Vacancy-defect modulated pathway of photoreduction of CO2 on single atomically thin AgInP2S6 sheets into olefiant gas

Abstract

Artificial photosynthesis, light-driving CO 2 conversion into hydrocarbon fuels, is a promising strategy to synchronously overcome global warming and energy-supply issues. The quaternary AgInP 2 S 6 atomic layer with the thickness of ~ 0.70 nm were successfully synthesized through facile ultrasonic exfoliation of the corresponding bulk crystal. The sulfur defect engineering on this atomic layer through a H 2 O 2 etching treatment can excitingly change the CO 2 photoreduction reaction pathway to steer dominant generation of ethene with the yield-based selectivity reaching ~73% and the electron-based selectivity as high as ~89%. Both DFT calculation and in-situ FTIR spectra demonstrate that as the introduction of S vacancies in AgInP 2 S 6 causes the charge accumulation on the Ag atoms near the S vacancies, the exposed Ag sites can thus effectively capture the forming *CO molecules. It makes the catalyst surface enrich with key reaction intermediates to lower the C-C binding coupling barrier, which facilitates the production of ethene.

Full text

ARTICLE
Vacancy-defect modulated pathway of
photoreduction of CO
2
on single atomically thin
AgInP
2
S
6
sheets into olefiant gas
Wa Gao1,9, Shi Li2,9, Huichao He 3, Xiaoning Li 4, Zhenxiang Cheng 4, Yong Yang 5, Jinlan Wang 2✉,
Qing Shen6, Xiaoyong Wang 1, Yujie Xiong7✉, Yong Zhou 1,8✉& Zhigang Zou1,8
Artificial photosynthesis, light-driving CO
2
conversion into hydrocarbon fuels, is a promising
strategy to synchronously overcome global warming and energy-supply issues. The qua-
ternary AgInP
2
S
6
atomic layer with the thickness of ~ 0.70 nm were successfully synthesized
through facile ultrasonic exfoliation of the corresponding bulk crystal. The sulfur defect
engineering on this atomic layer through a H
2
O
2
etching treatment can excitingly change the
CO
2
photoreduction reaction pathway to steer dominant generation of ethene with the yield-
based selectivity reaching ~73% and the electron-based selectivity as high as ~89%. Both
DFT calculation and in-situ FTIR spectra demonstrate that as the introduction of S vacancies
in AgInP
2
S
6
causes the charge accumulation on the Ag atoms near the S vacancies, the
exposed Ag sites can thus effectively capture the forming *CO molecules. It makes the
catalyst surface enrich with key reaction intermediates to lower the C-C binding coupling
barrier, which facilitates the production of ethene.
https://doi.org/10.1038/s41467-021-25068-7 OPEN
1Key Laboratory of Modern Acoustics (MOE), Institute of Acoustics, School of Physics, Jiangsu Key Laboratory of Nanotechnology, Eco-materials and
Renewable Energy Research Center (ERERC), National Laboratory of Solid State Microstructures, Collaborative Innovation Center of Advanced
Microstructures, Nanjing University, Nanjing, China. 2School of Physics, Southeast University, Nanjing, China. 3State Key Laboratory of Environmental
Friendly Energy Materials, Southwest University of Science and Technology, Mianyang, China. 4Institute of Superconducting & Electronic Materials,
Innovation Campus, University of Wollongong, Squires Way, North Wollongong, NSW, Australia. 5Key Laboratory of Soft Chemistry and Functional
Materials (MOE), Nanjing University of Science and Technology, Nanjing, China. 6University of Electrocommunication, Grad Sch Informatics and Engineering,
Chofu, Tokyo, Japan. 7Hefei National Laboratory for Physical Sciences at the Microscale, Collaborative Innovation Center of Chemistry for Energy Materials
(iChEM), School of Chemistry and Materials Science, University of Science and Technology of China, Hefei, Anhui, China. 8School of Science and
Engineering, The Chinese University of Hongkong (Shenzhen), Shenzhen, Guangdong, China.
9
These authors contributed equally: Wa Gao, Shi Li.
✉email: jlwang@seu.edu.cn;yjxiong@ustc.edu.cn;zhouyong1999@nju.edu.cn
NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturec ommunications 1
1234567890():,;
Content courtesy of Springer Nature, terms of use apply. Rights reserved

Photocatalytic conversion of CO
2
with H
2
O into solar fuels
would be like killing two birds with one stone in terms of
saving supplying energy and environment, which occurs
mostly on the surfaces of semiconductors through complicated
processes involving multi-electrons/protons transfer reactions1.
Photo-driving CO
2
hydrogenation into C
1
species have been well
achieved in the recent decade2, and our group has exploited a
series of promising photocatalysts to converse CO
2
to selectively
form specific hydrocarbons, such as Zn
2
GeO
4
ultrathin nanor-
ibbons for CH
4
3, atomically thin InVO
4
nanosheets for CO4, and
TiO
2
-graphene hybrid nanosheets for C
2
H
6
5and so on. However,
the controlled C–C coupling to produce high-value C
2
or C
2+
products still remains a great challenge. Olefiant gas (ethylene,
C
2
H
4
) is a chemical source of particular importance due to its
high demand in the chemical industry. C
2
H
4
is usually derived
from steam cracking of naphtha under harsh production condi-
tions (800–900 °C). It is definitely desirable for the realization of
C
2
H
4
synthesis through mild and environmentally benign
pathways6.
Transition metal thio/selenophosphates (TPS) is a broad class
of van der Waals layered structures with two sulfur or selenium
layers sandwiching a layer of metal ions and P
2
pairs and general
compositions of M
4
[P
2
X
6
]4−,[M
2+]
2
[P
2
X
6
]4−, and
M1+M3+[P
2
X
6
]4−, where M1+=Cu, Ag; M3+=Cr, V, Al, Ga,
etc. X =S, Se7. Those quaternary compounds exhibit mixed
electron–ionic conductivity, promising optical and thermoelectric
properties8. AgInP
2
S
6
is a typical TPS with a rhombohedral
structure and contains a sulfur framework with the octahedral
voids filled by Ag, In, and P–P triangular patterns. Each AgInP
2
S
6
monolayer consists of the [P
2
S
6
] anionic complex and two
metallic cations (Ag and In) located at the center of sulfur near-
octahedral polyhedrons connected one with the other by edges.
