Combined Intrinsic and Extrinsic Proton Conduction in Robust Covalent Organic Frameworks for Hydrogen Fuel Cell Applications

Article abstract of DOI:10.1002/anie.201913802

Developing new materials for the fabrication of proton exchange membranes (PEMs) for fuel cells is of great significance. Herein, a series of highly crystalline, porous, and stable new covalent organic frameworks (COFs) have been developed by a stepwise synthesis strategy. The synthesized COFs exhibit high hydrophilicity and excellent stability in strong acid or base (e.g., 12 m NaOH or HCl) and boiling water. These features make them ideal platforms for proton conduction applications. Upon loading with H₃PO₄, the COFs (H₃PO₄?COFs) realize an ultrahigh proton conductivity of 1.13×10^?1 S cm^?1, the highest among all COF materials, and maintain high proton conductivity across a wide relative humidity (40–100 %) and temperature range (20–80 °C). Furthermore, membrane electrode assemblies were fabricated using H₃PO₄?COFs as the solid electrolyte membrane for proton exchange resulting in a maximum power density of 81 mW cm^?2 and a maximum current density of 456 mA cm^?2, which exceeds all previously reported COF materials.

Full text of DOI:10.1002/anie.201913802

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
International Edition Chemie
www.angewandte.org
Accepted Article
Title: Predisposed Intrinsic and Extrinsic Proton Conduction in Robust
Covalent Organic Frameworks for Hydrogen Fuel Cell Application
Authors: Yi Yang, Xueyi He, Penghui Zhang, Yassin Andaloussi, Hailu
Zhang, Zhongyi Jiang, Yao Chen, Shengqian Ma, Peng
Cheng, and Zhenjie Zhang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.201913802
Angew. Chem. 10.1002/ange.201913802
Link to VoR: http://dx.doi.org/10.1002/anie.201913802
http://dx.doi.org/10.1002/ange.201913802

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
Predisposed Intrinsic and Extrinsic Proton Conduction in Robust
Covalent Organic Frameworks for Hydrogen Fuel Cell Application
[
a]
[c]
[a]
[e]
[f]
[c]
Yi Yang, Xueyi He, Penghui Zhang, Yassin H. Andaloussi, Hailu Zhang, Zhongyi Jiang, Yao
[
b]
[g]
[d]
[a][b][d]
Chen, Shengqian Ma, Peng Cheng, and Zhenjie Zhang*
Abstract: Developing new materials to fabricate proton exchange
membranes (PEMs) for fuel cells is of great significance. Herein, a
series of highly crystalline, porous and stable new covalent organic
frameworks (COFs) have been developed by a stepwise synthesis
strategy. A high density of azo groups accompanied by phenolic
hydroxyl groups are integrated into the COF structures, allowing azo
groups to serve as proton acceptors and load sites for added acids,
while phenolic hydroxyl groups serve as proton donors facilitating
proton conduction. The synthesized COFs exhibit high hydrophilicity
and excellent stability in strong acid or base (e.g. 12 M NaOH or HCl),
and boiling water. These features make them ideal platforms for
developing clean and renewable energy is in urgent demand.
Hydrogen fuel is an ideal candidate to address this challenge due
to its high energy density and clean combustion product (i.e.
water).2-4 Hydrogen fuel cell (HFC) technology, especially proton
exchange membrane (PEM) fuel cells, offer a clean and reliable
alternative energy source due to their high energy conversion
efficiency, zero emissions, and mild operating conditions.5
Currently, exploring new proton conducting materials that can
serve as solid electrolyte membranes of PEM fuel cells is a
primary focus in this field.6-11 At present, few materials offer the
high proton conductivity and long working life necessary for
commercial application, with the main exception being Nafion, a
perfluorinated sulfonated polymer. However, the high cost,
3 4
proton conduction applications. Upon loading with H PO , the COFs
(
H
3
PO
4
@COFs) not only realize an ultrahigh proton conductivity of
-
1
1
.13ꢀ×ꢀ10 ꢀS/cm, the highest among all COF materials, but also
synthetic challenges and restricted operating conditions hinders
further application.1
2-13
Therefore, developing alternative
maintains high proton conductivity across a wide relative humidity (40-
00%) and temperature range (20-80 °C). Furthermore, membrane
electrode assemblies are fabricated using H PO @COFs as the solid
1
materials to fabricate PEMs is of great significance and urgently
needed.
3
4
electrolyte membrane for proton exchange resulting in a maximum
Organic polymers have offered great promise for PEMs owing
to their potential for high porosity, good structure robustness and
facile membrane fabrication.14-20 However, traditional organic
polymers are amorphous and difficult to achieve well-defined
control of structure and pore environments (i.e. pore size, shape
and exposed functional groups), thereby leading to a lack of
thorough understanding of the structure–property relationships
involved.21-22 In the past decade, covalent organic frameworks
2
power density of 81 mW/cm and a maximum current density of 456
2
mA/cm , which exceeds all previously reported COF materials. This
work not only develops an approach to prepare COFs with multiple
bond linkages, but also sets new COF benchmarks for proton
conduction and PEM fuel cell performance.
