Twinned Growth of Metal-Free, Triazine-Based Photocatalyst Films as Mixed-Dimensional (2D/3D) van der Waals Heterostructures

Article abstract of DOI:10.1002/adma.201703399

Design and synthesis of ordered, metal-free layered materials is intrinsically difficult due to the limitations of vapor deposition processes that are used in their making. Mixed-dimensional (2D/3D) metal-free van der Waals (vdW) heterostructures based on

Full text of DOI:10.1002/adma.201703399

CommuniCation
2
D Materials
www.advmat.de
Twinned Growth of Metal-Free, Triazine-Based
Photocatalyst Films as Mixed-Dimensional (2D/3D)
van der Waals Heterostructures
Dana Schwarz, Yu Noda, Jan Klouda, Karolina Schwarzová-Pecková, Ján Tarábek,
ˇ
Ji ˇr í Rybá cˇ ek, Ji rˇ í Janoušek, Frank Simon, Maksym V. Opanasenko, Ji ˇr í Cejka,
Amitava Acharjya, Johannes Schmidt, Sören Selve, Valentin Reiter-Scherer,
Nikolai Severin, Jürgen P. Rabe, Petra Ecorchard, Junjie He, Miroslav Polozij,
Petr Nachtigall, and Michael J. Bojdys*
The quest to make metal-free 2D-mate-
Design and synthesis of ordered, metal-free layered materials is intrinsi-
cally difficult due to the limitations of vapor deposition processes that are
used in their making. Mixed-dimensional (2D/3D) metal-free van der Waals
rials “beyond graphene” is a great chal-
lenge to chemical bond formation and
material design, and only very recently the
“graphene family” grew by one to a total
(
vdW) heterostructures based on triazine (C N ) linkers grow as large area,
3 3
of six: graphene, hexagonal boron nitride,
boron carbonitride, fluorographene, gra-
phene oxide, and triazine-based graphitic
transparent yellow-orange membranes on copper surfaces from solution. The
membranes have an indirect band gap (E = 1.91 eV, Eg,elec = 1.84 eV) and
g,opt
2
−1
are moderately porous (124 m g ). The material consists of a crystalline
[1]
carbon nitride (TGCN), a compound
2
2
D phase that is fully sp hybridized and provides structural stability, and an
highlighted as an emerging competitor
2
[2]
for the miracle material graphene. When
amorphous, porous phase with mixed sp –sp hybridization. Interestingly, this
it comes to device applications, the lack
of chemical variability of the “graphene
family” becomes limiting. From these six,
graphene is the only material with a zero
band gap, and all others are wide band gap
semiconductors—with the exception of
TGCN (direct band gap between 1.6 and
2
D/3D vdW heterostructure grows in a twinned mechanism from a one-pot
reaction mixture: unprecedented for metal-free frameworks and a direct con-
sequence of on-catalyst synthesis. Thanks to the efficient type I heterojunc-
tion, electron transfer processes are fundamentally improved and hence, the
material is capable of metal-free, light-induced hydrogen evolution from water
without the need for a noble metal cocatalyst (34 µmol h−
The results highlight that twinned growth mechanisms are observed in the
realm of “wet” chemistry, and that they can be used to fabricate otherwise
challenging 2D/3D vdW heterostructures with composite properties.
1
−1
g without Pt).
2
.0 eV).
The evidently narrow range of proper-
ties within the “graphene family” can be
enhanced by combination of distinct 2D
Dr. D. Schwarz, Dr. Y. Noda, Dr. M. J. Bojdys
Faculty of Science
J. Janoušek
Faculty of Chemical Engineering
Department of Organic Chemistry
Charles University
Hlavova 8, 128 43 Prague 2, Czech Republic
E-mail: m.j.bojdys.02@cantab.net
Department of Analytical Chemistry
University of Chemistry and Technology Prague
Technická 3, 160 00 Prague 6, Czech Republic
Dr. F. Simon
Dr. Y. Noda, Dr. M. J. Bojdys
Institute of Organic Chemistry and Biochemistry of the CAS
Flemingovo nám. 2, 166 10 Prague 6, Czech Republic
Leibniz Institute of Polymer Research Dresden
Hohe Str. 6, 01069 Dresden, Germany
ˇ
Dr. M. V. Opanasenko, Prof. J. Cejka
J. Klouda, Dr. K. Schwarzová-Pecková
Faculty of Science
J. Heyrovský Institute of Physical Chemistry
Academy of Sciences of the Czech Republic
Department of Analytical Chemistry
Charles University
Hlavova 8, 128 43 Prague 2, Czech Republic
v.v.i., Dolejškova 2155/3, 182 00 Prague 8, Czech Republic
Dr. A. Acharjya, Dr. J. Schmidt, S. Selve
Technische Universität Berlin
Dr. J. Tarábek, Dr. J. Rybá cˇ ek, J. Janoušek
Institute of Chemistry
Institute of Organic Chemistry and Biochemistry of the CAS
Flemingovo nám. 2, 166 10 Prague 6, Czech Republic
Hardenbergstr. 40, 10623 Berlin, Germany
DOI: 10.1002/adma.201703399
Adv. Mater. 2017, 1703399
1703399 (1 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
layers into vdW heterostructures with unique functionalities.[3]
Here, we are not limited to vdW interactions between pure 2D
materials. Any 2D layer can interface with a large selection of
materials capable of vdW bonding to form mixed-dimensional
heterostructures for example with molecules and nanoparti-
that are created in the early stages of the reaction (Figure 1c).