Semiconducting AgInP
2
S
6
crystal possesses an appropriate
bandgap structure (E
g
=~2.4 eV), which is favored for visible
light absorption9. The low value of the effective mass of electrons
and the high value of the effective mass of holes facilitate accel-
erating the mobility dynamics of photogenerated electrons onto
the surface prior to holes10, which may enhance local electron
density, benefiting the photo-driving reduction reaction. The
centrosymmetry structure of AgInP
2
S
6
also enables the photo-
excited electrons to distribute on the surface of the layer crystal
uniformly11, which may remarkably reduce the energy barrier for
catalytic molecule activation, alter the catalytic reduction path-
way, and enhance yield and enrich species of products.
An atomically thin 2D structure is an ideal platform to provide
atomic-level insights into the structure-activity relationship12.
Firstly, the ultrathin structure allows the photo-generated carriers
to easily transfer from the interior to the surface with shortened
charge transfer distance, decreasing the bulk recombination.
Secondly, large surface exposure renders rich catalytic active sites.
Thirdly, transparency resulting from ultrathin thickness helps for
light absorption. The creation of vacancy defects in the ultrathin
structure can also additionally enrich the reaction intermediates,
resulting in low-coordinated atoms on the surface of the catalyst,
which are known to facilitate to the generation of multi-carbon
species from CO
2
photoreduction13,14.
Herein, we report the synthesis of the AgInP
2
S
6
single atomic
layer (abbreviated as SAL) of ~0.70 nm in thickness through a
facile probe sonication exfoliation of the corresponding bulk
crystal (abbreviated as BC). The sulfur vacancy (abbreviated as
V
S
) defects were introduced in the resulting SAL through an
etching process with H
2
O
2
solution (abbreviated as V
S
-SAL),
which was prospectively utilized for photocatalytic reduction of
CO
2
in the presence of water vapor. While BC and SAL dom-
inantly produce CO, the implemented defect engineering changes
the reaction pathway of the CO
2
photoreduction on V
S
-SAL,
which allows steering CO
2
conversion into C
2
H
4
with the yield-
based selectivity reaching ~73% and the electron-based selectivity
as high as ~89%, and the quantum yield of 0.51% at a wavelength
of 415 nm. Both DFT calculation and in situ FTIR spectra
demonstrate that the key step for the CO production on BC and
SAL follows a conventional hydrogenation process of CO
2
to
form *COOH, which further couples a proton/electron pair to
generate *CO. *CO easily liberates from the defect-free AgInP
2
S
6
surface with low absorption energy to become free CO gas. In
contrast, the introduction of V
S
in AgInP
2
S
6
causes the charge
accumulation on the Ag atoms near V
S
. Thus, the exposed Ag site
in V
S
-SAL can effectively capture the forming *CO, making the
catalyst surface enrich with key reaction intermediates to promote
C–C coupling into C
2
species with the low binding energy barrier.
This work may provide fresh insights into the design of an
atomically thin photocatalyst framework for CO
2
reduction and
establish an ideal platform for reaffirming the versatility of defect
engineering in tuning catalytic activity and selectivity.
Results
Structure characterization of the AgInP
2
S
6
related samples.BC
was synthesized through PVT in a two-zone furnace, which
displays bright yellowish-brown color (Supplementary Fig. 1a).
The SAL was produced through mechanical exfoliation in ethyl
alcohol solution through a probe sonication technique, which can
transfer high energy into layered materials and weaken the Van
der Waals forces between adjacent layers, resulting in effective
delamination. The well-defined Tyndall effect of the resulting
transparent solution of SAL indicates high monodispersity of the
ultrathin sheets (Supplementary Fig. 1b). Etching of SAL with
H
2
O
2
solutions allows to deliberately create V
S
on the surface of
SAL15.
The powder X-ray diffraction (XRD) pattern of BC and SAL
agrees with the simulated one from the crystal structure of ICSD
202185 well with the P
3
1c
space group (Supplementary Fig. 2)12,
and no impurity peaks were detected. The stronger SAL peak
intensity ratio of (002) to (112) relative to BC indicates that the
exfoliation of AgInP
2
S
6
occurs along [001] direction. The field
emission scanning electron microscopy (FE-SEM) image shows
that BC displays an angular shape with an apparent laminar
structure (Supplementary Fig. 3a, b). The energy dispersive
spectroscopy (EDS) spectra demonstrate the uniform spatial
distribution of Ag, In, P, and S (Supplementary Fig. 3c–f). The
TEM image of exfoliated SAL displays light contrast of the
extremely thin 2D structure (Fig. 1a). A magnified transmission
electron microscopy (TEM) image of a vertically standing sheet
shows the single layer with a thickness of ∼0.71 nm (Fig. 1b). A
typical edge-curling sheet as marked with an arrow also
particularly shows the thickness of ~0.72 nm of SAL (Fig. 1b),
well in agreement with the AgInP
2
S
6
monolayer along [002]
orientation [d
(002)
=6.68 Å]. The corresponding atomic force
microscopy (AFM) image of SAL also confirms ~0.66–0.73 nm
range in thickness (Fig. 1c and Supplementary Fig. 4 for more
images), demonstrating the single-atom layer feature. A high-
resolution TEM (HRTEM) image of SAL reveals that the
interplanar d-spacing between the well-defined lattice fringes
were examined 0.54 nm, which can be indexed to (010) (Fig. 1d).