(
COFs) have emerged as a new class of crystalline porous
Introduction
organic polymer materials and demonstrated great potential to
overcome the limits of conventional organic polymers owing to
their well-defined structures, high surface areas, fine-tunable pore
environments, and custom-design functionalities.23-35 These
advantages render COFs a promising platform for proton
conductivity and PEMs. Recently, significant progress has been
made by Dichtel,36 Banerjee,37 our group38 and others39 to
fabricate high-performance COF membranes. The Banerjee
Global energy demand is continuously increasing, but
nevertheless energy supply is still mostly derived from the
combustion of fossil fuels which unavoidably brings about severe
environment issues, such as air and water pollution.1 Thus,
[
a]
Y. Yang, P. Zhang, Prof. P. Cheng, Prof. Z. Zhang
Renewable energy conversion and storage center, College of
Chemistry, Nankai University, Tianjin, 300071, China
E-mail: zhangzhenjie@nankai.edu.cn
-2
group achieved a benchmark proton conductivity (7.8×10 S/cm)
for a freestanding COF membrane which further served as the
PEM of a membrane electrode assembly (MEA) and delivered the
[
[
[
[
[
[
b]
c]
d]
e]
f]
Prof. Y. Chen, Prof. Z. Zhang
State Key Laboratory of Medicinal Chemical biology, Nankai
University, Tianjin 300071, China
X. He, Prof. Z. Jiang
School of Chemical Engineering and Technology, Tianjin University,
Tianjin 300072, China
2
highest power output (maximum 24 mW/cm ) among all reported
40
COFs. Currently, there are two facile strategies to fabricate
proton conductive COF materials: i) fabricating intrinsic proton
conductive COFs via covalently incorporating proton donor
groups such as sulfonate, imidazole and phenolic hydroxyl groups
Prof. P. Cheng, Prof. Z. Zhang
Key Laboratory of Advanced Energy Materials Chemistry, Ministry of
Education, Nankai University, Tianjin 300071, China
Y. H. Andaloussi
Department of Chemical Sciences, Bernal Institute, University of
Limerick, Limerick V94 T9PX, Republic of Ireland
Prof. H. Zhang
Suzhou institute of Nano-Tech and Nano-Bionics, Chinese Academy
of Sciences, Suzhou 215123, China
Prof. S. Ma
41
into COF skeletons; ii) endowing COFs with extrinsic proton
3 4
conductivity via doping proton carriers such as H PO into COF
pores.42-46 For this strategy, COFs should possess functional
groups with high binding affinity to the proton carriers to prevent
–
leakage. For example, azo (N=N) groups can bind with H
anion from H PO
2 4
PO
3
2,47
4
via hydrogen bonds to facilitate proton
To maximize the conductivity performance of
conduction.4
g]
Department of Chemistry, University of South Florida, 4202 East
Fowler Avenue, Tampa, Florida 33620, United States
COFs, a feasible approach is to combine the above two strategies
together to fabricate COFs with both intrinsic and extrinsic proton
conductivity in one system.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
Figure 1. a) Illustration of the synthetic route of NKCOFs. b) The hexagonal structural of NKCOFs. c-f) The side views of NKCOF-1, -2, -3, and -4, respectively,
resulting from the eclipsed AA stacking.
In this study, we created a stepwise synthesis strategy to create
COFs with multiple bond linkages and functional groups that are
not easily produced via the traditional one-step synthesis strategy
Scheme 1. Illustration of traditional one-step synthesis strategy of COFs with
single bond linkage and our synthesis strategy to fabricate multifunctional COFs
with multiple bond linkages.
(
Scheme 1). A series of highly porous and robust COFs (NKCOF-
, -2, -3, -4, NKCOF= Nankai Covalent Organic Framework) with
1
an abundance of proton accepting and donating groups were
successfully synthesized via this stepwise synthesis strategy. The
resultant NKCOFs possess azo groups that allow the doping of
Results and Discussion
To demonstrate the strategy to design proton conductive COFs,
a series of NKCOFs with an abundance of azo and phenolic
hydroxyl groups were synthesized (Figure 1a). Taking NKCOF-1
as a representative example to explain the synthesis process, a
3 4
H PO , for extrinsic proton conductivity, while phenolic hydroxyl
groups present in NKCOFs serve as acids to directly donate
protons for intrinsic proton conductivity. Meanwhile, the azo
groups and phenolic hydroxyl groups endow NKCOFs with high
hydrophilicity, which promotes the absorption of water molecules
into COF pores and thereby facilitates the proton conducting
process.
3
C -symmetric precursor of Azo-NHBoc with three azo groups and
three phenolic hydroxyl groups was first synthesized from a
reaction of N-Boc-p-phenylenediamine with phloroglucinol under
mild conditions (yield: ~80%) (Figure S1-S3). Subsequently, Azo-
NHBoc monomer was reacted with 1,3,5-triformylphloroglucinol
(
TP) in a mixed solvent of mesitylene, 1,4-dioxane, 6 M acetic acid
and trifluoroacetic acid for 3 days at 120 °C. A black powder of
NKCOF-1 was harvested in a yield of 85%. Adopting identical
synthetic procedures and replacing TP with other aldehydes such
as
2,5-dihydroxyterephthalaldehyde
(DPA),
1,3,5-tri(4-
formylphenyl)benzene (TFB), and 1,3,5-tri(3-hydroxy-4-formyl-
ethynylphenyl)benzene (THEB) (Figure S4-S6) afforded NKCOF-
2, -3, and -4 (yield: 84%, 90%, 87%), respectively (Figure S7). It
is notable that this is the first example to directly react -NHBOC
with -CHO groups to synthesize new COFs. This synthetic
method was verified by synthesis of a model compound by
reaction of Azo-NHBoc with 2-hydroxybenzaldehyde (Figure S8-
S10).
