Plasmonic Enhanced Reactive Oxygen Species Activation on Low-Work-Function Tungsten Nitride for Direct Near-Infrared Driven Photocatalysis

Article abstract of DOI:10.1002/smll.202004557

Realizing near-infrared (NIR) driven photocatalytic reaction is one of the promising strategies to promote the solar energy utilization and photocatalytic efficiencies. However, effective reactive oxygen species (ROS) activation under NIR irradiation remains to be great challenge for nearly all previously reported photocatalysts. Herein, the cubic-phase tungsten nitride (WN) with strong plasmonic NIR absorption and low-work function (≈3.59?eV) is proved to be able to mediate direct ROS activation by both of experimental observation and theoretical simulation. The cubic WN nanocubes (NCs) are synthesized via the hydrothermal-ammonia nitridation process and its NIR-driven photocatalytic properties, including photocatalytic degradation, hydroxylation, and de-esterification, are reported for the first time in this work. The 3D finite element simulation results demonstrate the size dependent and wavelength tuned plasmonic NIR absorption of the WN NCs. The NIR-driven photocatalytic mechanism of WN NCs is proposed based on density functional theory (DFT) calculated electronic structure and facet dependent O₂ (or H₂O) molecular activation, radicals scavenging test, spin trapped electron paramagnetic resonance measurements, and ultraviolet photoelectronic spectrum (UPS). Overall, the results in this work pave a way for the application of low-work-function materials as highly reactive NIR photocatalyst.

Full text of DOI:10.1002/smll.202004557

Full PaPer
www.small-journal.com
Plasmonic Enhanced Reactive Oxygen Species Activation
on Low-Work-Function Tungsten Nitride for Direct
Near-Infrared Driven Photocatalysis
Weicheng Huang, Yan Gao, Jinxin Wang, Pengcheng Ding, Mei Yan, Fengmin Wu,
Jian Liu, Danqing Liu,* Chongshen Guo,* Bin Yang,* and Wenwu Cao
utilizing NIR light that takes up to about
Realizing near-infrared (NIR) driven photocatalytic reaction is one of the
promising strategies to promote the solar energy utilization and photocata-
lytic efficiencies. However, effective reactive oxygen species (ROS) activation
under NIR irradiation remains to be great challenge for nearly all previously
reported photocatalysts. Herein, the cubic-phase tungsten nitride (WN) with
strong plasmonic NIR absorption and low-work function (≈3.59 eV) is proved
to be able to mediate direct ROS activation by both of experimental observa-
tion and theoretical simulation. The cubic WN nanocubes (NCs) are synthe-
sized via the hydrothermal-ammonia nitridation process and its NIR-driven
photocatalytic properties, including photocatalytic degradation, hydroxylation,
and de-esterification, are reported for the first time in this work. The 3D finite
element simulation results demonstrate the size dependent and wavelength
tuned plasmonic NIR absorption of the WN NCs. The NIR-driven photocata-
lytic mechanism of WN NCs is proposed based on density functional theory
50% of solar spectrum. One of the rea-
sons is that the low energetic NIR photons
only can be utilized to generate free car-
riers by narrow bandgap semiconductors,
up-conversion luminescent material or
some localized surface plasmon resonance
(
LSPR) enhanced noble metal nanomate-
[3]
rials. These materials always suffer from
single-wavelength response, poor chemical
stability and potentials mismatching. In
many cases, the chemical potentials of
hole–electrons are not suitable to support
−
1
the generation of ROS (•OH, •O2 and O )
2
through individual transition for previous
[4]
materials. What is more, the NIR-driven
photocatalytic mechanism remains to be
ambiguous due to the absence of theoret-
ical investigation on the mechanism of NIR
absorbance and surface/interface ROS gen-
eration kinetics. Therefore, exploiting high-
efficiency NIR-reactive photocatalyst and
comprehensive understanding NIR-driven
(
DFT) calculated electronic structure and facet dependent O (or H O) mole-
2 2
cular activation, radicals scavenging test, spin trapped electron paramagnetic
resonance measurements, and ultraviolet photoelectronic spectrum (UPS).
Overall, the results in this work pave a way for the application of low-work-
function materials as highly reactive NIR photocatalyst.
[5]
photocatalytic mechanism are urgent.
