Exploitation of two-dimensional conjugated covalent organic frameworks based on tetraphenylethylene with bicarbazole and pyrene units and applications in perovskite solar cells

Article abstract of DOI:10.1039/d0ta02956d

In this work, two-dimensional conjugated covalent organic frameworks (2D-COF) based on a building block of tetraphenylethylene are successfully developed. Bicarbazole and pyrene moieties are respectively coupled with 4,4′,4′′,4′′′-(ethane-1,1,2,2-tetrayl)tetranilino (ETTA)via[4 + 4] solvothermal condensation conditions of 3,3′,6,6′-tetraformyl-9,9′-bicarbazole (Car-4CHO) and 1,3,6,8-tetrakis(4-formylphenyl)pyrene (TFPPy) to affordCar-ETTAandTFPPy-ETTACOFs. According to thermogravimetric analysis (TGA), powder X-ray diffraction (PXRD), and N₂adsorption and desorption measurements, bothCar-ETTAandTFPPy-ETTACOFs exhibit excellent thermal stability, highly crystalline structure, and high specific surface area, respectively. These findings are supported by force field andab initiocalculations. Furthermore, successful applications of these COFs in perovskite solar cells (PVSCs) are demonstrated owing to their well-conjugated properties and π-π interactions. While serving as interlayers in the devices, these COFs could effectively promote the interfacial charge dynamic to further optimize the resulting performance. Besides, certain interaction between COFs and perovskite also leads to an improved morphology and crystallinity of the perovskite layer, presenting defect passivation capability. As a result, we show that the performance of the COF-modified PVSC could be improved from 17.40 to 19.80%.

Full text of DOI:10.1039/d0ta02956d

View Article Online
View Journal
Journal of
Materials Chemistry A
Materials for energy and sustainability
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: M. G. Mohamed,
C. Lee, A. F. M. EL-Mahdy, J. Luder, M. Yu, Z. Li, Z. Zhu, C. Chueh and S. Kuo, J. Mater. Chem. A, 2020,
DOI: 10.1039/D0TA02956D.
Volume
Number 12
28 March 2018
Pages 4883-5230
6
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Journal of
Materials Chemistry A
Materials for energy and sustainability
rsc.li/materials-a
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2050-7488
COMMUNICATION
Zhenhai Wen et al.
An electrochemically neutralized energy-assisted low-cost
acid-alkaline electrolyzer for energy-saving electrolysis
hydrogen generation
rsc.li/materials-a

Please do not adjust margins
Journal of Materials Chemistry A
Page 1 of 13
View Article Online
DOI: 10.1039/D0TA02956D
ARTICLE
Exploitation of Two-Dimensional Conjugated Covalent Organic
Frameworks Based on Tetraphenylethylene with Bicarbazole and
Pyrene Units and Applications in Perovskite Solar Cells
Received 00th January 20xx,
Accepted 00th January 20xx
Mohamed Gamal Mohamed,a+ Chia-Chen Lee,b,c+ Ahmed F. M. EL-Mahdy,a+ Johann Lüder,d Ming-
Hsuan Yu,b Zhen Li,c Zonglong Zhu,c Chu-Chen Chueh,b,e* and Shiao-Wei Kuoa,f*
DOI: 10.1039/x0xx00000x
In this work, two-dimensional conjugated covalent organic frameworks (2D-COF) based on a building block of
tetraphenylethylene are successfully developed. Bicarbazole and pyrene moieties are respectively coupled with 4,4',4",4"'-
(ethane-1,1,2,2-tetrayl)tetranilino (ETTA) via [4+4] solvothermal condensation conditions of 3,3',6,6'-tetraformyl-9,9’-
bicarbazole (Car-4CHO) and 1,3,6,8-tetrakis(4-formylphenyl)pyrene (TFPPy) to afford Car-ETTA and TFPPy-ETTA COFs.
According to thermogravimetric analysis (TGA), powder X-ray diffraction (PXRD), and N2 adsorption and desorption
measurements, both Car-ETTA and TFPPy-ETTA COFs exhibit excellent thermal stability, highly crystalline structure, and high
specific surface area, respectively. These findings are supported by force field and ab initio calculations. Furthermore,
successful applications of these COFs in perovskite solar cells (PVSCs) are demonstrated owing to their well-conjugated
properties and π-π interactions. While serving as interlayers in the devices, these COFs could effectively promote the
interfacial charge dynamic to further optimize the resulting performance. Besides, certain interaction between COFs and
perovskite also leads to an improved morphology and crystallinity of the perovskite layer, presenting defect passivation
capability. As a result, we show that the performance of the COF-modified PVSC could be improved from 17.40 to 19.80%.
Introduction
heterogeneous catalysis,22 water splitting,23 drug delivery,24
solar cells,25 and chemical sensors.26
Covalent organic frameworks (COFs) have been considered
as a fascinating class of organic materials with high crystallinity
and ordered conjugation.1-6 According to their structural
arrangement, this kind of high crystalline material can be
divided into layered two-dimensional (2D) and three-
dimensional (3D) networks.7-10 In recent years, COFs have
attracted significant research interest in both academic and
industrial communities due to their well-defined porous
properties, tailorable electronic and structural design, ordered
skeletons, facile preparation through reversible covalent bond
formation between organic linkers, and excellent chemical and
thermal stability.11-17 Owing to these appealing properties,
these materials have been widely applied in many technological
fields and device prototypes, such as pseudocapacitors,18 gas
From the perspective of molecular design, to promote the
charge transfer, the donor–acceptor (D-A) structure is the key
feature in controlling the charge dynamic by providing a
bicontinuous heterojunction structure and self-sorted D and A
columnar arrays for charge separation and ambipolar pathways
for charge collection.27 2D COFs, which usually consist of two
different conjugated rigid organic molecular building blocks,
present a 2D stacking pattern with D-A arrangement that
permits charge migration along the framework and through π-
π interactions.28-30 Besides, according to the D-A design
principle, the bandgaps and energy levels are adjustable and the
absorption range can even span over a broad visible and near-
infrared region up to 1100 nm.31 Credited with the crystalline
and conjugated properties, COFs show a promising potential in
photovoltaic applications and could provide an unprecedented
strategy in charge dynamics within organic materials. The large
contact area of COFs attributed to porous structure provides a
perfect space that is large enough for prospective filler (i.e.
fullerene derivatives, perovskite nanocrystals) to promote the
charge collection between the two layers.28,33 Meanwhile, their
high dimensional (2D and 3D) geometry might turn over a new
page in the optoelectronic applications as compared to the 1D
conjugated counterparts since they can better ameliorate the
charge collect, phase separation, and the orientation issues of
the devices.
separation,19
proton
conduction,20
semiconductors,21
a. Department of Materials and Optoelectronic Science and Center of Crystal
Research, National Sun Yat-Sen University, Kaohsiung 80424, Taiwan
b. Department of Chemical Engineering, National Taiwan University, Taipei 10617,
Taiwan.
c. Department of Chemistry, City University of Hong Kong, Kowloon 999077, Hong
Kong.
d. Department of Materials and Optoelectronic Science, National Sun Yat-Sen
University, Kaohsiung 80424, Taiwan
e. Advanced Research Center for Green Materials Science and Technology, National
Taiwan University, Taipei 10617, Taiwan.
f. Department of Medicinal and Applied Chemistry, Kaohsiung Medical University,
Kaohsiung, Taiwan.
† These authors are equally contributed to this work.
Electronic Supplementary Information (ESI) available: [details of any supplementary
information available should be included here]. See DOI: 10.1039/x0xx00000x
Provided these significant properties, COFs have been
utilized in organic solar cells as the photoactive materials.34,35
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 2 of 13
ARTICLE
Journal Name
However, only inferior power conversion efficiency (PCE) has
been achieved so far owing to the narrow absorption and
unsatisfactory charge collection with pure COF layer.