Attributed to this stepwise synthesis strategy, various bond
linkages and functional groups can be introduced into the skeleton
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
of afforded COFs, which may otherwise be difficult to produce.
The formation of NKCOFs were confirmed by Fourier-transform
sized spheres comprised of nanometer-sized crystalline grains
(Figure 4b and S13). The high-resolution transmission electron
microscopy (HR-TEM) of NKCOFs showed distinct lattice fringes
(Figure 4c-4f and S14), indicative of the high crystallinity and 2-
dimensional layer structures of NKCOFs.
1
3
infrared (FT-IR) spectroscopy and solid-state
polarization magic angle spinning nuclear magnetic resonance
CP-MAS NMR). Compared with the corresponding aldehydes
C cross-
(
and Azo-NHBoc monomer, FT-IR spectra of the NKCOFs
showed the disappearance of characteristic absorption peaks of
-1
aldehyde C=O (1634-1682 cm ) stretching bands, and the
-
1
-1
disappearance of N-H (3306 cm ), C=O (1704 cm ) and C-H
(
-1
-1
2978 cm , 2931 cm ) stretching bands from Azo-NHBoc (Figure
2
a and S11). The appearance of characteristic signals for the two
-
distinct bond linkages of C=N (stretching bands at 1614-1687 cm
1)
cm ) finally confirmed the successful construction of the NKCOFs.
In addition, NKCOF-1 showed peaks at 1575 cm-1 and 1660 cm-1
assigned to the stretching bands of C=C and C=O of the enol-keto
tautomer form, respectively. NKCOF-4 showed a peak at 2158
-1
and N=N (stretching bands at 1453-1473 cm , and 1397-1402
-1
cm-1 assigned to stretching bands of the C≡C triple bond. The 13
CP-MAS NMR spectra further confirmed the formation of COFs
Figure 2b and S12). All NKCOFs possess a peak between 145-
65 ppm assigned to the carbon of the imine bonds. Additionally,
C
(
1
keto-enol tautomerism was observed in NKCOF-1 with the peaks
at 184 ppm assigned to the C=O (keto form), and 147 ppm, 102
ppm assigned to the C=C (enol form). NKCOF-4 showed a
characteristic peak at 91 ppm assigned to the C≡C triple bond.
Figure 3. a) PXRD patterns of NKCOF-1, and its structure. b) PXRD patterns
of NKCOF-2, and its structure. c) PXRD patterns of NKCOF-3, and its structure.
d) PXRD patterns of NKCOF-4, and its structure.
Figure 2. a) FT-IR spectra of NKCOF-1 compared with reactants. b) 13C CP-
MAS NMR spectra of NKCOF-1.
The crystalline and porous structure of NKCOFs were verified
by powder X-ray diffraction (PXRD) analysis, electron microscopy
images and nitrogen adsorption-desorption isotherms. As shown
in Figure 3a, the PXRD pattern of NKCOF-1 exhibits a major peak
at 4.56° with minor peaks at 8.34° and 27.34°, assigned to 100,
110 and 001 faces, respectively. The observed PXRD pattern of
NKCOF-2 exhibited an intense peak at 2.32° along with peaks at
3.98°, 4.80°, 6.26°, 14.86° and 27.22° which were assigned to the
100, 110, 200, 210, 220 and 001 reflections (Figure 3b). NKCOF-
3 showed a very intense peak at 3.37° and five peaks at 5.95°,
6.83°, 9.15°, 12.41° and 25.43° corresponding to the 100, 110,
200, 210, 220 and 001 reflections, respectively (Figure 3c).
NKCOF-4 showed five peaks at 3.14°, 5.46°, 6.30°, 8.34° and
7.04°, which were assigned to the 100, 110, 200, 210 and 001
reflections (Figure 3d). All experimental PXRD patterns of
NKCOFs reflected good crystallinity and were consistent with the
simulated patterns acquired from structural simulations according
to an AA eclipsed layer stacking. Scanning electron microscopy
2
Figure 4. a) SEM image of NKCOF-1. b) SEM image of NKCOF-2. c) HR-TEM
image of NKCOF-1. d) Partial enlarged detail of c). e) HR-TEM image of
NKCOF-2. f) Partial enlarged detail of e).
The permanent porosity of as-synthesized NKCOFs were
(
SEM) images revealed that NKCOF-1 exhibited a distinctive
assessed by N
in Figure 5a, all NKCOFs showed rapid N
comparatively low pressure range (P/P < 0.05). NKCOF-1
2
adsorption isotherms collected at 77 K. As shown
morphology like nanowires about 200 nm in diameter (Figure 4a).
The morphology of NKCOF-2, -3 and -4 presented as micrometer-
2
uptake at a
0
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
exhibited the type-I isotherm, indicative of its microporous nature.