These Cu NPs provide more statistical nucleation sites for the
formation of the kinetic product, the 3D TzG.
The TzF/TzG vdW heterostructure replicates the surface
morphology of the copper foil it was deposited on, and it retains
this shape once all the copper has been etched away. Thanks to
their 2D component, these TzF/TzG films show excellent chem-
ical and physical stability. The 3D component is interesting as
a photocatalyst, since it features an open, porous structure and
an intriguing donor–acceptor (D–A) motif: electron rich buta-
1,3-diyene (D) and electron poor triazines (A)—both have been
identified as promising structural motifs in photocatalytic water
[4]
[5]
cles (2D/0D), with ribbons and tubes (2D/1D), and as bulk
heterojunctions (2D/3D).[ Most commonly, 2D materials and
their heterostructures are obtained from (successive) chemical
6]
[
7]
(
CVD) and physical vapor deposition (PVD) processes. This
has two serious drawbacks: (1) Nucleation events are statistic
and severely limit homogeneity and size of the crystalline
domains of the obtained materials.[
7a,8]
(2) Vapor deposition
[
11]
techniques are incompatible with the overwhelming majority
of organic synthons, and the resulting materials (if any) do not
feature any predictable bonding patterns. Any changes to the
chemical composition of thus obtained materials require com-
plex postsynthetic treatment.[
splitting.
TzF/TzG films are uniquely positioned as metal-
free photocatalysts for water reduction, since to our knowledge,
there are only two other examples of photocatalysts that were
[
12]
obtained as free-standing films —both are based on poly-
9]
[13]
meric carbon nitride (PCN).
Here, we show that the synthetic limits of CVD and PVD
techniques can be overcome in a facile one-pot, “wet” chemistry
approach. We demonstrate the twinned growth of a 2D/3D vdW
We believe that this facile preparation method will promote
“wet” chemistry as a feasible avenue toward mixed-dimensional
vdW heterostructures, and that it will lead to a re-examination
of a number of coupling reactions that did not yield 2D crys-
talline organic materials in conventional synthetic protocols in
which a homogeneous catalyst is employed.
2
heterostructure that consists of an sp hybridized, crystalline
2
D layer—covalent triazine-based framework (TzF)—and an
2
sp –sp hybridized 3D, amorphous component—triazine-based
graphdiyene (TzG). The synthon for both components of the
heterostructure is 2,4,6-tris(4-ethynylphenyl)-1,3,5-triazine. We
chose copper foil as 2D template and dual-action catalyst. The
TzF/TzG heterostructure grows via a twinned mechanism with
copper as the sole, cheap catalyst in the reaction, and the role of
copper shifts as the reaction progresses. Initially, copper cata-
lyzes the formation of the 2D TzF at the flat copper interface via
copper mediated [2 + 2 + 2] cyclotrimerization (Figure 1a). In
the later stages of the reaction, copper-catalyzed dimerization of
the terminal alkynes leads to 3D TzG (Figure 1b). This “dimen-
sional dependence” in catalytic action is surprising because
conventional wisdom predicts a simple Glaser-type coupling
of acetylenes in the presence of copper that would lead exclu-
sively to TzG.[ We propose that the high local concentration
of surface-adsorbed synthons facilitates a multicomponent,
The synthon, 2,4,6-tris(4-ethynylphenyl)-1,3,5-triazine, was
obtained as described in the Supporting Information. In brief,
the triazine ring was formed by the treatment of 4-bromoben-
[
14]
zonitrile with trifluoromethanesulfonic acid. The trimethyls-
ilyl-protected acetylene group was attached to the molecule by
a Negishi coupling reaction. Finally, the molecule was depro-
tected using K CO in tetrahydrofuran/methanol before the on-
2
3
surface synthesis. The TzF/TzG film was grown on a common
copper foil using pyridine as a base and as solvent. Figure 1
shows the suggested reaction scheme for the twinned growth of
TzF and TzG. Synthons have to adsorb from solution to the cat-
alytically active surface in order to react. This increases the local
concentration of synthons at the pyridine/copper interface and
makes a three-component reaction far more likely than in solu-
tion. A feasible mechanism for the [2 + 2 + 2] cyclotrimerization
involves metalation (or metal coordination) of terminal ethynyl
groups and a stepwise insertion pathway that leads to extru-
sion of elemental copper (Figure 1a). Analogous mechanisms
have been suggested in the past for a variety of charged metal
10]
[
2 + 2 + 2] cyclotrimerization. In addition, the templating effect
of the catalytically active 2D surface leads to the formation of
thermodynamically more stable, conjugated 2D TzF. The kineti-
cally uninhibited growth of the 3D layer in later stages of the
reaction is promoted by nascent copper nanoparticles (Cu NPs)
[
15]
catalysts like Ni, and Cu/Zr catalysts in solution, yet not for
Cu(0). There is, however, evidence of the [2 + 2 + 2] cyclotri-
[
16]
merization of acetylene from the gas phase on Cu(111) and
[
17]
V. Reiter-Scherer, Dr. N. Severin, Prof. J. P. Rabe
Humboldt-Universität zu Berlin
Department of Physics
Physics of Macromolecules
Newtonstr. 15, 12489 Berlin, Germany
on Cu(110). Experimental and theoretical calculations show
that [2 + 2 + 2] cyclotrimerization of acetylenes proceeds via
two exothermic single-step reactions under such conditions.
An alternative explanation is that leached palladium from an
earlier Negishi-type coupling may assist such a cyclotrimeri-
Dr. P. Ecorchard
Materials Chemistry Department
[
18]
zation (Figure S1, Supporting Information).