The selected area electron diffraction shows an ordered array of
spots recorded from [001] zone axis (Fig. 1d, inset), confirming
that SAL is of single crystallinity and preferentially enclosed by
{002} top and bottom surfaces. The crystalline model of SAL
from top and side views was schematically illuminated in Fig. 1f.
With H
2
O
2
solution treatment for optimized 10 s, the sulfur
atoms, which locate outermost in SAL, can be partially etched
away from the surface to form V
S
. The generation of V
S
was
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-25068-7
2NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturecom munications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

confirmed with the electron paramagnetic resonance (EPR)
spectra (Supplementary Fig. 5). The Raman spectra show that
the peak intensity of both S–P–PandP–S–PforVs-SALwere
lowered, compared with those of SAL (Supplementary Fig. 6),
which additionally verifies the detected defect sites can be
assigned to V
S
16, rather than the open of S–M (metal) bond or
the possible insertion of O atoms.
No obvious difference of the XRD patterns between SAL and
Vs-SAL demonstrates no crystal structure change of the SAL
before and after H
2
O
2
etching treatment (Supplementary Fig. 2).
The TEM image also shows that the resulting V
S
-SAL
10
displays
no morphology change in ultrathin structure (Supplementary
Fig. 7). The corresponding EDS reveals that Ag, In, and P
contents were nearly stoichiometric 1:1:2 of AgInP
2
S
6
, expect S
element less than the stoichiometric ratio (Supplementary Fig. 8).
It indicates that H
2
O
2
treatment mainly leads to V
S
, and has no
etching effect on other moieties, which was also verified with the
following XPS and the X-ray absorption near edge structure
(XANES) spectra. The atomic resolution, aberration-corrected
high-angle annular dark-field scanning TEM (HAADF-STEM)
clearly reveals that a considerable number of V
S
were confined in
the sheet (Fig. 1e), in contrast to the few sporadic ones in SAL
(Supplementary Fig. 9).
Full XPS spectra demonstrate the presence of Ag, In, P, and S
(Supplementary Fig. 10a). The high-resolution S 2p spectrum of
BC shows the S 2p peak falling between 162 and 164 eV
(Supplementary Fig. 10b), revealing the −2 oxidation state of S.
The S 2ppeaks of SAL show dramatic low binding energy shift,
compared with BC, and V
S
-SAL
10
possesses further low-energy
shift. The former shift may originate from exfoliation-resulting
monolayerization17 and the latter from V
S
15. As the decrease of
binding energy indicates the enhanced electron screening effect
due to the increase of the electron concentration15,18, it implies
that the electron density around the S sites increases in the
sequence of BC, SAL, and V
S
-SAL
10
. It reveals that the residual S
atoms exist in an electron oversaturated form and possess high
electron density. No obvious change of binding energy of P
elements was observed (Supplementary Fig. 10c), further
demonstrating that the mechanical exfoliation and chemical
etching only damage sulfur atoms and have little effect on P
moiety. Weak O1s XPS peaks were observed for both SAL and
Vs-SAL
10
(Supplementary Fig. 10d), which more likely originate
from absorbed components from the ambiance. The almost same
intensity and location of O1s peak indicate no apparent oxidation
change before and after H
2
O
2
treatment. The pre-edge char-
acteristic of the XANES spectra of the S K-edges of three
AgInP
2
S
6
was shown in Fig. 2a, which could be fitted with
components of a spin−orbit split. The spectra indicate the
existence of main transitions energies between 2460 and 2500 eV,
which originates from the excitation of an electron from a 1S
Fig. 1 Morphological structure characterization of the fabricated SAL and V
S
-SAL
10
.TEM images of aSAL, bvertically standing, and b’laying single piece
SAL, cAFM image of SAL showing an average thickness of ~0.69 nm. dHRTEM image and the EDS. eHAADF-STEM image of V
S
-SAL
10
, in which the
atomically dispersed V
s
are highlighted with the yellow circles. fThe crystalline models of SAL from top and side views.
Fig. 2 XANES spectra of BC, SAL, and Vs-SAL10. a S and bP K-edge XANES spectra of BC, SAL, and Vs-SAL10.
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-25068-7 ARTICLE
NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturec ommunications 3
Content courtesy of Springer Nature, terms of use apply. Rights reserved

inner orbital to a higher-energy orbital as a result of interaction
with an X-ray. In comparison with BC, SAL shows a shift for S K-
edge peaks to the lower energy side. This can be explained by the
fact that the core electrons of S become more loosely bound after
mechanical exfoliation due to the increased screening of the
nuclear charge. Through V
S
engineering, the S K-edge of V
S
-
SAL
10
can have a further small move to the lower energy side
(Fig. 2a). Moreover, the K-edge peak of P between 2100 to
2250 eV exhibits almost no differences among BC, SAL, and V
S
-
SAL
10
(Fig. 2b), which is in good agreement with the above-
mentioned XPS results.
The UV–vis diffuse reflectance spectra show that the bandgap
of SAL was determined 2.66 eV, a little larger than that of BC
(2.31 eV) (Supplementary Fig. 11), exhibiting a strong quantum
size effect in the lateral direction. V
S
-SAL
10
displays a slightly
narrowed bandgap (2.57 eV) with respect to SAL. It derives from
that introduction of V
S
may tailor the electronic structure of SAL
through generating impurity states near the conduction band
(CB) edge, which can be overlapped and delocalized with the CB
minimum edge, leading to a reduced bandgap that may broaden
the light absorption edge19,20. The XPS spectra show that the
Ag
3d
peak of Vs-SAL
10
shifts to lower binding energy relative to
that of SAL (Supplementary Fig. 10e), confirming the valance
changes of Ag in Vs-SAL
10
. The VB change of AgInP
2
S
6
may lead
to the corresponding changes of its CB20. The Mott–Schottky
plots reveal that the CB edge of V
S
-SAL
10
upshifts by ~0.06 and
~0.26 eV, relative to that of SAL and BC, respectively, as
schematically illustrated in Supplementary Fig. 12. All BC, SAL,
and V
S
-SAL
10
were thus confirmed to possess suitable bandgaps
as well as the appropriate band edge positions for photocatalytic
CO
2
reduction under visible-light irradiation.