The sharp increase of adsorption for NKCOF-1 under high
pressure indicated the existence of macroporosity, likely
were pressed into thin pellets which exhibited good mechanical
stability without breaking or fragmentation during testing. In
anhydrous condition, the Nyquist plots of NKCOFs at 293 K and
353 K showed relatively low proton conductivity (Figure S24).
However, once in hydrous condition, a dramatic increase of
proton conductivity was observed. The proton conductivity was
measured as 7.08×10 S/cm, 3.09×10 S/cm, 1.20×10 S/cm,
and 5.43×10-3 S/cm for NKCOF-1, -2, -3, -4, respectively, at 353
K under 98% RH (Figure 6a, S25 and Table S1). This water
promotion phenomenon is consistent with literature which states
that water molecules play a crucial role in the proton conduction
process.42,45,49,50 Moreover, phenolic hydroxyl groups in NKCOFs
are able to provide protons in the proton conduction process. In
order to verify the contribution of the phenolic hydroxyl groups to
proton conductivity, TpPa-1,51 an analogue of NKCOF-1, was
selected as comparison (Figure S26). TpPa-1 was synthesized
4
8
generated by particle packing. All other NKCOFs exhibited the
typical type-IV isotherms, indicative of their mesoporous
structures. The Langmuir specific surface areas were calculated
2
2
2
2
-3
-3
-4
to be 1011 m /g, 1510 m /g, 1139 m /g and 2612 m /g for
NKCOF-1, -2, -3, -4, respectively (Figure S15-S18). Pore size
distribution calculated using density functional theory (DFT)
revealed that NKCOF-1, -2, -3, -4 possessed pore widths of 1.8,
3.3, 2.4, and 2.8 nm, respectively, which are in good agreement
with the pore size established from the structural analysis and
simulations (Figure 5b). The high surface areas and highly
ordered nanopores make NKCOFs excellent platforms for loading
and transferring substances.
via
a reported Schiff base reaction between TP and p-
phenylenediamine (PA) using solvothermal synthesis. Due to the
keto-enol tautomerism, phenolic hydroxyl groups mostly exist as
the keto form in TpPa-1. As such, TpPa-1 possesses fewer
phenolic hydroxyl groups than NKCOFs. As a result, the proton
conductivity of TpPa-1 (9.10×10-5 S/cm) was much lower than
NKCOFs at 353 K under 98% RH (Figure S27 and Table S1) and
as much as two orders of magnitude lower than the analogous
NKCOF-1. These results highlight the increased proton
conduction of NKCOFs due to intrinsic conductivity.
2
Figure 5. a) N adsorption (solid symbols) and desorption (open symbols)
isotherms of NKCOFs at 77 K. b) Pore size distribution of NKCOFs. c) PXRD
patterns of pristine NKCOF-3 (red) and after treatment for 2 days in boiling water
3 4 2
(orange), 12 M HCl (violet), 12 M H PO (cyan), and 12 M NaOH (purple). d) N
adsorption (solid symbols) and desorption (open symbols) isotherms of
NKCOF-3 (red) and after treatment for 2 days in boiling water (orange), 12 M
HCl (violet), 12 M H PO (cyan), and 12 M NaOH (purple).
3 4
Thermogravimetric analysis (TGA) was performed to determine
thermal stability and confirm the absence of guest molecules
inside the pores. The TGA curves of activated NKCOFs revealed
that these COFs are thermal stability up to around 260 °C (Figure
S19). The chemical stability of NKCOFs were examined by
exposing NKCOFs to various harsh conditions. It was found that
NKCOFs demonstrated excellent stability in acid, base, boiling
water and various organic solvents. Notably, NKCOF-1, -3 and -
Figure 6. a) Nyquist plots of NKCOF-1 measured under 98 % RH at different
temperatures. b) Nyquist plots of H
at different temperatures. c) Arrhenius plots for NKCOF-1 (blue) and
PO @NKCOF-1 (red). d) Proton conductivity of H PO @NKCOF-1 measured
at 313K under different relative humidity.
3 4
PO @NKCOF-1 measured under 98 % RH
H
3
4
3
4
4
(
N
were stable in concentrated HCl (12 M), concentrated H
12 M) and concentrated NaOH (12 M) as verified by PXRD and
sorption data (Figure 5c, 5d and S20-S23). The high structure
3 4
PO
As reported in literature, azo groups exhibit a binding affinity to
42,47
2
proton carriers such as H PO .
3
4
Thus, in order to further
improve the proton conductivity of NKCOFs, a feasible approach
is to incorporate H PO as proton carriers due to its high proton
content, low volatility (>158 °C), high mobility and conductivity.
After simply immersing NKCOFs into 5 M H PO for 12 h and then
robustness and chemical stability of NKCOFs provide a powerful
guarantee for their practical application.
3
4
5
2
NKCOFs possess a high density of azo and phenolic hydroxyl
groups. These functional groups make NKCOFs intrinsic proton
conductive materials. Electrochemical impedance spectroscopy
3
4
washing thoroughly with water until the eluent reaches pH = 7, the
(
EIS) was applied to measure the proton conductivities of
final loading of phosphoric acid calculated via TGA were 8.1 wt %,
NKCOFs in both hydrous and anhydrous conditions. NKCOFs
2.0 wt %, 4.7 wt %, and 4.0 wt % for H PO4@NKCOF-1, -2, -3, -
3
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
4
, respectively (Figure S28-S31). Energy Dispersive X-Ray (EDX)
spectroscopy mapping experiment further confirmed the
existence of H PO in COFs as indicated by the existence of well-
dispersed P element (Figure S32-S35). PXRD data revealed that
NKCOFs loaded with H PO (H PO @NKCOFs) retained their
crystallinity (Figure S36) attributed to the high acid resistance of
NKCOFs. N sorption data revealed that H PO @NKCOFs
showed reduced surface areas, due to the successful
incorporation of H PO into NKCOFs’ pores (Figure S37-S40).