However, pal-
Institute of Inorganic Chemistry AS CR
ladium concentrations in synthons and in TzF/TzG films are
below the detection limit of all employed analytical techniques
v.v.i., 250 68 Rˇ ež, Czech Republic
Dr. J. He, M. Polozij, Prof. P. Nachtigall
Faculty of Science
Department of Physical and Macromolecular Chemistry
Charles University
(
(
(
energy-dispersive x-ray spectroscopy (EDX), elemental analysis
EA), inductively coupled plasma atomic emission spectroscopy
ICP-OES), X-ray photoelectron spectroscopy (XPS); Figure S3
Hlavova 8, 128 43 Prague 2, Czech Republic
and Tables S2 and S3, Supporting Information, and Figure 2h),
and the formation of Cu NPs over the course of the reaction is
a strong indicator that copper species have an active part in the
The ORCID identification number(s) for the author(s) of this article
can be found under https://doi.org/10.1002/adma.201703399.
Adv. Mater. 2017, 1703399
1703399 (2 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 1. Reaction scheme for the twinned growth of the TzF/TzG vdW heterostructure on copper foil. a) Synthons (2,4,6-tris(4-ethynylphenyl)-1,3,5-
[
10]
triazine) are coadsorbed on the copper catalyst. Two alkenyl-copper groups react with a terminal ethynyl group (under possible transmetalation).
Reductive elimination of copper metal affords aryl bridged 2D TzF and free copper atoms that agglomerate to copper nanoparticles (Cu NPs) over the
course of the reaction. b) Synthons undergo a Glaser-type oxidative coupling reaction via a copper(I)–acetylide intermediate.[ This reaction yields the
buta-1,3-diyne bridged 3D TzG. c) Initially, nucleation and growth of 2D TzF happens at the copper/liquid interface. Individual Cu atoms are knocked
out of the surface and agglomerate to copper nanoparticles (NPs). In later stages of the reaction, nucleation happens at the dispersed Cu NPs and
leads to the growth of the kinetic product, 3D TzG.
11]
Adv. Mater. 2017, 1703399
1703399 (3 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 2. Composition and morphology of TzF/TzG vdW heterostructures. a) Schematic illustrating postsynthetic processing of TzF-TzG films grown
on copper foil. Photographs of b) TzF/TzG films on copper foil, c) delamination in H PO , d) free-standing TzF/TzG films, and e) exfoliated TzF films
3
4
at the CH Cl /H O interface after ultrasonication. f) Scanning electron microscope (SEM) image of TzF/TzG heterostructures on copper foil shows the
2
2
2
transition from TzF layers to TzG polymer as film thickness increases. g) TEM image of an exfoliated TzF flake. The inset shows the extent of single-
crystalline domains. h) XPS data for TzF/TzG heterostructures before and after removal of Cu NPs. i) 13C CP-MAS ssNMR of TzF/TzG heterostructures
and TzF films. The triazine carbon signal (marked i) appears at 170 ppm, and the quaternary diyene carbons (marked iv) at ≈75 ppm. Phenyl peaks are
marked as ii and iii. Spinning side-bands are denoted with an asterisk (*). j) FT-IR spectra of TzF/TzG heterostructures and TzF films. Terminal unreacted
C CH bonds are visible at 1, while the characteristic triazine breathing mode is seen at 2. k) Raman spectra of TzF/TzG heterostructures at 532 nm
−
1
are dominated by background fluorescence, yet the characteristic diyene stretch can be discerned at ≈2200 cm (marked as 3).
formation of 2D TzF (Figure S1, Supporting Information, and
Figure 2). 2D TzF grows in an ordered, surface-templated mor-
phology for a number of layers (Figure 2f,g), and polymeric,
disordered aggregates are only observed for thicker films
(Figure 2f and Figures S4 and S5, Supporting Information).
Continued growth of TzF passivates the catalyst surface, and
Adv. Mater. 2017, 1703399
1703399 (4 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
the dispersed Cu NPs become the dominant sites of nucleation
and growth (Figure 1c). Now, however, the reaction is mainly
diffusion controlled: the knocked-out copper atoms and Cu NPs
are fully solvated by pyridine, and multicomponent reactions
are statistically less likely to occur in solution than on a sur-
the low scattering contrast of the (mostly disordered) 2D–3D
heterostructures and turbostratic disorder known for other lay-
ered system.
[
1,25]
Scanning force microscopy (SFM) measurements on exfoli-
ated TzF/TzG heterostructures confirm a hexagonal unit-cell
(Figure 3d and Figure S11 and Table S6, Supporting Informa-
tion). Attribution of the distances observed via SFM to the ideal
bonding pattern are not trivial, though, due to the possibility of
guest inclusion and surface restructuring at ambient imaging
conditions (Supporting Information). Hence, we employ high-
resolution transmission electron microscopy (HR-TEM) and
selected area electron diffraction (SAED) to investigate the crys-
talline structure of 2D TzF. We obtain neat 2D TzF either from
ultrasound treatment (Figure 1a), or via mechanical cleaving
of TzF/TzG heterostructures with dicing tape (Supporting
Information).