Meanwhile, based on their porous characteristic, the COFs were
further blended with phenyl-C61-butyric acid methyl ester
(PCBM) to enhance the charge dissociation and transporting
capabilities.33 Although some improvements were made, the
overall PCEs were still below 1%. Instead of being used as the
photoactive component, Chen et al. developed a two-in-one
molecular design to successfully facilitate the growth of 2D-
COFs on ITO and employed them as the interface modifier in
perovskite solar cells (PVSCs).32 This pioneering exploration
opens up the potential applications of 2D-COFs in the
photovoltaic community. Meanwhile, Zhao et al. recently
invented a series of spirobifluorene-based 3D-COFs and used
them as additive in PVSCs to yield an improved PCE to 18.0%.25
With suitable energy-level relationship, when the 3D-COFs is
excited by light, electrons are generated at the LUMO-localized
acceptor and transported to perovskite, while holes are
produced at the HOMO-localized donor and accept electrons
excited by nearby perovskite. Based on this phenomenon, the
3D-COF addition in the perovskite layer facilitates charge
transfer and simultaneously reduce charge recombination
within perovskite regions, resulting in a higher short-circuit
current (Jsc) and fill factor (FF) of the derived device.25 This result
signifies a new direction of applying COFs in the photovoltaic
devices; despite this, the applications of COFs in photovoltaics
remain rare until now and their promising properties which can
be relevant for many technological fields also need further
explorations.
View Article Online
DOI: 10.1039/D0TA02956D
Scheme 1. Synthesis of Car-ETTA COF and TFPPy-ETTA
COF.
synthesized by [4+4] solvothermal condensation condition of
Car-4CHO and TFPPy both with ETTA in the presence of 1,4-
dioxane/mesitylene/acetic acid (6M) for Car-ETTA COF and n-
butanol/1,2-dichlorobenzene/acetic acid (6M) for TFPPy-ETTA
o
COF at 120 C for 72 h, as displayed in Scheme 1. Both yellow
powders of Car-ETTA and TFPPy-ETTA COFs are not soluble in
common organic solvents including DMF, NMP, acetone, and
THF. The building monomers of Car-4CHO, TFPPy, and ETTA
were prepared according to the previously reported methods
(
Scheme S1-S5
)
and all spectroscopic analyses were
summarized in the experimental parts (Figures S1-S8).38-42 As
seen in the 1H-NMR spectra, the ETTA shows characteristic
signals at 4.82 and 6.26-6.58 ppm for the NH2 group and
aromatic protons (Figure S1) while Car-4CHO displays the
proton signals at 10.20 ppm for the aldehydic group (Figure S7).
Further, the 13C-NMR spectrum of ETTA displays five
characteristic signals of carbon atoms (113.74-146.47 ppm)
To well unroll the typical properties of COFs, in this work, we
synthesized two kinds of 2D COFs, Car-ETTA and TFPPy-ETTA.
We first investigate the structure-property relationship of these
two COFs in both aspects of experiment and theoretical
calculation. Further, we demonstrated their efficacy of serving
as an interlayer in PVSCs to enhance PCE, which can be
attributed to their eminent charge transporting ability as a
result of well-conjugated property and π-π interactions. It is also
revealed that their porous characteristic can facilitate the
nucleation and crystal growth of perovskite to possibly reduce
(
Figure S2) while that of Car-4CHO exhibits the typical C=O at
192.21 ppm (Figure S8). All these results indicate the successful
syntheses of ETTA and Car-4CHO monomers in high purity.
The chemical structures of Car-ETTA and TFPPy-ETTA COFs
were examined by using Fourier transform infrared (FTIR) and
solid state 13C NMR spectroscopy. As the FTIR spectra presented
in Figure S9, ETTA shows typical absorption bands at 3358 and
3420 cm-1 for the amino NH group and Car-4CHO displays
characteristic bands at 2842, 2752, and 1695 cm-1 for the
aldehydic (HC=O) group. The FTIR spectrum of TFPPy shows the
typical signals at 2824, 2731, and 1700 cm-1 due to C-H and C=O
stretching of HC=O units (Figure S10). The FTIR spectra of Car-
ETTA and TFPPy-ETTA COFs are presented in Figure S9 and
Figure S10, which clearly reveal the successful formation of the
imine linkage with the feature absorption of C=N stretching
band located at 1620 cm-1 and the bands of aromatic (C=C)
the charge recombination at the interface arising from defective
37
states.36,
Notably, we manifested that the different
synthesized units can tailor the energy levels of the derived
COFs to modulate the band alignment across associated
interfaces that leads to different open-circuit voltage (Voc).
Combing all these advantages, these COF interlayers enable the
PVSCs to deliver enhanced PCEs up to 19.80%, outperforming
the performance (17.40%) of the control device without COF
interlayers.
Results and Discussion
Synthesis and characterization
We first constructed and synthesized two new 2D imine- groups centered at 1595 and 1512 cm-1. Figure S11 displays the
linked Car-ETTA and TFPPy-ETTA COFs and investigated their solid state 13C NMR spectra of Car-ETTA and TFPPy-ETTA COFs,
structure-property relationship in both aspects of experiment wherein the carbon resonance signal of C=N is respectively
and simulations. Car-ETTA and TFPPy-ETTA COFs were both appeared at 158.6 ppm for Car-ETTA COF and at 156.4 ppm for
2 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 13
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
signals at 5.87°, 8.88°, 11.83°, and 19.70°, corresponding to the
View Article Online
DOI: 10.1039/D0TA02956D
(110), (210), (220), and (001) planes. The d-spacing between the
(110) plane (d110) and the π-stacking interlayer distance
between 2D sheets of Car-ETTA and TFPPy-ETTA COFs are then
calculated by Bragg equation. The respectively obtained values
are 1.26 nm and 4.23 Å for Car-ETTA COF and 1.50 nm and 4.50
Å for TFPPy-ETTA COF.
To better understand the structures of the studied COFs,
their PXRD patterns are also simulated from the Pawley
refinement (Figure 1a, red curve). As seen, the simulated PXRD
patterns are very similar to the experimental PXRD pattern
(
Figure 1a, black curve), as evidenced by their difference
pattern (Figure 1c, green curve). In addition, the XRD pattern
derived from the AA-eclipsed stacking models of both Car-ETTA
COF and TFPPy-ETTA COF (Figure 1a, blue curve, and Figure 1b-
1c) also well matches with the experimental patterns; whereas,
the pattern derived from their corresponding AB’ staggered
stacking model (Figure 1a, aqua curve, and Figures S13-S14
)
deviates largely from the experimental results. The refined unit
cell of the Car-ETTA COF is calculated to be a = 19.30 Å, b = 17.00
Å, and c = 4.60 Å with α = β = γ = 90o (residuals Rwp = 7.15% and
Rp = 5.10%) (Table S2). While, the refined unit cell of the TFPPy-
ETTA COF is a = 21.20 Å, b = 21.20 Å, and c = 4.50 Å with α = β =
γ = 90o (residuals Rwp = 4.40% and Rp = 3.24%) (Table S3).
To elucidate the permanent porosities properties of both
COFs, their respective Brunauer-Emmett-Teller (BET) surface
areas are determined by nitrogen adsorption and desorption
measurements at 77 K, as presented in Figure 2a. Both Car-ETTA
COF and TFPPy-ETTA COF show type I isothermal behavior with
a sharp uptake at low-pressure region, which is attributed to
microporosity properties. We found that the SBET = 829 m2 g-1
with a specific pore volume of 0.46 cm3 g-1 for Car-ETTA COF and
the SBET = 1156 m2 g-1 with a specific pore volume of 0.72 cm3 g-
1 for TFPPy-ETTA COF. In addition, the pore sizes for Car-ETTA
COF and TFPPy-ETTA COF were estimated by nonlocal density
functional theory and the results were 1.29 and 2.23 nm as
displayed in Figure 2b, respectively. From the BET results,
TFPPy-ETTA COF had a higher surface area and larger pore sizes
than Car-ETTA COF, which suggests a larger mesopore with
Figure 1. (a) PXRD pattern of the Car-ETTA and TFPPy-ETTA
COFs: experimental (black), Pawley refined (red),
difference between the experimental and calculated (FF)
data (green), calculated for the AA-stacking model (blue)
and AB-stacking model (aqua). (b) and (c) Top and side view
of the AA-stacking model of Car-ETTA and TFPPy-ETTA
COFs, respectively.