Additionally, SEM images showed that H PO @NKCOFs retained
networks of hydrogen bonds with functional groups and additives
(e.g. azo, phenolic hydroxyl and H
3 4 3 4
PO ) in H PO @NKCOFs to
3
4
provide a highway for proton hopping transfer.
3
4
3
4
2
3
4
3
4
3
4
the particle morphology as the pristine NKCOFs (Figure S41).
EIS results revealed that the proton conductivity of
H
3
PO
4
@NKCOFs were negligible under anhydrous conditions
-2
-2
(
Figure S42). However, values up to 4.69×10 S/cm, 1.12×10
-4
-2
S/cm, 2.06×10 S/cm, and 5.47×10 S/cm were measured for
PO @NKCOF-1, -2, -3, -4, respectively, at 293 K under 98 %
RH. When increasing the testing temperature, we found that the
proton conductivity of H PO @NKCOFs increased (Figure 6b,
S43-S45 and Table S2) up to 1.13×10 S/cm, 4.28×10 S /cm,
.12×10-2 S/cm, and 7.71×10-2 S/cm at 353 K under 98% RH.
Notably, H PO
@NKCOF-1 sets up a new record (1.13×10-1
S/cm) for all reported COF materials, which exceeds the current
H
3
4
3
4
-1
-2
1
3
4
Figure 7. a) Scheme of the PEM fuel cells using H
electrolyte membranes of MEA. b) Lifetime for OCV of H
Fuel cell polarization curves (navy) and power density curves (red) of
3
PO
4
@NKCOFs as solid
3
PO @NKCOF-1. c)
4
-2
record (7.8×10 S/cm for PTSA@TpAzo at 353 K under 95%
RH)40 and even comparable to the commercial Nafion (~1×10-1
H PO @NKCOF-1 measured at 60 °C using single H /O cell assembly under
3
4
2
2
S/cm at 353 K under 98% RH)12,53 (Table S3). In addition, when
100% RH. d) Cyclic stability of OCV (navy) and power density (red) of
PO @NKCOF-1.
H
3
4
keeping H
days at 323 K under 98% RH, proton conductivity showed no
reduction, implying no leakage of H PO and high stability. Pellet
3 4
PO @NKCOF-1 under continuous assessment for two
3
4
Encouraged by the high
PO @NKCOFs, we further investigated their performances as
solid electrolyte membranes for PEM fuel cell under H /O
operating conditions. Thin pellets were made of H PO @NKCOFs
for use as the electrolyte membranes. Carbon papers coated with
0 wt % Pt/C catalyst were used as the anode and the cathode
proton conductivity of
stability was further demonstrated by soaking in water for 24 h
which showed no evidence of the powder dispersing (Figure S46).
Furthermore, TGA data revealed that the loading of phosphoric
acid was maintained before and after EIS measurements (Figure
S47), further indicating the strong binding of phosphoric acid in
NKCOFs. The stability of all COF materials after EIS
measurements was verified by the PXRD and N sorption data,
2
indicative of their intact structures (Figure S48-S51).
In order to investigate the proton conduction mechanism,
NKCOF-1 and H PO @NKCOF-1 were examined further. There
3 4
are two main mechanisms known for proton transport: the
Grotthuss and the vehicular mechanisms, which can be identified
primarily by activation energy (vehicular mechanisms >0.4 eV;
Grotthuss mechanisms <0.4 eV).22,50 As shown in Figure 5c, the
a
3 4
activation energy (E ) of NKCOF-1 and H PO @NKCOF-1
calculated by Arrhenius plots were 0.24 eV and 0.14 eV,
H
3
4
2
2
3
4
6
electrodes, and the Pt loading was kept as 0.5 mg/cm on each
electrode. All of above were assembled into membrane electrode
assemblies (MEAs) (Figure S56). Operating temperature was set
at 60 °C and the flow rate of humid H
mL/min. The afforded MEAs exhibited open circuit voltages
OCVs) of 0.978 V, 0.953 V, 0.928 V, and 0.906 V for
PO @NKCOF-1, -2, -3, -4, respectively. The high OCV of
MEAs indicated high proton conductivity of H PO @NKCOFs and
good mechanical properties of the pellets without H crossover.