High-resolution TEM images feature a clearly visible, hexag-
onal bonding motif (Figure 3a and Figure S5, Supporting Infor-
mation). The corresponding SAED image along the (00l) direc-
tion of the multilayered TzF material (Figure 2b and Figure S8,
Supporting Information) corresponds to a hexagonally ordered
structure with crystalline domain sizes up to 1 µm (Figure 2g,
inset). SAED images calculated for the density functional theory
(DFT) optimized, multilayer TzF structure find the best match
for AB-stacked 2D TzF with unit-cell parameters of a = 14.750 Å,
b = 14.776 Å, and α = 120.172° (Figure 3c,e and Figure S8, Sup-
porting Information), which is in excellent agreement with the
2D hexagonal lattice parameters reported for structurally analo-
gous covalent triazine-based frameworks (CTFs).
face. Oxidative coupling via a copper(I)–acetylide intermediate
becomes the preferred reaction pathway,[
10a]
and the resulting
product is buta-1,3-diyne bridged 3D TzG (Figure 1b). Vertical
growth of the TzF/TzG films is capped at ≈6 µm—a thickness
that is reached and retained after 48 h (Figure S4d, Supporting
Information). It is likely that this maximum thickness corre-
lates with the diffusion limit of the Cu NPs. In principle, the
backbone of TzG is very rigid, and one would expect that the
2
strongly directing sp –sp bonding motif leads to an ordered,
2
D product. However, long range order is not observed for any
porous networks that were obtained from bulk, homogeneous
reactions,[ and interdigitation of these polymers may occur if
pore sizes are sufficiently large.[
19]
20]
TzF/TzG heterostructures can be easily detached from
the copper support for further analysis and processing using
a gentle wash with phosphoric acid (1 m), yet removal of the
embedded Cu NPs needs a more thorough treatment with con-
centrated nitric acid (Table S1, Supporting Information, and
Figure 2a–d). XPS shows that the Cu NPs are indeed either
Cu metal or Cu(I) oxides (Figure 2h) as we see no evidence of
Cu(II) satellite peaks at ≈945 eV.[ Residual quantities of para-
magnetic Cu(II) species (≈0.6 nmol per 1 mg of TzF/TzG) can
however be detected using electron paramagnetic resonance
21]
(EPR) (Supporting Information). The chemical and structural
composition of TzF/TzG stays intact during this etching pro-
cess as confirmed by 13C crosspolarization magic angle spin-
ning (CP-MAS) solid-state NMR (ssNMR) (Figure 2i). Both, the
triazine carbon (≈170 ppm) and the diyene (75–85 ppm) signals
remain unchanged.
The diffuse reflectance UV–vis spectrum shows that TzF/
TzG flakes have a discernible absorption edge at ≈560 nm
(Figure 4a). That corresponds to a direct optical band gap of
2.24 eV or an indirect optical band gap of 1.91 eV according
to the Kubelka–Munk function (Figure 4b). Photoluminescence
(PL) spectroscopy was used to further investigate the interac-
tion of TzF/TzG heterostructures. There are two overlapping
peaks observed at excitation transition energies of 2.05 and
2.34 eV, corresponding to wavelengths of 605 and 530 nm,
respectively (Figure 4b). The peak at 2.34 eV (530 nm) is attrib-
uted to the exciton in 2D TzF and is in good agreement with PL
We are able to separate the TzF/TzG heterostructures using
high powered ultrasonication (max. 980 W) in a medium of
[22]
CH Cl and water (Figure 2e). Flakes of TzF are then collected
2
2
at the CH Cl /water interface, and they show highly crystalline,
2
2
hexagonal domains on the order of micrometers (Figure 2g).
Fourier transform infrared (FT-IR) spectra of TzF/TzG hetero-
structures show the characteristic triazine breathing mode at
[
25]
profiles of structurally analogous CTF materials (Figure 4).
The stronger peak at 2.05 eV (605 nm) is attributed to the 3D
TzG phase.
−
1
8
05 cm and evidence of terminal, unreacted C CH bonds
−1
at ≈3300 cm that originate from the 3D TzG phase (Figure 2k)
Figure S6, Supporting Information). Complementary Raman
(
We investigate the electronic band gap of TzF/TzG using
cyclic voltammetry (CV) (Figure 4d and Figure S12, Supporting
Information). For bulk heterojunctions of organic semiconduc-
tors the band gap energy corresponds to the energy difference
between the highest occupied molecular orbital (HOMO) and
spectra of TzF/TzG films are dominated by a fluorescent back-
ground, yet they show clear evidence of C C stretching in an
2
−1
[23]
sp –sp hybridized network at ≈2200 cm (Figure 2j).
Powder X-ray diffraction (PXRD) profiles of TzF/TzG hetero-
structures show a diagnostic, low-angle peak that corresponds
to the (100) in-plane reflection of a hexagonal unit cell with 2D
lattice parameters a = b = 14.7 Å (Figure 3e,f). This is in agree-
ment with the unit-cell parameters of structurally analogous
[
26]
lowest unoccupied molecular orbital (LUMO).
The electro-
chemically obtained band gap value is 1.84 eV, which is in good
correlation with the optical band gap of 1.91–2.24 eV obtained
from the UV–vis absorption edge and with the excitation transi-
tion energies observed via PL spectroscopy (2.05 and 2.34 eV).