Compared with traditional semiconduc-
tive photocatalysts, metallic photocatalysts
with higher carrier density, excellent conductivity, and low-work
function open the door for the aimed NIR-responsive photo-
catalysis. Until now, many theoretical calculations and experi-
mental results have revealed that low-work-function catalysts,
such as Ti C T MXene (Φ = 3.4 eV), halogen-doped monolayer
1
. Introduction
Employing photocatalyst to convert the solar energy into the
available chemical energy or achieve environmental purification
has been widely investigated since the rising concerns on energy
3
2 X
[1]
crisis and environmental pollution. Until now, most of reported
g-C N (Φ = 3.3–4.15 eV), LaCu0.67S1.33 (Φ = 3.5 eV) and MgO
3 4
photocatalysts are ultraviolet–visible responsive,[ failing in
2]
(Φ = 3.7 eV), can server as good electron donor and facilitate
W. Huang, J. Wang, P. Ding, Prof. J. Liu, Prof. B. Yang, Prof. W. Cao
School of Instrumentation Science and Engineering
Harbin Institute of Technology
Prof. F. Wu
Department of Physics
School of Science
Harbin University of Science and Technology
Harbin 150080, China
Harbin 150080, P. R. China
E-mail: binyang@hit.edu.cn
Y. Gao, Dr. M. Yan, Prof. C. Guo
School of Chemistry and Chemical Engineering
Harbin Institute of Technology
Dr. D. Liu
School of Chemical and Environmental Engineering
Harbin University of Science and Technology
Harbin 150040, China
Harbin 150080, P. R. China
E-mail: chongshenguo@hit.edu.cn
E-mail: danqingliu76@163.com
Prof. W. Cao
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/smll.202004557.
Department of Mathematics and Materials Research Institute
The Pennsylvania State University
Pennsylvania, PA 16802, USA
DOI: 10.1002/smll.202004557
Small 2020, 2004557
2004557 (1 of 8)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
the escaping of the electron from the surface of material, and
photocatalyis application[10] In this work, the theoretical cal-
culation of DFT and finite element analysis (FEA) are carried
out to investigate the electronic structure and optical prop-
erty of the cubic phase (WN-cub) firstly. Then, the NIR-driven
photocatalytic performance of cubic phase WN is systematically
studied. Finally, a possible plasmonic enhanced ROS activation
and NIR-driven photocatalytic mechanism of low-work-function
WN NCs are proposed based on the DFT calculation of crystal
facet dependent ROS activation and ROS detection results.
[6]
thus promote ROS activation potentially. Meanwhile, some
metallic candidates such as Co1.11Te , MoO , CuS nanosheet,
2
2
and MoN have been successfully applied for the Vis/NIR-driven
photocatalytic CO reduction, water splitting or organic com-
2
[7]
pound degradation. Small or zero bandgap, quasi-continuous
occupation electronic state and high carrier density of metallic
materials are benefit for the NIR-induced consecutive plas-
monic interband/intraband electronic transition, which can
pump the free carriers to suitable energy state and initiate
photocatalytic process.[ Meanwhile, under the light irradiation,
the photoexcited hot electrons are more readily to escape from
surface of materials for the low-work-function metallic photo-
catalysts, being in favor of efficient electrons transfer between
photocatalyst and the lowest unoccupied molecular orbitals of
absorbed molecules (O and H O). Resultantly, the ROS can be
8]
2
. Results and Discussion
In this work, we predicted that the WN species would be a
promising candidate as directly NIR-driven photocatalyst. We
first carry out the first-principle calculations of DFT to predict
the electronic structure and optical properties of cubic WN.
(Details of the calculation are available in the Supporting Infor-
mation) The relaxed lattice parameters are a = b = c = 4.347 Å
for WN-cub, which are in good accordance with previous
2
2
activated with low energy barrier and it makes the utilization
of low energetic NIR light in the photocatalytic reaction to be
possible.
The metallic WN with high free electron density and good
conductivity is quite similar to these metallic photocatalysts,
and thus the strong plasmonic enhanced NIR absorption
and direct NIR-driven photocatalytic performance are highly
expected.[ However, to the best of our knowledge, applica-
tion of WN as NIR-driven photocatalyst has not caught much
attentions yet. Previously, Wang et al. operate an overall-water-
spitting photocatalysis with black WN under 765 nm, which
enlighten us to employ real NIR light above 800 nm for the
[
11]
results. In the Figure 1a and Table S1 (Supporting Informa-
tion), the enthalpy of WN-cub is calculated −2204.6 eV, indi-
cating that it is thermodynamically stable phase as expected.
The sliced EDD map with quadruple symmetry (Figure 1b)
9]
[
12]
correspond to the (1 1 0) crystal facet of WN-cub. According
to the Mulliken charge analysis (Table S1, Supporting Infor-
mation) and the EDD map, it is easily to find that WN-cub
shows strong charge transfer between W and N atoms
Figure 1. DFT calculated electronic structure and optical properties of WN-cub. a) The optimized 3D structure. b) Electron density difference (EDD)
distribution sliced along (1 1 0) facets, wherein red and blue color represent charge accumulation and depletion respectively. c) Total density of state
(DOS) and the corresponding projected DOS for W and N atoms. d,e) Calculated dielectric function and optical absorption spectra of WN from visible
to NIR region.