TFPPy-ETTA COF. Both the FTIR and solid state 13C NMR spectra
confirm the successful syntheses of the targeted Car-ETTA and
TFPPy-ETTA COFs.
Crystallinity analysis with force-field calculations
The thermal stability of both synthesized COFs is then
examined by thermogravimetric analysis (TGA) measured under
a N2 atmosphere. As seen in Figure S12, both COFs display
excellent thermal stabilities. Car-ETTA COF possesses a high
decomposition temperature (Td10) of 467 oC with a char yield of
59 % while the Td10 of TFPPy-ETTA COF is 580 oC with a char yield
81% (Table S1). Next, we performed powder X-ray diffraction
(PXRD) and computational simulations (classical force field (FF)
simulation43 in combination with XRD simulation on the fitted
unit cell; the results are compared to DFT, for details see section
of Computational Details to elucidate the crystalline nature and
unit cell parameters of Car-ETTA and TFPPy-ETTA COFs. As
depicted in Figure 1a, the experimental PXRD patterns display
the characteristics PXRD signals at 6.98°, 10.37°, 13.15°, and
20.96° for Car-ETTA COF, corresponding to the (110), (210),
(220), and (001) planes. While TFPPy-ETTA COF shows PXRD
Figure 2. (a) N2 gas adsorption and desorption isotherms
and (b) pore size distribution curve of Car-ETTA and TFPPy-
ETTA COFs. (c, e) FE-SEM and (d, f) TEM images of (c, d)
Car-ETTA and (e, f) TFPPy-ETTA COFs, respectively.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 4 of 13
ARTICLE
Journal Name
microporous exists inside TFPPy-ETTA COF. Such discrepancy
lies in the different structure of the bridging unit, wherein the
more planar and the spacing between bridging units of TFPPy
also increases the pore size. The morphology of both COFs is
examined by field emission scanning and transmission electron
microscopy (FE-SEM and TEM) as presented in Figures 2c-2d
and Figures 2e-2f. The FE-SEM images showed that a uniform
spherical shaped aggregation morphology is observed for Car-
ETTA COF while entangled nanofibers are observed for TFPPy-
ETTA COF. Their corresponding TEM images are shown in
View Article Online
DOI: 10.1039/D0TA02956D
Figures 2d and 2f
. Car-ETTA COF possesses spherical
morphology while TFPPy-ETTA COF shows microtubular
morphology.
After elucidating the structural properties of these studied
COFs, we are interested in exploring their potential in recent
rising photovoltaic technologies, PVSCs, motivated by their
layered structure. Based on our previous work and related well-
conjugated property and π-π interactions within the 2D works
from other research groups, it was unveiled that introducing
additional mesoporous scaffolds at the perovskite/CTL interface
can promote the crystallinity of the perovskite film grown on the monomer model. While the values of the HOMOs are in fair
top.36, 37 Thus, we herein envisage to explore the roles of both agreement with the reported experimental values, the larger
COFs as a surface modifier of charge-transporting layer (CTL) in differences observed in the LUMO values can be explained with
PVSC devices.
the limitations of DFT (e.g. contracted band gap), size effect,
Prior to the test in the device, we first characterize the and the missing bulk contributions. Compared to the Car-4CHO
bandgap and energy levels of Car-ETTA COF and TFPPy-ETTA and ETTA molecules, the gap values of the monomers are
COF. The bandgap of both COFs is determined through their decreased, while the TFPPy molecule has a similar value. Details
corresponding Tauc plots (Figure S15a-b) and their work are given in Figure S16. The frontier orbitals are delocalized in
function is measured using a photoelectron spectrometer the molecules. The Car-ETTA and TFPPy-ETTA monomers show
model, AC-2 (Figure S15c-d). The respective highest occupied a localization of the equivalent frontier states. The localization
molecular orbital (HOMO) level for Car-ETTA and TFPPy-ETTA is of the orbitals on different parts of the molecules and a small
estimated to be -5.49 and -5.54 eV and the respective lowest but non-vanishing overlap of the occupied and the unoccupied
unoccupied molecular orbital (LUMO) level is -3.39 and -3.54 eV. frontier states may suggest possible electron transfer paths in
It thus can be concluded that both COFs are more suitable to the 2D layers of the COF as indicated in the Figure 3 by green
serve the surface modifier of hole-transporting layer (HTL) in arrows. The transfer can happen between subunits as well as
PVSCs, especially for the inverted structure (Figure 8a, more across the COF among different units. However, an accurate
details will be discussed later), to better mediate the energy- description of electron transfer process requires computational
level alignment at the HTL/perovskite interface.44-46
techniques that are beyond ground state based DFT calculations
and are beyond the scope of this article.
Ab initio calculations
Figure 4 shows the structures of monolayer and bulk phase
To gain insights at the atomic scale, first principle structures for both COFs for comparison. For the bulk phases,
calculations based on Density Functional Theory were the formation energies of the COFs in different stacking
performed on monomers, monolayers, and bulk phases for the configurations are given in Table 1. Details of the different
two prepared COFs. The optimized structure of the Car-ETTA geometries of the bulk phases are given in the computational
and the TFPPy-ETTA monomers with passivated ends are shown details. The formation energies of the bulk phase with respect
in Figure 3 together with the orbital shapes of HOMO and LUMO to individual monolayers were calculated according to Eq. (1)
of the individual molecules. Both Car-ETTA and TFPPy-ETTA (see below). Car-ETTA has a formation energy of -1.4 eV and
monomers are non-flat, which can cause a non-flat structure in TFPPy-ETTA of -1.7 eV (Table 1). For all cases, the energy
the 2D layer as confirmed by the FF calculations. The electronic difference between differently stacked geometries is small (in
levels are also indicated in Figure 3. The Car-ETTA monomer has the order of 100 meV). Car-ETTA has a preferred AB1 eclipsed
eigenvalues of HOMO and LUMO at -5.88 and -0.59 eV resulting stacking, i.e. alternating layers are shifted by (0.5, 0.5) lateral
in a fundamental gap of 5.3 eV. The fundamental gap has a unit vectors, but with the same relative orientation, although
similar value (5.3 eV) in TFPPy-ETTA mainly due to a slightly the energy difference of 50 meV to the simple unit cell with
deeper LUMO at -1.53 eV and a HOMO of ca. -6.82 eV. In
agreement with the experimental values, TFPPy-ETTA has a
lower HOMO than Car-ETTA, while the absolute value for TFPPy-
ETTA is overestimated by the DFT calculation potentially due to
4 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 13
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/D0TA02956D
Car-ETTA, two dihedral rotations between diphenylmethylene
and Carbazol-9-ide subunits can cause the out-of-plane
displacements. In the single-molecule calculations, the dihedral
angle was close to 90° between the Carbazol-9-ide-like subunits
in Car-4CHO and 57.3° in ETTA. In the Car-ETTA monolayer as
well as the bulk phase calculations, this dihedral angle
drastically reduces to a value of ca. 52.3 and 46.3°, respectively,
in Carbazol-9-ide. The dihedral angle in diphenylmethylene
subunit is also decreased when compared to the single
molecule calculation. The values reach ~ 12-18° in the
monolayer and bulk phase. Smaller values are prevented by the
steric hindrance. Similar trends can be seen in TFPPy-ETTA, in
which these dihedral angles also change to small values of less
than 10°. However, the dihedral angles in TFPPy subunits
remain almost unchanged (ca. 53°) in monolayer and bulk
formation. Moreover, it is noticeable that the bulk phase has
generally smaller dihedral angles than the monolayer phases
and the chemical surrounding has only a slight influence on the
structure of each layer. For the most stable configurations, we
estimate pore areas from the computed atomic structure
(neglecting H atoms) and an area is spanned to frame the void
regions. The computed pore sizes in Car-ETTA and TFPPy-ETTA
COFs are 123.7 Å2 and 158.8 Å2, respectively, shown in Figure a
and 5b, qualitatively confirming the experimental pore volume
given above.
vertically eclipsed stacking (AA, e.g., similar to the FF simulation
above) is minimal. TFPPy-ETTA has an energy difference of 90
meV between the vertically stacked configuration (AA) (Table 1
)
and the most stable configuration being an eclipsed AB stacking
(i.e. AB3), in which the B layer is rotated and shifted.