DFT calculations predict (1) a band gap of 1.95 eV for 3D TzG
and 2.54 eV for 2D TzF (Figure 4e) and (2) that the band gap
of 3D TzG is completely contained in the band gap of 2D TzF,
and the calculations agree well with the profile of the valence
band region observed in XPS (Figure S10, Supporting Informa-
tion). The optical and electronic band gap that we observe for
[
24]
covalent triazine-based networks, CTF-1 (a = b = 14.547 Å),
and P1M and P4M (a = b = 14.574 Å).[ Stacking distances
of individual TzF layers are in the range of 3.4–3.6 Å as seen
from the broad peak around 25°–27° 2θ (Figure 3e,g), which
is in good agreement with layered aromatic systems such as
graphite (3.35 Å). However, the PXRD peaks are too broad and
weak to deduce further structural information. This is due to
25]
Adv. Mater. 2017, 1703399
1703399 (5 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 3. Structure analysis of the crystalline 2D TzF component of TzF/TzG heterostructures. a) High-resolution TEM image of 2D TzF reveals a
hexagonal bonding motif and b) the corresponding electron diffraction pattern indicating prominent diffraction peaks with red, dotted circles as guide
for the eye. c) Simulated electron diffraction pattern of AB-stacked TzF shows the same set of diffraction peaks (indicated with red, dotted circles).
d) Representative SFM friction image of freshly cleaved TzF layers and the corresponding Fourier transformation (inset) shows a hexagonal structure
with parameters a = b = 0.54 ± 0.01 Å (range of 0.50–0.60 Å), and α = 60° ± 2°. e) The powder X-ray diffraction pattern of as-synthesized TzF/TzG is
shown in black. Simulated diffraction patterns for the ideal, hexagonal unit-cell of 2D TzF structure with AB- (a = b = 14.574 Å, c = 6.917 Å, space-group
P21/M in blue) and AA-stacking (a = b = 14.574 Å, c = 3.459 Å, space-group P-6M2 in green) are displayed below. The copper (111) peak is indicated
by an asterisk (*). f) The AB packing motif of 2D TzF (with the lower-lying layer indicated in yellow) is a stable structure obtained at DFT level. The
in-plane unit-cell with parameters a = 14.750 Å, b = 14.776 Å, and α = 120.172 is indicated as a black, dotted line with the characteristic (100) repeat
marked in red. g) Stacking between layers of 2D TzF gives rise to the characteristic (002) repeat marked in red.
[
13]
the TzF/TzG heterostructure is closer to the band gap energy of
D TzG. This is indicative of a type I band alignment between
and without platinum (Pt) cocatalyst. The rate of hydrogen
−
1
−1
−1 −1
3
evolution is 35 µmol h
g
with Pt and 34 µmol h
g
TzF and TzG—a direct effect of the twinned nature of this vdW
heterostructure (Figure 4e).
without Pt cocatalyst (Supporting Information). After photo-
catalysis, FT-IR spectra of recovered TzF/TzG show no signifi-
cant change in material composition (Figure S17, Supporting
Information). In contrast, neat 2D TzF flakes show no meas-
urable hydrogen evolution (Supporting Information). Indeed,
the HOMO of 3D TzG has a better match with the hydrogen
reduction potential compared to 2D TzF (Figure 4e), and the
results from this photocatalytic study nicely corroborate our
previous band gap assignment as the straddled, type I band gap
of ≈1.8–2.0 eV for TzF/TzG vdW heterostructures. Hence, we
can attribute the photocatalytic activity of TzF/TzG to the 3D,
porous TzG component.
The argon sorption isotherms for TzF/TzG vdW hetero-
structures show that the flakes have a specific surface area of
2
−1
1
24 m
g
calculated according to the Brunauer–Emmett–
Teller model (Figure 4f). The adsorption–desorption branches
of the isotherm show a pronounced hysteresis loop character-
istic for micro/mesoporous materials,[ and pore size distribu-
tion analysis shows pore diameters of 3–4 nm (Figure S14, Sup-
porting Information). It is noteworthy that the surface area of
27]
2
−1
exfoliated TzF flakes decreases to 37 m g . This is equivalent
to the external surface area of flakes of highly oriented, pyro-
lytic graphite and indicates that the pore channels of 2D TzF
are indeed blocked by AB-type stacking (Figure 3).
In other polymeric photocatalysts, Pt usually mediates the
transfer of electrons from the photoactive framework to the
[
28]
Next, we probe the activity of TzF/TzG films as heteroge-
neous, metal-free photocatalysts in light-promoted hydrogen
evolution from water. We perform the hydrogen evolution
experiments by irradiating a suspension of TzF/TzG flakes in
a water:acetonitrile mixture (1:1) at room temperature with
visible light (395 nm cut-off filter). Triethanolamine is the
sacrificial electron donor, and the reaction is performed with
sites of water reduction. Here, addition of Pt cocatalyst does
not lead to a significant increase of the rate of photocatalytic
hydrogen evolution. We propose that the crystalline 2D TzF
component is responsible for efficient, noble-metal free elec-
tron transfer from the 3D TzG photocatalyst. To this end, we
examine spin localization in TzF/TzG using EPR and electron
mobility in the vdW heterostructure using direct conductivity
Adv. Mater. 2017, 1703399
1703399 (6 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
Figure 4. Optical, electronic, and sorption properties of TzF/TzG heterostructures. a) Solid-state UV–vis diffuse-reflectance spectrum of TzF/TzG vdW
heterostructures and b) the corresponding Kubelka–Munk plot assuming a direct (2.24 eV, green) and indirect (1.91 eV, orange) band gap. c) Photolu-
minescence emission and excitation spectrum of TzF/TzG. The two overlapping peaks are attributed to excitation transitions within TzG (i) and TzF (ii).
–
1
d) Cyclic voltammogram of a TzF/TzG vdW heterostructure on copper foil as the working electrode in a solution of NaClO (0.1 mol L ) in acetonitrile.
