Li, Y. Ma, Q. Wei, W. Qiu, H. Guo, X. Shi, P. Zhang, A. M. Asiri, L.
Chen, B. Tang, X. Sun, Adv. Energy Mater. 2018, 8, 1801357; e) H. Cheng, L.-
X. Ding, G.-F. Chen, L. Zhang, J. Xue, H. Wang, Adv. Mater. 2018, 30, 1803694;
f) C. Lv, C. Yan, G. Chen, Y. Ding, J. Sun, Y. Zhou, G. Yu, Angew. Chem., Int.
Ed. 2018, 57, 6073-6076; g) D. Wu, H. Wang, H. Huang, R. Zhang, L. Ji, H.
Chen, Y. Luo, J. You, D. Tang, Z. Zhang, X. Sun, Chem. Commun. 2019, 55,
7546-7549; h) Y. Luo, G.-F. Chen, L. Ding, X. Chen, L.-X. Ding, H. Wang, Joule
2019, 3, 279-289; i) H.-M. Liu, S.-H. Han, Y. Zhao, Y.-Y. Zhu, X.-L. Tian, J.-H.
Zeng, J.-X. Jiang, B. Y. Xia, Y. Chen, J. Mater. Chem. A 2018, 6, 3211-3217; j)
J. Lv, S. Wu, Z. Tian, Y. Ye, J. Liu, C. Liang, J. Mater. Chem. A 2019, 7, 12627-
12634; k) X. Zhang, Q. Liu, X. Shi, A. M. Asiri, Y. Luo, X. Sun, T. Li, J. Mater.
Chem. A 2018, 6, 17303-17306; l) J. Zhao, L. Zhang, X.-Y. Xie, X. Li, Y. Ma, Q.
Liu, W.-H. Fang, X. Shi, G. Cui, X. Sun, J. Mater. Chem. A 2018, 6, 24031-
24035; m) W. Fu, Y. Cao, Q. Feng, W. R. Smith, P. Dong, M. Ye, J. Shen,
Nanoscale 2019, 11, 1379-1385; n) H. Wang, H. Yu, Z. Wang, Y. Li, Y. Xu, X.
Li, H. Xue, L. Wang, Small 2019, 15, 1804769; o) X. Wu, L. Xia, Y. Wang, W.
Lu, Q. Liu, X. Shi, X. Sun, Small 2018, 14, 1803111; p) W. Guo, Z. Liang, J.
Zhao, B. Zhu, K. Cai, R. Zou, Q. Xu, Small Methods 2018, 2, 1800204; q) J.-X.
Yao, D. Bao, Q. Zhang, M.-M. Shi, Y. Wang, R. Gao, J.-M. Yan, Q. Jiang, Small
Methods 2019, 3, 1800333.
[22] A. Martínez-de la Cruz, U. Amador, J. Rodríguez-Carvajal, F. García-
Alvarado, J. Solid State Chem. 2005, 178, 2998-3003.
6
This article is protected by copyright. All rights reserved.

10.1002/anie.202002029
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
−1
Efficient electrochemical Nitrogen reduction reaction (NRR): Both High NH3 yielding rate (24.7 μg h−1 mg ) and Faradic efficiency
cat.
(20.0%) achieving on the nanoreactor of Fe doped W18O49 nanowires @ carbon fiber papers at −0.15 V vs. reversible hydrogen
electrode. Fe atoms not only efficiently increase oxygen vacancies of W18O49, but optimize the nitrogen adsorption energy, and
facilitates the electrocatalytic NRR.
7
This article is protected by copyright. All rights reserved.