Surface Silylation of Hybrid Benzidinium Lead Perovskite and its Influence on the Photocatalytic Activity

Article abstract of DOI:10.1002/cctc.201901681

Surface coating of benzidinium lead iodide perovskite has been successfully accomplished by silanization with three different silylating agents, obtaining samples with average thickness from 2 to 6 nm as revealed by transmission electron microscopy. The obtained (organo)silica-coated hybrid perovskites exhibit enhanced hydrophobic character and, therefore, increased stability against moisture. However, its photocatalytic activity towards the cis-to-trans isomerization of stilbene diminishes as a function of the coating thickness, although a notable activity for this photocatalytic reaction is still observed.

Full text of DOI:10.1002/cctc.201901681

DOI: 10.1002/cctc.201901681
Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Surface Silylation of Hybrid Benzidinium Lead Perovskite
and its Influence on the Photocatalytic Activity
[a]
[a]
[a]
Yong Peng, Josep Albero, and Hermenegildo García*
Surface coating of benzidinium lead iodide perovskite has been
successfully accomplished by silanization with three different
silylating agents, obtaining samples with average thickness
from 2 to 6 nm as revealed by transmission electron micro-
scopy. The obtained (organo)silica-coated hybrid perovskites
exhibit enhanced hydrophobic character and, therefore, in-
creased stability against moisture. However, its photocatalytic
activity towards the cis-to-trans isomerization of stilbene
diminishes as a function of the coating thickness, although a
notable activity for this photocatalytic reaction is still observed.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Introduction
PbI chains, that are held by Coulombic and hydrogen bridge
6
interactions with the diammonium organic cations, see
Scheme 1. The empty space between the chains is occupied by
solvent DMF molecules in the single crystal material. Although
PbI Bz exhibited some range of stability in organic apolar
Hybrid organic lead perovskites have drastically modified the
landscape of photovoltaic cells by reaching very high efficien-
[1]
cies over 23%, overcoming the values that were achieved
3
0.5
[
2]
with dye sensitized solar cells over decades. The operation
mechanism of photovoltaic cells have many common funda-
mental steps with photocatalytic transformations, including
light absorption, charge separation and carrier migration to the
solvents, the material was still instable in the presence of
moisture and other polar solvents.
Continuing with this research line aimed at implementing
hybrid perovskites in photocatalysis, in the present manuscript,
we report the modification of PbI Bz perovskite surface by
[3]
surface. However, in contrast to the impact that hybrid
organic perovskites have had in the realm of solar cells, the lack
of stability has precluded the use of this hybrid perovskites in
3
0.5
coating with silylating reagents. There are sufficient data in the
literature showing that surface silylation can protect materials
by modifying the hydrophobic-hydrophilic properties of the
[4]
photocatalysis.
[8]
In an attempt to develop and explore optoelectronic
properties of hybrid lead organic perovskites in photocatalysis,
we have reported recently the preparation of 1D benzidinium
surface. For instance, the surface silica coating of inorganic
lead halide perovskites (CsPbX ; X=Cl, Br, I) has been reported
3
not only to improve the perovskite stability, but also to enhance
the quantum efficiency and to narrow the photoluminescence
+
+ [5]
lead iodide perovskite (PbI Bz , Bz: NH À C H À C H À NH ),
3
0.5
3
6
4
6
4
3
[9]
the rationale for the preparation of this hybrid perovskite was
to show the possibility to modulate optoelectronic properties of
the material by proper selection of the organic ammonium
linewidth in light emitting devices. It will be shown below
that, while silylation of the most common methylammonium
lead iodide perovskite was unsuccessful, coating of PbI Bz
3
0.5
[6]
cation. In particular, biphenyls have been widely used in
photoinduced electron transfer reactions due to the stability of
the corresponding radical ions that can act as electron transfer
particles can be successfully achieved by using different
conventional trialkoxysilane agents. It will be presented that the
photocatalytic activity of the silica-coated samples is still
maintained after coating the hybrid perovskite surface with the
thin silica layer. The photocatalytic activity of the silica-coated
material is maintained for 3 consecutive runs.
[7]
mediators.