Moreover, the OCV maintained stability during the testing period
2 2
and O were both 50
(
H
3
4
3
4
2
(
(
>4 hours) (Figure 7b and S57). As shown in Figure 7c
respectively (Figure 6c). Thus, both NKCOF-1 and
polarization curves), H
3
PO
4
@NKCOF-1 membrane achieved a
H
3
PO
protons “Hopping” along the hydrogen bond network. The
activation energy for PO @NKCOF-2, -3 and -4 were
4
@NKCOF-1 undergo the Grotthuss mechanism with
2
maximum power density of 81 mW/cm , much higher than other
reported crystalline porous organic framework materials, and a
H
3
4
2
maximum current density of 456 mA/cm . H
3
PO
4
@NKCOF-2, -3,
calculated to be 0.24 eV, 0.40 eV and 0.08 eV, respectively,
corresponding to Grotthuss transport mechanism. The activation
2
and -4 possessed maximum power density of 45 mW/cm , 24
2
2
mW/cm , and 56 mW/cm , respectively, and had maximum
3 4
energies of H PO @NKCOF-1 and -4 are notable for being lower
2
2
2
current density of 231 mA/cm , 133 mA/cm , and 269 mA/cm
Figure S58, table S4). The maximum power density and the
5
4
than that of Nafion (0.22 eV). In addition, we found that upon
increasing the relative humidity, the proton conductivity of
(
maximum current density were seen to positively correlate with
materials’ proton conductivity. The cycle stability of the MEAs with
H PO @NKCOFs membranes were examined, displaying that
3 4
the OCV and the maximum power density remained stable for
three cycles (Figure 7d). While powder-compressed pellets were
successful in forming a PEM membrane that prevented fuel
3 4
H PO @NKCOF-1 increased (Figure 6d and S52-S53), indicative
of the crucial role of water in the conductive process. To explore
the in-depth reason, hydrophobic angle measurements (Figure
S54) and water vapor adsorption (Figure S55) were tested and
revealed high hydrophilicity for both NKCOFs and
3 4
H PO @NKCOFs. Thus, water molecule could form infinite
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
crossover, industrial applications would benefit from a self-
standing membrane that may be continuously produced.
Therefore, we fabricated large-sized, self-standing, flexible COF
membranes (NKCOF-1-M, -2-M, -3-M, -4-M) (Figure S59-S60)
[4]
D. A. J. Rand, R. M. Dell, Angew. Chem. Int. Ed. 2008, 47, 5880-5881.
M. Z. Jacobson, W. G. Colella, Science 2005, 308, 1901-1905.
W. Gao, G. Wu, M. T. Janicke, D. A. Cullen, R. Mukundan, J. K. Baldwin,
E. L. Brosha, C. Galande, P. M. Ajayan, K. L. More, A. M. Dattelbaum,
P. Zelenay, Angew. Chem. Int. Ed. 2014, 53, 3588-3593.
[
[
5]
6]
via a method reported by the Banerjee’s group.40
loaded into the NKCOF-Ms via a similar process as with NKCOFs.
The afforded PO @NKCOF-Ms showed high proton
conductivity, of which H PO @NKCOF-1-M was the highest at
.41×10-2 S/cm (Figure S61 and Table S5). After assembling
3 4
H PO was
[
[
7] M. A. Hickner, H. Ghassemi, Y. S. Kim, B. R. Einsla, J. E. McGrath, Chem.
Rev. 2004, 104, 4587-4612.
H
3
4
8]
M. Sadakiyo, T. Yamada, H. Kitagawa, J. Am. Chem. Soc. 2009, 131,
9906-9907.
3
4
6
[9]
P. G. M. Mileo, K. Adil, L. Davis, A. Cadiau, Y. Belmabkhout, H. Aggarwal,
G. Maurin, M. Eddaoudi, S. Devautour-Vinot, J. Am. Chem. Soc. 2018,
140, 13156-13160.
H
3
PO
4
@NKCOF-1-M into an MEA, a maximum power density of
2
70 mW/cm (Figure S62) was achieved, slightly lower than that of
[
[
10] S. Hu, M. Lozada-Hidalgo, F. Wang, A. Mishchenko, F. Schedin, R. R.
Nair, E. W. Hill, D. W. Boukhvalov, M. I. Katsnelson, R. A. W. Dryfe, I. V.
Grigorieva, H. A. Wu, A. K. Geim, Nature 2014, 516, 227-230.
11] L. Mogg, G. Hao, S. Zhang, C. Bacaksiz, Y. Zou, S. J. Haigh, F. M. Peeters,
A. K. Geim, M. Lozada-Hidalgo, Nat. Nanotech. 2019, 14, 962–966.
12] K. A. Mauritz, R. B. Moore, Chem. Rev. 2004, 104, 4535-4585.
13] A. Kraytsberg, Y. Ein-Eli, Energy Fuels 2014, 28, 7303-7330.
14] H. Zhang, P. Shen, Chem. Rev. 2012, 112, 2780-2832.
15] R. F. Service, Science 2006, 312, 35.
3 4
H PO @NKCOF-1 (thin pellet).
Conclusion
[
[
[
[
[
In conclusion, we created a stepwise synthesis strategy to
prepare COFs with multiple bond linkages resulting in the
synthesis of a series of highly robust COFs (NKCOFs) with
abundant proton accepting and donating groups. NKCOFs not
only possessed high crystallinity and surface areas, but also
displayed outstanding stability in various harsh conditions such as
treating with organic solvents, boiling water, strong acid or base
16] M. Adamski, T. J. G. Skalski, B. Britton, T. J. Peckham, L. Metzler, S.
Holdcroft, Angew. Chem. Int. Ed. 2017, 56, 9058-9061.
[
17] J. Miyake, R. Taki, T. Mochizuki, R. Shimizu, R. Akiyama, M. Uchida, K.
Miyatake, Sci. Adv. 2017, 3, 1-8.
[18] Z. Zhou, S. Li, Y. Zhang, M. Liu, W. Li, J. Am. Chem. Soc. 2005, 127,
8024-18025.
1
(
e.g. 12 M HCl or NaOH). These structural features and
outstanding stability endow NKCOFs with high potential for proton
conduction application. Indeed, NKCOFs loaded with H PO
new proton conductivity record
1.13ꢀ×ꢀ10 ꢀS/cm at 80ꢀ°C under 98% RH) among all reported
[
19] N. Asano, M. Aoki, S. Suzuki, K. Miyatake, H. Uchida, M. Watanabe, J.