Small 2020, 2004557
2004557 (2 of 8)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
−
(
0.72 e for WN-cub). The calculated total DOS and band struc-
Figure 2b–d, all the peaks are well-matched with the standard
ture (Figure S1, Supporting Information) both demonstrate the
metallic behavior of WN-cub as the continuously electronic
occupation states across the fermi level.[13] According to pre-
vious studies, the DOS and band structure can be divided into
B , CB and B band, where B band is occupied by the strong
PDF card and verify the successful synthesis of WO •H O
3
2
NCs, WO NCs, WN NCs step by step. The scanning electron
3
microscopy (SEM) (Figure 3a–c) and transmission electron
microscopy (TEM) (Figure 3d–f) are employed to reveal the
morphological features of the samples. The WO •H O NCs are
-
1
1
-1
3
2
hybridized W 5d and N 2p states, CB band is the conduction
seen with sharp edges, and statistical particle size from SEM
image is about 166.0 nm (Figure S3, Supporting Information).
The subsequent thermal process makes all the edges become
blunt and the particles size increases to 225.3 nm eventually
(Figure S4, Supporting Information). To further investigate
the crystallographic structure of the samples, the high resolu-
tion TEM and select area electronic diffraction (SAED) are per-
formed. The clear crystal fringes in Figure 3g,i,k are measured
as 0.378, 0.334, and 0.234 nm, which could be well indexed
as the (200), (021), and (111) crystal facet of the orthorhombic
WO •H O, monoclinic WO and cubic WN, respectively. What
is more, the corresponding SAED (Figure 3h,j,l) with sym-
metrical and theoretical self-consistent diffraction pattern also
shows the high crystallinity of the samples. Thus, we could con-
firm that the cubic WN NCs are successfully synthesized with
suitable size, phase, and crystallinity.
band and partially occupied by W 5d states, and B band is
1
the empty band above the Fermi level (Figure 1c).[
10,14]
What is
more, the work function of WN-cub is about 3.47 eV (Figure S2,
Supporting Information), being much lower than most of pre-
viously reported photocatalysts.[
6a,c]
Therefore, the lower work
function of WN-cub foreseeing its superiority potential on
electron donation and ROS activation.[ Finally, the dielec-
tric function and optical absorption coefficient are calculated
15]
(Figure 1d,e). The real part of the dielectric function presents a
negative value window between 600 and 2100 nm for WN-cub,
and the image part dielectric function of WN-cub shows a peak
at 1500 nm, revealing the significant NIR-plasmonic behavior
the metallic WN.[
3
2
3
16]
According to above DFT calculation results, the metallic
cubic WN possesses low-work function and strong plasmonic
assisted NIR absorption, endowing it with more favorable elec-
tronic property as promising NIR-driven photocatalyst. There-
fore, the cubic phase WN and its NIR-driven photocatalytic
performance are systematically investigated in this work later.
WN NCs are fabricated via a three-step procedure, as shown in
Figure 2a. The green WO •H O NCs are first synthesized by
To further investigate the chemical nature of WN NCs, the X-ray
photoelectronic spectra (XPS) are conducted. The full range survey
in Figure 3m shows the clear elemental peaks of W (4s, 4p, 4d,
and 4f), N1s, O1s, and C1s orbits, and no other impurity elements
are evidently observed. The dominant peak in the high resolution
XPS spectra of N 1s (Figure S5, Supporting Information) locates
at 397.4 eV, which can be assigned to the WN bonding of WN.
NO peaks at 399.8 may originate from the superficially adsorbed
H O. What is more, the high-resolution core-level of W in the
3
2
a hydrothermal method, wherein the tartaric acid and Na WO
2
4
solution are used as raw materials.[ Followed by annealing of
17]
WO •H O NCs, the yellow WO NCs are formed. Finally, the
3
2
3
2
4f
black WN NCs are obtained via high temperature nitridation
range of 32.3 to 39.3 eV (Figure 3n) can be fitted into the quad-
ruple peaks, among which the doublet peaks at low binding energy
region (35.8 and 33.3 eV) are corresponding to the WN chemical
of WO NCs under NH atmosphere (see Supporting Informa-
3
3
[18]
tion for the details). According to X-ray diffraction analysis in
Figure 2. a) Schematic illustration of synthetic process of the WN NCs. XRD pattern and 3D crystal structure of b) orthorhombic WO •H O, c) monoclinic
3
2
WO and d) cubic WN.