It was expected, therefore, that benzidinium dications
should favor charge separation in the hybrid perovskite. More-
over, due to the size of Bz dication, the 3D structure of the
2
+
+
most widely used methylammonium (MA ) lead iodide (PbI MA)
3
changed to 1D. Single crystal characterization of novel
benzidinium perovskite shows that the material is constituted
by PbI octahedra sharing the faces, forming an array of parallel
6
[
a] Y. Peng, Dr. J. Albero, Prof. H. García
Instituto Universitario de Tecnología Química CSIC-UPV
Universitat Politècnica de València
Avda. de los Naranjos s/n
Valencia 46022 (Spain)
E-mail: hgarcia@qim.upv.es
Supporting information for this article is available on the WWW under
https://doi.org/10.1002/cctc.201901681
3
Scheme 1. (a) Structure of PbI Bz0.5 perovskite with red and green spheres
indicating I and Pb atoms, respectively, and (b) chemical structure of the
three silylating reagents employed in this study (ODTMS, TEOS and FTS).
This manuscript is part of the Special Issue “Photocatalysis: From Solar
Power to Sustainable Chemical Production”, which is part of the wider
project “Building A New Energy Economy with Catalysis”.
ChemCatChem 2019, 11, 1–8
1
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Results and Discussion
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Surface modification by silylating reagents is based on the
[
10]
presence of hydroxyl groups on the surface. Although ideally
organic hybrid perovskites should not contain hydroxyl groups,
contacting these materials with the ambient moisture results in
a strong water adsorption that eventually has been proposed to
À
replace I ions by hydroxyl groups with concomitant evolution
of HI. In fact, the presence of adsorbed water has been
proposed as the main cause for the observed deterioration of
photovoltaic efficiency in perovskite solar cells upon exposure
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
[11]
to moist ambient.
The leading hypothesis of the present study is that also for
PbI Bz some hydroxyl groups could be present on the
3
0.5
[12]
Figure 1. X-ray diffraction patterns of PbI Bz (A), PbI Bz after dispersion in
hexane for 7 h (B) and SHP1 (C). Inset: enlargement X-ray diffraction pattern
corresponding to the red dash frame.
3
0.5
3
0.5
perovskite surface due to environmental humidity, and these
hydroxyl groups would react with the silylating reagents
forming a thin silica coating on the hybrid perovskite crystal.
In order to confirm this hypothesis, high resolution XP
spectrum of O 1s peak of PbI Bz sample and the best
3
0.5
[5]
deconvolution into individual components was recorded and it
is presented as Figure S1 in supplementary information. The
presence of À OH groups in the Pb surface cannot be clearly
distinguished in the Pb 4f 7/2 spectrum because of the overlap
observed in the related work, and it was attributed to the
change in the 1D structure of hybrid perovskite crystals caused
by the removal of DMF molecules from the intracrystalline
space due to the dissolution of DMF in organic solvents, such as
hexane or toluene. The same hypothesis can be assumed here,
since the material after silylation is still a crystalline material
and the XRD patterns exactly match with the patterns derived
from sample that presumably has lost the DMF molecules. It
should be commented that the quality of the crystals after
silylation is not enough to perform structural elucidation by
single crystal XRD.
2
+
À
of the Pb bonded to I signal at 138.1 eV. However, the Pb 4f
2
+
5
/2 shows two components related to Pb and Pb(OH) at
2
1
42.88 eV and 144.03 eV, respectively. Additionally, the O 1s
spectrum shows the presence of four different components
centered at 533.39 eV, 531.65 eV, 529.9 eV and 528.20 eV. The
components at higher binding energies (533.39 and 531.65 eV)
have been previously attributed to weakly bound oxygen
À
species, such as OH and H O, as consequence of the
Preservation of the optoelectronic properties of the three
silica-coated hybrid perovskites were assessed by diffuse
reflectance UV-vis absorption spectroscopy presented in Fig-
ure 2. Only minor changes in the absorption onset of the hybrid
perovskite were recorded after silylation. Specifically, the
absorption onset showed a small red-shift after silylation from
431 nm to 434 nm. Accordingly, the Tauc plots showed a minor
2
[13]
interaction with water in outer regions. The components at
lower binding energies has been assigned to strongly bounded
lattice oxygen (À COOÀ and PbÀ OH).
