Integration of an N-Heterocyclic Carbene Precursor into a Covalent Triazine Framework for Organocatalysis

Article abstract of DOI:10.1002/chem.201804373

The successful incorporation of a thermally fragile imidazolium moiety into a covalent triazine framework resulted in a heterogeneous organocatalyst active in carbene-catalyzed umpolung reaction. The structural integrity of the imidazolium moiety was confirmed by combining solid-state NMR and XPS experiments.

Full text of DOI:10.1002/chem.201804373

A Journal of
Accepted Article
Title: Integration of an N-heterocyclic carbene precursor into a covalent
triazine framework for organocatalysis.
Authors: Erik Troschke, Khoa Dang Nguyen, Silvia Paasch, Johannes
Schmidt, Georg Nickerl, Irena Senkovska, Eike Brunner, and
Stefan Kaskel
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: Chem. Eur. J. 10.1002/chem.201804373
Link to VoR: http://dx.doi.org/10.1002/chem.201804373
Supported by

10.1002/chem.201804373
Chemistry - A European Journal
COMMUNICATION
Integration of an N-heterocyclic carbene precursor into a covalent
triazine framework for organocatalysis.
E. Troschke, [a] K. D. Nguyen, [a] S. Paasch, [b] J. Schmidt, [c] G. Nickerl, [a] I. Senkovska, [a] E. Brunner, [b]
and S. Kaskel*[a]
Abstract: The successful incorporation of
a
thermally fragile
N-heterocyclic carbenes (NHC), representing a well-known
imidazolium moiety into a covalent triazine framework resulted in a
heterogeneous organocatalyst active in carbene-catalysed
Umpolung reaction. The structural integrity of the imidazolium moiety
was confirmed by combining solid-state NMR and XPS experiments.
homogeneous organocatalyst system, turned out to be an
auspicious candidate to be introduced into a CTF.[19] The
successful integration of an imidazolium monomer as NHC-
precursor into a CTF would result in a robust and functional
catalyst but is just rarely described. In contrast to syntheses of
MOFs[20–22] or EOFs,[23] an ionothermal trimerization of nitriles
requires temperatures of up to 400 °C. Therefore, chemically
fragile monomers bearing a NHC-precursor will not withstand
these harsh conditions and well-defined functionalities would
simply decompose.
This issue should be taken into critical consideration when
discussing the few examples of CTFs with embedded cationic
imidazolium moieties (imid-CTF).[24–26] To generate the
intermediary carbene species, a proton has to be abstracted
from the imidazolium cation precursor system prior to its
targeted application. In previous work, imid-CTFs were shown to
capture CO2[24] and to hydrogenate CO2 to formate by the aid of
a coordinated iridium catalyst.[25] However, the best results were
obtained by CTFs synthesised at 400 °C or above. In contrast,
thermogravimetric analysis revealed a substantial decomposition
of the respective monomers starting at 300 °C already. Thus, a
complete structural integrity of these materials remains unclear.
Moreover, the presented applications give at least indirect
evidence of an intact structure.
Porous organic polymers (POPs) have stimulated considerable
interest due to preferable characteristics like their modular
construction principle.[1] For instance, covalent triazine
frameworks (CTFs) are obtained through trimerization of
aromatic nitriles in molten ZnCl2.[2] CTFs provide pronounced
porosity combined with both chemical and thermal stability.[3,4]
Furthermore, a postsynthetic chemical modification of these
polymers with e.g. metal complexes is possible, thus rendering
CTFs suitable for heterogeneous catalytic applications.[5]
To utilise CTFs as heterogeneous catalysts, three different
pathways have been pursued lately. The most facile approach is
applying CTF materials as robust heterogeneous and porous
catalyst support for catalytically active metal nanoparticles as
demonstrated for palladium[6,7] and nickel.[8] However, these
approaches do not take any advantage of particular structural
benefits like open coordination sites of the CTF material itself.
