Facile synthesis and theoretical conformation analysis of a triazine-based double-decker rotor molecule with three anthracene blades

Article abstract of DOI:10.1002/chem.201400364

The facile synthesis of a rotor-shaped compound with two stacked triazine units, which are symmetrically connected by three anthracene blades through oxygen linkers, is presented. This new double-decker, which is a potential monomer for two-dimensional polymerization, was synthesized by using readily available, cheap building blocks, exploiting the known selectivity difference for the nucleophilic substitution of cyanuric chloride. The crystal structure of a C_3h symmetric rotor-shaped compound with 9,10-dihydroanthracene blades, which is a direct precursor to the targeted monomer, and the crystal structure of the new double-decker with the desired C_3h symmetry, are also reported. The synthetic efforts were preceded by a computational analysis, which was triggered by the question of conformational stability of the potential monomer. Two stable conformers could be found, and the barrier for the transition path in the gas phase between these conformers was determined by quantum chemical calculations. Exploratory Born-Oppenheimer molecular-dynamics simulations revealed a strong influence of solvent-solute interactions on the stability of the conformers, which resulted in an energetic preference of the C_3h symmetric conformation of the double-decker.

Full text of DOI:10.1002/chem.201400364

DOI: 10.1002/chem.201400364
Full Paper
&
Cyclophanes
Facile Synthesis and Theoretical Conformation Analysis of
a Triazine-Based Double-Decker Rotor Molecule with Three
Anthracene Blades
Max J. Kory,[a] Maike Bergeler,[b] Markus Reiher,*[b] and A. Dieter Schlꢀter*[a]
Abstract: The facile synthesis of a rotor-shaped compound
with two stacked triazine units, which are symmetrically con-
nected by three anthracene blades through oxygen linkers,
is presented. This new double-decker, which is a potential
monomer for two-dimensional polymerization, was synthe-
sized by using readily available, cheap building blocks, ex-
ploiting the known selectivity difference for the nucleophilic
substitution of cyanuric chloride. The crystal structure of
a C3h symmetric rotor-shaped compound with 9,10-dihy-
droanthracene blades, which is a direct precursor to the tar-
geted monomer, and the crystal structure of the new
double-decker with the desired C3h symmetry, are also re-
ported. The synthetic efforts were preceded by a computa-
tional analysis, which was triggered by the question of con-
formational stability of the potential monomer. Two stable
conformers could be found, and the barrier for the transition
path in the gas phase between these conformers was deter-
mined by quantum chemical calculations. Exploratory Born–
Oppenheimer molecular-dynamics simulations revealed
a strong influence of solvent–solute interactions on the sta-
bility of the conformers, which resulted in an energetic pref-
erence of the C3h symmetric conformation of the double-
decker.
Introduction
Cyanuric chloride can be utilized for the facile synthesis of di-
verse triazine derivatives. It shows a remarkable difference in
selectivity of its three chlorine atoms toward the attack of a nu-
cleophile and allows the introduction of a broad range of nu-
cleophiles with high conversions. Much of what is known
today about this type of chemistry dates back to the contribu-
tions of the Stamford Research Laboratories of the American
Cyanamid Co. during the 1950s.[1] Nowadays, cyanuric chloride
is industrially employed for the synthesis of, for example, pesti-
cides, herbicides, dyes, and cross-linking agents. It has also
been used as a cheap precursor for large-scale dendrimer syn-
thesis and supramolecular structures, as well as for various
functional-group transformations.[2a,b,3] In the context of the
present work, which addresses the synthesis of a three-bladed
monomer, the application of this symmetric trifunctional com-
ponent with bifunctional nucleophiles is of particular impor-
tance. Such reactions have led to a host of cyclic products,[4a,b]
which have one remaining chloride per triazine unit in
Figure 1. Structure of monomer 1 in its rotor-shaped C3h conformation.
common. Surprisingly, there were no attempts reported, in
which these last remaining active groups were exploited for
further cyclizations with yet another bifunctional nucleophile
to give bicyclic double-decker-type compounds with two tri-
azine units placed on top of each other, such as in compound
1 (Figure 1).[5] Our group is active in the synthesis of two-di-