[
14]
Herein, perovskite silylation was carried out using trime-
thoxy(octadecyl)silane (ODTMS, material named as SHP1), as
well as tetra ethyl orthosilicate (TEOS, materials named as SHP2)
and triethoxy perfluorodecyl silane (FTS, material SHP3). The
structure of the silylating agents are illustrated in Scheme 1 (b).
The reaction was carried out under inert atmosphere by
suspending the hybrid perovskite in a toluene solution at 50°C
containing the desired silylating reagent. These conditions are
compatible with the preservation of the structure of the hybrid
perovskite as determined by powder XRD.
According to the thermogravimetric analysis (see Figure S2
in supporting information) and in situ XRD patterns, monitoring
the crystalline structure at different temperatures (Figure S3 in
supporting information), the initial hybrid perovskite crystallin-
ity is maintained in the coating process. On the contrary, higher
temperatures or the use of other solvents can cause change in
the crystal phase or even decomposition of the hybrid
perovskite. It was observed, however, that the XRD pattern of
pristine PbI Bz changes after the silylation step. Figure 1
3
0.5
shows a comparison between the XRD patterns of the hybrid
benzidinium perovskite before and after silylation, together
with the XRD pattern of benzidnium perovskite exposed to
hexane for 7 h. This change in the XRD pattern was already
Figure 2. Diffuse reflectance UV-Vis spectra absorption of PbI
3
Bz0.5 (black)
and SHP1 (red). Inset: Tauc plot of the corresponding PbI Bz (black) and
3
0.5
g
SHP1 (red) with intercept used to calculate E (dashed blue lines).
ChemCatChem 2019, 11, 1–8
www.chemcatchem.org
2
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
À 1
bandgap change from 2.78 to 2.76 eV. The small shift means
that after silylation, the hybrid perovskite holds the optical
properties of the parent material and that the negligible
changes in the absorption profile may arise from the organic-
inorganic interaction at the interlayer.
vibration band from 3250 till 2250 cm due to the acid À XÀ H
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
(X: N and O) bonds present in the organic cation. Upon
silylation, the presence of new vibration corresponding to
À 1
SiÀ OH silanol groups at 3500 cm as well as CÀ H stretching
À 1
À 1
vibrations at 2950 cm and 2916 cm were clearly visible, as
À 1
FTIR spectroscopy allows following the changes in the
functional groups of the hybrid perovskite after silylation.
Figure 3 shows the FTIR spectra of the samples prepared in this
study. Initially, the FTIR spectrum of PbI Bz shows a very broad
well as the characteristic SiÀ O at about 1050 cm . The position
and shape of the SiÀ OH band indicates the occurrence of
hydrogen bridges, since isolated silanol groups exhibit higher
frequency. It is likely that ambient moisture is responsible for
the hydrolysis of alkoxy silyl groups. The weak intense SiÀ OÀ Si
vibration band also indicates that condensation of silanol
groups is not taking place in a large extent.
3
0.5
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
High resolution XP spectra of C1s, O1s, Si2p and Pb 4f peaks
and the best deconvolution into individual components of
SHP1 are shown in Figure 4. As it can be observed there, the
XPS C 1s peak (Figure 4 a) shows the presence of three main
components attributed to C=C (78%) (284.6 eV, blue), CÀ O
(19%) (285.6 eV, orange) and CÀ Si (3%) (283.1 eV, green). The O
1
s spectrum (Figure 4 b) presents a main component attributed
to O atoms single bonded to aliphatic C appearing at 532.2 eV
85.2%), as well as two smaller components related to O double
bonded to C (9.9%) and SiÀ O (4.9%) bonds at 527.9 and
29.5 eV, respectively. The Si 2p spectrum (Figure 4 c) shows a
(
5
main component related to the SiÀ C bond from silane groups
located at 102.2 eV (86.9%), although a second component
centered at 103.8 eV (13.1%) assigned to SiÀ O can be also seen.
The presence of À OH groups in the Pb surface cannot be clearly
3
Figure 3. FTIR spectra of parent PbI Bz0.5 (black) and SHP1 (red).
Figure 4. High resolution XPS peaks for C 1s (a), O 1s (b), Si 2p (c) and Pb 4f (d) recorded for SHP1. The best deconvolution into individual components for
each peak is also shown.