A more developed method makes use of pyridinic nitrogen
atoms in the framework which enable coordination of different
metal complexes. Palkovits et al. demonstrated coordination of
platinum(II) to a CTF based on a 2,6-dicyanopyridine to convert
methane into methanol.[5] Very recently, vanadium complexes
were immobilized in acetylacetone based CTFs and provided
excellent activity in a Mannich-type reaction.[9] Using both noble
and non-noble metal modified CTFs, many efforts have been
devoted to illustrate the electrocatalytic activity towards oxygen
reduction reaction,[10,11] hydrogen oxidation,[12] and alcohol
oxidation.[13]
To overcome the delicate balance between decomposition of the
imidazolium functionality and its transformation into a CTF, a
finely adjusted synthetic protocol was developed. In
consequence, the imid-CTF provides intact imidazolium
functionalities as verified by applying the material as
heterogeneous organocatalyst in carbene-catalysed Umpolung
reaction (Scheme 1).
Bypassing the need for a coordinated metal species, a myriad of
metal-free alternatives have emerged. For example, different
metal-free CTF photocatalysts for the hydrogen evolution
reaction were intensively studied.[14–17] An alternative approach,
presented by Roeser et al., applies the basic nitrogen sites of
CTFs to perform a ring-opening reaction to generate cyclic
carbonates.[18]
However, none of the aforementioned examples can be
considered as a purely CTF-based organocatalyst that benefits
from a monomer bearing a catalytically active functionality.
Scheme 1. Synthesis of imid-CTF and application as heterogeneous
organocatalyst using a base (Diazabicycloundecene = DBU) to deprotonate
the imidazolium moieties and generate active carbene sites.
Owing to the assumed thermal lability of the imidazolium-based
monomer (for monomer synthesis: see ESI), thermogravimetric
analysis (TGA) in argon was performed in order to estimate
suitable synthetic conditions for the new imid-CTF. These
experiments revealed a substantial mass loss of 70 w% between
350°C and 400 °C (Fig. S20, ESI). Annealing at 400 °C for ca.
40 h is usually employed for conventional CTF syntheses.[2] With
[a]
[b]
[c]
Department of Chemistry, Institute of Inorganic Chemistry,
Technische Universität Dresden, Bergstrasse 66, D-01062 Dresden,
Germany. E-mail: Stefan.Kaskel@tu-dresden.de
Faculty of Chemistry and Food Chemistry, Chair of Bioanalytical
Chemistry, Technische Universität Dresden, Bergstrasse 66, D-
01062 Dresden, Germany.
Department of Chemistry, Functional Materials, Technische
Universität Berlin, Hardenbergstrasse 40, 10623 Berlin, Germany.
respect to these findings, it was necessary to apply
a
significantly reduced reaction temperature of only 300 °C. To
compensate the reduced reaction temperature and to achieve a
This article is protected by copyright. All rights reserved.

10.1002/chem.201804373
Chemistry - A European Journal
COMMUNICATION
high degree of polymerization, the reaction time was increased
to 168 h. To this end a mixture of the imidazolium monomer and
ZnCl2 (5 eq.) was transferred into a quartz ampoule and
consequently sealed under vacuum. The ampoule was heated to
300 °C (for temperature program see ESI), cooled and opened.
After grinding of the resulting black monolith, the crude product
was washed with aqueous HCl, water and DMF in order to
remove residual salt and organic impurities (see ESI for
synthetic details).
nitrogen atoms (BE of ≈ 401.7 eV, only one signal can be
detected due to resonance) provided clear evidence of an
(partially) intact structure bearing cationic imidazolium
moieties.[29] Interestingly, an additional signal at
a BE of
≈ 400.0 eV could be observed that most probably stems from a
by-product as a consequence of decomposition processes (see
also ESI Fig. S13 for N 1s XPS spectrum of the monomer and
further explanation).
Due to the reduced reaction temperature (and thus, a low
degree of carbonisation), a well-defined material is obtained as
reflected by discrete vibrations in the FT-IR spectrum (Fig. 1c).