Am. Chem. Soc. 2006, 128, 1762-1769.
3
4
[
[
20] J. A. Mader, B. C. Benicewicz, Macromolecules 2010, 43, 6706-6715.
21] E. B. Trigg, T. W. Gaines, M. Maréchal, D. E. Moed, P. Rannou, K. B.
Wagener, M. J. Stevens, K. I. Winey, Nat. Meter. 2018, 17, 725-731.
22] X. Meng, H. Wang, S. Song, H. Zhang, Chem. Soc. Rev. 2017, 46, 464-
480.
(
(
3 4
H PO @NKCOFs) set a
-1
COFs. Finally, we successfully fabricated membrane electrode
assemblies and applied them in real PEM fuel cells. Notably,
[
H
3
PO
4
@NKCOF-1 set a new benchmark for all COF materials, a
[23] A. P. Cote, A. I. Benin, N. W. O’ckwig, M. O’Keeffe, A. J. Matzger, O. M.
Yaghi, Science 2005, 310, 1166-1170.
2
maximum power density of 81 mW/cm and a maximum current
density of 456 mA/cm . This work provides important guidance on
2
[24] S. Das, P. Heasman, T. Ben, S. Qiu, Chem. Rev. 2017, 117, 1515-1563.
[
[
[
[
25] C. S. Diercks, O. M. Yaghi, Science 2017, 355, 923-930.
26] R. P. Bisbey, W. R. Dichtel, ACS Cent. Sci. 2017, 3, 533-543.
27] M. S. Lohse, T. Bein, Adv. Funct. Mater. 2018, 28, 1705553.
28] X. Chen, K. Geng, R. Liu, K. Tan, Y. Gong, Z. Li, S. Tao, Q. Jiang, D.
Jiang, Angew. Chem. Int. Ed. 2019, DIO: 10.1002/anie.201904291.
designing multifunctional COFs, and the synthesis strategy
reported in this study is of broad scope, likely applicable to other
COF systems. Moreover, this study not only creates new records
for superprotonic conductivity, but also achieves presently the
best performance of PEM fuel cells for COFs.
[29] Q. Sun, Y. Tang, B. Aguila, S. Wang, F. Xiao, P. K. Thallapally, A. M. Al-
Enizi, A. Nafady, S. Ma, Angew. Chem. Int. Ed. 2019, 58, 1-7.
[
[
30] F. Beuerle, B. Gole, Angew. Chem. Int. Ed. 2018, 57, 4850-4878.
31] a) J. Zhang, X. Han, X. Wu, Y. Liu, Y. Cui, J. Am. Chem. Soc. 2017, 139,
8
277-8285. b) K. C. Ranjeesh, R. Illathvalappil, V. C. Wakchaure,
Goudappagouda, S. Kurungot, and S. S. Babu, ACS Appl. Energy Mater.
018, 1, 6442-6450.
Acknowledgements
2
The authors acknowledge the National Natural Science
Foundation of China (21971126).
[32] a) L. Wang, J. Zhou, Y. Lan, S. Ding, W. Yu, W. Wang, Angew. Chem.
Int. Ed. 2019, 58, 1-6. b) K. C. Ranjeesh, L. George, V. C. Wakchaure,
Goudappagouda, R. N. Devi, S. S. Babu, Chem. Commun., 2019, 55,
1
627-1630.
Dedication
[
33] S. Jiang, S. Gan, X. Zhang, H. Li, Q. Qi, F. Cui, J. Lu, X. Zhao, J. Am.
Chem. Soc. 2019, 141, 14981-14986.
Dedicated to the 100th anniversary of Nankai University
[34] a) Y. Hu, N. Dunlap, S. Wan, S. Lu, S. Huang, I. Sellinger, M. Ortiz, Y.
Jin, S. Lee, W. Zhang, J. Am. Chem. Soc. 2019, 141, 7518-7525. b) E.
Vitaku, W. R. Dichtel, J. Am. Chem. Soc. 2017, 139, 12911-12914.
Keywords: Covalent organic framework • multiple bond linkage
[
35] D. A. Pyles, W. H. Coldren, G. M. Eder, C. M. Hadad, P. L. McGrier, Chem.
Sci. 2018, 9, 6417-6423.
•
Membrane • Proton conduction • Fuel cell
[
36] M. Matsumoto, L. Valentino, G. M. Stiehl, H. B. Balch, A. R. Corcos, F.
Wang, D. C. Ralph, B. J. Marin, W. R. Dichtel, Chem. 2018, 4, 308-317.
[1]
[2]
[3]
N. Armaroli, V. Balzani, Angew. Chem. Int. Ed. 2007, 46, 52-66.
[37] (a) K. Dey, M. Pal, K. Rout, S. Kunjattu H, A. Das, R. Mukherjee, U. K.
Kharul, R. Banerjee, J. Am. Chem. Soc. 2017, 139, 13083-13091. (b) A.
Halder, M. Ghosh, M. A. Khayum, S. Bera, M. Addicoat, H. S. Sasmal, S.
W. Lubitz, W. Tumas, Chem. Rev. 2007, 107, 3900-3903.
M. J. Shultz, M. Kelly, L. Paritsky, J. Wagner, J. Chem. Educ. 2009, 86,
1051-1053.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
Karak, S. Kurungot, R. Banerjee, J. Am. Chem. Soc. 2018, 140, 10941-
10945.