ChemCatChem 2019, 11, 1–8
www.chemcatchem.org
3
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Figure 5. HRTEM image of SHP1 with 1:0.5 (a and b) and 1:5 (c) ODTMS/perovskite mass ratio.
distinguished in the Pb 4f 7/2 spectrum because of the overlap
these hybrid methylammonium perovskites. On the contrary,
PbI Bz perovskite was successfully coated not only with
ODTMS, but also with TEOS and FTS (see Figure S9 in
supporting information), pointing out to a general method-
ology for coating hybrid perovskites, particularly the one
containing benzidinium organic ligands considered here.
2
+
À
of the Pb bonded to I signal at 138.1 eV.
3
0.5
2
9
Solid-state Si NMR spectroscopy was recorded for the
SHP1 sample. The spectrum is presented as Figure S4 in
supporting Information. Peaks at À 49, À 59 and À 67 ppm were
3
2
1
recorded and attributed to T , T and T Si atoms in the silane
groups, confirming the presence of these moieties in the hybrid
perovskite. Thus, all the available spectroscopic data are in
agreement with the presence of a silica coating on the material
surface.
It is worth commenting that the obtained silica-coated SHP-
n hybrid perovskites exhibit enhanced hydrophobic character
as can be observed in Figure S10 in supplementary information.
The photographs of the materials SHP1, SHP2, SHP3 and
PbI Bz placed on pure water show that while SHP1 and SHP3
In fact, the presence of the silica coating on the hybrid
perovskite crystals could be clearly observed in HRTEM that also
allows us determining the thickness of the silica layer. Samples
with different average thickness were prepared in the case of
silica-coated PbI Bz perovskite by performing the silylation
3
0.5
materials were floating in water for long periods (days), SHP2
presented only partial stability in water, presenting some
material precipitation. The parent PbI Bz precipitated immedi-
ately in contact with the water. Additionally, the Z-potential
values of the parent benzidine lead iodide perovskite, SHP1,
SHP2 and SHP3 were measured, obtaining values of À 11.3, 44.1,
40.2 and 47.8, respectively. These values indicate that while the
parent perovskite exhibits a slightly negative surface, after
coating the surface Coulombic charge is reversed and becomes
positive, independently of the silica precursor employed.
3
0.5
3
0.5
step using different ODTMS to perovskite mass ratios in the
silylation step. Figure 5 shows representative HRTEM images of
SHP1 with different thickness. Elemental mapping by EDS (see
Figure S5 in the supporting information) confirmed the pres-
ence of Si completely coating the perovskite crystals surface.
EDX analysis on the surface and at the center of a sample
revealed that the Si proportion at the edges is much higher
than the amount at the center for the silica capped perovskite
Regarding the photocatalytic activity of PbI Bz , it was
shown that this hybrid perovskite can promote the photo-
3
0.5
[
5]
(
see Figure S6 in supporting information). This location of the
catalytic cis-to-trans isomerization of stilbene (Scheme 2). It
was, therefore, of interest to determine if after capping with the
silica layer the hybrid SHP-n perovskites are still active to
promote this photocatalytic reaction.
Figure 6 (a) shows the temporal evolution of the photo-
catalytic cis-to-trans isomerization as function of the different
silylating agents (TEOS, ODTMS and FTS). As it can be seen
there, cis-to-trans isomerization occurs in the presence of the
silica-capped perovskites, independently of the silylating
silica layer agrees with the lighter contrast of the thin layers on
the crystals observed in TEM compared to the darker contrast
of the interior of the particles due to the presence of heavy Pb
element. It should be noted that HRTEM images show that the
thickness of the layers is not constant throughout the particle
and changes depending on the area monitored were observed.
Figure S7 in the supporting information shows the silica thick-
ness average determined by measuring a statistical relevant
number of samples as a function of the concentration of
ODTMS used in the preparation of these samples.
