Altering Hydrogenation Pathways in Photocatalytic Nitrogen Fixation by Tuning Local Electronic Structure of Oxygen Vacancy with Dopant

Article abstract of DOI:10.1002/anie.202104001

To avoid the energy-consuming step of direct N≡N bond cleavage, photocatalytic N₂ fixation undergoing the associative pathways has been developed for mild-condition operation. However, it is a fundamental yet challenging task to gain comprehensive understanding on how the associative pathways (i.e., alternating vs. distal) are influenced and altered by the fine structure of catalysts, which eventually holds the key to significantly promote the practical implementation. Herein, we introduce Fe dopants into TiO₂ nanofibers to stabilize oxygen vacancies and simultaneously tune their local electronic structure. The combination of in situ characterizations with first-principles simulations reveals that the modulation of local electronic structure by Fe dopants turns the hydrogenation of N₂ from associative alternating pathway to associative distal pathway. This work provides fresh hints for rationally controlling the reaction pathways toward efficient photocatalytic nitrogen fixation.

Full text of DOI:10.1002/anie.202104001

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Altering Hydrogenation Pathways in Photocatalytic Nitrogen
Fixation by Tuning Local Electronic Structure of Oxygen Vacancy
with Dopant
Authors: Yanan Bo, Haiyun Wang, Yunxiang Lin, Tian Yang, Run Ye,
Yu Li, Canyu Hu, Pengye Du, Yangguang Hu, Zhi Liu, Ran
Long, Chao Gao, Bangjiao Ye, Li Song, Xiaojun Wu, and
Yujie Xiong
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.202104001
Link to VoR: https://doi.org/10.1002/anie.202104001

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
Altering Hydrogenation Pathways in Photocatalytic Nitrogen
Fixation by Tuning Local Electronic Structure of Oxygen Vacancy
with Dopant
Yanan Bo,[a, b, +] Haiyun Wang,[a, +] Yunxiang Lin,[a, +] Tian Yang,[c, d] Run Ye,[e] Yu Li,[f] Canyu Hu,[a] Pengye
Du,[a] Yangguang Hu,[a] Zhi Liu,[c, g] Ran Long,[a] Chao Gao,*[a] Bangjiao Ye,[e] Li Song,[a] Xiaojun Wu,*[a]
Yujie Xiong*[a, b]
Dedication ((optional))
[a]
Y. Bo, H. Wang, Dr. Y. Lin, Dr. C. Hu, P. Du, Y. Hu, Prof. R. Long, Prof. C. Gao, Prof. X. Wu, Prof. Y. Xiong
Hefei National Laboratory for Physical Sciences at the Microscale, Collaborative Innovation Center of Chemistry for Energy Materials (iChEM), School of
Chemistry and Materials Science, National Synchrotron Radiation Laboratory, and CAS Center for Excellence in Nanoscience Institution
University of Science and Technology of China
Hefei, Anhui 230026, China Fax: (+) 86-551-63606657
E-mail: yjxiong@ustc.edu.cn; xjwu@ustc.edu.cn; gaoc@ustc.edu.cn
Y. Bo, Prof. Y. Xiong
[b]
[c]
Institute of Energy
Hefei Comprehensive National Science Center
350 Shushanhu Rd., Hefei, Anhui 230031, China
T. Yang, Prof. Z. Liu
State Key Laboratory of Functional Materials for Informatics, Shanghai Institute of Microsystem and Information Technology
Chinese Academy of Sciences
Shanghai 200050, China
[d]
[e]
T. Yang
University of Chinese Academy of Sciences
Beijing 100049, China
Dr. R. Ye, Prof. B. Ye
State Key Laboratory of Particle Detection and Electronics
University of Science and Technology of China
Hefei, Anhui, 230026, China
[f]
Dr. Y. Li
Institute for New Energy Materials & Low-Carbon Technologies
Tianjin University of Technology
Tianjin 300384, China
[g]
Prof. Z. Liu
School of Physical Science and Technology
ShanghaiTech University
Shanghai 201203, China
[+] These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of the document.
Abstract: To avoid the energy-consuming step of direct N≡N bond
cleavage, photocatalytic N2 fixation undergoing the associative
pathways has been developed for mild-condition operation. However,
it is a fundamental yet challenging task to gain comprehensive
understanding on how the associative pathways (i.e., alternating vs.
distal) are influenced and altered by the fine structure of catalysts,
which eventually holds the key to significantly promote the practical
implementation. Herein, we introduce Fe dopants into TiO2
nanofibers to stabilize oxygen vacancies and simultaneously tune
their local electronic structure. The combination of in-situ
characterizations with first-principles simulations reveals that the
modulation of local electronic structure by Fe dopants turns the
hydrogenation of N2 from associative alternating pathway to
associative distal pathway. This work provides fresh hints for
rationally controlling the reaction pathways toward efficient
photocatalytic nitrogen fixation.
Introduction
The traditional industrial ammonia synthesis by the Haber-
Bosch process (N2 + 3H2 = 2NH3) generally performs the
dissociative nitrogen reduction pathway, in which the direct
cleavage of N≡N triple bonds takes place prior to the addition of
hydrogen atoms. To cleave the ultra-stable N≡N triple bonds
(with a bond energy of 941 kJ/mol), the Haber-Bosch process
requires high energy input (673-873k, 15-25 MPa).[1-3]
Photocatalytic N2 fixation (2N2 + 6H2O = 4NH3 + 3O2) is an
attractive alternative NH3 synthesis method to the Haber-Bosch
process, because, instead of using H2 under the harsh conditions,
it employs inexhaustible solar energy and water to convert N2
into NH3 at ambient temperature and pressure. Essentially, such
photocatalytic nitrogen reduction can be carried out under
ambient condition just because it undergoes the associative
This article is protected by copyright. All rights reserved.