Photocatalytic performance toward CO
2
photoreduction.The
photocatalytic CO
2
conversion was carried out in the presence of
water vapor under simulated solar irradiation (Fig. 3). CO was
detected the major product for BC and SAL (Fig. 3a, b). BC shows
the CO yield of 2.44 μmol g−1for the first hour and a trace amount
of CH
4
of 0.63 μmol g−1(Fig. 3a). The photogenerated holes in
the VB oxidize H
2
O to produce hydrogen ions by the reaction of
H
2
O→1/2O
2
+2H++2e−
. CO is formed by reacting with two
protons and two electrons (CO2þ2eþ2Hþ!CO þH2Oð1Þ),
and CH
4
formation through accepting eight electrons and eight
protons (CO2þ8eþ8Hþ!CH4þ2H2Oð2Þ). SAL exhibits
6.9 and 14.3-time enhancement of production of CO and CH
4
relative to BC, reaching 17.1 and 9.0 μmol g−1for the first hour,
respectively (Fig. 3b). A small amount of H
2
was also generated as a
typical competitive reaction with CO
2
reduction (Supplementary
Fig. 13). The prerogative of atomic ultrathin geometry of SAL
may be mainly responsible for the enhanced photocatalytic
activity besides larger surface area, allowing charge carriers to
move from interior to the surface quickly to conduct catalysis,
avoiding the recombination in the body. A small amount of
C
2
H
4
was also detected for SAL with a yield of 5.3μmol g−1.C
2
H
4
is generated by accepting 12 electrons and 12 protons
(2CO2þ12eþ12Hþ!C2H4þ4H2Oð3Þ). With the H
2
O
2
etching process, excitingly, C
2
H
4
excitingly becomes the main pro-
duct for V
S
-SAL
10
with a yield of 44.3 μmol g−1(Fig. 3c). The cal-
culated yield-based selectivity reaches ~73%, and the electron-based
selectivity is as high as ~89%21 (Fig. 3e). Meanwhile, CO and CH
4
minority products were also traced with the yields of 10.9 and
5.6 μmol g−1, respectively, both less than the case of SAL. It indicates
that the surface of V
S
-SAL
10
preferentially promotes the C
1
inter-
mediates to C–C couple into C
2
product rather than liberate them
into free CO and CH
4
gases. The quantum yield of V
S
-SAL
10
was
measured 0.51% at a wavelength of 415 nm using monochromatic
light (see the details in SI). The etching process time was found
determinative for the dominant production of C
2
H
4
.TheEPR
measurement shows that the signal intensity gradually increases
with prolonging etching time from 5 to 15 s (Supplementary Fig. 5),
indicating being raised a number of V
S
in V
S
-SAL. Elongation of the
etching time from 5 to 10 s was favorable for increasing the yield of
C
2
H
4
(Supplementary Fig. 14). However, a much long etching time
of 15 s decreases activity negatively, which may be due to that an
excess of V
S
defects may accelerate the recombination of
Fig. 3 Photocatalytic CO
2
reduction performance. Photocatalytic gases evolution amounts as a function of light irradiation times of aBC, bSAL, and
cV
S
-SAL. dPhotocatalytic activity for the first hour. eTable illustration for the yield and electron-based selectivities of photocatalytic CO
2
conversion.
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-25068-7
4NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturecom munications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

photogenerated carriers22. The reduction experiment of CO
2
per-
formed in the dark or absence of the photocatalyst shows no
appearance of CO and hydrocarbon products, proving that the
reduction reaction of CO
2
is driven by light under photocatalyst. A
blank experiment with the identical condition and in the absence of
CO
2
shows no appearance of C
2
H
4
,CO,andCH
4
, proving that the
carbon source was completely derived from input CO
2
.Anisotope
labeling experiment using 13CO
2
confirms that the produced C
2
H
4
originates from the input CO
2
(Supplementary Fig. 15). The O
2
production was also detected using the similar isotope H
2
18Otracer
control experiment (Supplementary Fig. 15). It should be mentioned
that after 12 h light irradiation, increased tendency of the generation
of the hydrocarbon products over the present photocatalyst slowed
down. It may be assigned to the potential carbon deposition as
intermediates covering the active sites of the photocatalyst during
the photoreduction process. The problem may be resolved through
post washing treatment to recover the catalytic activity to a certain
extent, as shown in Fig. S16.
Mechanism of the photocatalytic performance of the V
S
-SAL.