3
Small 2020, 2004557
2004557 (3 of 8)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 3. Morphology and crystal structure analysis on the samples. SEM, TEM, HRTEM, and SAED of orthorhombic a,d,g,h) WO •H O NCs;
3
2
b,e,i,j) monoclinic WO NCs; and c,f,k,l) cubic WN NCs NCs. m) Full XPS and n) high resolution deconvoluted W 4f XPS.
3
bond (W3 4f5/2 and W3+ 4f7/2),[19] while the shoulder peaks at the
+
3D FEA is performed to simulate the local electric field distribu-
tion and the absorption cross-section σ (see Supporting
high binding energy range (38.0 eV and 35.0 eV) may come from
the trace superficially adsorbed H O or oxygen (Figure S6, Sup-
porting Information).
Then, the optical properties and the NIR-driven photo-
catalytic performance of the WN NCs are evaluated. Figure 4a
shows the optical absorbance of the WN NCs. The black powder
of WN NCs (inset of Figure 4a) shows strong photoabsorption
over the full tested spectra range, especially in the visible–NIR
region, being in accordance with our previous DFT prediction.
To further verify the plasmonic nature of the metallic WN, the
absorbance
[
7d]
Information). According to the statistical particle size from
SEM image (Figure S4, Supporting Information), the 3D FEA
model is built by taking particle size of 100–350 nm into con-
sideration. As revealed by computation results in Figure 4b–e,
the irradiated WN may generate localized surface plasmon
resonance (LSPR) hot spots and the homogeneity of the hot
spots could be significantly improved with wavelength incre-
ment. What is more, the peaks of the localized electric field
2
enhancement |E/E | (Figure 4f) and the absorption cross-section
0
Figure 4. Optical property and NIR-driven photocatalytic investigation of WN NCs. a) UV–Vis–NIR diffuse reflection spectrum (the inset photograph
shows the black powder of WN NCs); FEA simulation of WN NCs: b–e) The xz-plan |E/E | distribution for single particle (particle size D = 225 nm)
0
under λ = 800, 1200, 1600, and 2000 nm monochromatic light irradiation; f) the calculated particle size dependent of |E/E | distribution and g) the
0
absorption cross-sections. h) The NIR light excited transition photocurrent at 1 V versus Ag/AgCl in 0.1 ꢀ Na SO electrolyte; WN NCs used as NIR-
2
4
driven photocatalyst for photocatalytic i) degradation of MB, j) hydroxylation of TPA, and k) de-esterification of H DCFDA.
2
Small 2020, 2004557
2004557 (4 of 8)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
(Figure 4g) red-shift obviously to NIR region as particle size
nonluminous TPA and H DCFDA could be oxidized by ROS
2
increasing. Hence, the broadband photoabsorption of WN
NCs in Figure 4a could be interpreted as the collective LSPR
effect from different sized WN NCs. Upon NIR irradiation,
the enhancement of the localized electromagnetic (EM) field
coupling may happen at the interfaces of WN NCs, and then
form “hot spots” to promote the generation of the energetic
hot electrons (or holes).[1] Thereafter, these plasmonic-excited
hot electrons (holes) may further transfer to the material inter-
face and participate in photochemical reactions. Given this, the
NIR excited photoelectronic effect is then studied on WN NCs
by photocurrent measurements. As shown in the Figure S7
and form photoluminescent 2,5-dihydroxyterephthalic acid
(DTPA) and 2′,7′-dichlorofluorescein (DCF). The DTPA and
DCF could be excited by 320 and 510 nm light and give lumines-
cence at 425 and 520 nm, respectively. As the results shown
in Figure 4j,k, the photoluminescence intensity of DTPA and
DCF increases with the NIR irradiation time, demonstrating
the high photocatalytic activity of WN NCs in hydroxylation and
de-esterification reaction as well. Such encouraging NIR-driven
photocatalytic performances of WN NCs show promising appli-
cations in many ROS-mediated chemical reactions.
The previous results of DFT calculations, characterizations,
FEA simulation and photocatalytic performance have revealed
the excellent direct NIR-driven photocatalytic properties of
the brand-new metallic WN NCs. However, the mechanism
behind the experimental results has not yet been disclosed. It
is well known that the generation of ROS is the prerequisite
to initiate photocatalytic reactions, and the activation of O2
[
7d]
(Supporting Information) and 4h, the photocurrent under NIR
light is higher than that in dark condition in the full test region
and the transient photocurrent signals exhibit evident light-
−
2
tuned behavior with net photocurrent response of ≈2.5 µA cm
upon NIR irradiation, indicating that the NIR-excited hot elec-
tron–hole pairs could be effectively separated.[
14d]
The strong
LSPR photoabsorption and robust photocurrent response of
WN NCs are beneficial for the generation of ROS during the
photocatalytic process. Consequently, above results forecast the
WN NCs as direct NIR-driven photocatalyst.