The electronic structure, especially the HOMO-LUMO gap,
can drastically change when going from monomer to the bulk
phase. The energy difference between valence band maximum
and conduction band minimum reduces to ca. 1.7 eV for Car-
ETTA and ca 2.0 eV for TFPPY-ETTA. For the monolayers, these
For the most stable configurations, that are the shifted Car-
ETTA (AB1) and the shifted and rotated TFPPy-ETTA (AB3), the
computed lattice constants (angles) are 19.6, 16.4 and 9.73 Å
(86.9°, 110.0° and 90.0°) and 21.9, 21.9 and 9.21 Å (87.3°, 87.7°
and 90.1°) for Car-ETTA and TFPPy-ETTA, respectively. These
values agree with the results of the experimental and FF
calculations above. Only the c vector is about twice because
current model includes two layers in one-unit cell. Compared
to the monolayer with lateral unit vectors of 19.4 and 16.3 Å
and (89.4°, 89.0°) for Car-ETTA and 22.9 and 20.5 Å (88.5°, 89.9°)
for TFPPy-ETTA, the bulk phases exhibit little difference in those
vectors indicating weak interactions between COF layers.
Furthermore, the calculations indicate that van der Waals
forces dominate the interaction between COF layers resulting in
minimal distances of ca 3 Å and an average height of 4.6 to 5 Å
between neighboured COF planes.
In general, each COF layer is not flat but shows slight out-of-
plane displacements resulting in a total height (out of molecular
plane) difference within a COF layer of ca. 3 Å. These
displacements are caused by dihedral rotations of molecular
subunits. Details of the dihedral angles are given in Table S4. In
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 6 of 13
ARTICLE
Journal Name
energy gaps are 1.7 and 1.4 eV for Car-ETTA and TFPPy-ETTA, result in good agreement with the measured XVRieDw AsritgicnleaOl.nliInne
DOI: 10.1039/D0TA02956D
respectively. In the bulk phase, the values of the fundamental contrast, the single-layer unit cell (vertical stacking AA) has in
energy gaps are much closer to the experimental values for the both COFs a relatively wide separation of a double peak. This
HOMO-LUMO gap of ca. 2 eV for both COFs. Here, the indicates in combination with the computed formation energies
theoretical values are contracted, which is typical for the mean- that the COF has a preferred AB1 stacking, i.e., there are two
field approach.
Besides these calculated results, experimental XRD
layers in the unit cell.
For TFPPy-ETTA, all configurations but the single-layer have
measurements in combination with simulated XRD signals excellent agreement with the experiment. The main deviation
provide a reliable way to confirm the atomic structure of COFs, in the simulated XRD signal is seen in the AA case, where a
especially in cases where the energy difference between relatively wide double peak is formed. The here simulated
computed structure is relatively small as seen for different COF double peak is too narrow to be revealed by the experiment
configurations. Figure 6. Comparison of measured and that has a more substantial broadening of the peaks. In
computed XRD signals for (a) Car-ETTA (AB1) and (b) TFPPy- comparison to the peak width of Car-ETTA, it can be expected
ETTA (AB3), the inserts give the comparison of different that a wide separation of the double peak would result in a
configurations as given in Table 1.Figure a and 6b compare the broader experimental peak for TFPPy-ETTA, and therefore the
measured and computed XRD signals for Car-ETTA and TFPPy- AA configuration appears unlikely. Among the remaining AB
ETTA, respectively. For all computed configurations good to stackings, the energy separation is at least 270 meV between
very good agreement with the experimental main peak the most stable AB stacking configuration and similar cases.
positions at 7° and 5.9° for Car-ETTA and TFPPy-ETTA were Hence, AB3 is confirmed as experimental structure for TFPPy-
obtained. Car-ETTA shows excellent agreement for the shifted ETTA, also indicates that two layers are in the unit cell.
and rotated AB stacking (AB3). However, the shifted
configuration (AB1) is by more than 200 meV more stable and COF application in perovskite solar cells
has good agreement with the experimental XRD measurement.
With these insightful understanding at the atomic scale of
The apparent double peak in AB1 is very narrow which can still the synthesized COFs, we next study their applications in PVSCs.
As discussed earlier, both COFs are more suitable to serve the
surface modifier of HTL in inverted PVSCs to better mediate the
energy-level alignment at the HTL/perovskite interface (Figure
8a).44-46 Besides, it is important noting that a hydrophobic
surface has been proven to effectively promote the vertical
grain growth of perovskite film owing to the reinforced vertical
diffusion of precursor solution.4-6 To clarify the advantage of the
hydrophobic property of both COFs, we then investigate the
crystallinity of the perovskite film grown on top of their
modified HTLs.
We first examine the morphology of the perovskite films
grown on the COF-modified hydrophilic NiOx HTL, which is
6 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 13
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
commonly used in the inverted PVSCs.47-49 Figure 7a presents the COF-modified PTAA HTL is blue-shifted. GVeineweArartilclyle,Ontlhinee
the surface morphology of these films taken by the field- spontaneous radiative recombination leDaOdIs: 1t0o.10a39r/eDla0tTiAv0e29r5e6dD-
emission gun scanning electron microscope (FE-SEM). As can be shifted PL peak whereas the passivation of surface defects
clearly seen, the grain size of the perovskite film becomes larger results in the blue-shifted PL peak.50 The observed blue-shift in
when depositing on the COF-modified hydrophilic NiOx HTL. PL peak thus suggests that introducing COF at the
Being similar to the cases reported in the literature, introducing perovskite/PTAA interface can passivate the interfacial defects
these
hydrophobic mesoporous
scaffolds at
the and promote the film crystallinity to a certain degree by
perovskite/NiOx interface seems to effectively promote the providing an additional nucleation scaffold.36, 37 Based on these
crystallinity of the perovskite film.36,
The improved findings, we next ascertain the interactions between these COFs
37
crystallinity of the perovskite film is also confirmed by its and perovskite by using X-ray photoelectron spectroscopy (XPS),
corresponding X-ray diffraction (XRD) patterns, as displayed in and the whole XPS survey of the studied perovskite films are
Figure S17. The full width at half maximum (FWHM) value of the presented Figure S19a. As seen in Figure 7d, the binding energy
film deposited on Car-ETTA- and TFPPy-ETTA-modified NiOx HTL of Pb-4f peak of the film deposited on Car-ETTA COF is
is 0.274° and 0.287°, respectively, which is smaller than the negatively shifted compared to the pristine and TFPPy-ETTA-
value (0.296°) of the pristine film. Thus, this result indicates that modified films. This large shift can be attributed to the more
introducing COF-based mesoporous scaffolds at the lone pairs on Car-ETTA, which provides more intense
perovskite/NiOx interface effectively promote the crystallinity interactions with the perovskite through the Lewis acid-base
of the perovskite film, being like the previous cases of MOF- interactions.51 Besides, the N-1s spectra shown in Figure S19b
based mesoporous materials reported in the literature.36, 37
However, in order to fabricate high-performance PVSCs, we the studied COFs and the perovskite.