DFT simulations were performed to explore the V
S
-mediated
catalytic selectivity mechanism toward CO and C
2
H
4
on
AgInP
2
S
6
.CO
2
molecules are initially adsorbed on the catalyst
surface where H
2
O molecules dissociate into hydroxyl and
hydrogen ions at the same time. The free-energy profile for the
photocatalytic CO
2
-to-hydrocarbon process with the lowest-
energy pathway on the perfect AgInP
2
S
6
surface was calculated,
as shown in Fig. 4. The key step for CO production is the
hydrogenation of CO
2
to form *COOH, and the free-energy
change of the step is 0.48 eV. Subsequently, the reaction inter-
mediate (*COOH) further couples a proton/electron pair to
generate CO and H
2
O molecules. The adsorption energy of
−0.07 eV of the produced *CO on the defect-free AgInP
2
S
6
surface implies the physical adsorption on the catalyst (Supple-
mentary Fig. 17a). It means that *CO molecules can easily
liberate from BC and SAL to become free CO gas, allowing high
CO catalytic selectivity. Additional parts of *CO were con-
tinuously reduced by the incoming electrons and the successive
protonation process to transform into CH
4
20,23. While the charge
density of the valence band (VB) for pristine AgInP
2
S
6
is evenly
located on all the S and Ag atoms, contrastingly, the charge
density of the VB is mainly located on the Ag atoms near the V
S
for V
S
-AgInP
2
S
6
, (Supplementary Fig. 18). That is to say, the
presence of V
S
in V
S
-AgInP
2
S
6
causes the charge enrichment on
the Ag atoms near the V
S
, which would benefit for stabilizing the
reaction intermediates. For V
S
-SAL, V
S
can act as a trap for the
*CO molecule, that is, the *CO molecule can chemically adsorb
at exposed Ag sites with an adsorption energy of −0.25 eV (CO
can only physically adsorb on the exposed P and In sites with a
distance of 2.56 and 3.20 Å, See Supplementary Fig. 17b–d). The
higher CO onset desorption temperature on V
S
-SAL
10
than SAL
affirms the stronger absorption (Supplementary Fig. 19). The
absorbed *CO can be further protonated to successively form a
series of key reaction intermediates with unsaturated coordina-
tion, which was confirmed with in situ FTIR measurement
(Supplementary Fig. 20). The other *CO molecules produced on
the surface diffuses toward V
S
and couple with those reaction
intermediates to produce C
2
H
4
. The C
2
H
4
free energy diagrams
are summarized in Fig. 4c, while the corresponding C–C coupling
barriers are presented in Fig. 4b. The different C-C coupling
energy barriers were evaluated for three unsaturated reaction
intermediates (*COH, *CHOH, and *CH
2
) (Fig. 4b). The cou-
pling energy barrier with a value of 0.84 eV (*CO–CHOH) is
lower than that of other coupling pathways (*CO–COH, 1.01 eV
and *CO–CH
2
, 1.84 eV), hence the C
2
H
4
will be produced via
CO–CHOH coupling and hydrogenation. The whole free energy
diagram shows that the process of *CO to *COH is regarded as
the potential determining step (0.86 eV). It should be especially
emphasized that the detected small amount of C
2
H
4
on SAL
possibly originates from the potential existence of the tiny
number of V
S
in SAL, resulting from mechanically detaching
Fig. 4 Theoretical investigations. a Gibbs free energy diagrams for CO
2
reduction to CO over perfect AgInP
2
S
6
.bThree kinds of possible C–C coupling
pathways over AgInP
2
S
6
containing V
s
.cGibbs free energy diagrams for CO reduction to C
2
H
4
over AgInP
2
S
6
with V
s
. The insets show the corresponding
optimized geometries for the reaction intermediates during the CO
2
reduction process. Sulfur, phosphorus, indium, silver, carbon, oxygen, and hydrogen
atoms are yellow, purple, lilac, gray, black, red, and white, respectively.
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-25068-7 ARTICLE
NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturec ommunications 5
Content courtesy of Springer Nature, terms of use apply. Rights reserved

sulfur atoms from SAL during the probe sonication exfoliation
process. The reaction process for the reduction of CO
2
into C
2
H
4
,
CO, and CH
4
over V
S
-SAL under light illumination is thus
proposed in Supplementary Fig. 21. To confirm CO as an
important intermediate for the C
2
H
4
formation, CO as the
starting reactant substituting for CO
2
was also conducted for the
similar photocatalytic performance. The result reveals that a
considerable amount of C
2
H
4
was indeed detected (Supplemen-
tary Fig. 22). In addition, a small amount of ethane (C
2
H
6
) and
propylene (C
3
H
6
) were also produced. It indicates that CO as
starting reactants may be further favorable for C–C, even C
2
–C
coupling.
Surface photovoltage spectroscopy (SPV) was employed to
study the separation and transport behavior of photoinduced
charge carriers of the studied AgInP
2
S
6
. More negative SPV signal
change reflects a higher concentration of photogenerated
electrons before and after light illumination. All BC, SAL, and
Vs-SAL
10
show the SPV response under light illumination (Fig. 5
and Supplementary Fig. 23), corresponding to band-to-band
transition. The SAL and BC exhibit 20–30 mV and 5–10 mV
negative change before and after light illumination, respectively.
More negative SPV signal change of SAL than BC exactly
demonstrates that the atomically thin structure enables to
alleviate the bulk electron–hole recombination to achieve high-
concentration accumulation of photogenerated electrons on the
surface. The V
S
-SAL
10
display obviously dramatic change of
50–60 mV, indicating that introduction of V
S
can further favor
the carrier separation and allow much increment of electron
concentration on the surface. The excess surviving electrons are
not only the necessary prerequisite to photoconversion of CO
2
,
but also can promote CO
2
adsorption and activation on the
surface of the photocatalyst.
Photoluminescence (PL) decay profiles show that the SAL
(~1.32 ns) possesses a longer PL lifetime than BC (~0.40 ns)
(Supplementary Fig. 24), demonstrating that the atomically thin
structure can indeed shorten the transfer distance of the carriers
and decrease recombination chance of electron and hole in the
body. V
S
-SAL
10
exhibits the longest PL lifetime (~1.50 ns),
confirming that the surface V
S
can serve as surface separation
centers for charge carriers and further promote the charge
separation, therefore offering more opportunities for photocata-
lytic CO
2
reduction. Transient photocurrent shows that the
photocurrent intensity of SAL was enhanced with a steadily
repeating course due to promoted charge separation, compared
with BC (Supplementary Fig. 25a). The highest photocurrent
intensity of V
S
-SAL
10
implies that the V
S
also makes an effective
contribution to saving carriers. Electrochemical impedance
spectra reveal that V
S
-SAL
10
manifests the smallest semicircle
in Nyquist plots (Supplementary Fig. 25b), suggesting the lowest
charge-transfer resistance, which permits fast transport of
photoinduced charge.
Discussion
In summary, single atomically thin AgInP
2
S
6
layers were suc-
cessfully synthesized through a facile probe sonication exfoliation
Fig. 5 SPV characterization. Height images of aV
S
-SAL
10
,bSAL, and cBC. The SPV images dV
S
-SAL
10
,eSAL, and fBC in (a–c), respectively, are
differential images between potential images under light and in the dark. All scale bars represent 0.5 μm. The surface photovoltage change by subtracting
the potential under dark conditions from that under illumination (SPV, ΔCPD =CPD dark −CPD light) of hV
S
-SAL
10
,iSAL, and jBC.