4
−
1
−1
−1
The scan rate is 50 mV s and the potential is given versus Ag/0.01 mol L AgNO /1 mol L NaClO in acetonitrile. The electronic band gap of
3
4
1
.84 eV is given as the difference of the onsets of the anodic and cathodic currents. e) Plot of orbital coefficients for the frontier orbitals of 3D TzG (left)
and 2D TzF (right) as derived from DFT calculations. The relative locations of HOMO and LUMO for the TzF/TzG heterojunction (type I) are shown in
the center with the redox potential of water indicated at pH 0 and 7. f) Argon sorption isotherms for TzF/TzG (red) and TzF (blue) measured at 88 K.
measurements. The linewidth of the EPR signal and spin con-
centration are directly related to the spin motion and environ-
ment.[ For TzF/TzG films we observe a symmetric, narrow
EPR signal (g factor 2.0036, linewidth 0.65 mT) comparable
to the EPR parameters of conductive films of poly(3,4-eth-
ylenedioxythiophene): polystyrene sulfonate (PEDOT:PSS),
poly(phenylenevinylene), polyaniline (PANI), and poly(methyl
methacrylate) (PMMA) (g values of 2.0030–2.0035 and linewidth
At first glance, TzF/TzG shows a low overall activity for
photocatalytic hydrogen evolution compared with some Pt-
modified benchmark materials, e.g., covalent triazine frame-
29]
−
1
−1
−1 −1 [32]
work (CTF-1, 1072 µmol h g ), melon (720 µmol h g ),
[
33]
and other functional carbon nitride materials.
However,
TzF/TzG is comparable with the first reported heptazine-
based, metal-free photocatalyst, polymeric carbon nitride
−
1
−1 [13]
(40 µmol h g ). Further, it is also one of the few photocata-
lysts that operates well in the absence of a noble metal cocata-
lyst. Here, our twinned TzF/TzG vdW heterostructure shows
a substantially improved electron transfer performance com-
[30]
of 0.42–0.80 mT) (Figure 5a). This indicates that the spins are
delocalized over a few rings of the aromatic backbone. Judging
from the linewidth and the expected low-degree of hydrogen
bonding within TzF/TzG, we expect that polaron mobility of
TzF/TzG films is comparable to undoped PEDOT:PSS and
[
11c,28]
pared to conventional microporous materials.
In summary, the use of copper foil as catalyst and template
for the growth of triazine-based polymers allows us to demon-
strate the twinned growth of a 2D and a 3D material in a one-
pot reaction. Importantly, the presented strategy of “on-catalyst”
synthesis from solution opens the door to the fabrication of
otherwise challenging 2D/3D vdW heterostructures. The sim-
plicity of the synthetic process may be extended to the entire
field of organic materials obtained via Glaser-type coupling and
may lead to a host of vertically stacked, mixed-dimensional van
der Waals heterostructures.
[
31]
PANI:PMMA films.
We have established so far that the TzF/TzG vdW hetero-
structure forms a type I heterojunction with a useful band gap
of ≈1.8–2.0 eV. TzF/TzG possesses some degree of microporo-
sity due to its amorphous, open 3D TzG phase, and a crystal-
line, stable backbone of 2D TzF. Thanks to the modular nature
of this vdW heterostructure, we are able to attribute both the
porosity and the photocatalytic activity to the buta-1,3-diyne
bridged 3D TzG.
Adv. Mater. 2017, 1703399
1703399 (7 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
SFM tests and analyzed the data. P.E. performed the ultrasonic
exfoliation experiments of TzF/TzG. J.H., M.P., and P.N. performed
computational modeling and analysis of the data. M.J.B. conceived
and designed the experiments, contributed reagents/materials/analysis
tools, analyzed the data, and wrote the paper.
Conflict of Interest
The authors declare no conflict of interest.
Keywords
2
D
materials, conjugated microporous polymers, donor–acceptor,
membranes, triazine
Received: June 18, 2017
Revised: July 31, 2017
Published online:
[
[
1] G. Algara-Siller, N. Severin, S. Y. Chong, T. Bjorkman, R. G. Palgrave,
A. Laybourn, M. Antonietti, Y. Z. Khimyak, A. V. Krasheninnikov,
J. P. Rabe, U. Kaiser, A. I. Cooper, A. Thomas, M. J. Bojdys, Angew.
Chem., Int. Ed. Engl. 2014, 53, 7450.
Figure 5. Spin localization and electron mobility in TzF/TzG heterostruc-
tures. a) Experimental EPR spectrum of untreated TzF/TzG vdW hetero-
structures (black), simulated EPR spectrum (cyan), and the simulated
radical component (green). The radical component has a g factor of
2] a) Chem. Ind. 2014, 78, 24; b) M. J. Bojdys, Macromol. Chem. Phys.
2
016, 217, 232.
2
.0036 and a linewidth 0.65 mT. b) I–V profile obtained in air at room
temperature showing the dependence of device current (A) on the bias
voltage (V) vertically through the TzF/TzG heterostructure on copper.
Device current at peak voltage of 10 V is in the range of 180–200 µA.
c) Schematic representation of TzF/TzG device. Glass was coated with
a 5 nm layer of chromium to ensure adhesion of copper (Cu, 100 nm)
and gold electrode (Au). The TzF/TzG vdW heterostructure was deposited
from solution before the Au contacts.
[3] a) A. K. Geim, I. V. Grigorieva, Nature 2013, 499, 419;
b) A. Mishchenko, J. S. Tu, Y. Cao, R. V. Gorbachev, J. R. Wallbank,
M. T. Greenaway, V. E. Morozov, S. V. Morozov, M. J. Zhu,
S. L. Wong, F. Withers, C. R. Woods, Y. J. Kim, K. Watanabe,
T. Taniguchi, E. E. Vdovin, O. Makarovsky, T. M. Fromhold,
V. I. Fal’ko, A. K. Geim, L. Eaves, K. S. Novoselov, Nat. Nanotechnol.
2
014, 9, 808.