also exhibits clear shifts, indicating certain interactions between
herein would like to fabricate the PVSCs by using poly[bis(4-
To examine the real efficacy of these COFs as an interlayer
phenyl)(2,4,6-trimethylphenyl)amine (PTAA) as the HTL. in PVSCs, we herein fabricate inverted p-i-n PVSCs with a device
Different to NiOx HTL, PTAA is much more hydrophobic; configuration
of
indium
tin
oxide
therefore, the increase of the grain size in the perovskite film (ITO)/PTAA/COFs/(FAPbI3)0.83(MAPbBr3)0.17(CsPbI3)0.05/PCBM/B
grown on top of PTAA is not that obvious (Figure S18). CP/Ag. Figure 8a display the energy-level diagram of the studied
Regardless, these COFs still present certain surface modification device and its schematic illustration of COFs in the device. We
effects. As shown in the photoluminescence (PL) spectra in first access the charge transfer efficiency across the
Figure 7b-7c, the PL peak of the perovskite films deposited on perovskite/PTAA interface and perovskite/COF interface using
Figure 8. (a) The energy-level diagram (The unit for the energy level is eV) and schematic illustration of the studied PVSCs. (b)
Time-resolved PL spectra of the pristine perovskite film and the films deposited on studied COFs. (c) J-V characteristics (close
circle: measured under forward scan; open circle: measured under reverse scan), (d) EQE curves, and (e) dark currents of the
studied PVSCs.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 8 of 13
ARTICLE
Journal Name
time-resolved photoluminescence (TRPL). The results are
presented in Figure 8b, where the lifetimes (τ) are fitted with
the double-exponential model, and the detailed parameters
are summarized in Table S4. Generally, τ1 is regarded as time
from excitation to recombination at the surface while τ2 is
associated with a similar process but the recombination
happens in the bulk perovskite film. As shown, the average τ
(τavg) of the pristine perovskite film is 1475.8 ns and this value
will respectively reduce to 969.0 and 535.7 ns when the
perovskite film respectively deposits on TFPPy-ETTA COF and
Car-ETTA COF, revealing certain hole extraction capability of
both COFs. Note that Car-ETTA COF seems to have a better
hole extraction ability than TFPPy-ETTA COF. This can be
View Article Online
DOI: 10.1039/D0TA02956D
attributed to the higher-lying HOMO level of Car-ETTA COF current of the derived devices, which is almost one order less
relative to that of perovskite, which provides a larger than that of the control device. This result unveils that the
thermodynamic driving force for charge transfer. Owing to the reticulating porous structure of a COF scaffold indeed modified
better film formation and hole transport capability, the τavg of the charge transfer at the interface where charge loss is
the perovskite film deposited on PTAA can be largely reduced considered to be the main constituent limiting device
to 172.0 ns. However, the τavg of the perovskite film can be performance.
further reduced to 147.0 ns and 123.1 ns when it respectively
To further probe the charge dynamic in these devices, the
deposits on the TFPPy-ETTA-/Car-TTTA-modified PTAA layer. maximum exciton generation rate (Gmax) and charge collection
This result clearly manifests the advantageous role of COFs in probability (P) are investigated by plotting the photocurrent
improving the charge transfer efficiency across the density (Jph)-effective voltage (Veff) curves. The Jph is defined as
perovskite/PTAA interface, which benefits from their well- Jph = JL − JD, where JL and JD is the photocurrent measured under
conjugated property and π-π interactions within the 2D layered 1 sun AM 1.5G spectrum and dark current measured in a totally
structure.
dark condition, respectively. The Veff is defined as Veff = V0 − Vbias,
The photovoltaic performance of the fabricated devices is which V0 is the voltage while JL = JD and Vbias is the applied
measured under AM 1.5 G solar irradiance (100 mWcm-2). voltage bias. Generally, Jsat can be defined as the current when
Figure 8c compare their current density–voltage (J–V) all the generated excitons are dissociated into free carriers. And
characteristics and the relevant photovoltaic parameters are Gmax can be calculated by the equation of Jsat = eGmaxL, where e
summarized in Table 2. The performance of the fabricated PVSC is the elementary charge and L is the thickness of the perovskite
is clearly enhanced after inserting the COFs into the film (500 nm). The greater Gmax principally represents the higher
perovskite/PTAA interface. The TFPPy-ETTA-/Car-TTTA- utilized efficiency of incident photons of the device. From the
modified PVSCs can deliver a respective PCE of 19.72 and results shown in Figure S20, the Gmax of the control, TFPPy-
19.79% (measured under forward scan), outperforming the PCE ETTA-modified, and Car-ETTA-modified devices is calculated to
(17.40%) of the control device. Both COF-modified devices yield be 2.38 × × ×
1027, 2.62 1027, and 2.61 1027 s−1 m−3, respectively.
higher Jsc than the control device, which can result from the This result shows that the COF-modified devices possess higher
improved quality of the perovskite films and facilitated charge utilized efficiency of incident photons compared to the control
transfer as discussed earlier. Besides, both COF-modified PVSC device, confirming their higher Jscs. Besides, P can be
devices also give a slightly higher Voc than that of the control calculated by the ratio of Jph/Jsat, which can be referred as the
device. This enhancement can be attributed to the reduced charge collection capability of CTLs in the device. Taking Veff
=
charge recombination at the perovskite/PTTA interface and the 0.2 V for instance, the obtained P value for TFPPy-ETTA- and
deeper-lying HOMO levels of COFs. Notably, the Voc of TFPPy- Car-ETTA-modified devices is 90.6 and 90.3%, respectively,
ETTA-modified PVSC is larger than that of Car-TTTA-modified largely exceeding the value (74.9%) of the control device. It
PVSC owing to the deeper-lying HOMO level of the former again indicates the enhanced charge collection capability after
(
Figure 8a).
inserting the COF scaffolds. All the above results not only clearly
Figure 8d shows the external quantum efficiency (EQE) demonstrate the potential of COFs to serve as the interface
spectrum of each top-performing device, wherein all of them modifier in PVSCs but also highlight the importance of their
possess similar photo-response profiles suggesting the well light structural design in affecting the resultant photovoltaic
absorption of the perovskite layer in these devices. The parameters.
photocurrent integrated from the EQE spectrum is in a well
congruence with the value obtained in the J-V measurement,
validating the accuracy of the performance. Figure 8e presents
Conclusions
the dark current of the fabricated device. The dark current of a
device has been known to correlate with the charge
recombination at associated interface. As seen, both TFPPy-
ETTA-/Car-TTTA-modified PTAA layers can inhibit the dark
In summary, two kinds of highly crystalline 2D COFs, Car-ETTA
and TFPPy-ETTA, were successfully synthesized through [4+4]
imine solvothermal condensations of Car-4CHO and TFPPy with
the core building unit, ETTA. Both Car-ETTA and TFPPy-ETTA
8 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 13
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
COFs possess highly crystalline structure and well thermal then annealed at 100 °C for 10 minutes in glove-bVoiexw. ATrhticelenO, ntlhinee
DOI: 10.1039/D0TA02956D
COF interlayers were spin-coated onto NiOx or PTAA layers at
stability with a respective high degradation temperature (Td10%
)
of 467 and 580 oC. Besides, they have a high surface area of 829 4000 rpm. Afterwards, (FAPbI3)0.83(MAPbBr3)0.17(CsPbI3)0.05 films
and 1156 m2 g-1, respectively. Furthermore, these two COFs, as were spin-coated onto the HTL or COF-modified HTL at 5000
the 2D porous conjugated materials, are unveiled to benefit the rpm for 30 s. During the spin-coating process, 150 µL of
crystal growth of the perovskite film grown on top and to chlorobenzene used as anti-solvent was dropped onto the
facilitate the charge transfer across the associated interface. As center of film before the end of spin-coating, followed by
a result, we manifest the performance of the COF-modified annealing at 100 °C for 30 min. The PC61BM ETLs were spin-
PVSC could be improved from 17.40 to ~19.80%. In addition, we coated onto the perovskite films at 1000 rpm for 30 s. Finally,
show that the energy levels of the derived COFs can be tailored the BCP layer (6 nm) and Ag electrode (820 nm) was evaporated
by the constituent units, which highlights the importance of the under high vacuum (< 4 × 10-6 Torr). The effective illumination
structural design in affecting the resultant photovoltaic area was defined as 0.13 cm2.
parameters.