and H O molecular is the prerequisite to generate ROS. Then,
2
the absorption and activation of O and H O molecular on the
2
2
typical (100), (110), and (111) crystal facets of cubic WN (noted
as WN-100, WN-110 and WN-111, Figure S8, Supporting Infor-
mation) are studied by DFT calculation. The WN-100 shows
the lowest surface energy (Table S2, Supporting Information),
In this work, NIR-driven photocatalytic activity is first
evaluated by photodegradation of methylene blue (MB) with
a 200 W NIR lamp (see Supporting Information for experi-
mental details). The photocatalytic behavior of C /C with time
[
10]
proving that the (100) crystal facet is the most thermostable.
However, the WN-110 seems to be more preferable to absorb
t
0
is given in Figure 4i. About 48% of MB is degraded by WN
NCs within 4 h in comparison with control group received NIR
irradiation only, which is superior to previously reported NIR
photocatalysts of up-conversion or polaron transfer.[ Besides,
the NIR-driven photocatalytic hydroxylation and de-esterifica-
tion performance of colorless aromatic compounds, such as
terephthalic acid (TPA) and 2′,7′-dichlorodihydrofluorescein
the O and H O molecular as the adsorption energies of both
2
2
H O and O are smallest as compared with other crystal facets
2
2
(Table S3 and Figures S9 and S10, Supporting Information).
5]
The charge transfer between the adsorbate (O and H O mole-
2
2
cular) and the WN slabs is investigated by the Bader charge
[
20]
Δq. As shown in Figure 5a–c, the total charge changes (Δq) of
−
H O molecular in WN-100, WN-110, and WN-111 are -0.014 e ,
2
−
−
diacetate (H DCFDA), are also taken into assessment. The
-0.043 e and +0.0065 e , respectively. The positive Δq (H O)
2
2
Figure 5. Theoretical calculation and experimental evidence for demonstrating facet dependent H O and O molecule activation and the NIR-triggered
2
2
ROS formation on WN. a–c) EDD distribution of different WN facet after H O adsorption. d) ESR spectra of WN NCs in the present of DMPO for
2
1
•
OH detection. e–g) EDD distribution of different WN facet after O adsorption. h) ESR spectra of WN NCs in the present of TEMP for O detection.
2
2
Bader charge Δq presents the EDD of the H O or O molecule, positive value indicates electrons transfer from WN surface to the absorption molecule,
vice versa.
2
2
Small 2020, 2004557
2004557 (5 of 8)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
value in WN-111 forecasts the poor H O activation ability in
The sequential electronic state occupying across the Fermi
level also indicates the metallic nature of WN (Figure 6a).
The VB-XPS spectrum in Figure S15 (Supporting Informa-
tion) shows two intense peaks in 0–4 and 4–12 eV, which are
denoted to the CB and B-1 band, respectively. The measured
work function and relative band position of CB and B-1 are
quiet agreement with the DFT calculation results showed in
Figure S2 (Supporting Information) (3.47 eV) and Figure 1c.
2
(
111) surface. The oxidation process of H O to •OH requires
2
the photocatalyst to facilitate the electron transfer from H O
2
to the catalysts. Obviously, WN-110 shows most negative value
and resultant best activation ability on H O. Meanwhile, the
2
total charge changes (Δq) of O molecular in WN-100, WN-110
2
−
−
and WN-111 (Figure 5e–g) are determined as +0.61 e , +0.91 e ,
−
and +0.25 e , respectively. By capturing electrons from the
slab, the bonding length of O is prolonged (Table S3, Sup-
Resultantly, the chemical potential of the E can be calculated
2
f
−
porting Information), corresponding to catalytic activation of
as −0.91 eV (vs NHE), being lower than O /•O2 of −0.046 eV.
2
[21]
O molecular. Similarly, WN-110 also accounts for the best
On the other hand, the chemical potential of B band is about
2
-1
activation of O2 ability as evidenced by highest total charge
change and bonding length. From above results, we can con-
clude that the WN-110 slab presents the strongest electron
transfer between O (H O) molecular and WN, which is benefi-
cial to the ROS generation and the subsequent surface catalytic
reaction.
−4.2 eV relative to the Fermi level. So the B-1 band relative to
−
NHE is about 3.33 eV, being higher than OH /•OH of 1.99 eV.