1

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
pathways that no longer need to directly cleave the N≡N triple
Herein, we report the fabrication of iron-doped TiO2 nanofibers
via simple electrospinning toward photocatalytic fixation of N2 to
NH3 under ambient atmosphere. The doped Fe atoms have
similar radii to intrinsic Ti atoms so that the substitution of Fe
dopants for Ti atoms takes place to enable localized charge
compensation and thus stabilize the neighbored oxygen
vacancies. The TiO2 nanofibers with Fe-induced oxygen
vacancies possess different local electronic structure from the
TiO2 nanofibers having oxygen vacancies without any doped
heteroatoms, and as such, we are able to reveal the role of
dopants-modulated local electronic structure in determining the
reaction pathways for photocatalytic N2 reduction. Based on in-
situ characterizations in combination with theoretical calculations,
we reveal that tuning the local electronic structure of oxygen
vacancies by doped Fe can tailor the local electron density of
adsorbed N≡N, lower the energy barrier of rate-limiting step, and
consequently enable the control of hydrogenation in favorable
pathways. As a result, efficient and selective photocatalytic
reduction of N2 to NH3 is achieved for the Fe-doped TiO2
nanofibers in pure water without any sacrificial agent. This work
offers a promising approach to modulate the local electronic
structure toward efficient photocatalytic nitrogen fixation, as well
as highlights the role of dopants-modulated local electronic
structure in controlling reaction pathways.
bonds, significantly reducing the required energy input. In this
regard, it is
a fundamental yet challenging theme to
comprehensively understand how the nitrogen reduction
pathways (i.e., associative alternating vs. associative distal)
could be controlled and altered in a favorable way, eventually
enabling efficient photocatalytic N2 fixation.
The catalytic behavior involving reaction pathways is, to a
great extent, determined by the local electronic structure of
catalytic sites. In terms of controlling catalytic sites, oxygen
vacancies have been extensively investigated for oxide catalysts
in catalytic nitrogen reduction,[4-11] highlighting their importance
as catalytic sites. For this reason, we focus on the surface
oxygen vacancies using titanium dioxide (TiO2) as a catalyst
model, and aim to elucidate the mystery how their local electronic
structure impacts on reaction pathways. To modulate the local
electronic structure of oxygen vacancies, we decide to adopt the
strategy of doping heteroatoms that has been proven as a simple
and effective way toward such a purpose.[10,
noteworthy advantage for heteroatom doping is to improve the
stability of catalysts containing oxygen vacancies. In most cases,
oxygen vacancies are created via annealing in
temperature and oxygen-deficient atmosphere;[15] however, they
generally tend to recombine with oxygen and disappear once
exposed to air at room temperature,[16] which would severely
impede our investigation on the local electronic structure of
oxygen vacancies. Such a limitation can be overcome when we
choose to create local oxygen vacancies by introducing
heteroatoms during the synthesis process. The heteroatoms on
the surface can serve as charge-compensating species to accept
electrons and thus effectively stabilize oxygen vacancies. As
such, heteroatom doping can provide an excellent platform for
investigating the role of dopants in modulating the local
electronic structure and the impact of local electronic structure
adjacent to oxygen vacancy on the hydrogenation pathways in
photocatalytic nitrogen fixation, which remain unexplored in the
past research for oxygen vacancy-induced NRR.
12-14]
Another
a high-
Results and Discussion
Generally, the successful incorporation of heteroatoms as
dopants highly depends on its relative radii compared to host
atomic radii. If the radius of the heteroatoms is too large to enter
crystal lattice, the heteroatoms would be loaded on the surface
in the form of oxide.[12] When the atomic radius is similar or
smaller, heteroatom may replace the host atom or be inserted
into the host crystal lattice. In this work, we choose typical TiO2
as our model catalyst and Fe as the dopant inspired by biological
Figure 1. a) Schematic illustration for the electrospinning setup. b) TEM, c) SEM and d) HRTEM images of as-prepared 5-FTNFs. The inset graph in panel c shows
the sectional view of as-prepared nanofibers. e) Scanning TEM image and (f-h) EDS elemental mapping.
This article is protected by copyright. All rights reserved.
2

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
nitrogenase ferroprotein as well as considering the suitable
radius of Fe atom compared to Ti atom.[17] To fabricate the model
catalyst with high controllability, we employ electrospinning for
the materials preparation in our work. Electrospinning is a
universal and simple method to fabricate nanofibers at large
scale with controllable morphology and flexible composition,
which is economically attractive and compatible with industrial
requirements for future practical application.[18] The flexible
composition enlightens us to modulate electronic structure via
altering pristine composition with dopants.[19-20] To this end, a
series of Fe-doped TiO2 nanofibers with different doping
and S5). Moreover, Co-doped TiO2 and Ni-doped TiO2 exhibit
inferior activity for photocatalytic NRR as compared with Fe-
doped TiO2, and only give slight improvement in ammonia
production rates as benchmarked against pristine TiO2 (Figure
S6), suggesting the specific role of doped Fe in enhancing the
photocatalytic NRR activity.
concentrations are prepared through
a simple scalable
electrospinning method (Figure 1a) followed by annealing in air.