Characterization: Proton and carbon nuclear magnetic
resonance (1H and 13C NMR) spectra were recorded using an
INOVA 500 instrument with DMSO-d6 and CDCl3 as solvents and
tetramethylsilane (TMS) as the external standard. Chemical
Experimental Section
Materials: Carbazole (98%), pyrene (98%), 4,4’- shifts are provided in parts per million (ppm). FTIR spectra were
diaminobenzophenone (98%), 4-formybenzeneboronic acid recorded using a Bruker Tensor 27 FTIR spectrophotometer and
(98%), tetrakis(triphenylphosphine)palladium(0) [Pd(pph3)4], the conventional KBr plate method; 32 scans were collected at
dichloromethane, anhydrous magnesium sulfate (MgSO4), a resolution of 4 cm-1. Solid state nuclear magnetic resonance
bromine solution (Br2), and N-bromosuccinimide were (SSNMR) spectra were recorded using a Bruker Avance 400
purchased from Acros. Potassium permanganate (KMnO4), NMR spectrometer and a Bruker magic angle spinning (MAS)
potassium carbonate (K2CO3), N-formylpiperidine (99%), and probe, running 32,000 scans. Cross-polarization with MAS
mesitylene (99%) were obtained from Sigma. Tin (Sn), acetic (CPMAS) was used to acquire 13C NMR spectral data at 75.5 MHz.
acid (99.8%), hydrochloric acid (37%), n-butanol, 1,2- The CP contact time was 2 ms; 1H decoupling was applied
dichlorobenzene, acetone, n-butyl lithium 2.5 M in hexane, and during data acquisition. The decoupling frequency
1,4-dioxane were ordered from Alfa Aesar. 1,3,6,8- corresponded to 32 kHz. The MAS sample spinning rate was 10
tetrabromopyrene
[Pyrene-4Br],
and
1,3,6,8-Tetra(4- kHz. Elemental analysis (EA) was performed using an Elementar
formylphenyl)pyrene (TFPPy) were synthesized according to the vario EL III apparatus. Powder Xray diffraction (PXRD) was
40
literature previously reported method.39,
PbI2 powder (> carried out with a Siemens D5000 using monochromated Cu/Kα
99.99%) and Nickel(II) 2,4-pentanedionate (95%) were (λ = 0.1542 nm). The sample was spread in a thin layer on the
purchased from Alfa Aesar. PC61BM (99.5%), and BCP (>99.5%) square recess of an XRD sample holder. Field emission scanning
were obtained from UniRegion. PbBr2 (99.9%) was purchase electron microscopy (FE-SEM) was conducted using a JEOL JSM-
from TCI. HC(NH2)2I (FAI) were purchase from Greatsolar. 7610F scanning electron microscope. Samples were treated via
CH3NH3Br (MABr) and CsI were purchase from Xi’an Polymer Pt sputtering for 100 s prior to observation. BET surface area
Light Technology Corporation. The detailed synthetic and porosimetry measurements of the prepared samples (ca.
procedures and characterization of studied COFs are described 20–100 mg) were performed using a Micromeritics ASAP 2020
in the supporting information.
Surface Area and Porosity analyzer. Nitrogen isotherms were
Solution preparation: The precursor solution of NiOx was generated through incremental exposure to ultrahigh-purity N2
prepared by dissolving nickel(II) 2,4-pentanedionate in ethanol (up to ca. 1 atm) in a liquid nitrogen (77 K) bath. TGA was
with a concentration of 25.7 mg/ml and with the addition of 10 performed using a TA Q-50 analyzer under a flow of N2
µl HCl. The poly(bis(4-phenyl)(2,4,6-trimethylphenyl)amine) atmosphere. The samples were sealed in a Pt cell and heated
(PTAA) HTL was prepared by dissolving in toluene with a from 40 to 800 °C at a heating rate of 20 °C min-1 under a flow
concentration of
processing the COF interlayers were prepared by dispersing modeling was performed using Reflex, a software package for
COFs in DMF with concentration of 0.5 mg/ml. crystal determination from XRD patterns, implemented in MS
2
mg/ml. The precursor solutions for of N2 atmosphere at a flow rate of 50 mL min-1. Molecular
a
(FAPbI3)0.83(MAPbBr3)0.17(CsPbI3)0.05 was prepared by mixing FAI modeling (v. 4.4, Accelrys). Unit cell dimensions were first
(1.25 M), PbI2 (1.5 M), MABr (1.25 M), PbBr2 (1.35 M), CsPbI3 manually determined from the observed XRD peak positions
(1M) in a mixed solvent of DMF:DMSO = 5:1 (v/v). PC61BM was using the coordinates. The UV-vis spectrum was recorded using
dissolved in chlorobenzene (CB) with a concentration of 20 a Hitachi U-4100 UV-visible spectrophotometer. The static-
mg/ml.
photoluminescence (PL) emission spectrum was measured by
Device fabrication: ITO-glasses were cleaned by DI water, FLS980 spectrofluorometer (Edinburgh) system and the time-
acetone, and isopropanol (IPA), followed by the plasma resolved PL was measured by time-correlated single photo
treatment for 10 minutes. For NiOx bottom layer, they were counting (TCSPC), fitting with the function of R(t) = B1exp(−t/τ1)
spin-coated onto the ITO-glasses at 5000 rpm for 10 s and then + B2exp(−t/τ2). J-V curves were measured under AM1.5G (100
annealed at 325 °C for 40 min in air. Additionally, the PTAA HTLs mW cm-2) illumination by a SS-F5-3A simulator, Enlitech,
were spin-coated onto the ITO-glasses at 6000 rpm for 30 s and recorded with a computer-controlled Keithley 2400 source. The
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 9
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 10 of 13
ARTICLE
Journal Name
EQE spectrum was performed with QE-R, Enlitech Co., Ltd, configurations were investigated following the resVuieltwsAortifcltehOenlFinFe
DOI: 10.1039/D0TA02956D
AM1.5G reference spectrum and corrected by a single crystal Si calculations. These configurations include vertical AA and three
photovoltaic cell. The x-ray photoelectron spectroscopy (XPS) different AB stackings with eclipsed geometries, which are
were measured with ULVAC-PHI PHI 5000 Versaprobe II. The FE- structures that have overlapping void regions along the c unit
SEM images were taken by NOVA NANO SEM 450 and Philips cell vector leading to the formation of void channels. Here,
XL30 FEG. The energy levels were measured with photoelectron eclipsed structures are considered because their simulated XRD
spectroscopy by a photoelectron spectrometer model AC-2 patterns show good agreement with the experimental XRD
(Riken Keiki) And the XRD spectrum was measured by the spectrum, while those of other tested structures did not.
X’PERT X-ray diffractometer.
Besides a simple vertical stacking (AA), the other configurations
were generated by shifting layers with respect to each other by
(0.5,0.5,1.0) unit cell vectors (AB1), a rotation of one layer by
90° around a rotation axis parallel to the c vector (i.e. parallel to
Computational Details
Classical FF simulations of COF bulk phases were performed the stacking direction) (AB2), and combinations of both
in the Materials Studio 7.0 suite.43 One initial guess for each of structure modifications (AB3). In should be noted that the
the crystalline COFs was constructed by hand. The setups resulting AB stackings are eclipsed in contrast to the staggered
contained one COF layer in a unit cell. Atomic positions, and unit AB’ stacking investigated by FF calculations and that the shifting
cell shapes and sizes were motivated by the results of the single by (0.5,0.5,1.0) unit cell vectors causes the Car and ETTA
molecules, i.e. Car-4CHO, ETTA, and TFPPy. These are the subunits to be vertically stacked along c direction in alternating
starting points for the FF optimizations employing the Forcite layers for Car-ETTA, and TFPPy and ETTA subunits in TFPPy-ETTA.
module optimizing atomic positions and the unit cells via energy The bulk formation energies with respect to a free-standing
minimization. The cell optimizations were conducted via a monolayer are computed as the total energy difference
volume relaxation, i.e. the unit cell shape is maintained. between an isolated COF layer and their bulk phase normalized
Interatomic forces were obtained with a universal force field.52 by the number of layers in the unit cell, i.e.
Moreover, atom-based van der Waals forces, which are
퐸푚 푙푎푦푒푟−푚 퐸푚표푛표푙푎푦푒푟
important for layered structures, and electrostatic
contributions were assigned by the Ewald method with a
threshold of 0.0001 kcal mol-1. Energy and force convergence
퐸푓 =
(1)
푚
where 퐸푓 is the formation energy, 퐸푚 푙푎푦푒푟 is the total energy
obtained from DFT calculations of the unit cell with m layers,
and 퐸푚표푛표푙푎푦푒푟 is the total energy of the single layer. Negative
values indicate a release of energy.
thresholds were set to 0.0001 kcal mol-1 and 0.005 kcal mol-1 Å-
1
,
respectively. Convergence was achieved within 500 iterations.