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-25068-7
6NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturecom munications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

of BC. The atomically thin structure of SAL, relative to BC,
enables more charge carriers to mobile from the interior onto the
surface and surviving accumulate onto the active sites to improve
the photocatalytic activity. While SAL exhibits obvious conver-
sion efficiency with CO as the major product, the presence of V
S
in V
S
-SAL changes the CO
2
photoreduction pathway to allow the
dominant generation of C
2
H
4
. This work not only paves an
effective approach for selectively producing multi-carbon pro-
ducts from CO
2
photoreduction but also provides a new insight
for catalyst design through vacancy defect engineering.
Methods
Synthesis of BC, SAL, and V
S
-SAL. The AgInP
2
S
6
crystals have been synthesized
by physical vapor transport (PVT) in a two-zone furnace. Stoichiometric amounts
of high-purity elements (mole ratio Ag: In: P: S =1:1:2:6, around 1 g in total) were
sealed into a quartz ampoule with the pressure of 1 × 10−4Torr inside the ampoule.
The length of the quartz ampoule was about 15–18 cm with a 13 mm external
diameter. The ampoule was kept in a two-zone furnace (680 →600 °C) for
1 week11. After the furnace was cooled down to room temperature, the AgInP
2
S
6
crystalline powders could be found inside the ampoule (Supplementary Fig. 1a).
SAL was prepared by sonication-assisted liquid exfoliation processes from synthetic
AgInP
2
S
6
crystalline powders. For the detail, AgInP
2
S
6
bulk powder was ground
carefully, and then dispersed in ethanol solution. After continuous ultrasonification
for 12 h with a probe-type cell crusher (~1200 W), the solution was conducted for
static settlement, and the supernatant was taken to ultrasonic dissection further.
Through 4000 × gfor 10 min centrifugation, the samples peeled insufficiently were
removed, and the supernatant was collected through an additional 12,000 × gfor
20 min centrifugation to obtain the AgInP
2
S
6
monolayer. The derived SAL has
dispersed in the water again for subsequent liquid nitrogen refrigeration and being
dried in a vacuum freezing dryer at a pressure below 20 Pa for 2 days. The residual
ethanol can be considered to be totally removed.
SAL was immersed in H
2
O
2
solutions with the of concentrations 0.1 mol/L
inside which SAL was allowed to react with H
2
O
2
for 5, 10, and 15 s, referred to V
S
-
SAL
5
,V
S
-SAL
10
, and V
S
-SAL
15
, respectively, at 25 °C. All the obtained samples
were carefully washed and dried before use.
Characterizations. XRD (Rigaku Ultima III, Japan) was used to investigate the purity
information and crystallographic phase of the as-prepared powder samples. The XRD
pattern was recorded by using Cu-ka radiation (λ=0.154178 nm) at 40 kV and 40 mA
with a scan rate of 10° min−1. The morphology was characterized by the FESEM (FEI
NOVA NANOSEM 230). The TEM and HRTEM images were taken on a JEM 200CX
TEM apparatus. X-ray photoelectron spectroscopy (XPS; K-Alpha, Thermo Fisher
Scientific) was standardized according to the binding energy of the adventitious C 1s
peak at 284.8 eV, which was used to inspect the chemical states. A UV–vis spectro-
photometer (UV-2550, Shimadzu) was hired to record the UV-visible diffuse reflec-
tance spectra and switched to the absorption spectrum on the basis of the
Kubelka−Munk connection at room temperature. In situ FTIR spectra were measured
with synchronous illumination Fourier transform infrared spectroscopy on Bruker IFS
66V FT spectrometer. The PL decay profile was described by the single-particle
confocal fluorescence spectroscopy measurement (PicoHarp300). SPV was detected
through AFM (Asylum Research, MFP-3D-SA, USA) analysis with the photo-assisted
(a 405 nm laser excitation) Kelvin probe force microscopy. Photoelectrochemical
measurements were detected by a CHI660E electrochemical workstation using a
standard three-electrode system in 1 mM NaSO
4
solution. Soft X-ray absorption
spectra (XAS) were collected from the Soft X-ray Spectroscopy beamline at the
Australian Synchrotron (AS, Australia), part of ANSTO.
For the electrochemistry measurement, the AgInP
2
S
6
catalyst ink was prepared
by dispersing 10 mg of as-prepared catalysts in 1 mL of ethanol under sonication.
Then, 50 μL of the ink was evenly spread onto a piece of pretreated FTO within a
1cm
2area and dried at room temperature. The catalysts were thus attached to
FTO. The solid-state current–voltage (J–V) test curves exhibit Ohmic
characteristics (Supplementary Fig. 26), confirming the formation of ohmic back
contact between samples and FTO. The working area of the electrode is as large as
1cm
2. The scan rate was 5 mV s−1. The reference electrode was the saturated Ag/
AgCl electrode, and a Pt foil was employed as the counter electrode. The 0.5M
Na
2
SO
4
aqueous solution was used as the electrolyte.
Measurement of photocatalytic activity. For the photocatalytic reduction of CO
2
,
4–5 mg of sample was uniformly dispersed on the glass reactor with an area of 4.2 cm2.
A 300 W Xenon arc lamp was used as the light source of the photocatalytic reaction.
The volume of the reaction system was about 460 ml. Before the irradiation, the system
was vacuum-treated several times, and then the high purity of CO
2
gas was followed
into the reaction setup for reaching ambient pressure. Totally, 0.4 mL of deionized
water was injected into the reaction system as a reducer. The as-prepared photo-
catalysts were allowed to equilibrate in the CO
2
/H
2
O atmosphere for several hours to
ensure that the adsorption of gas molecules was complete. During the irradiation,
about 1mL of gas was continually taken from the reaction cell at given time intervals
for subsequent CO, CH
4
,andC
2
H
4
concentration analysis by using a gas chroma-
tograph (GC-2014C, Shimadzu Corp., Japan).