X-ray diffraction (XRD, Figure S1) confirms that the
introduction of Fe does not change the original phase or form a
new phase. The diffraction patterns of all the samples match well
with that of rutile phase TiO2 (JCPDS No. 88-1172). The intensity
of diffraction peaks gradually decreases and the peak position
gradually shifts toward larger diffraction angles with the increase
of Fe3+ percentage, indicating that iron has been successfully
doped into TiO2 and contracted the lattice.[21-22] Moreover,
excessive doping (10%) of iron forms a new Fe2TiO5 phase
(Figure S2). Transmission electron microcopy (TEM) image
shows that one-dimensional TiO2 nanofibers with a diameter of
about 200 nm are well obtained through the electrospinning, and
similar nanofibers are observed for the samples with different
amounts of Fe doping (Figure S3). Upon Fe doping (5%), unique
hollow structure is formed to increase surface area as observed
by TEM (Figure 1b), scanning electron microscopy (SEM, Figure
1c) and Brunauer-Emmett-Teller (BET) specific surface areas
(Figure S4, Table S1) on account of different ionic thermal
diffusion rates between Fe3+ and Ti4+ during the annealing
process.[23] To confirm the presence and distribution of Fe in the
as-synthesized sample, elemental energy-dispersive X-ray
spectroscopy (EDS) mapping (Figure 1d-h) is employed to
characterize the sample, revealing that Fe distributes uniformly
on the nanofibers. In accordance with inductively-coupled
plasma mass spectroscopy (ICP-MS), EDS analysis reveals that
the ratio of Fe to Ti is basically conformed to feeding ratio (Table
S2). According to high resolution TEM (HRTEM, Figure 1d), the
as-prepared Fe-doped TiO2 NFs are mostly exposed with (110)
crystal planes. Namely, the Fe-doped TiO2 NFs samples are
noted as x-FTNFs (x = 0.625, 1.25, 2.5 and 5), where x refers to
the molar percentage of doped Fe element.
Upon successfully obtaining the Fe-doped TiO2 nanofibers, we
are now in a position to examine the performance of FTNFs with
different Fe contents as a catalyst for photocatalytic N2 reduction.
The photocatalytic experiments are performed under the
irradiation of full-spectrum light in nitrogen-saturated water
without any sacrificial agent, and the concentration of produced
NH4+ is quantified by ion chromatography. As indicated in Figure
2a, the pristine TiO2 (0-FTNFs) exhibits a low nitrogen reduction
ability, while the NH3 production rate for 5-FTNFs (5% Fe doping)
is about 5.3 times higher than that of pristine TiO2 NFs (12.1 vs.
64.2 μmol g-1 h-1), suggesting that the activity for nitrogen
reduction has been significantly improved by doping Fe. With the
increasing doping amounts, the ammonia production rates are
improved correspondingly; however, the introduction of
excessive Fe (10%) leads to a decline in ammonia production
rates due to the formation of a new phase of Fe2TiO5 (as
indicated by XRD pattern and NRR performance in Figure S2
Figure 2. a) Photocatalytic NRR performance for TiO2 nanofibers with different
Fe-doping amounts. b) Photocatalytic ammonia production rates in the first 3
hours for cyclic tests of 5-FTNFs. c) Time-dependent production of NH3 for
TiO2 nanofibers with different Fe-doping amounts. d) TOF-MS (left panel) and
1H-NMR (right panel) of the products formed in different reaction atmospheres
(
14N2 or 15N2).
To further examine the stability of the Fe-doped TiO2
nanofibers, the used catalysts are collected and re-dispersed in
fresh water for cycling tests. After the five cycles each of which
lasts 3 h, the recycled catalyst well retains the activity (Figure 2b).
The X-ray photoelectron spectroscopy (XPS) characterization
(Figure S7) shows that the catalyst after reaction does not
undergo noticeable change in comparison with the fresh one,
indicating an excellent reusability and stability of the Fe-doped
TiO2 nanofibers for photocatalytic N2 reduction. Furthermore, the
ammonia product for all the TiO2 nanofibers with different Fe-
doping amount is continuously generated over time and keeps
the consistent trend (Figure 2c), demonstrating the reliability and
stability of N2 fixation. To exclude any possible NOx and NH3
contaminants, we have conducted control experiments in the
absence of photocatalyst, without illumination or in the Ar
atmosphere (Figure S8 and Table S3). No NH3 production can
be detected, providing solid evidence for the absence of any NOx
or NH3 contaminants both in feed gas and photocatalysts.
Moreover, the indophenol blue colorimetric method (Figure S9)
is also adopted to confirm the actual amounts of evolved NH3
measured by ion chromatography. We also evaluate the
wavelength-dependent apparent quantum efficiency (AQE) by
different monochromatic light irradiation, which matches well with
the light absorption capacity of the sample (Figure S10). To
further verify the nitrogen source of produced NH3, isotope
labeling photocatalytic N2 reduction is performed by using high-
purity 15N2 as the reactant, and the product is tested by time-of-
flight mass spectrometry (TOF-MS, left panel in Figure 2d) and
1H nuclear magnetic resonance (1H-NMR) spectroscopy (right
This article is protected by copyright. All rights reserved.