It should be noted the optimizations were split into three stages
to achieve convergence. A Quasi-Newton algorithm was used
for initial and intermediate optimizations with loser
convergence parameters and the final structures were obtained
with a conjugate gradient algorithm and the convergence
thresholds mentioned above.
Conflicts of interest
There are no conflicts to declare.
Density functional theory (DFT) calculations including single
point and structure optimizations of single molecules were
performed with the Gaussian package.53 The exchange-
correlation effects were described by the CAM-B3LYP
functional,54 a range-separated modification of the B3LYP
functional, and the electronic states were represented by the 6-
31G(d,p) basis set on all atoms. Calculations for COFs in
monolayer and bulk phase configurations were performed with
the Quantum Espresso code.55,56 The PBE functional57 and the
PAW method58 were used to describe the exchange-correlation
effects and electronic states, respectively, and an additional
correction for the missing van der Waals interaction was
included by the Grimme D3 method.59 The cutoff of the kinetic
energy was 600 eV, and Monkhorst-Pack k-point meshes of
2x2xn and 2x3xn were used for TFPPy-ETTA and Car-ETTA
calculations, respectively, to sample the Brillouin zone (where n
is in stacking direction and n=1 for a monolayer, n=10 for a bulk
phase unit cell with one layer and n=5 for a bulk phase unit cell
containing two COF layers).
Acknowledgements
The authors thank the financial support from the Ministry of
Science and Technology in Taiwan (MOST 108-2638-E-002-003-
MY2, 108-2218-E-110-013-MY3, and 108-2221-E-002-026-
MY3). C.-C. C also thank the financial supports from the
Featured Area Research Center Program within the framework
of the Higher Education Sprout Project by the Ministry of
Education (108L9006) and the Ministry of Science and
Technology in Taiwan (MOST 108-3017-F-002-002) and Top
University Project from National Taiwan University (109L7846).
J. L. acknowledges financial support from MOST 108-2112-M-
110-001-MY2 and is grateful to the National Center for High-
performance computing for computer time and facilities.
Notes and references
1
S. Lin, C. S. Diercks, Y. B. Zhang, N. Kornienko, E. M. Nichols, Y.
Zhao, A. R. Paris, D. Kim, P. Yang, O. M. Yaghi and C. J. Chang,
Science 2015, 349, 1208-1213.
Besides a monolayer configuration in which the vertical
distance between layers was more than 20 Å to avoid cross-talk
between unit cells in vertical c direction, four layered
2
N. Huang, X. Chen, R. Krishna and D. Jiang, Angew. Chem., Int.
10 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 11 of 13
Journal of Materials Chemistry A
Please do not adjust margins
Journal Name
ARTICLE
Ed. 2015, 54, 2986-2990.
Nanoscale 2018, 10, 4708-4717.
D. Bessinger, L. Ascherl, F. Auras and T. Bein, J. Am. Chem. Soc.
2017, 139, 12035-12042.
Y. Li, Q. Chen, T. Xu, Z. Xie, J. Liu, X. Yu, S. Ma, T. Qin and L.
Chen, J. Am. Chem. Soc. 2019, 141, 13822-13828.
M. Dogru, M. Handloser, F. Auras, T. Kunz, D. Medina, A.
Hartschuh, P. Knochel and T. Bein, T. Angew. Chem. Int. Ed.
2013, 52, 2920-2924.
View Article Online
DOI: 10.1039/D0TA02956D
3
4
Y. Zheng, R. Zou and Y. Zhao, Adv. Mater., 2016, 28, 2855-2873. 31
A. F. M. EL-Mahdy, Y. H. Hung, T. H. Mansoure, H. H. Yu, Y. S.
Hsu, K. C. W. Wu and S. W. Kuo, J. Taiwan Inst. Chem. Engrs. 32
2019, 103, 199-208.
A. F. M. El-Mahdy, C. Young, J. Kim, J. You, Y. Yamauchi and S. 33
W. Kuo, ACS Appl. Mater. Interfaces 2019, 11, 9343-9354.
H. L. Li, X. L. Feng, X. H. Cui, Y. X. Ma, S. Y. Ding and W. Wang,
5
6
J. Am. Chem. Soc. 2017, 139, 6042−6045.
Q. Gao, X. Li, G. H. Ning, H. S. Xu, C. Liu, B. Tian, W. Tang and
K. P. Loh, Chem. Mater. 2018, 30, 1762-1768.
H. R. Abuzeid, A. F. M. EL-Mahdy, M. M. M. Ahmed and S. W.
Kuo, Polym. Chem., 2019, 10, 6010-6020
P. Côté, I. Benin, N. W. Ockwig, M. Keefe, A. J. Matzger and O.
M. Yaghi, Science 2005, 310, 1166-1170.
34
35
M. Calik, F. Auras, L. M. Salonen, K. Bader, I. Grill, M.
Handloser, D. D. Medina, M. Dogru, F. Lobermann, D. Trauner,
A. Hartschuh and T. Bein, J. Am. Chem. Soc. 2014, 136, 17802-
17807.
J. Guo, Y. Xu, S. Jin, L. Chen, T. Kaji, Y. Honsho, M. A. Addicoat,
J. Kim, A. Saeki, H. Ihee, S. Seki, S. Irle, M. Hiramoto, J. Gao
and D. Jiang, Nat Commun. 2013, 4, 2736-2744.
C. C. Lee, C. I. Chen, Y. T. Liao, K. C. Wu and C. C. Chueh, Adv.
Sci. 2019, 6, 1801715-1801724.
7
8
9
10
11
12
13
14
Q. Jiang, Y. Li, X. Zhao, P. Xiong, X. Yu, Y. Xu and L. Chen, J. 36
Mater. Chem. A 2018, 6, 17977-17981.
H. Ding, J. Li, G. Xie, R. Chen, Z. Peng, C. Yang, B. Wang, J. Sun 37
and C. Wang, Nat. Chem. 2018, 9, 5234-5241.
S. Y. Ding and W. Wang, Chem. Soc. Rev. 2013, 42, 2013, 548- 38
568.
S. Das, P. Heasman, T. Ben and S. Qiu, Chem. Rev. 2017, 117, 39
1515-1563.
A. F. M. EL-Mahdy, C. H. Kuo, A. Alshehri, C. Young, Y. 40
Yamauchi, J. Kim and S. W. Kuo, J. Mater. Chem. A 2018, 6,
19532-19541.
D. Shen, A. Pang, Y. Li, J. Dou and M. Wei, Chem. Commun.
2018, 54, 1253-1256.
S. Feng, H. Xu, C. Zhang, Y. Chen, J. Zeng, D. Jiang and J. X.
Jiang, Chem. Commun. 2017, 53, 11334-11337.
P. J. Waller, Y. S. AlFaraj, C. S. Diercks, N. N. Jarenwattananon
and O. M. Yaghi, J. Am. Chem. Soc. 2018, 140, 9099-9103.
A. F. M. EL-Mahdy, M. G. Mohamed, T. H. Mansoure, H. H. Yu,
T. Chen and S. W. Kuo, Chem. Commun., 2019, 55, 14890-
14893.
15
16
17
W. Cai, C. C. Chu, G. Liu and Y. X. J. Wáng, Small 2015, 11, 41
4806-4822.
H. Lyu, C. S. Diercks, C. Zhu and O. M. Yaghi, J. Am. Chem. Soc. 42
2019, 141, 6848-6852.
Y. Liu, C. S. Diercks, Y. Ma, H. Lyu, C. Zhu, S. A. Alshmimri, S. 43
Alshihri and O. M. Yaghi, J. Am. Chem. Soc. 2019, 141, 677-
S. Xu, X. Bai, J. Ma, M. Xu, G. Hu, D. T. James and L. Wang,
Anal. Chem. 2016, 88, 157853-7857.
F. Xu, S. Yang, X. Chen, Q. Liu, H. Li, H. Wang, B. Wei and D.
Jiang, Chem. Sci. 2019, 10, 6001-6006.
BIOVIA Materials Studio 7.0, Dassault Systemes Biovia,
Cambridge, UK, 2015
683.
44
S. Wang, T. Sakurai, W. Wen and Y. Qi, Adv. Mater. Interfaces
2018, 5, 1800260.