The absence of the strong C≡N vibration (2236 cm-1) in the
polymer points on a complete conversion of the monomer, since
new vibrations at 1530 cm-1 and 1378 cm-1 suggest the formation
of the triazine core.[2]
Further verification of an intact structure is also implied by
elemental analysis (EA). For the imid-CTF, a molar C/N ratio of
6.7 (ideal C/N: 5.3) and a molar C/H ratio of 1.2 (ideal C/H: 1.1)
was determined (see ESI Table S1). In contrast to these almost
ideal values, thermal decomposition would be accompanied by
increased C/N and C/H ratios. Powder X-ray diffraction (PXRD)
shows interlayer (001) stacking as displayed by a broad
reflection at 22.1° (for PXRD data, see ESI). According to TGA
in air, a residual mass as low as 2.2 w% could be detected after
combustion which can be ascribed to residual ZnCl2 within the
sample. Furthermore, the material is thermally stable under
argon up to 300 °C which matches the range of expectation
based on the monomer stability. Scanning electron microscopy
(SEM) pictures reveal a rough and lamellar surface of the
particles (Fig. S18, ESI).
To determine the porosity of the imid-CTF, nitrogen
physisorption measurements (77 K) were performed initially. The
material is nonporous against nitrogen (Fig. S15, ESI). This
might be a result of the low reaction temperature, thus inducing
no chemical activation of ZnCl2, which triggers both the
trimerization reaction but also serves as a porogen at higher
temperatures. Limitations in using nitrogen physisorption, as a
result of a denser state after desolvation and cooling to 77 K,
may arise for ultramicroporous materials or highly flexible
frameworks.[30] In consequence, CO2 adsorption at 273 K was
performed in order to enhance the mobility of the adsorptive and
framework to unveil potential microporosity.[30] The resulting
The successful integration of the imidazolium moiety into the
1
polymer was confirmed by means of solid-state H magic angle
spinning (1H MAS) NMR, X-ray photoelectron spectroscopy
(XPS) and FT-IR spectroscopy. Comparing the solid-state 1H
MAS NMR spectra of the monomer and the resulting imid-CTF
confirms – despite differences in the line width due to a more
disordered structure in the polymer – that the imidazolium
moiety within the covalent triazine framework remains mostly
intact (Fig. 1a). The spectra show the presence of methyl groups
(1 – 4 ppm) as well as of the protons in the aromatic regime (7 –
10 ppm) and most importantly the signal of the acidic
imidazolium proton at a higher chemical shift of 12.2 ppm.
Abstraction of this proton would lead to a free carbene which
acts as catalytically active species for the targeted application.[27]
However, thermal treatment of imidazolium compounds is often
associated with dimerization and/or elimination reactions of the
imidazolium proton.[28] Regarding the signal areas of the methyl
groups and the aromatic protons in the solid-state 1H MAS NMR
spectrum, no distinct decrease in the resulting polymer is
evident. The signal intensity of the imidazolium proton in the
imid-CTF appears to be reduced slightly, which might stem from
broadened signals caused by a less ordered polymer structure
and/or a limited degree of decomposition.
XPS was anticipated to be
a
powerful tool for the
characterisation of CTF materials, since these materials are
usually black and insoluble solids thus limiting the use of optical
characterisation techniques. High resolution nitrogen N 1s
spectra were recorded to distinguish between the two nitrogen
species which should be present within the imid-CTF (Fig. 1b).
Deconvolution of the N 1s spectra revealed the existence of
pyridinic nitrogen with a binding energy (BE) of ≈ 398.4 eV which
reflects nitrogen atoms from the triazine cores of the polymeric
network.[29] More importantly, the presence of imidazolium
Figure 1. a) 1H solid-state MAS NMR spectra of imidazolium-monomer and corresponding imid-CTF; b) Deconvoluted N 1s XPS spectrum of imid-CTF, gray
dashed line: additional signal probably caused by a by-product as a result of decomposotion, black dashed line: counts, green solid line: envelope; and c) FT-IR
spectra of imidazolium-monomer and corresponding imid-CTF.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804373
Chemistry - A European Journal
COMMUNICATION
isotherm reveals a significant uptake of 2.05 mmol g-1 at 1 bar
(Fig. 2a). Moreover, the desorption branch shows a hysteresis
loop in the whole pressure range, indicating a flexible character
of the polymeric framework.[31] The high affinity of CO2 towards
the charged network originates from quadrupole interactions.
However, these findings do not allow for an exact pore size
analysis, since strong interactions and dynamics lead to
erroneous correlations of pore filling pressure and the
corresponding pore size.[30] However, CO2 adsorption indicates
the presence of small but accessible ultramicropores. We
performed vapour sorption experiments at 298 K (see Fig. 2b) to
Scheme 2. Application of imid-CTF as heterogeneous catalyst in carbene-
catalysed conjugated Umpolung reaction.
further evaluate the materials’ suitability for
a catalytic
application in the liquid phase. Due to interactions with different
solvents, a significant swelling of the imid-CTF was anticipated
caused by its flexible character. For MeCN and EtOH as
adsorptives, thus very polar and small molecules, a continuous
uptake over the whole pressure range at 298 K was observed,
suggesting
a constant swelling of the framework upon
adsorption. This effect arises from strong interactions of the
charged and nitrogen-rich polymer with these polar solvents.
Especially interactions with MeCN (bearing nitrile groups) induce
efficient swelling of the imid-CTF which contains analogue
structural motifs. In contrast, vapour adsorption applying toluene
displays very weak interactions as reflected in a negligible
uptake at p/p0 = 1 and almost zero uptake in the low pressure
regime (< p/p0 = 0.2).
Figure 3. a) Comparison of yields using different solvents in heterogeneous
catalysis; b) Kinetic data and heterogeneity testing of imid-CTF.
Table 1. Results of catalytic testing and comparison of imid-CTF with samples
received at higher temperature and homogeneous analogues.
catalyst
cycle
yield (GC)[a]
dr(like : unlike)[b]
1
2
3
4
5
81
79
81
76
76
1.1:1
1.2:1
1.2:1
1.2:1
1.2:1
imid-CTF
imid-CTF_320
imid-CTF_330
Monomer
1
1
1
1
7.9
0.7
1.3:1
1.3:1
1.4:1
1.3:1
Figure 2. a) CO2 physisorption isotherm (273 K) of imid-CTF; b) Vapour
sorption isotherms (MeCN, EtOH, toluene) of imid-CTF.
100
100
IMesxHCl
[a] Determined by GC. [b] Determined by GC and confirmed by 1H-NMR.
After verifying the structural integrity and the sorption properties
of the imid-CTF, the catalytic activity of the material was
evaluated choosing conjugated Umpolung as a prototypical test
reaction (Scheme 2).[27] This reaction takes advantage of an
in situ generated free carbene species (by using the base DBU
to deprotonate the imidazolium moieties) which catalyses the
reaction of an ,-unsaturated cinnamaldehyde with trifluoro-
acetophenone. To explore the feasibility of the imid-CTF for this
reaction, and thus to prove the existence of accessible carbene
sites, several solvents (with high yields in the homogeneous
reaction)[32] were tested for the reaction (Fig. 3a). These
experiments confirm the results obtained from vapour sorption.
For the reactions in MeCN and EtOH, thus polar and small
solvents, yields of 81 % and 63 % after 24 h were obtained. In
contrast, for t-BuOH (polar, but bulky) and aromatic solvents like
toluene and ortho-dichlorobenzene just very low yields of less
than 10 % were found. Based on these results, MeCN was
chosen for the follow-up experiments. The heterogeneity of the
reaction was confirmed by filtering the catalyst off before
reaching complete conversion. No significant product formation
could be observed, thus excluding a leaching of the catalyst
system (Fig. 3b). These results disconfirm recent findings in
literature, proposing the existence of an intact imidazolium motif
in CTF materials produced at 400 °C. Finally, the heterogeneous
system was compared with its homogeneous counterparts. As
expected, the reaction kinetics (for kinetic data, see ESI) as well
as the overall yield were slightly higher applying two different
This article is protected by copyright. All rights reserved.