It should be commented that we tried to perform also
surface silylation for the most common methylammonium lead
perovskite with iodine or bromine ions using ODTMS as
silylating agent, but the silylation failed in all cases as revealed
by HRTEM (see Figure S8 in supporting information). This failure
Scheme 2. Proposed mechanism for the photocatalytic cis-to-trans stilbene
isomerization through the intermediacy of the corresponding equilibrated
radical cations.
to silylate the surface of PbX MA solids could be attributed to
3
lack of a sufficient density of hydroxyl groups on the surface of
ChemCatChem 2019, 11, 1–8
www.chemcatchem.org
4
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Figure 6. a) Photocatalytic cis-to-trans isomerization of stilbene using as photocatalysts the parent PbI
3
Bz0.5 perovskite (blue triangles) or SHP1 (red dots), SHP2
(
black squares), SHP3 (green triangles) and in the absence of photocatalysts (yellow triangles). b) Influence of the thickness on the photocatalytic cis-to-trans
stilbene isomerization using parent PbI Bz0.5 (red dots) or 2 nm (black squares), 4 nm (green triangles) and 5 nm (blue triangles) thick SHP1 samples Reaction
3
conditions: 0.25 mmol of cis-stilbene dissolved in 1.5 mL of toluene using 2 wt% (4.1 mg) photocatalyst. Prior to irradiation, the reactor was purged with
argon. The visible light irradiation (λ>450 nm) was carried out using a 300 W Xe lamp output adapted with 450 nm filter.
reagent used to form the coating. It was, however, observed
that the photocatalytic activity of silylated materials decreased
respect to the uncoated material. An example of GC plot
recorded during the photocatalytic cis-to-trans stilbene isomer-
ization using SPH1 is presented in Supplementary Information
as Figure S11 to illustrate the difference in retention time for
the two stereoisomers.
A control experiment in the absence of any photocatalyst
has been carried out, and the result is presented in Figure 6a.
As can be observed, the photocatalytic reaction did not take
place in the absence of the photocatalysts or in its presence in
the dark.
These results can be easily interpreted assuming that the
surface of parent perovskite is free to interact with cis-stilbene,
which undergoes photoinduced electron transfer isomerization.
The presence of the porous silica layer introduces some
diffusing limitations to the access of cis-stilbene to the active
sites of the material and, for this reason, the photocatalytic
stability. The results of the successive reuses as well as the
structure and morphological characterization after reaction are
presented as Figures S12, S13 and S14 in Supplementary
Information. As can be seen in Figure S12, the photocatalytic
activity of SPH1 did not decrease significantly during three
consecutive runs, exhibiting a notable stability under the
reaction conditions. The XRD of the reused SHP1 photocatalysts
(Figure S13) shows the same peaks than the fresh sample,
indicating that the crystalline structure has been preserved
during the three uses. The morphology of perovskite particles
decreased somewhat in size after reuse (Figure S14 and Fig-
ure S15). However, compared with the parent perovskite, the
silica coated sample exhibited excellent stability against de-
aggregation during the photocatalytic reaction (Figure S14).
Therefore, the silica coating has a positive effect protecting the
perovskite from degradation under these reaction conditions.
reaction becomes slower. It is worth noticing that the photo- Conclusions
catalytic activity of the different coated perovskites did not
present drastic differences among them under the present
reaction conditions. This could indicate either that the porosity
of the silica layers is similar or that the photocatalytic isomer-
ization occurs on the external surface and electrons and holes
are tunneling through this interface.
In order to further investigate the influence of the silica
coating thickness in the photocatalytic reaction the temporal
evolution of the photocatalytic cis-to-trans isomerization of
stilbene was studied for SHP1 exhibiting different average
thickness. The results are presented in Figure 6 (b). It seems
however that the thickness of the layer in the range from 2 to
The present study has shown the possibility to coat the external
surface of the hybrid organic lead perovskite PbI Bz by
3
0.5
silylation with different silylating reagents obtaining layers
below 6 nm. The silylation process depends on the density of
hydroxyl groups on the surface of the hybrid perovskite.
Although the reaction rate was somewhat lower using the silica
coating, the photocatalytic activity of the silica-coated perov-
skites was still remarkable. Considering that silylation is a well-
known approach to modify the surface properties of materials,
the present study opens the door for the control of the
photocatalytic properties of perovskites, particularly those that
influence their stability in different solvents.
5
nm screened in the present study do not make additional
influence on the activity.