−
Consequently, •OH and •O2 can be generated by NIR plas-
monic excitation, which can further turn into other ROS, such
2
2
1
as •HO and O . At this point, it can be confirmed that the
2
2
−
+
LSPR-excited hot carriers (e and h ) and ROS can be effec-
tively generated under NIR light radiation, and WN-110 crystal
facet gives the highest photocatalytic activity during sur-
face chemical reaction (as the schematic diagram shown in
Figure 6b) by combining all the theoretical and experimental
results. The plausible mechanism (Figure 6c) and relevant
chemical reactions related to the NIR-driven photocatalytic
property of WN NCs can be summary as following:
We then conduct photocatalytic degradation of MB in
presence of ROS scavengers of EDTA-2Na and isopropanol
+
(
IPA) for quenching •OH and hole (h ), respectively. As the
results shown in Figure 4i, the C /C neither increases nor
t
0
keeps steady when the ROS scavengers is added into the
+
photocatalytic system, indicating the •OH and h have been
generated and played important role during the photocata-
lytic process. Moreover, the ROS are directly probed by the
DMPO and TEMP spin-trapping electron spin resonance
∗
−
+
WN + hv
(
NIR
)
→ WN + e + h
(1)
(2)
(3)
(4)
(5)
(6)
(
ESR) measurement. The typical 1:2:2:1 quadruple resonant
peak (Figure 5d) and 1:1:1 triplex resonant peak (Figure 5h)
in the range of 3480–3530 Gs could be evidently observed
after 20 min NIR irradiation, demonstrating the forma-
∗
−
−
WN( + O + e → WN
110
)
2
(
110
)
+ •O2
1
[4]
tion of •OH and O under NIR excitation with WN NCs.
2
∗
+
+
To further understand the generation of ROS, we use the
colorless nitrotetrazolium blue chloride (NBT) solution as
model to evaluate the generation of superoxide.[ As shown
in Figures S11 and S12 (Supporting Information), the con-
centration of NBT shows significant decrement within 2 h
photocatalytic reaction and the NIR-driven photocatalytic
NBT degradation rate could maintain about 7% after four
cycle runs with no crystal phase transition and morphology
changing on WN NCs (Figures S13 and S14, Supporting
Information), indicating the excellent ROS generation and
good stability of WN NCs as well.
WN( + H O+ h → WN
)
+ H O
110
)
2
(
110
2
22]
+
+
H O → H + •OH
2
−
+
•O
2
+ H + → •HO
2
−
−
1
•O
2
+ •OH → OH + O
2
The energy level matching between WN NCs mediated
electron–hole pairs and redox potential of ROS generation
(1) NIR light excites the collective LSPR effect in WN NCs and
plenty of “hot electron-hole pairs” are generated by the plas-
monic interband/intraband quasi-continuous transition
(Equation 1).
−
−
(
O /•O , OH /•OH) is another important issue we should
2 2
confirm. We believe that the ability of WN to reduce O and
2
oxidize H O for producing superoxide radical and hydroxyl
(2) The most reactive photocatalytic surface of WN 110 prefers to
adsorb O and H O molecular. Then, the activation of O and
2
radical simultaneously originates from the typical metallic like
band structure. First, to further confirm the metallic property
of the cubic WN NCs, we also test the electrical resistivity of
WN NCs vis 4-terminal method. The resistivity of WN at room
temperature (300 K) is about 2.5 × 10− Ω m, which is smaller
than the previous reports, revealing the high electron conduc-
tivity and mobility in the WN nanoparticles as well. Then, the
electronic structures of the as prepared WN NCs were char-
acterized by UPS and XPS valance band spectra. The inter-
cepts in the high (17–19 eV) and low (−5–3 eV) binding energy
region are demined to be 17.61 and −0.65 eV. Accordingly,
the work function of WN NCs can be calculated as 3.59 eV.
2
2
2
H O molecular can be achieved by charge transfer between
2
the molecular and WN (Equations 2 and 3).
+
(3) The activated H O turns into •OH on the WN 110 surface
2
3
−
(Equation 4). The activated O turns into •O , and subse-
2
2
+
quently reacts with H to form •HO on the WN surface
(Equation 5). The •O2 may further react with •OH to gener-
2
−
1
ate O (Equation 6).
2
(4) All the ROS generated by WN NCs participate in the NIR-
driven photocatalytic reactions, including degradation of
MB (or NBT), hydroxylation of TPA and de-esterification of
H DCFDA.
2
Small 2020, 2004557
2004557 (6 of 8)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
Figure 6. NIR-driven photocatalytic mechanism investigation of WN NCs. a) UPS valance band spectra. b) Schematic diagram of the NIR-excited holes
and electrons in WN NCs. c) Illustration of NIR-driven photocatalytic process.
Young Scholar (Grant No. JC2017001), and Harbin Applied Technology
3
. Conclusion
Research and Development Project (2017RAXXJ003). W.H. extends
his gratitude to Mr. Zhiqiang Yu, Mr. Xiaobin Hao, and Dr. Chengyue
Sun for their strong support and enthusiastic help in 3D modeling and
experimental implementation.