3

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
panel in Figure 2d). In TOF-MS, the produced 15NH3
corresponding to the m/z = 18 in mass spectrometry is originated
from 15N2. In 1H-NMR spectroscopy, the doublet pattern with the
dopants in TiO2, the Fe 2p1/2 and 2p3/2 peaks of tri-valent Fe(III)
at 724.0 eV and 711.0 eV for 5-FTNFs shift to higher binding
energy as compared with 0.625-FTNFs (Figure 3b).[10] Such
shifts of Ti and Fe peaks in the opposite direction indicate that
the local electron density of Fe decreases due to the formation
of Fe-O bond while the Fe-induced oxygen vacancy increases
the local electron density of Ti, which is also proven by differential
charge density (Figure S16). Such lopsided local charge
distribution would affect the local electron density of adsorbed N2
molecules and thus contribute to polarizing the adsorbed N2
molecules for better activation, as evidenced by the following
DFT calculation.
coupling constant of JN-H = 72 Hz corresponding to 15NH4 is
observed, while no triplet pattern with the coupling constant of
JN-H = 52 Hz corresponding to 14NH4 can be detected. The
above results confirm that the NH3 product totally evolves from
N2 rather than any contamination.
+
+
X-ray absorption fine structure (XAFS) spectroscopy is further
performed to probe the variation of local electronic structure
caused by introducing Fe into TiO2 lattice (Figure 3c, 3d and S17-
S19). Generally, rutile titanium dioxide has a octahedral structure
with six-fold coordination.[25, 26] As shown in Figure 3c, Ti K-edge
X-ray absorption near-edge structure (XANES) spectra for 0-
FTNFs and 5-FTNFs possess three typical pre-edge peaks
associated with the rutile phase (denoted as A1, A2, and A3),
corresponding to quadruple-allowed 1s→3d transitions, dipole-
allowed Ti s→p transitions with Ti 3d (t2g) and Ti 3d (eg)
characters, respectively.[26] The reduced intensity of A2 peak for
5-FTNFs compared to 0-FTNFs is caused by severe Ti site
distortions after Fe incorporation. The Ti site distortion can be
also testified by Fourier transformed extended XAFS (EXAFS)
spectra (Figure 3d), in which 5-FTNFs possess shorter equator
bond length and longer apical bond length than 0-FTNFs.
Essentially, such a distortion is associated with the contracted
crystal lattice due to the addition of Fe, as confirmed by XRD that
diffraction peaks shift to larger 2θ angles (Figure S1). Moreover,
Fe K-edge XANES spectra of 5-FTNFs (Figure S18) display
characteristic peaks distinguished from those of Fe foil and
Fe2O3, excluding the exsistence of metallic Fe or Fe2O3 phase.[27-
28] In the meantime, the addition of Fe reduces the intensity of Ti-
O peak in Figure 3d, which is consistent with the smaller
coordination numbers of 5-FTNFs than 0-FTNFs (4.98 vs. 6.00,
Table S4). In the 5-FTNFs, the unsaturated coordinated Ti atoms
are stabilized by Fe dopants through charge compensation,[10]
and as such, the oxygen vacancies are stabilized to provide
abundant active sites. Indeed, the concentration of oxygen
vacancies in 5-FTNFs is dramatically higher than that in 0-FTNFs,
as evidenced by positron annihilation spectrometry (Figure S20
and Table S5). All these evidences indicate that heteroatom Fe
has been successfully incorporated into TiO2 lattice and
simultaneously modulated the local electronic structure nearby
oxygen vacancies.
Figure 3. High-resolution a) Ti 2p and b) Fe 2p XPS spectra for 0-FTNFs and
5-FTNFs. c) Ti K-edge XANES spectra for extended edge and d) magnitude of
k3-weighted Fourier transforms of Ti K-edge EXAFS spectra for 0-FTNFs and
5-FTNFs.
As discussed above, the Fe doping largely promotes the
activity of pristine TiO2 nanofibers for nitrogen reduction. To
elucidate the origin of this superior activity, we have to examine
the influence of Fe doping on light utilization, charge separation
and Fe-induced local defect structure, which are the potential
factors for altering photocatalytic performance. Our
characterizations on band structures and charge separation
(Figure S11-S14) indicate that doping Fe into TiO2 lattice can
promote the light utilization in visible region and facilitates the
charge separation and transfer, which enables to provide
abundant photogenerated electrons for N2 reduction. However,
according to the photocurrents, the improvement of light
utilization and charge separation does not sufficiently enhance
the photocatalytic performance to such an extent. As such, we
look into the local electronic structure of active sites that could
play a crucial role in determining the catalytic performance.
Surface oxygen vacancies are well accepted as catalytic sites for
the catalytic reduction of nitrogen. We thus closely examine the
modulated local electronic structure nearby oxygen vacancy
along with Fe doping. To reveal the local electronic structure,
XPS measurements are conducted to analyze the valence state
variation (Figure 3 and S15). The XPS spectra for 0-FTNFs show
characteristic Ti 2p1/2 and 2p3/2 peaks of tetra-valent Ti(IV) at
458.8 and 462.2 eV, respectively.[24] In comparison, these peaks
for the 5-FTNFs shift to lower binding energy by about 0.24 eV
(Figure 3a). In the meantime, with the increased amount of Fe
In order to further understand the role of tuning local electronic
structure for nitrogen reduction reaction, various spectral
characterizations are performed to examine the interactions of
catalyst with N2 molecules. The electron paramagnetic
resonance (EPR) spectrum of 5-FTNFs under the He
atmosphere shows a clear signal near g = 2.003 (Figure 4a),
indicating the presence of abundant oxygen vacancies although
the preparation process involves annealing in air. In contrast, no
obvious signal of oxygen vacancies is observed in the EPR
spectra of 0-FTNFs prepared via the same air-annealing (Figure
S21a) as the annealing in air generally removes the oxygen
vacancies. For this reason, the oxygen vacancies in 0-FTNFs
appear after post annealing in anoxic Ar atmosphere (denoted
as 0-FTNFs-Ar). This indicates that the Fe dopants are capable
This article is protected by copyright. All rights reserved.