I. Gelmetti, N. F. Montcada, A. Pérez-Rodríguez, E. Barrena,
C. Ocal, I. García-Benito, A. Molina-Ontoria, N. Martín, A.
Vidal-Ferran and E. Palomares, Energy Environ. Sci. 2019, 12,
1309-1316.
18
19
20
21
D. P. Cao; J. H. Lan; W. C. Wang and B. Smit Angew. Chem., Int.
Ed. 2009, 48, 4730-4733.
H. S. Yang; Y. Du; S. Wan, G. D. Trahan, Y. H. Jin and W. Zhang,
Chem. Sci. 2015, 6, 4049-4053.
H. Xu; S. Tao; Jiang and D Jiang, Nat. Mater. 2016, 15, 722-
726.
M. Dogru, M. Handloser, F. Auras, T. Kunz, D. Medina, A.
45
46
C. C. Chueh, C. Z. Li and A. K. Y. Jen, Energy Environ. Sci. 2015,
8, 1160-1189.
Hartschuh, P. Knochel and T. Bein, Angew. Chem., Int. Ed. 47
2013, 52, 2920-2924.
H. Xu, J. Gao and D. Jiang, Nat. Chem. 2015, 7, 905-912.
P. P. Pachfule, A. Acharjya, J. Roeser, T. Langenhahn, M. 48
Schwarze, R. Schomacker, A. Thomas and J. Schmidt, J. Am.
C. H. Tsai, N. Li, C. C. Lee, H. C. Wu, Z. Zhu, L. Wang, W. C,
Chen, H. Yan and C. C. Chueh, J. Mater. Chem. A 2018, 6,
12999-13004.
C. Bi, Q. Wang, Y. Shao, Y. Yuan, Z. Xiao and J. Huang, Nat
Commun 2015, 6, 7747.
22
23
Chem. Soc. 2018, 140, 1423-1427.
L. Bai, S. Z. F. Phua, W. Q. Lim, A. Jana, Z. Luo, H. P. Tham, L.
Zhao, Q. Gao and Y. Zhao, Chem. Commun., 2016, 52, 4128- 50
4131.
49
C. C. Lee, C. I. Chen, C. T. Fang, P. Y. Huang, Y. T. Wu and C.
C. Chueh, Adv. Funct. Mater. 2019, 29, 1808625.
Y. Shao, Z. Xiao, C. Bi, Y. Yuan and J. Huang, Nat Commun. 2014,
5, 5784-5791.
24
25
26
27
28
29
C. Wu, Y. Liu, H. Liu, C. Duan, Q. Pan, J. Zhu, F. Hu, X. Ma, T. Jiu, 51
Z. Li and Y. Zhao, J. Am. Chem. Soc. 2018, 140, 10016-10024.
Q. Gao, X. Li, G. H. Ning, K. Leng, B. Tian, C. Liu, W. Tang, H. S. 52
Xu and H. K. P. Loh, Chem. Commun. 2018, 54, 2349-2352.
H. T. Turan, O. Kucur, B. Kahraman, S. Salman and V. Aviyente, 53
Phys. Chem. Chem. Phys. 2018, 20, 3581-3591.
H. Bildirir, V. G. Gregoriou, A. Avgeropoulos, U. Scherf and C.
L. Chochos, Mater. Horiz. 2017, 4, 546-556.
S. Jin, X. Ding, X. Feng, M. Supur, K. Furukawa, S. Takahashi,
M. Addicoat, M. E. El-Khouly, T. Nakamura, S. Irle, S. Fukuzumi,
A. Nagai and D. Jiang, Angew. Chem. Int. Ed. 2013, 52, 2017-
2021.
J. W. Lee, H. S. Kim and N. G. Park, Acc. Chem. Res 2016, 49,
311-319.
A. K. Rappe, C. J. CasewitK. S. Colwell, W. A. Goddard III and
W. M. Skiff, J. Am. Chem. Soc. 1992, 114, 10024-10035.
Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B.
Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G.
Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M.
Caricato, A. Marenich, J. Bloino, B. G. Janesko, R. Gomperts,
B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L.
Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi,
J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe,
V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T.
30
S. Park, M. S. Kim, W. Jang, J. K. Park and D. H. Wang,
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 11
Please do not adjust margins

Please do not adjust margins
Journal of Materials Chemistry A
Page 12 of 13
ARTICLE
Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell,
Journal Name
View Article Online
J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. Bearpark, J.
J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith, R.
Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C.
Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M.
Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K.
Morokuma, O. Farkas, J. B. Foresman, and D. J. Fox, Gaussian,
Inc., Wallingford CT, 2016.
DOI: 10.1039/D0TA02956D
54
55
T. Yanai, D. P. Tew and N. C. Handy, Chem. Phys. Lett. 2004,
393, 51.
P. Giannozzi, S. Baroni, N. Bonini, M. Calandra, R. Car, C.
Cavazzoni, D. Ceresoli, G. L. Chiarotti, M. Cococcioni, I. Dabo,
A. Dal Corso, S. de Gironcoli, S. Fabris, G. Fratesi, R. Gebauer,
U. Gerstmann, C. Gougoussis, A. Kokalj, M. Lazzeri, L. Martin-
Samos, N. Marzari, F. Mauri, R. Mazzarello, S. Paolini, A.
Pasquarello, L. Paulatto, C. Sbraccia, S. Scandolo, G. Sclauzero,
A. P. Seitsonen, A. Smogunov, P. Umari, and R. M.
Wentzcovitch, J. Phys. Condens. Matter 2009, 21, 395502.
P. Giannozzi, O. Andreussi, T. Brumme, O. Bunau, M.
Buongiorno Nardelli, M. Calandra, R. Car, C. Cavazzoni, D.
Ceresoli, M. Cococcioni, N. Colonna, I. Carnimeo, A. Dal Corso,
S. de Gironcoli, P. Delugas, R. A. DiStasio, A. Ferretti, A. Floris,
G. Fratesi, G. Fugallo, R. Gebauer, U. Gerstmann, F. Giustino,
T. Gorni, J. Jia, M. Kawamura, H.-Y. Ko, A. Kokalj, E. Küçükbenli,
M. Lazzeri, M. Marsili, N. Marzari, F. Mauri, N. L. Nguyen, H.-
V. Nguyen, A. Otero-de-la-Roza, L. Paulatto, S. Poncé, D. Rocca,
R. Sabatini, B. Santra, M. Schlipf, A. P. Seitsonen, A. Smogunov,
I. Timrov, T. Thonhauser, P. Umari, N. Vast, X. Wu and S. Baroni,
J. Phys. Condens. Matter 2017, 29, 465901.
56
57
J. P. Perdew, K. Burke and M. Ernzerhof, Phys. Rev. Lett. 1996,
77, 3865.
58
59
P. E. Blöchl, Phys. Rev. B. 1994, 50, 17953.
S. Grimme, J. Antony, S. Ehrlich and H. Krieg, J. Chem. Phys.
2010, 132, 154104.
12 | J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 13 of 13
Journal of Materials Chemistry A
View Article Online
Exploitation of Two-Dimensional Conjugated Covalent Organic Frameworks Based on
DOI: 10.1039/D0TA02956D
Tetraphenylethylene with Bicarbazole and Pyrene Units and Applications in Perovskite
Solar Cells
Mohamed Gamal Mohamed,a+ Chia-Chen Lee,b,c+ Ahmed F. M. EL-Mahdy,a+ Johann
Lüder,a,d Ming-Hsuan Yu,b Zhen Li,c Zonglong Zhu,c Chu-Chen Chueh,b,e* and Shiao-Wei
Kuoa,f*
Keywords: Covalent organic frameworks; perovskite solar cells; interfacial materials
Two conjugated covalent organic frameworks (COFs) are successfully synthesized in this
work, presenting decent thermal stability, high crystallinity, and conjugated properties, which
is supported by classical force field and ab initio calculations. They are further introduced as
the interface modifier in perovskite solar cells (PVSCs) and are demonstrated to promote the
perovskite morphology and associated charge transfer to effectively enhance the resultant
photovoltaic performance of the derived PVSCs.
1