In summary, we have confirmed that WN NCs can be utilized
as NIR-driven photocatalyst through the combination of DFT
calculation and many experimental evidences in this work. The
full spectra absorption and good NIR photocurrent response
originated from the LSPR effect of the metallic WN. The 3D
FEA results demonstrated the size dependent and wavelength
tuned plasmonic behavior as well. The DFT calculated facet
dependent O and H O molecular activation revealed that the
Conflict of Interest
The authors declare no conflict of interest.
2
2
most reactive catalytic facet was the (110) surface. The UPS
results proved the metallic nature and the suitable chemical
redox potential for generating ROS by WN NCs. Therefore, the
successful direct NIR-driven photocatalytic degradation of MB,
Keywords
DFT calculation, finite element analysis, nanocubes, NIR photocatalysts,
tungsten nitride
hydroxylation of TPA and de-esterification of H DCFDA were
2
realized in this work.
Received: July 28, 2020
Revised: September 23, 2020
Published online:
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
[
1] L. Zhou, D. F. Swearer, C. Zhang, H. Robatjazi, H. Zhao,
L. Henderson, L. Dong, P. Christopher, E. A. Carter, P. Nordlander,
N. J. Halas, Science 2018, 362, 69.
Acknowledgements
This work was financially supported by the National Natural Science
Foundation of China (Nos. 51572056, 51572059
Heilongjiang Provincial Natural Science Foundation for Distinguished
[
2] a) K. Wenderich, G. Mul, Chem. Rev. 2016, 116, 14587; b) S. Shen,
J. Chen, M. Wang, X. Sheng, X. Chen, X. Feng, S. S. Mao, Prog.
Mater. Sci. 2018, 98, 299; c) L. Lin, Z. Yu, X. Wang, Angew. Chem.,
Int. Ed. 2019, 58, 6164.
& 51602083),
Small 2020, 2004557
2004557 (7 of 8)
© 2020 Wiley-VCH GmbH

www.advancedsciencenews.com
www.small-journal.com
[
3] a) X. Wu, Y. Li, G. Zhang, H. Chen, J. Li, K. Wang, Y. Pan, Y. Zhao,
Y. Sun, Y. Xie, J. Am. Chem. Soc. 2019, 141, 5267; b) Y. Sang, Z. Zhao,
M. Zhao, P. Hao, Y. Leng, H. Liu, Adv. Mater. 2015, 27, 363; c) J. Li,
X. Wu, W. Pan, G. Zhang, H. Chen, Angew. Chem., Int. Ed. 2018, 57,
[10] Y. Wang, T. Nie, Y. Li, X. Wang, L. Zheng, A. Chen, X. Gong, H. Yang,
Angew. Chem., Int. Ed. 2017, 56, 7430.
[11] Y. Benhai, W. Chunlei, S. Xuanyu, S. Qiuju, C. Dong, J. Alloys
Compd. 2009, 487, 556.
4
2
91; d) J. Wu, Z. Zhang, B. Liu, Y. Fang, L. Wang, B. Dong, Sol. RRL
018, 2, 1800039; e) Q. Zhang, X. Liu, L. Tan, Z. Cui, X. Yang, Z. Li,
[12] K. Balasubramanian, S. Khare, D. Gall, Phys. Rev. B 2016, 94, 174111.
[13] M. Kavitha, G. Sudha Priyanga, R. Rajeswarapalanichamy,
K. Iyakutti, J. Phys. Chem. Solids 2015, 77, 38.
[14] a) X. Xu, C. Randorn, P. Efstathiou, J. T. S. Irvine, Nat. Mater. 2012,
11, 595; b) X. Jiao, K. Zheng, Z. Hu, Y. Sun, Y. Xie, ACS Cent. Sci.
2020, 6, 653; c) L. Liang, X. Li, J. Zhang, P. Ling, Y. Sun, C. Wang,
Q. Zhang, Y. Pan, Q. Xu, J. Zhu, Y. Luo, Y. Xie, Nano Energy 2020,
69, 104421; d) S. Xu, S. Gong, H. Jiang, P. Shi, J. Fan, Q. Xu, Y. Min,
Appl. Catal., B 2020, 267, 118661.
Y. Liang, S. Zhu, K. W. K. Yeung, X. Wang, Y. Zheng, S. Wu, Appl.
Catal., B 2019, 258, 117980; f) W. Gao, B. Tian, W. Zhang, X. Zhang,
Y. Wu, G. Lu, Appl. Catal., B 2019, 257, 117908; g) J. Li, J. Wang,
G. Zhang, Y. Li, K. Wang, Appl. Catal., B 2018, 234, 167.
4] Y. Nosaka, A. Y. Nosaka, Chem. Rev. 2017, 117, 11302.