[
4] a) P. Sehnal, I. G. Stará, D. Šaman, M. Tichý, J. Míšek, J. Cva cˇ ka,
L. Rulíšek, J. Chocholoušová, J. Vacek, G. Goryl, M. Szymonski,
I. Císa rˇ ová, I. Starý, Proc. Natl. Acad. Sci. USA 2009, 106, 13169;
b) G. Konstantatos, M. Badioli, L. Gaudreau, J. Osmond,
M. Bernechea, F. P. G. de Arquer, F. Gatti, F. H. L. Koppens,
Nat. Nanotechnol. 2012, 7, 363.
Supporting Information
Supporting Information is available from the Wiley Online Library or
from the author.
[
5] M. G. Gong, T. A. Shastry, Y. Xie, M. Bernardi, D. Jasion, K. A. Luck,
T. J. Marks, J. C. Grossman, S. Q. Ren, M. C. Hersam, Nano Lett.
2014, 14, 5308.
Acknowledgements
Prof. Arne Thomas is acknowledged for fruitful discussions. Dr. Ivana
Sloufová is acknowledged for Raman measurements, Dr. Miroslav
Stepanek is acknowledged for solid-state fluorescence measurements,
Šárka Pšondrová is acknowledged for IR measurements, Yaroslav
Kochergin is acknowledged for thermogravimetric analysis (TGA) and
Stanislava Mat eˇ jková is acknowledged for ICP-OES. The authors thank
Dr. Nikolai Makukhin for discussions, and Dr. Simona Hybelbauerova
and Dr. Martin Dra cˇ ínský for their support in solid-state NMR
[6] D. Jariwala, T. J. Marks, M. C. Hersam, Nat. Mater. 2017, 16, 170.
[7] a) R. S. Weatherup, B. Dlubak, S. Hofmann, ACS Nano 2012, 6,
9996; b) Y. J. Gong, J. H. Lin, X. L. Wang, G. Shi, S. D. Lei, Z. Lin,
X. L. Zou, G. L. Ye, R. Vajtai, B. I. Yakobson, H. Terrones,
M. Terrones, B. K. Tay, J. Lou, S. T. Pantelides, Z. Liu, W. Zhou,
P. M. Ajayan, Nat. Mater. 2014, 13, 1135.
[
8] D. Jariwala, V. K. Sangwan, L. J. Lauhon, T. J. Marks, M. C. Hersam,
ACS Nano 2014, 8, 1102.
9] S. S. Chen, Q. Z. Wu, C. Mishra, J. Y. Kang, H. J. Zhang, K. J. Cho,
W. W. Cai, A. A. Balandin, R. S. Ruoff, Nat. Mater. 2012, 11, 203.
ˇ
measurements. M.J.B. thanks the Czech Science Foundation (GA CR) for
junior grant funding (CAMs – 16-21151Y) and the European Research
Council (ERC) for funding under the Starting Grant scheme (BEGMAT
[
–
678462). M.J.B., J.C., and P.N. further acknowledge the Charles
[
10] a) G. Eglinton, A. R. Galbraith, Chem. Ind. 1956, 737; b) G. X. Li,
Y. L. Li, H. B. Liu, Y. B. Guo, Y. J. Li, D. B. Zhu, Chem. Commun.
University Centre of Advanced Materials (CUCAM) (OP VVV Excellent
Research Teams, Project No. CZ.02.1.01/0.0/0.0/15_003/0000417).
D.S. conceived and designed the experiments, performed the synthetic
experiments, analyzed the data, and wrote the paper. J.K. and K.S.-P.
performed the electrochemical (CV) tests and analyzed the data.
J.T. performed the EPR tests and analyzed the data. Y.N., J.J., and J.R.
performed the electrical device assembly and conductivity tests. F.S.
and J.S. performed XPS experiments and analyzed the data. M.V.O. and
J.C. performed Ar sorption tests and analyzed the data. A.A. performed
photocatalytic tests and analyzed the data. S.S. and M.J.B. performed the
HR-TEM and analyzed the data. V.R.-S., N.S., J.P.R., and Y.N. performed
2
010, 46, 3256; c) K. S. Sindhu, G. Anilkumar, RSC Adv. 2014, 4,
2
7867.
[
11] a) H. Möhwald, V. Bliznyuk, S. Kirstein, Synth. Met. 1993, 61,
9
1; b) Y. J. Li, L. Xu, H. B. Liu, Y. L. Li, Chem. Soc. Rev. 2014, 43,
2
572; c) V. S. Vyas, F. Haase, L. Stegbauer, G. Savasci, F. Podjaski,
C. Ochsenfeld, B. V. Lotsch, Nat. Commun. 2015, 6, 9.
[12] a) J. S. Zhang, M. W. Zhang, L. H. Lin, X. C. Wang, Angew. Chem.,
Int. Ed. Engl. 2015, 54, 6297; b) J. J. Duan, S. Chen, M. Jaroniec,
S. Z. Qiao, ACS Nano 2015, 9, 931.
Adv. Mater. 2017, 1703399
1703399 (8 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim

www.advancedsciencenews.com
www.advmat.de
[
[
[
13] X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson,
K. Domen, M. Antonietti, Nat. Mater. 2009, 8, 76.
14] T. Nishikubo, A. Kameyama, C. Saito, J. Polym. Sci., Part A: Polym.
Chem. 2000, 38, 3604.