The external quantum efficiency (EQE). The quantum yield was calculated
according to the below equation
EQ¼NðelectronÞ=NðphotonÞ
¼½NðCOÞ´2þNðCH4Þ´8þNðC2H4Þ´12=NðphotonÞ´100%ð4Þ
where N(electron) signifies two electrons are required to produce one molecule CO
in unit time. The N(photon) is figured out according to the equation:
NðphotonÞ¼½light intensity ´illumination area ´time=½average single photon energy ´NA
ð5Þ
Light-emitting diodes (LEDs) provides the monochromatic incident light with
identical conditions. The light intensity of LEDs with 415 nm wavelength is
10.5 mW/cm2, the illumination area is controlled to 4.91 cm2,N
A
is the Avogadro
constant, and the average single photon energy is calculated according to the
equation:
EðphotonÞ¼hc=λð6Þ
in which his the Planck constant, c indicates the speed of light, and λis the
wavelength.
Computational details. The density functional theory (DFT) calculations were
made with the Vienna Ab Initio Simulation Package24,25 code. The exchange-
correlation interactions and the ion–electron interactions were solved by the
Perdew–Burke–Ernzerhof functionals26,27 and the projector-augmented wave
method28, respectively. The monolayer AgInP
2
S
6
was a model with a 2 × 2 super-
cell. A plane-wave cutoff of 450 eV was adopted and the maximal force on all-atom
was below 0.02 eV/Å. The distance between periodic units in the vertical direction
was larger than 16 Å. The DFT-D2 method of Grimme29 was used in all calcula-
tions to accurately describe long-range Van der Waals (vdW) interactions. The
climbing-image nudged elastic band (CI-NEB) method30 incorporated with spin-
polarized DFT was used to locate the minimum-energy path. The intermediate
images of each CI-NEB simulation were relaxed until the perpendicular forces were
smaller than 0.1 eV/Å.
The free energies of each reaction intermediates were determined according to G=
E+ZPE −TS. (7) The electronic energies (E) can be directly obtained from DFT
computations. The zero-point energy (ZPE) and entropy correction (TS) were
calculated from vibration analysis by standard methods. The computational hydrogen
electrode model31 was used to treat the free energy change of each reaction step
involving a proton–electron pair transfer. In this model, the free energy of a proton-
electron pair at 0V vs. RHE is equal to half of the free energy of a hydrogen molecule.
Data availability
The data that support the findings of this study are available from the corresponding
author upon reasonable request.
Received: 8 February 2021; Accepted: 14 July 2021;
References
1. Tu, W., Zhou, Y. & Zou, Z. Photocatalytic conversion of CO
2
into renewable
hydrocarbon fuels: state-of-the-art accomplishment, challenges, and prospects.
Adv. Mater. 26, 4607–4626 (2014).
2. Chen, G. et al. From solar energy to fuels: recent advances in light-driven C-1
chemistry. Angew. Chem. Int. Ed. 58, 17528–17551 (2019).
3. Liu, Q. et al. High-yield synthesis of ultralong and ultrathin Zn
2
GeO
4
nanoribbons toward improved photocatalytic reduction of CO
2
into renewable
hydrocarbon fuel. J. Am. Chem. Soc. 132, 14385–14387 (2010).
4. Han, Q. T. et al. Convincing synthesis of atomically thin, single-crystalline
InVO
4
sheets toward promoting highly selective and efficient solar conversion
of CO
2
into CO. J. Am. Chem. Soc. 141, 4209–4213 (2019).
5. Tu, W. G. et al. An in situ simultaneous reduction-hydrolysis technique for
fabrication of TiO
2
-graphene 2D sandwich-like hybrid nanosheets: graphene-
promoted selectivity of photocatalytic-driven hydrogenation and coupling
CO
2
into methane and ethane. Adv. Funct. Mater. 23, 1743–1749 (2013).
6. Jiang, W. et al. Pd-modified ZnO–Au enabling alkoxy intermediates formation
and dehydrogenation for photocatalytic conversion of methane to ethylene. J.
Am. Chem. Soc. 143, 269–278 (2021).
7. Bai, W. et al. Parasitic ferromagnetism in few-layered transition-metal
chalcogenophosphate. J. Am. Chem. Soc. 142, 10849–10855 (2020).
8. Vysochanskii, Y. et al. Ferroelectric and semiconducting properties of Sn
2
P
2
S
6
crystals with intrinsic vacancies. Ferroelectrics 418, 124–133 (2011).
NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-25068-7 ARTICLE
NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturec ommunications 7
Content courtesy of Springer Nature, terms of use apply. Rights reserved

9. Babuka, T., Glukhov, K., Vysochanskii, Y. & Makowska-Janusik, M.
Structural, electronic, vibration and elastic properties of the layered AgInP
2
S
6
semiconducting crystal-DFT approach. RSC Adv. 8, 6965–6977 (2018).
10. Dziaugys, A., Banys, J., MacUtkevic, J. & Vysochanskii, J. Conductivity
spectroscopy of new AgInP
2
S
6
crystals. Integr. Ferroelectr. 103,52–59 (2008).
11. Wang, X. et al. Second-harmonic generation in quaternary atomically thin
layered AgInP
2
S
6
crystals. Appl. Phys. Lett.109 (2016).
12. Huang, B., McGuire, M. A., May, A. F., Xiao, D., Jarillo-Herrero, P. & Xu, X.
D. Emergent Phenomena and Proximity Effects in Two-Dimensional Magnets
and Heterostructures. Nat. Mater. 19, 1276–1289 (2020).
13. Wang, L. M. et al. Surface Strategies for Catalytic CO
2
Reduction: from Two-
Dimensional Materials to Nanoclusters to Single Atoms. Chem. Soc. Rev. 48,
5310–5349 (2019).