5] L. Wang, X. Xu, Q. Cheng, S. X. Dou, Y. Du, Small 2019, 1904107,
https://doi.org/10.1002/smll.201904107.
[
[
[
6] a) B. Zhu, J. Zhang, C. Jiang, B. Cheng, J. Yu, Appl. Catal., B 2017,
[15] Z. Kou, T. Wang, H. Wu, L. Zheng, S. Mu, Z. Pan, Z. Lyu, W. Zang,
S. J. Pennycook, J. Wang, Small 2019, 15, 1900248.
2
07, 27; b) T. N. Ye, Y. Lu, J. Li, T. Nakao, H. Yang, T. Tada, M. Kitano,
H. Hosono, J. Am. Chem. Soc. 2017, 139, 17089; c) T. Jaouen,
B. Hildebrand, P. Aebi, G. Delhaye, S. Tricot, B. Lépine, G. Jézéquel,
P. Schieffer, J. Phys. Chem. C 2017, 121, 4363; d) C. Peng, P. Wei,
X. Li, Y. Liu, Y. Cao, H. Wang, H. Yu, F. Peng, L. Zhang, B. Zhang,
K. Lv, Nano Energy 2018, 53, 97.
7] a) Y. Xu, J. Mo, G. Xie, D. Ding, S. Ding, X. Wang, C. Li, Chem.
Commun. 2019, 55, 6862; b) Z. Hu, G. Liu, X. Chen, Z. Shen,
J. C. Yu, Adv. Funct. Mater. 2016, 26, 4445; c) X. Li, L. Liang, Y. Sun,
J. Xu, X. Jiao, X. Xu, H. Ju, Y. Pan, J. Zhu, Y. Xie, J. Am. Chem. Soc.
[16] a) A. Habib, F. Florio, R. Sundararaman, J. Opt. 2018, 20, 064001;
b) Y. P. Zhu, J. K. El-Demellawi, J. Yin, S. Lopatin, Y. Lei, Z. Liu,
X. Miao, O. F. Mohammed, H. N. Alshareef, Adv. Mater. 2020, 32,
1908392.
[17] X. Su, F. Xiao, Y. Li, J. Jian, Q. Sun, J. Wang, Mater. Lett. 2010, 64,
1232.
[18] C. Zhang, S. Wang, Z. Chen, J. Fan, Z. Zhong, X. Zhang, Nanoscale
2019, 11, 2027.
[19] H. Jin, H. Zhang, J. Chen, S. Mao, Z. Jiang, Y. Wang, J. Mater. Chem.
A 2018, 6, 10967.
[
2
019, 141, 423; d) W. Huang, H. Meng, Y. Gao, J. Wang, C. Yang,
D. Liu, J. Liu, C. Guo, B. Yang, W. Cao, J. Mater. Chem. A 2019, 7,
8538; e) J. Ran, H. Wang, H. Jin, C. Ling, X. Zhang, H. Ju, L. Jing,
[20] H. Liu, P. Chen, X. Yuan, Y. Zhang, H. Huang, L. Wang, F. Dong,
Chin. J. Catal. 2019, 40, 620.
1
J. Wang, R. Zheng, S. Qiao, J. Mater. Chem. A 2018, 6, 23278.
8] X. Zhang, X. Li, M. E. Reish, D. Zhang, N. Q. Su, Y. Gutierrez,
F. Moreno, W. Yang, H. O. Everitt, J. Liu, Nano Lett. 2018, 18, 1714.
9] H. Wang, E. J. Sandoz-Rosado, S. H. Tsang, J. Lin, M. Zhu,
G. Mallick, Z. Liu, E. H. T. Teo, Adv. Funct. Mater. 2019, 29,
[21] X. Sun, X. Luo, X. Zhang, J. Xie, S. Jin, H. Wang, X. Zheng, X. Wu,
Y. Xie, J. Am. Chem. Soc. 2019, 141, 3797.
[22] a) N. Tian, H. Huang, Y. He, Y. Guo, T. Zhang, Y. Zhang, Dalton
Trans. 2015, 44, 4297; b) Y. Zhao, Y. Wang, X. Liang, H. Shi, C. Wang,
J. Fan, X. Hu, E. Liu, Appl. Catal., B 2019, 247, 57; c) J. Guo, Y. Liu,
Y. Hao, Y. Li, X. Wang, R. Liu, F. Li, Appl. Catal., B 2018, 224, 841.
[
[
1
902663.
Small 2020, 2004557
2004557 (8 of 8)
© 2020 Wiley-VCH GmbH