Finally, SHP1 was reused three times as photocatalyst for
the cis-to-trans isomerization of stilbene in order to check its
ChemCatChem 2019, 11, 1–8
www.chemcatchem.org
5
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Experimental Section
atmosphere. The final product was isolated by centrifugation and
dried in air for further use.
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
Materials
MilliQ water was obtained using an IQ 7000 purifying system.
Absolute ethanol (analytical grade), anhydrous tetrahydrofuran
Characterization
Powder X-ray diffraction (XRD) patterns were collected with a
Shimadzu XRD-7000 diffractometer by using Cu Kα radiation (λ=
(
99.99%, THF), diethyl ether (puriss. p.a. ACS reagent dried 0.05%
GC 0.0075% water), hexane (HPLC grade), toluene (chromasolv for
HPLC 99.9%) and anhydrous N,N-dimethylformamide 99.8% (DMF)
were purchased from Scharlau. Hydrazobenzene (<10% azoben-
zene), tetradecane (�99%), aqueous hydriodic acid (HI, 55 wt%),
aqueous hydrochloric acid (HCl, 37 wt%), lead diiodide (99.999%
1
.5418 Å), operating at 40 kV and 40 mA at a scanning speed of 10°
per min in the 2–90° 2θ range. Diffuse reflectance UV/Vis
absorption spectra in the range of 200–800 nm were collected in a
Varian Cary 5000 spectrophotometer. Fourier-transformed infrared
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
À 1
(
FTIR) spectra were recorded in the 4000 to 400 cm range using a
trace metals basis) (PbI ), tetraethyl orthosilicate (TEOS, reagent
2
Thermo Nicolet 6700 FTIR spectrophotometer with ATR accessory
instrument (Thermo scientific, USA). Thermogravimetric measure-
grade 98%), cis-stilbene (96%) and trans-stilbene (96%) were
purchased from Sigma Aldrich. Octadecyl trimethoxy silane
ments (TGA) were performed under N stream with a Metler Toledo
2
(ODTMS, 97%) and perfluoro decyl triethoxy silane ((FTS, 98%)
À 1
TGA/SDTA 851E analyzer at a heating rate of 10 Kmin . HRTEM
were purchased from ABCR. NaOH micro pearls (98%) were
purchased from ACROS. All solvents and reagents were used
without further purification.
images were obtained with a JEM 2100F JEOL electronic micro-
scope operating at 200 kV combined with an EDX detector (Oxford
Instruments) coupled to the DF-STEM measurement system for the
element distribution analysis. SEM images were acquired with a
JEOL JSM 6300 apparatus equipped with an X-MAX detector of
OXFORD INSTRUMENTS. Zeta potential was obtained by the Nano
Zetasizer Malvern (ZEN3600), sample for analysis was dispersed in
toluene with a concentration of 20 ppm. Gas chromatography (GC)
analysis were performed with a Varian CP-3800 apparatus equipped
Synthesis
Benzidine Preparation
Benzidine preparation was performed according to our previous
1
[5]
with a Carbowax column (15 m×0.32 mm×0.25 μm). H NMR
report. Commercial hydrazobenzene (1 g) was dissolved in
ethanol/water (60 mL/40 mL) solution with stirring, followed by
adding 10 mL concentrated HCl (~37 wt%). After 10 min reaction,
the solution was concentrated under reduced pressure evaporation
by about one third of the initial volume and benzidine precipitate
was obtained by neutralizing the solution with 1 M NaOH. The final
spectra were recorded with a Bruker AV300 (300 MHz) spectrom-
1
eter, Chemical shifts of H signals are reported in ppm using the
solvent peak as internal standard (Ethanol-D6: 5.29 ppm). Data are
reported as follows: chemical shift, multiplicity (d=doublet, t=
triplet), integral, coupling constants (Hz) and assignment. Hydrazo-
1
3
benzene H NMR: (300 MHz, Ethanol-D6) δ=7.08–7.13 (t, 4H; J =
obtained product was isolated by filtration and washed with H O/
EtOH. The purity of benzidine was confirmed by H NMR and IR
spectroscopy (Figure S16 and 17, respectively).