15] a) T. Takahashi, Z. F. Xi, A. Yamazaki, Y. H. Liu, K. Nakajima,
M. Kotora, J. Am. Chem. Soc. 1998, 120, 1672; b) B. E. Straub,
C. Gollub, Chem. Eur. J. 2004, 10, 3081; c) X. Hong, D. Holte,
D. C. G. Gotz, P. S. Baran, K. N. Houk, J. Org. Chem. 2014, 79,
[25] S. Ren, M. J. Bojdys, R. Dawson, A. Laybourn, Y. Z. Khimyak,
D. J. Adams, A. I. Cooper, Adv. Mater. 2012, 24, 2357.
[26] E. Bundgaard, F. C. Krebs, Sol. Energy Mater. Sol. Cells 2007,
91, 954.
[27] a) Y. Liao, J. Weber, B. M. Mills, Z. Ren, C. F. J. Faul, Macromol-
ecules 2016, 49, 6322; b) J. X. Jiang, F. Su, H. Niu, C. D. Wood,
N. L. Campbell, Y. Z. Khimyak, A. I. Cooper, Chem. Commun. 2008, 486.
[28] J. Kiwi, M. Gratzel, Nature 1979, 281, 657.
1
2177.
[29] E. Houze, M. Nechtschein, Phys. Rev. B 1996, 53, 14309.
[30] K. D. Gourley, C. P. Lillya, J. R. Reynolds, J. C. W. Chien, Macromol-
ecules 1984, 17, 1025.
[
[
[
[
16] G. Kyriakou, J. Kim, M. S. Tikhov, N. Macleod, R. M. Lambert,
J. Phys. Chem. B 2005, 109, 10952.
17] H. Oberg, Y. Nestsiarenka, A. Matsuda, J. Gladh, T. Hansson,
L. G. M. Pettersson, H. Ostrom, J. Phys. Chem. C 2012, 116, 9550.
18] N. E. Braml, L. Stegbauer, B. V. Lotsch, W. Schnick, Chem. Eur. J.
[31] a) K. H. Wu, Y. S. Lai, C. C. Shih, G. P. Wang, C. C. Yang, Polym.
Compos. 2008, 29, 902; b) O. Bubnova, Z. U. Khan, H. Wang,
S. Braun, D. R. Evans, M. Fabretto, P. Hojati-Talemi, D. Dagnelund,
J. B. Arlin, Y. H. Geerts, S. Desbief, D. W. Breiby, J. W. Andreasen,
R. Lazzaroni, W. M. M. Chen, I. Zozoulenko, M. Fahlman,
P. J. Murphy, M. Berggren, X. Crispin, Nat. Mater. 2014, 13, 190.
[32] K. Schwinghammer, B. Tuffy, M. B. Mesch, E. Wirnhier,
C. Martineau, F. Taulelle, W. Schnick, J. Senker, B. V. Lotsch, Angew.
Chem., Int. Ed. Engl. 2013, 52, 2435.
2
015, 21, 7866.
19] a) J. X. Jiang, F. Su, A. Trewin, C. D. Wood, N. L. Campbell,
H. Niu, C. Dickinson, A. Y. Ganin, M. J. Rosseinsky, Y. Z. Khimyak,
A. I. Cooper, Angew. Chem., Int. Ed. Engl. 2007, 46, 8574;
b) M. Trunk, A. Herrmann, H. Bildirir, A. Yassin, J. Schmidt,
A. Thomas, Chem. Eur. J. 2016, 22, 7179.
[
[
[
20] Y. Liao, J. Weber, C. F. Faul, Chem. Commun. 2014, 50, 8002.
21] B. T. Li, X. Luo, Y. R. Zhu, X. J. Wang, Appl. Surf. Sci. 2015, 359, 609.
22] a) H. Benes, R. K. Donato, P. Ecorchard, D. Popelkova, E. Pavlova,
D. Schelonka, O. Pop-Georgievski, H. S. Schrekker, V. Stengl,
RSC Adv. 2016, 6, 6008; b) V. Stengl, J. Henych, M. Slusna,
P. Ecorchard, Nanoscale Res. Lett. 2014, 9, 14.
[33] a) F. K. Kessler, Y. Zheng, D. Schwarz, C. Merschjann, W. Schnick,
X. Wang, M. J. Bojdys, Nat. Rev. Mater. 2017, 2, 17030; b) X. Wang,
K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson,
K. Domen, M. Antonietti, Nat. Mater. 2009, 8, 76; c) H. Yu, R. Shi,
Y. Zhao, T. Bian, Y. Zhao, C. Zhou, G. I. N. Waterhouse, L.-Z. Wu,
C.-H. Tung, T. Zhang, Adv. Mater. 2017, 29, 1605148; d) J. Chen,
D. Zhao, Z. Diao, M. Wang, S. Shen, Sci. Bull. 2016, 61, 292;
e) H. Ou, L. Lin, Y. Zheng, P. Yang, Y. Fang, X. Wang, Adv. Mater.
2017, 29, 1700008; f) L. Lin, H. Ou, Y. Zhang, X. Wang, ACS Catal.
2016, 6, 3921.
[
[
23] J. Wang, S. Zhang, R. Liu, R. Du, H. Xu, Z. Liu, J. Zhang, Z. Liu,
Phys. Chem. Chem. Phys. 2014, 16, 11303.
24] P. Kuhn, M. Antonietti, A. Thomas, Angew. Chem., Int. Ed. Engl.
2
008, 47, 3450.
Adv. Mater. 2017, 1703399
1703399 (9 of 9)
© 2017 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim