Angewandte Chemie International Edition
10.1002/anie.201707272
COMMUNICATION
of signals of all CDHF intermediates are rather low. The similar
behavior of CDHF process was also observed for other precursor
molecules. Thus, in the course of zipping of the P1 precursor the
formation of only small amounts of monocyclized intermediate
In summary, we have demonstrated the possibility of
regioselective HF-zipping of m-oligophenylenes to the
“preprogrammed”
nanostructures
via truly domino-like
cyclodehydrofluorination. We show that the unique nature of the
CDHF process allows to perform this challenging transformation
in completely controlled manner. The high efficiency of the
approach has been demonstrated by the effective transformation
of five precursors to the target nanographenes by formal “rolling-
up” of oligophenylene chains. Notably, all NGs were obtained in
highly pure form without involving any additional purification
procedures. Taking into account difficulties connected with
purification of virtually insoluble unsubstituted NGs, which is
highly crucial for electronic applications, the CDHF approach
appears to be superior over existing aryl-aryl coupling techniques.
Moreover, the CDHF-zipping approach could provide a facile way
for bottom-up fabrication of nanographenes and nanoribbons
directly on insulating metal-oxide surfaces,[ which is crucial for
generation of truly-working electronic nanodevices. Finally, since
each cyclization step can be unambiguously predefined by the
proper design of the precursor the presented technique opens
access to highly interesting carbon-based nanostructures which
are difficult to obtained by existing alternative methods.
(
P1a) were detected. The HPLC analysis of the reaction (before
completion) shows that major components of the reaction mixture
are starting compound and the final product. This indicates that
the first cyclization step has a notably higher activation energy
than the second one. As a consequence, the intermediate P1a
undergoes quick transformation to the final product. The same
tendency was observed for P2, P3, and P4 precursors. This
observation is in a good agreement with the previously made
assumption stating that first cyclization is less favorable due to
steric strain caused by the presence of the phenyl group in the
bay-region. Since the mechanism of the CDHF is not fully
understood it was not possible to calculate activation energies for
individual CDHF steps. Therefore, we have performed DFT-
based thermochemistry analysis of double, triple and quadrupole
zipping on the example of P1, P2, and P3. For this purpose, we
have evaluated the reaction enthalpy for each individual step and
35]
compared them with
transformation of fluorinated o-terphenyl to triphenylene was
a
model reference reaction. The
chosen as
a model, for which smooth cyclization was
experimentally observed already at 150 °C.[23] Taking the
enthalpy of the model reaction (HMODEL) we adjust relative
Experimental Section
energy changes (H-HMODEL
)
for all products and all
intermediates (Figure 4). Since all CDHF belong to the same
family of reactions (chemically related processes undergoing
transformation through similar transition states) the correlation
between H and activation energies is expected. Our calculations
show that for the whole series of precursors the first CDHFs are
indeed thermodynamically less favorable in comparison to the
cyclization of the reference compound (o-terphenyl).
Consequently, higher activation barriers are expected for these
steps which is in good agreement with experiment showing low
degree of conversion at 150 °C. The enthalpy of following
cyclization steps was found to be similar to the reference reaction,
whereas the last CDHF is significantly more exothermic. This
explains the low content of intermediates in the reaction mixture
and presence of the starting compound.
General Procedure: Al O -Mediated HF Elimination. Typically, a glass
ampule was charged with -Al O (500 mg) which was initially activated by
2 3
2
3
annealing at 250 °C for 10 min in air. Afterwards, it was activated by
annealing under vacuum (10-2 mbar) for 10–15 min while the temperature
was increased from 250 °C to 550 °C in intervals of 100 °C. After cooling
to room temperature the ampule was filled with argon and the respective
fluoroarene (5 mg) was mixed with activated aluminium oxide. After
repetitive evacuation of the ampule (10-2 mbar) it was sealed. The
condensation was carried out at 200 °C for the indicated time in an oven
while the ampule was rotated. Al O was either extracted with hot o-
2
3
dichlorobenzene or by Soxhlet extraction with o-DCB. The crude product
mixture was analyzed by HPLC. Pure product was obtained by
precipitation with MeOH.
Acknowledgements
K.Y.A. and A.K.S. are grateful to the Deutsche
Forschungsgemeinschaft for financial support (project AM407).
Keywords: Nanographene• C-F activation • Aryl-Aryl coupling •
Domino reaction
[
1]
K. S. Novoselov, A. K. Geim, S. V. Morozov, D. Jiang, Y. Zhang, S. V.
Dubonos, I. V. Grigorieva, A. A. Firsov, Science 2004, 306, 666–669.
A. K. Geim, K. S. Novoselov, Nat. Mat. 2007, 6, 183–191.
[
[
[
2]
3]
4]
M. J. Allen, V. C. Tung, R. B. Kaner, Chem. Rev. 2010, 110, 132–145.
A. Narita, X.-Y. Wang, X. Feng, K. Muellen, Chem. Soc. Rev. 2015, 44,
6616–6643.
[
5]
A. H. Castro Neto, F. Guinea, N. M. R. Peres, K. S. Novoselov, A. K.
Geim, Rev. Mod. Phys. 2009, 81, 109–162.
[
[
6]
7]
J. Wu, W. Pisula, K. Muellen, Chem. Rev. 2007, 107, 718–747.
M. Grzybowski, K. Skonieczny, H. Butenschon, D. T. Gryko, Angew.
Chem. Int. Ed. 2013, 52, 9900–9930.
[
[
8]
9]
P. Rempala, J. Kroulik, B. T. King, J. Org. Chem. 2006, 71, 5067–5081.
L. Zhai, R. Shukla, S. H. Wadumethrige, R. Rathore, J. Org. Chem. 2010,
Figure 4. Graphical representation of the DFT-based thermochemistry analysis
of the CDHF zipping process for precursors P1, P2 and P3. The model
reference reaction is shown in inset.
75, 4748–4760.
This article is protected by copyright. All rights reserved.