2
3
1
7.726 Hz, ArÀ H), 6.84–6.87(d, 4H; J =8.322 Hz, ArÀ H), 6.66–6.71(t,
3
2H; J =7.267 Hz, ArÀ H).Benzidine 1H NMR: (300 MHz, Ethanol-D6)
3
3
δ=6.75–6.79 (d, 4H; J =8.41 Hz, ArÀ H), 7.26–7.30(d, 4H; J =
8
.42 Hz, ArÀ H).
Benzidinium Diiodide Preparation
The benzidinium diiodide was obtained according to our previous
Photocatalytic Reaction
[5]
report. Specifically, 0.92 g of benzidine (5 mmol) was dissolved in
The photoinduced isomerization reactions were performed under
visible light irradiation (λ>450 nm) using a 300 W Xe lamp
equipped with 450 nm filter. Specifically, cis-stilbene (0.25 mmol)
was dissolved in 1.5 mL of toluene and 0.15 mmol tetradecane was
added to the above solution as the internal standard to quantify
the amount of cis-stilbene. 2 wt% (4.1 mg) perovskite was added as
photocatalyst. Prior to irradiation, the reactor was purged with
argon. The course of cis-stilbene conversion was followed periodi-
cally by injecting aliquots of the reaction mixture in the GC. The
only product observed was trans-stilbene.
2
0 mL THF under stirring at 0°C (ice bath). Then 1.6 mL concen-
trated HI aqueous solution (55 wt%) was added dropwise to the
solution. After 2 h reaction, the benzidinium diiodide salt was
precipitated by adding 20 mL of cold diethyl ether into the solution.
The final product was filtered and washed with diethyl ether and
characterized by FTIR spectroscopy (Figure S17).
Benzidinium Lead Iodide Perovskite PbI Bz Preparation
3
0.5
Preparation of benzidinium lead iodide perovskite was carried out
following our previous report. PbI2 (184 mg, 0.4 mmol) and
benzidinium diiodide (88 mg, 0.2 mmol) were dissolved in 4 mL
[
5]
DMF under stirring. Then, 10 mL toluene were added to the Acknowledgements
precursor solution to precipitate the perovskite. The obtained
hybrid benzidinium lead perovskite fine powder was isolated by
filtration and washed with diethyl ether, then, dried under vacuum.
Financial support from the Spanish Ministry of Economy and
Competitiveness (Severo Ochoa, and RTI2018-890237-CO2-R1) and
the Generalitat Valenciana (Prometeo 2017/083) is gratefully
acknowledged. Yong Peng also thanks the Universitat Politecnica
de Valencia for a predoctoral scholarship.
Perovskite Surface Silylation
Surface silylation was performed in toluene under argon atmos-
phere. Perovskite (200 mg) was dispersed in toluene (20 ml) by
sonication in a 50 ml flask, followed by adding a desired amount of
silylating reagent (specifically, 1:1 mass ratio for TEOS and FTS, Conflict of Interest
1:0.5, 1:1.5 and 1:5 mass ratio for ODTMS were used). Then the
dispersion was kept stirring overnight at 50 C under inert
°
The authors declare no conflict of interest.
ChemCatChem 2019, 11, 1–8
www.chemcatchem.org
6
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
Keywords: hybrid perovskite
coating · silylation · photocatalysis
·
hydrophobicity
·
surface
[8] a) P. A. Zapata, Y. Huang, M. A. Gonzalez-Borja, D. E. Resasco, J. Catal.
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
2
013, 308, 82–97; b) J. Bu, H.-K. Rhee, Catal. Lett. 2000, 65, 141–145; c) Y.
Kuwahara, K. Maki, Y. Matsumura, T. Kamegawa, K. Mori, H. Yamashita, J.
Phys. Chem. C 2009, 113, 1552–1559.
9] F. Zhang, Z.-F. Shi, Z.-Z. Ma, Y. Li, S. Li, D. Wu, T.-T. Xu, X.-J. Li, C.-X. Shan,
G.-T. Du, Nanoscale 2018, 10, 20131–20139.
[
[
1] Q. Jiang, Y. Zhao, X. Zhang, X. Yang, Y. Chen, Z. Chu, Q. Ye, X. Li, Z. Yin,
J. You, Nat. Photonics 2019.
[10] W. Yoshida, R. P. Castro, J.-D. Jou, Y. Cohen, Langmuir 2001, 17, 5882–
5888.
[2] K. Kakiage, Y. Aoyama, T. Yano, K. Oya, J.-i. Fujisawa, M. Hanaya, Chem.
Commun. 2015, 51, 15894–15897.
[
[
11] a) T. A. Berhe, W.-N. Su, C.-H. Chen, C.-J. Pan, J.-H. Cheng, H.-M. Chen,
M.-C. Tsai, L.-Y. Chen, A. A. Dubale, B.-J. Hwang, Energy Environ. Sci.
2016, 9, 323–356; b) G. Niu, X. Guo, L. Wang, J. Mater. Chem. A 2015, 3,
[
3] a) R. Saravanan, F. Gracia, A. Stephen, Nanocomposites for Visible Light-
induced Photocatalysis (Eds.: M. M. Khan, D. Pradhan, Y. Sohn), Springer
International Publishing, Cham, 2017, pp. 19–40; b) J. Bisquert, D.
Cahen, G. Hodes, S. Rühle, A. Zaban, J. Phys. Chem. B 2004, 108, 8106–
8970–8980.
12] B. Cheng, T.-Y. Li, P.-C. Wei, J. Yin, K.-T. Ho, J. R. D. Retamal, O. F.
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
Mohammed, J.-H. He, Nat. Commun. 2018, 9, 5196.
8
118; c) P. V. Kamat, Acc. Chem. Res. 2017, 50, 527–531.
[13] a) A. B. Velichenko, R. Amadelli, E. A. Baranova, D. V. Girenko, F. I.
Danilov, J. Electroanal. Chem. 2002, 527, 56–64; b) O. Shmychkova, T.
Luk’yanenko, R. Amadelli, A. Velichenko, J. Electroanal. Chem. 2014, 717–
[
4] a) C. C. Boyd, R. Cheacharoen, T. Leijtens, M. D. McGehee, Chem. Rev.
2019, 119, 3418–3451; b) N. Aristidou, C. Eames, I. Sanchez-Molina, X.
Bu, J. Kosco, M. S. Islam, S. A. Haque, Nat. Commun. 2017, 8, 15218.
5] Y. Peng, J. Albero, E. Álvarez, H. García, Sustainable Energy Fuels 2019.
6] a) D. Cortecchia, S. Neutzner, A. R. Srimath Kandada, E. Mosconi, D.
Meggiolaro, F. De Angelis, C. Soci, A. Petrozza, J. Am. Chem. Soc. 2017,
718, 196–201.
[
[
[14] S. Pradhan, A. Stavrinadis, S. Gupta, Y. Bi, F. Di Stasio, G. Konstantatos,
Small 2017, 13, 1700598.
1
39, 39–42; b) N. Kawano, M. Koshimizu, Y. Sun, N. Yahaba, Y. Fujimoto,
T. Yanagida, K. Asai, J. Phys. Chem. C 2014, 118, 9101–9106; c) J. Albero,
H. García, J. Mater. Chem. C 2017, 5, 4098–4110.
[7] a) A. Ronlan, J. Coleman, O. Hammerich, V. D. Parker, J. Am. Chem. Soc.
974, 96, 845–849; b) M. R. Talipov, A. Boddeda, Q. K. Timerghazin, R.
Rathore, J. Phys. Chem. C 2014, 118, 21400–21408.
Manuscript received: September 4, 2019
Accepted manuscript online: September 8, 2019
Version of record online: ■■■, ■■■■
1
ChemCatChem 2019, 11, 1–8
www.chemcatchem.org
7
© 2019 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
��
These are not the final page numbers!

Full Papers
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
FULL PAPERS
The surface of benzidinium lead
iodide perovskite has been success-
fully coated by a simple silylation
method. The photocatalytic cis-to-
trans stilbene isomerization takes
place under visible light even with a
Y. Peng, Dr. J. Albero, Prof. H. García*
1 – 8
Surface Silylation of Hybrid Benzidi-
nium Lead Perovskite and its
Influence on the Photocatalytic
Activity
2
to 5 nm coating. This methodology
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
can be carried out using different si-
lylating reagents in order to enhance
the hydrophobic properties and the
stability against moisture of hybrid
perovskite materials.