4

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
of inducing the formation of abundant oxygen vacancies no
matter whether the post annealing is performed in air. The 0-
FTNFs-Ar with substantially more oxygen vacancies exhibits an
inferior NRR activity in contrast to 5-FTNFs (Figure S21b-d,
Table S6), suggesting that the Fe dopants play an important role
in facilitating the NRR process. The role of Fe dopants in
promoting the formation of oxygen vacancies is further confirmed
by DFT calculations, which indicate that the introduction of Fe
into TiO2 can reduce the formation energy for oxygen vacancy (-
3.56 eV for Fe-doped TiO2 vs. -1.83 eV for pristine TiO2, Table
S7). Essentially, such Fe-induced oxygen vacancies are formed
through the replacement of Ti(IV) by Fe(III), and the oxygen
vacancies exist stably because Fe(III) acts as an electron
compensation.
reaction after high-purity N2 is introduced into the analysis
chamber. In the case of 5-FTNFs (Figure 4b, top), after
illumination of 5 min, a broader band around 402.0 eV ascribed
to adsorbed N2 molecule [N2(a)] shows up clearly and can be
resolved into two peaks, revealing the states of two different
nitrogen atoms in a discrepant chemical environment.[32] Along
with the illumination of 10-30 min, the band of absorbed N2
molecule varies between 400-402 eV, suggesting that the
adsorbed N2 molecule has been hydrogenated to N2Hy (y=1 or
2).[33] As hydrogenation continuously proceeds, the signal of N=N
is gradually weakened and two peaks emerge at 395.3 and 397-
399 eV that are attributed to N≡ and NHy (y=1 or 2), respectively.
[34] This indicates that the N=N species evolves into N≡ and NHy
during the subsequent reaction steps. Apparently, no signal for
N≡ is observed at the initial stage of the reaction, excluding the
possibility of dissociative pathway. Furthermore, the observation
of N≡ and NHy intermediates during the reaction confirms that
the 5-FTNFs takes the associative distal pathway for N2
reduction as illustrated in Figure 4c.
We further examine the interaction of catalyst with N2
molecules by purging N2 into the testing tube. The presence of
N2 reduces the signal intensity of oxygen vacancies, illustrating
that oxygen vacancies could interact with N2. Upon illumination,
the signal intensity of oxygen vacancies further declines,
indicating that the oxygen vacancies are involved in N2
reduction.[29-30] To reveal the relationship between oxygen
vacancy and NRR activity, we compare the oxygen vacancy
contents of 0-FTNFs-Ar and 5-FTNFs. The raletive oxygen
vacancy contents are quantitatively analyzed by O 1s XPS
spectra (Figure S21b and c), in which the oxygen vacancy
content for 0-FTNFs-Ar is nearly double that of 5-FTNFs.[7]
However, 0-FTNFs-Ar with much more oxygen vacancies
exhibits an inferior NRR activity in contrast to 5-FTNFs (Figure
S21d), suggesting that the Fe dopants play an important role in
facilitating the NRR process. Taken together, the EPR
characterizations demonstrate that the doped Fe can induce the
formation of oxygen vacancies, which bring about abundant
active sites for N2 adsorption involved in the NRR process.
Beyond that, the doped Fe can exert significant influence on
NRR process contributing to a higher activity even with lower
oxygen vacancy content, as discussed in the following section.
During the NRR process, the N2 adsorption is followed by
chemical reactions. N2 is a symmetric nonpolar molecule so that
its binding with a certain atom of catalyst surface will lead to the
redistribution of spatial charge and weakening of N≡N triple bond.
In this case, the local electronic structure nearby active sites
would largely influence the adsorption and polarization of N2
molecule and thus alter the reaction pathways for N2
hydrogenation. To reveal such impact on the reaction pathways
for N2 hydrogenation, in-situ near ambient pressure X-ray
photoelectron spectroscopy (NAP-XPS) is performed for the
reaction system using 5-FTNFs as a catalyst in reference to 0-
FTNFs-Ar. Both the catalysts contain oxygen vacancies but only
5-FTNFs is incorporated with Fe dopants. We collect the time-
dependent NAP-XPS spectra for N 1s (Figure 4b), which gives
the information of N-containing intermediates. Gaseous nitrogen
exhibits higher binding energy for its ultra-stable triple bond
(941.7 kJ/mol) contrasting to double bond and single bond. In
general, the binding energies of N-containing species decline in
the sequence of gaseous N2 > chemical absorbed N2 > N2Hx
(x=1-4) > NHy (y=1-3).[31] More specifically, along with the
hydrogenation process, the binding energies of N-containing
species increase, i.e. N ≡ < -NH- < -NH2 < NH3. Our time-
dependent NAP-XPS shows no obvious signal under the
ultrahigh vacuum (UHV), and then the signal of gaseous N2
appears at 405.3 eV and exists concomitantly during the whole
Figure 4. a) EPR spectra of 5-NFTFs in the conditions of He atmosphere under
dark, N2 atmosphere under dark and N2 atmosphere under light irradiation,
respectively. b) In-situ NAP-XPS spectra of 5-FTNFs and 0-FTNFs-Ar
collected in the conditions of ultrahigh vacuum under dark, N2 atmosphere
under dark and N2 atmosphere under light irradiation, respectively. c)
Schematic pathway for N2 reduction on heterogeneous catalysts: dissociative
pathway (top), associative alternating pathway (middle) and associative distal
pathway (bottom). d) In-situ DRIFT spectra of 5-NFTFs collected in N2
atmosphere under dark and irradiation.