[
[
38] Z. Wang, Q. Yu, Y. Huang, H. An, Y. Zhao, Y. Feng, X. Li, X. Shi, J. Liang,
F. Pan, P. Cheng, Y. Chen, S. Ma, Z. Zhang, ACS Cent. Sci. 2019, 5,
1352-1359.
39] (a) P. Shao, J. Li, F. Chen, L. Ma, Q. Li, M. Zhang, J. Zhou, A. Yin, X.
Feng, B. Wang, Angew. Chem. Int. Ed. 2018, 57, 16501-16505. (b) H.
Wang, Z. Zeng, P. Xu, L. Li, G. Zeng, R. Xiao, Z. Tang, D. Huang, L.
Tang, C. Lai, D. Jiang, Y. Liu, H. Yi, L. Qin, S. Ye, X. Rena, W. Tang,
Chem. Soc. Rev., 2019, 48, 488-516. (c) C. Zhang, B. Wu, M. Ma, Z.
Wang, Z. Xu, Chem. Soc. Rev., 2019, 48, 3811-3841.
[
[
40] H. S. Sasmal, H. B. Aiyappa, S. N. Bhange, S. Karak, A. Halder, S.
Kurungot, R. Banerjee, Angew. Chem. Int. Ed. 2018, 57, 10894-10898.
41] (a) Y. Peng, G. Xu, Z. Hu, Y. Cheng, C. Chi, D. Yuan, H. Cheng, D. Zhao,
ACS Appl. Mater. Interfaces. 2016, 8, 18505-18512. (b) K. Ranjeesh, R.
Illathvalappil, S. Veer, J. Peter, V. Wakchaure, Goudappagouda, V. Raj,
S. Kurungot, S. S. Babu, J. Am. Chem. Soc. 2019, 141, 14950-14954.
42] S. Chandra, T. Kundu, S. Kandambeth, R. BabaRao, Y. Marathe, S. M.
Kunjir, R. Banerjee, J. Am. Chem. Soc. 2014, 136, 6570-6573.
[
[
[
43]
H. Ma, B. Liu, B. Li, L. Zhang, Y. Li, H. Tan, H. Zang, G. Zhu, J. Am.
Chem. Soc. 2016, 138, 5897-5903.
44] C. Montoro, D. Rodríguez-San-Miguel, E. Polo, R. Escudero-Cid, M. Ruiz-
Gonzalez, J. A. R. Navarro, P. Ocon, F. Zamora, J. Am. Chem. Soc. 2017,
139, 10079-10086.
[
[
[
45] M. Zheng, A. Aylin, M. A. Katherine, Chem. Mater. 2019, 31, 819-825.
46] H. Xu, S. Tao, D. Jiang, Nat. Mater. 2016, 15, 722-727.
47] (a) L. Halasz, K. Lukic, H. Vancik, Acta Cryst. 2007, C63, o61-64. (b) A.
M. Sanchez, M. Barra, R. H. de Rossi, J. Org. Chem. 1999, 64, 1604-
1
609.
Z. Zhang, H. Nguyen, S. A. Miller, S. M. Cohen, Angew. Chem. Int. Ed.
015, 54, 6152-6157.
49] P. Ramaswamy, N. E. Wong, G. K. H. Shimizu, Chem. Soc. Rev. 2014,
3, 5913-5932.
50] G. Xing, T. Yan, S. Das, T. Ben, S. Qiu, Angew. Chem. Int. Ed. 2018, 57,
-6.
[
[
[
[
[
[
[
48]
2
4
1
51] S. Kandambeth, A. Mallick, B. Lukose, M. V. Mane, T. Heine, R. Banerjee,
J. Am. Chem. Soc. 2012, 134, 19524-19527.
52] L. Vilciauskas, M. E. Tuckerman, G. Bester, S. J. Paddison, K. D. Kreuer,
Nat. Chem. 2012, 4, 461-466.
53] W. Kong, W. Jia, R. Wang, Y. Gong, C. Wang, P. Wu, J. Guo, Chem.
Commun. 2019, 55, 75-78.
54] X. Wu, X. Wang, G. He, J. Benziger, J. Polym. Sci., Part B: Polym. Phys.
2011, 49, 1437-1445.
This article is protected by copyright. All rights reserved.

Angewandte Chemie International Edition
10.1002/anie.201913802
RESEARCH ARTICLE
RESEARCH ARTICLE
Herein, we created
a
stepwise
Yi Yang, Xueyi He, Penghui Zhang,
Yassin H. Andaloussi, Hailu Zhang,
Zhongyi Jiang, Yao Chen, Shengqian
Ma, Peng Cheng, and Zhenjie Zhang*
synthesis strategy to prepare a series
of highly robust COFs with ultrahigh
proton conduction. These COFs
obtained
conductivity (1.13 ×ꢀ10 ꢀS/cm) after
doping with PO and were
a
record-high proton
-1
Page No. 1– Page No. 7
H
3
4
,
Predisposed Intrinsic and Extrinsic
Proton Conduction in Robust
Covalent Organic Frameworks for
Hydrogen Fuel Cell Application
assembled MEAs to output
maximum power of 81 mW/cm.
a
This article is protected by copyright. All rights reserved.