14. Chen, S. et al. Oxygen vacancy associated single-electron transfer for
photofixation of CO
2
to long-chain chemicals. Nat. Commun. 10, 788 (2019).
15. Wang, X. et al. Single-atom vacancy defect to trigger high-efficiency hydrogen
evolution of MoS
2
.J. Am. Chem. Soc. 142, 4298–4308 (2020).
16. Eckmann, A., Felten, A., Verzhbitskiy, I., Davey, R. & Casiraghi, C. Raman
study on defective graphene: effect of the excitation energy, type, and amount
of defects. Phys. Rev. B - Condens. Matter Mater. Phys. 88, 035426 (2013).
17. Liang, H. et al. Hydrothermal continuous flow synthesis and exfoliation of
NiCo layered double hydroxide nanosheets for enhanced oxygen evolution
catalysis. Nano Lett. 15, 1421–1427 (2015).
18. Zhu, M. et al. Metal-free photocatalyst for H
2
evolution in visible to near-
infrared region: black phosphorus/graphitic carbon Nitride. J. Am. Chem. Soc.
139, 13234–13242 (2017).
19. Zhang, N. et al. Refining defect states in W
18
O
49
by Mo doping: a strategy for
tuning N
2
activation towards solar-driven nitrogen fixation. J. Am. Chem. Soc.
140, 9434–9443 (2018).
20. Li, X. et al. Selective visible-light-driven photocatalytic CO
2
Reduction to CH
4
mediated by atomically thin CuIn
5
S
8
layers. Nat. Energy 4, 690–699 (2019).
21. Das, S. et al. Core-shell structured catalysts for thermocatalytic, photocatalytic,
and electrocatalytic conversion of CO
2
.Chem. Soc. Rev. 49, 2937–3004 (2020).
22. Wendumu, T. B., Seifert, G., Lorenz, T., Joswig, J. O. & Enyashin, A. Optical
properties of triangular molybdenum disulfide nanoflakes. J. Phys. Chem. Lett.
5, 3636–3640 (2014).
23. Gao, G., Jiao, Y., Waclawik, E. R. & Du, A. Single atom (Pd/Pt) supported on
graphitic carbon nitride as an efficient photocatalyst for visible-light reduction
of carbon dioxide. J. Am. Chem. Soc. 138, 6292–6297 (2016).
24. Kresse,G.&Furthmüller,J.Efficient iterative schemes for ab initio total-energy
calculations using a plane-wave basis set. Phys. Rev. B 54, 11169–11186 (1996).
25. Kresse, G. & Joubert, D. From ultrasoft pseudopotentials to the projector
augmented-wave method. Phys. Rev. B 59, 1758–1775 (1999).
26. Perdew, J. P. et al. Atoms, molecules, solids, and surfaces: applications of the
generalized gradient approximation for exchange and correlation. Phys. Rev. B
46, 6671–6687 (1992).
27. Perdew, J. P. & Wang, Y. Accurate and simple analytic representation of the
electron-gas correlation energy. Phys. Rev. B 45, 13244–13249 (1992).
28. Blöchl, P. E. Projector augmented-wave method. Phys. Rev. B 50, 17953–17979
(1994).
29. Grimme, S. Semiempirical GGA-type density functional constructed with a
long-range dispersion correction. J. Comput. Chem. 27, 1787–1799 (2006).
30. Henkelman, G., Uberuaga, B. P. & Jónsson, H. A. Climbing image nudged
elastic band method for finding saddle points and minimum energy paths. J.
Chem. Phys. 113, 9901–9904 (2000).
31. Nørskov, J. K. et al. Origin of the overpotential for oxygen reduction at a fuel-
cell cathode. J. Phys. Chem. B 108, 17886–17892 (2004).
Acknowledgements
The authors wish to acknowledge the support of the National Key R&D Program of
China (2018YFE0208500), 973 programs (2017YFA0204800), NSF of China
(21972065, 21773114, and 21773027), the Fundamental Research Funds for the
Central University (020414380167), NSF of Jiangsu Province (No. BK20171246), the
Hefei National Laboratory for Physical Sciences at the Microscale (KF2020006), the
Program for Guang-dong Introducing Innovative and Entrepreneurial Team
(2019ZL08L101) and The University Development Fund (UDF01001159). Prof. Ran
Long and Dr. Wenqing Zhang of USTC (China) were greatly acknowledged for in situ
FT-IR measurements.
Author contributions
Y.Z., Y.X. and Z.Z instructed this work. L.S. and J.W. carried out the DFT calculation.
W.G., H.H., X. Li, Z.C., Y.Y. and Q.S. performed the experiments and co-wrote this
paper. X.W. contributed to the PL spectrum measurement.
Competing interests
The authors declare no competing interests.
Additional information
Supplementary information The online version contains supplementary material
available at https://doi.org/10.1038/s41467-021-25068-7.
Correspondence and requests for materials should be addressed to J.W., Y.X. or Y.Z.
Peer review information Nature Communications thanks Andrew Bocarsly, Xuxu Wang
and the other, anonymous, reviewer for their contribution to the peer review of this
work. Peer reviewer reports are available.
Reprints and permission information is available at http://www.nature.com/reprints
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons license, and indicate if changes were made. The images or other third party
material in this article are included in the article’s Creative Commons license, unless
indicated otherwise in a credit line to the material. If material is not included in the
article’s Creative Commons license and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this license, visit http://creativecommons.org/
licenses/by/4.0/.
© The Author(s) 2021
ARTICLE NATURE COMMUNICATIONS | https://doi.org/10.1038/s41467-021-25068-7
8NATURE COMMUNICATIONS | (2021) 12:4747 | https://doi.org/10.1038/s41467-021-25068-7 | www.nature.com/naturecom munications
Content courtesy of Springer Nature, terms of use apply. Rights reserved

1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com