10.1002/chem.201804373
Chemistry - A European Journal
COMMUNICATION
[4]
[5]
[6]
[7]
[8]
[9]
P. Kuhn, A. Forget, J. Hartmann, A. Thomas, M. Antonietti, Adv. Mater.
2009, 21, 897–901.
imidazolium salts (monomer, IMesxHCl) (Table 1). However, the
heterogeneous catalysis using solid imid-CTF as catalyst
provides significant benefits, in particular recyclability and
stability. We further investigated the recyclability of the
heterogeneous imid-CTF catalyst. The yields did not vary
significantly over 5 cycles, ranging from 75% – 81%. Obviously
no significant structural collapse or pore blocking causes an
activity loss. Control experiments were conducted, applying
catalyst samples prepared at higher temperatures of 320 °C and
330 °C, respectively (henceforth referred to as imid-CTF_320
and imid-CTF_330, see also ESI Table S1 and Fig. S16 for
further characterisation). Given the fact that the imidazolium
moiety will decompose upon thermal treatment, these samples
should provide drastically reduced catalytic activity. These
assumptions are strongly supported by poor yields of only 8 %
(imid-CTF_320) and 0.7 % (imid-CTF_330). Synthesising imid-
CTFs at temperatures of 330 °C or above results essentially in
negligible catalytic activity, which confirms the proposed thermal
decomposition characteristics and sensitivity of the imidazolium
moiety.
R. Palkovits, M. Antonietti, P. Kuhn, A. Thomas, F. Schüth, Angew.
Chem., Int. Ed. 2009, 48, 6909–6912.
C. E. Chan-Thaw, A. Villa, P. Katekomol, D. Su, A. Thomas, L. Prati,
Nano Lett. 2010, 10, 537–541.
Z. Wang, C. Liu, Y. Huang, Y. Hu, B. Zhang, Chem. Commun. 2016, 52,
2960–2963.
Z. Li T. He, L. Liu, W. Chen, M. Zhang, G. Wu, P. Chen, Chem. Sci.
2017, 8, 781–788.
H. S. Jena, C. Krishnaraj, G. Wang, K. Leus, J. Schmidt, N. Chaoui, P.
Van Der Voort, Chem. Mater. 2018, 30, 4102–4111.
[10] K. Kamiya, R. Kamai, K. Hashimoto, S. Nakanishi, Nat. Commun. 2014,
5, 5040.
[11] K. Iwase, T. Yoshioka, S. Nakanishi, K. Hashimoto, K. Kamiya, Angew.
Chem., Int. Ed. 2015, 54, 11068–11072.
[12] R. Kamai, K. Kamiya, K. Hashimoto, S. Nakanishi, Angew. Chem., Int.
Ed. 2016, 55, 13184–13188.
[13] S. Yamaguchi, K. Kamiya, K. Hashimoto, S. Nakanishi, Chem.
Commun. 2017, 53, 10437–10440.
[14] K. Schwinghammer, S. Hug, M. B. Mesch, J. Senker, B. V. Lotsch,
Energy Environ. Sci. 2015, 8, 3345–3353.
[15] A. Bhunia, D. Esquivel, S. Dey, R. Fernández-Terán, Y. Goto, S.
Inagaki, P. Van Der Voort, C. Janiak, J. Mater. Chem. A 2016, 4,
13450–13457.
In summary, we have described the immobilisation of an
undamaged charged imidazolium moiety into a microporous
covalent triazine framework (imid-CTF).
synthetic protocol enabled the structural preservation of the
thermally labile imidazolium motif, which was verified by an
A finely adjusted
[16] C. B. Meier, R. S. Sprick, A. Monti, P. Guiglion, J.-S. M. Lee, M. A.
Zwijnenburg, A. I. Cooper, Polymer 2017, 126, 283–290.
[17] S. Kuecken, A. Acharjya, L. Zhi, M. Schwarze, R. Schomäcker, A.
Thomas, Chem. Commun. 2017, 53, 5854–5857.
1
in-depth structural characterisation, applying solid-state H MAS
[18] J. Roeser, K. Kailasam, A. Thomas, ChemSusChem 2012, 5, 1793–
1799.
NMR, XPS and FT-IR spectroscopy. The obtained imid-CTF is
active as a heterogeneous catalyst for conjugated Umpolung
reaction. More importantly, catalytic testing of the imid-CTF
provided clear evidence of an intact structure, since the
catalytically active carbene species is generated by proton
abstraction from the imidazolium moiety.
[19] M. N. Hopkinson, C. Richter, M. Schedler, F. Glorius, Nature 2014, 510,
485–496.
[20] K. Oisaki, Q. Li, H. Furukawa, A. U. Czaja, O. M. Yaghi, J. Am. Chem.
Soc. 2010, 132, 9262–9264.
[21] M. B. Lalonde, O. K. Farha, K. A. Scheidt, J. T. Hupp, ACS Catal. 2012,
2, 1550–1554.
[22] G.-Q. Kong, X. Xu, C. Zou, C.-D. Wu, Chem. Commun. 2011, 47,
11005–11007.
Acknowledgements
[23] M. Rose, A. Notzon, M. Heitbaum, G. Nickerl, S. Paasch, E. Brunner, F.
Glorius, S. Kaskel, Chem. Commun. 2011, 47, 4814–4816.
[24] K. Park, K. Lee, H. Kim, V. Ganesan, K. Cho, S. K. Jeong, S. Yoon, J.
Mater. Chem. A 2017, 5, 8576–8582.
We acknowledge Dr. Ilka Kunert for performing the
thermogravimetric analysis, Sebastian Ehrling for performing
SEM measurements and Philipp Lange for performing the
elemental analysis.
E. T. has received funding from the Federal Ministry of
Education and Research (BMBF), support code 03XP0030
(“StickLiS”).
[25] G. H. Gunasekar, K. Park, V. Ganesan, K. Lee, N.-K. Kim, K.-D. Jung,
S. Yoon, Chem. Mater. 2017, 29, 6740–6748.
[26] T.-T. Liu, R. Xu, J.-D. Yi, J. Liang, X.-S. Wang, P.-C. Shi, Y.-B. Huang,
R. Cao, ChemCatChem 2018, 10, 2036–2040.
[27] C. Burstein, F. Glorius, Angew. Chem., Int. Ed. 2004, 43, 6205–6208.
[28] F. E. Hahn, M. C. Jahnke, Angew. Chem. 2008, 120, 3166–3216.
[29] D. Y. Osadchii, A. I. Olivos Suarez, A. V. Bavykina, J. Gascon,
Langmuir 2017, 33, 14278–14285.
Keywords: covalent triazine frameworks • microporous
materials • porous polymers • organocatalysis • N-heterocyclic
carbenes
[30] M. Thommes, K. Kaneko, A. V. Neimark, J. P. Olivier, F. Rodriguez-
Reinoso, J. Rouquerol, K. S. W. Sing, Pure Appl. Chem. 2015, 87,
1051–1069.
[31] E. Troschke, S. Grätz, T. Lübken, L. Borchardt, Angew. Chem., Int. Ed.
2017, 56, 6859–6863.
[1]
[2]
A. Thomas, Angew. Chem., Int. Ed. 2010, 49, 8328–8344.
P. Kuhn, M. Antonietti, A. Thomas, Angew. Chem., Int. Ed. 2008, 47,
3450–3453.
[32] K. Hirano, I. Piel, F. Glorius, Adv. Synth. Catal. 2008, 350, 984–988.
[3]
P. Kuhn, A. Forget, D. Su, A. Thomas, M. Antonietti, J. Am. Chem. Soc.
2008, 130, 13333–13337.
This article is protected by copyright. All rights reserved.

10.1002/chem.201804373
Chemistry - A European Journal
COMMUNICATION
COMMUNICATION
A porous imidazolium-based
covalent triazine framework enables
heterogeneous organocatalysis.
E. Troschke, K. D. Nguyen, S. Paasch,
J. Schmidt, G. Nickerl, I. Senkovska, E.
Brunner, and S. Kaskel*
Page No. – Page No.
Integration of an N-heterocyclic
carbene precursor into a covalent
triazine framework for
organocatalysis.
This article is protected by copyright. All rights reserved.