As a comparison, the 0-FTNFs-Ar catalyst contains oxygen
vacancies but no Fe dopants. Upon illumination, the signal for
adsorbed N2 appears at 402.8 eV (Figure 4b, bottom). As the
This article is protected by copyright. All rights reserved.
5

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
illumination prolongs, the signal for N2Hy (y=1-4) species
emerges at 401-402 eV and eventually evolves into NH3 at 399.8
eV. During the nitrogen reduction reaction on 0-FTNFS-Ar, the
peaks for N≡ (395.3) and NHy (y=1 or 2, 397-399 eV) fragments
do not emerge throughout the process. This manifests that the
0-FTNFs-Ar takes the associative alternating pathway for N2
reduction (Figure 4c). The NAP-XPS results here provide a direct
evidence that the addition of Fe dopants in TiO2 can turn the
hydrogenation of N2 from the associative alternating pathway to
associative distal pathway.
performed to examine the evolution of N2 molecule on catalyst
surface. According to EXAFS results, we first calculate various
possible structures and select the most stable structure as a slab
model as illustrated in Figure S22.For ease of description, we
denote the active sites of Fe-free 0-FTNFs-Ar and Fe-doped 5-
FTNFs samples as Ti-Ovs-Ti and Fe-Ovs-Ti sites, respectively.
In both cases, N2 adsorption takes an end-on configuration.
However, the adsorption in Fe-Ovs-Ti takes place at the Fe sites
nearby oxygen vacancies (Figure S23), while the adsorption
sites in Ti-Ovs-Ti are the Ti atoms at the oxygen vacancies
(Figure S24). We then simulate the Gibbs free energies of both
the associative distal pathway and associative alternating
pathway at two different sites (i.e., Ti-Ovs-Ti and Fe-Ovs-Ti)
(Figure 5, S25and S26). In the case of Ti-Ovs-Ti sites, the first
hydrogenation step of NRR (N2+H++e-=N2H*) proceeds easily
with a downhill Gibbs free energy change (∆G) of -0.357 eV, but
the next hydrogenation step has an uphill energy barrier for the
formation of both NHNH* (0.251 eV) in the associative
alternating pathway and NH2* (0.431 eV) in the associative distal
pathway (Table S8). Considering the lower uphill energy barrier,
the N2 molecule at the Ti-Ovs-Ti site prefers to be hydrogenated
in the associative alternating pathway for N2 reduction. In terms
of Fe-Ovs-Ti sites, the first hydrogenation step for the formation
of N2H* has an uphill energy barrier of 0.171 eV. Afterwards, the
subsequent hydrogenation process becomes much easier with
gradual downhill Gibbs free energy in the associative distal
pathway. In comparison, there is an uphill energy barrier for the
formation of NH2NH2* in the associative alternating pathway.
To verify the reaction mechanism upon N2 reduction, in-situ
diffuse reflectance infrared Fourier-transform spectroscopy
(DRIFTS) is also employed to monitor the intermediates during
the photocatalytic N2 reduction process. The in-situ DRIFT
spectra of 5-FTNFs (Figure 4d) show that the signal of N-bearing
intermediates arises and is gradually boosted with time under
illumination. The broad band at 3555 cm−1 corresponds to the
ν(N-H) stretching mode, while the sharp band at 1641 cm−1 can
be assigned to the σ(N-H) bending mode. Furthermore, the weak
bands at 1464 and 2839 cm−1 are attributed to the characteristic
+
adsorption of NH4 , while the bands at 1552 and 2957 cm-1 are
ascribed to the characteristic absorption of NH3 molecule.[6, 30]
These signals of N intermediates are in good accordance with in-
situ NAP-XPS characterization, providing a convictive proof for
the reaction pathway.
To elucidate how the tuning of local electronic structure by Fe
dopants impacts on the reaction pathway of N2 hydrogenation in
NRR, density functional theory (DFT) calculations are further
Figure 5. Calculated Gibbs free energy profiles of NRR process at Ti-Ovs-Ti (red) and Fe-Ovs-Ti sites (blue) with associative distal and associative alternating
pathways and the corresponding optimized geometric structure of intermediates adsorbed on the substrate.
This article is protected by copyright. All rights reserved.
6

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
Therefore, the N2 adsorbed at the Ti-Ovs-Fe site tends to be
hydrogenated in the associative distal pathway. Indeed, we have
detected the corresponding N-containing intermediates for Fe-
doped 5-FTNFs (Ti-Ovs-Fe sites) and for Fe-free 0-FTNFs-Ar
(Ti-Ovs-Ti sites) during the reaction in above in-situ
characterizations, which provides a solid evidence to confirm the
calculation results. Furthermore, the limiting-potential of NRR at
the Ti-Ovs-Fe sites (N2 + H+ + e-=N2H*, 0.171 eV) is substantially
lower than that of Ti-OVs-Ti sites (NNH + H+ + e-=NNH2*, 0.251
eV). This implies that the Fe-doped catalyst can proceed
hydrogenation more easily enhancing the photocatalytic activity
in NRR.
91963108, 21703220, 11875248), CAS Key Research Program
of
Frontier
Sciences
(QYZDB-SSW-SLH018),
CAS
Interdisciplinary Innovation Team, DNL Cooperation Fund, CAS
(DNL201922), and Youth Innovation Promotion Association CAS.
XAFS measurements were performed at the beamline BL14W1
of the Shanghai Synchrotron Radiation Facility (SSRF), China. A
portion of this work was performed on the Steady High Magnetic
Field Facilities, High Magnetic Field Laboratory, CAS. In situ
DRIFTS was collected at the Infrared Spectroscopy and
Microspectroscopy Endstation (BL01B), and TOF-MS
measurements were performed on synchrotron radiation
ultraviolet photoionization mass spectroscopy (SR-PIMS,
BL04B) in the National Synchrotron Radiation Laboratory (NSRL)
in Hefei, China. We are thankful for the support from the USTC
Center for Micro- and Nanoscale Research and Fabrication.
Then a fundamental question naturally arises what factor
alters the hydrogenation pathways in NRR. The schematic
illustration for reaction pathways (Figure 4c) gives us a clue. In
the associative distal pathway, the hydrogenation first takes
place on one N atom to form one NH3, and then the
hydrogenation proceeds on another N atom. In sharp contrast,
the hydrogenation in the associative alternating pathway should
proceed on both two N atoms simultaneously. Thus the key to
altering the reaction pathways is to differentiate the two N atoms
in a N2 molecule. To look into the difference of two N atoms at
the Fe-Ovs-Ti site, we acquire the differential charge diagram
(Figure S27) that can reflect the polarization of adsorbed N2
molecule, which shows that the electrons can be preferentially
accumulated on the N atom bonded with Fe site. Such an
imbalanced charge distribution differentiates the two N atoms of
adsorbed N2, initiating the associative distal pathway at the Fe-
Ovs-Ti site. Taken together, the computational studies and in-
situ characterizations highlight that the introduction of Fe
dopants into catalyst is an efficient approach to create active
sites and modulate local electronic structure, which dramatically
alter the reaction pathways and substantially enhance the
efficiency for N2-to-NH3 conversion.
Conflict of interest
The authors declare no conflict of interest.
Keywords: Local electronic structure • Active sites • Oxygen
vacancy • Photocatalytic nitrogen fixation • Hydrogenation
pathway
[1]
[2]
[3]
J. W. Erisman, M. A. Sutton, J. Galloway, Z. Klimont, W. Winiwarter,
Nat. Geosci. 2008, 1, 636-639.
X. Xue, R. Chen, C. Yan, P. Zhao, Y. Hu, W. Zhang, S. Yang, Z. Jin,
Nano Res. 2019, 12, 1229-1249.
C. Chen, X. Zhu, X. Wen, Y. Zhou, L. Zhou, H. Li, L. Tao, Q. Li, S. Du,
T. Liu, D. Yan, C. Xie, Y. Zou, Y. Wang, R. Chen, J. Huo, Y. Li, J. Cheng,
H. Su, X. Zhao, W. Cheng, Q. Liu, H. Lin, J. Luo, J. Chen, M. Dong, K.
Cheng, C. Li, S. Wang, Nat. Chem. 2020, 12, 717-724.
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.
[4]
[5]
[6]
[7]
[8]
[9]
C. Lv, Y. Qian, C. Yan, Y. Ding, Y. Liu, G. Chen, G. Yu, Angew. Chem.
Int. Ed. 2018, 57, 10246-10250.
S. Wang, X. Hai, X. Ding, K. Chang, Y. Xiang, X. Meng, Z. Yang, H.
Chen, J. Ye, Adv. Mater. 2017, 29, 1701774.
Conclusion
C. Li, T. Wang, Z. J. Zhao, W. Yang, J. F. Li, A. Li, Z. Yang, G. A. Ozin,
J. Gong, Angew. Chem. Int. Ed. 2018, 57, 5278-5282.
H. Wang, D. Yong, S. Chen, S. Jiang, X. Zhang, W. Shao, Q. Zhang,
W. Yan, B. Pan, Y. Xie, J. Am. Chem. Soc. 2018, 140, 1760-1766.
S. Bai, N. Zhang, C. Gao, Y. Xiong, Nano Energy 2018, 53, 296-336.
In conclusion, we have developed a doping strategy to
modulate the local electronic structure of surface vacancy for
controlling the photocatalytic nitrogen fixation in a favorable
hydrogenation pathway. We have revealed that the doped Fe
atoms in TiO2 nanofibers can modulate the local electronic
structure nearby oxygen vacancy and act as binding sites. Such
an effect in turn facilitates the adsorption and polarization of N2,
lowers the energy barrier of rate-limiting step for nitrogen
reduction, and consequently alters the hydrogenation of N2 from
[10] Y. Tong, H. Guo, D. Liu, X. Yan, P. Su, J. Liang, S. Zhou, J. Liu, G. Q.
Lu, S. X. Dou, Angew. Chem. Int. Ed. 2020, 59, 7356-7361.
[11] J. Qian, S. Zhao, W. Dang, Y. Liao, W. Zhang, H. Wang, L. Lv, L. Luo,
H.-Y. Jiang, J. Tang, Adv. Sustain. Syst. 2021, 5, 2000282.
[12] Y. Zhou, F. Che, M. Liu, C. Zou, Z. Liang, P. De Luna, H. Yuan, J. Li, Z.
Wang, H. Xie, H. Li, P. Chen, E. Bladt, R. Quintero-Bermudez, T.-K.
Sham, S. Bals, J. Hofkens, D. Sinton, G. Chen, E. H. Sargent, Nat.
Chem. 2018, 10, 974-980.
the associative alternating pathway to
a more favorable
associative distal pathway, thus contributing to improved
performance for NH3 evolution in photocatalytic nitrogen fixation.
This work sheds a light on the rational design of catalytic sites
toward efficient photocatalytic nitrogen fixation through
modulating local electronic structure, and provides
comprehensive understanding on how the dopants-modulated
local electronic structure could alter reaction pathways.
[13] J. Zhang, X. Tian, M. Liu, H. Guo, J. Zhou, Q. Fang, Z. Liu, Q. Wu, J.
Lou, J. Am. Chem. Soc. 2019, 141, 19269-19275.
[14] T. Wu, X. Zhu, Z. Xing, S. Mou, C. Li, Y. Qiao, Q. Liu, Y. Luo, X. Shi, Y.
Zhang, X. Sun, Angew. Chem. Int. Ed. 2019, 58, 18449-18453.
[15] J. Yang, Y. Guo, R. Jiang, F. Qin, H. Zhang, W. Lu, J. Wang, J. C. Yu,
J. Am. Chem. Soc. 2018, 140, 8497-8508.
[16] Q. Wu, Q. Zheng, R. van de Krol, J. Phys. Chem. C 2012, 116, 7219-
7226.
[17] D. Sippel, M. Rohde, J. Netzer, C. Trncik, J. Gies, K. Grunau, I.
Djurdjevic, L. Decamps, S. L. A. Andrade, O. Einsle, Science 2018, 359,
1484-1489.
Acknowledgements
[18] L. Persano, A. Camposeo, C. Tekmen, D. Pisignano, Macromol. Mater.
Eng. 2013, 298, 504-520.
This work was supported by the National Key R&D Program of
China (2017YFA0207301), NSFC (21725102, 91961106,
[19] J. Xue, T. Wu, Y. Dai, Y. Xia, Chemical Reviews 2019, 119, 5298-5415.
This article is protected by copyright. All rights reserved.
7

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
[20] D. Joly, J.-W. Jung, I.-D. Kim, R. Demadrille, J. Mater. Chem. C 2016,
4, 10173-10197.
[21] J. Yang, H. Bai, Y. Guo, H. Zhang, R. Jiang, B. Yang, J. Wang, J. C. Yu,
Angew. Chem. Int. Ed. 2021, 60, 927-936.
[22] H. He, D. Sun, Q. Zhang, F. Fu, Y. Tang, J. Guo, M. Shao, H. Wang,
ACS Appl. Mater. Interfaces 2017, 9, 6093-6103.
[23] C. Lv, C. Yan, G. Chen, Y. Ding, J. Sun, Y. Zhou, G. Yu, Angew. Chem.
Int. Ed. 2018, 57, 6073-6076.
[24] J. Yang, Y. Guo, R. Jiang, F. Qin, H. Zhang, W. Lu, J. Wang, J. C. Yu,
J. Am. Chem. Soc. 2018, 140, 8497-8508.
[25] Q. Wu, R. van de Krol, J. Am. Chem. Soc. 2012, 134, 9369-9375.
[26] N. Jiang, D. Su, J. C. H. Spence, Phys. Rev. B 2007, 76, 214117.
[27]
Z. Zhang, Q. Wu, G. Johnson, Y. Ye, X. Li, N. Li, M. Cui, J. D. Lee, C.
Liu, S. Zhao, S. Li, A. Orlov, C. B. Murray, X. Zhang, T. B. Gunnoe, D.
Su, S. Zhang, J. Am. Chem. Soc. 2019, 141, 16548-16552.
[28] M. Açıkgöz, P. Gnutek, C. Rudowicz, Chemical Physics Letters
2012524, 49-55.
[29] Y. Zhao, Y. Zhao, R. Shi, B. Wang, G. I. N. Waterhouse, L. Z. Wu, C. H.
Tung, T. Zhang, Adv Mater 2019, 31, e1806482.
[30] H. Hirakawa, M. Hashimoto, Y. Shiraishi, T. Hirai, J. Am. Chem. Soc.
2017, 139, 10929-10936.
[31] C. Hu, X. Chen, J. Jin, Y. Han, S. Chen, H. Ju, J. Cai, Y. Qiu, C. Gao,
C. Wang, Z. Qi, R. Long, L. Song, Z. Liu, Y. Xiong, J. Am. Chem. Soc.
2019, 141, 7807-7814.
[32] M. L. Wagner, L. D. Schmidt, Surf. Sci. Lett. 1991, 257, A565.
[33] J. Zheng, Y. Lyu, M. Qiao, J. P. Veder, R. D. Marco, J. Bradley, R. Wang,
Y. Li, A. Huang, S. P. Jiang, S. Wang, Angew. Chem. Int. Ed. 2019, 58,
18604-18609.
[34] J. L. Bischoff, F. Lutz, D. Bolmont, L. Kubler, Surf. Sci. 1991, 251, 170-
174.
This article is protected by copyright. All rights reserved.
8

10.1002/anie.202104001
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
Modulating the local electronic structure nearby oxygen vacancy
by doping enables the control of N2 hydrogenation from the
associative alternating pathway to a more favorable associative
distal pathway, thus contributing to improved performance for
NH3 evolution in photocatalytic nitrogen fixation.
This article is protected by copyright. All rights reserved.
9