mensional polymers (2DP), which are sheet-like covalent mac-
romolecules with internal periodicity.[6] Such two-dimensional
polymers were recently produced by a “single-crystal ap-
proach”,[7] and attempts based on an “interfacial approach”
also appear promising.[8] To achieve lateral growth, monomers
[a] M. J. Kory, Prof. A. D. Schlꢀter
Laboratory of Polymer Chemistry, ETH Zꢀrich
Vladimir-Prelog-Weg 5, 8093 Zꢀrich (Switzerland)
E-mail: ads@mat.ethz.ch
[b] M. Bergeler, Prof. M. Reiher
Laboratory of Physical Chemistry, ETH Zꢀrich
Vladimir-Prelog-Weg 2, 8093 Zꢀrich (Switzerland)
E-mail: markus.reiher@phys.chem.ethz.ch
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201400364.
Chem. Eur. J. 2014, 20, 6934 – 6938
6934
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
containing three photoreactive anthracene groups were devel-
oped. The reactive groups are arranged in a C3-symmetric way
to be capable of connecting to neighboring monomers by an-
thracene [4+4]-cycloaddition dimerization.[9] To exploit this
known chemistry for the generation of sheets, the monomers
have to crystallize in layers, in which they are packed face-to-
face (ftf), so that all the bonding sites get involved in polymeri-
zation, thus to bring about “infinite” regular lateral growth. Al-
though this strategy proved feasible and led to the first exam-
ples of 2DP with all-carbon network structures,[7b] the mono-
mers developed to date require multistep syntheses involving
experimental procedures not compatible with technical-scale
production.
a theoretical study on the relative stability of the conforma-
tions of this new double-decker monomer. For this purpose,
we calculated the electronic energies of possible conformers
of 1 and explored the transition path between the most stable
conformers in the gas phase. In a second step, the influence of
solvent interactions was taken into account by studying the
solvated conformers in Born–Oppenheimer molecular-dynam-
ics (BOMD) simulations.
Results and Discussion
Before entering into a discussion of the synthesis sequence
leading to target monomer 1 (Scheme 1), a comment on con-
formational issues is appropriate. Because the sequence was
devised, it was evident that each of the compounds 5, 6, and
1 can adopt different conformations. It was thus not clear
whether in an unfavorable case, compound 1 could be ob-
tained at all, and if so, whether it would possibly assume a con-
formation favorable for the intended application in lateral
polymerization. Therefore, we performed a quantum chemical
analysis of three plausible conformations of 1 in the gas phase,
which revealed two stable conformers similar in energy, one of
which with the desired C3h symmetry. Because the isolated C3h-
symmetric conformer turned out not to be the lowest-energy
structure, the analysis was extended to include solvent–solute
interactions with explicit dimethyl sulfoxide (DMSO) solvation
as an example. Surrounded by DMSO molecules, the desired
Limited monomer accessibility is considered as a major ob-
stacle in the intended development of 2DP, and we therefore
sought for new monomers, which would combine the appro-
priate crystallization behavior with simple and cost-efficient
synthesis. In this connection, the triazine fragment caught our
attention, and we set out to synthesize rotor-shaped com-
pounds, such as 1, in which two triazine units are linked by
photoreactive anthracene blades to a C3h symmetric double-
decker structure. Herein, we report the highly efficient, easy to
perform synthesis of 1 in five steps from cheap, commercially
available chemicals without any column-chromatographic
work-up. It gave 0.5 g of 1 and 3 g of its threefold 9,10-dihy-
drogenated precursor 6 (Scheme 1). We also present the
single-crystal structures of 1 and 6 and discuss the conforma-
tions of both compounds, considering packing issues as well.
In rigid geometries, such as 1, high activation barriers for
the flipping motion of the blades are anticipated. However, it
was not known whether the desired C3h symmetry of this po-
tential monomer could be achieved at all, and whether it
would, if adopted, stay intact at elevated temperatures, which
are typically required for crystallization of compounds with
rigid scaffolds. Therefore, the synthetic effort was flanked by
C3h conformation of 1 is in fact energetically more favored over
the other one (see below). This then was efficient motivation
to enter experimentally into the sequence. The synthesis of
rotor 1 started from commercially available 1,8-dihydroxyan-
thraquinone 2, which was first converted into the hydrogenat-
ed anthracene derivative 3. This sequence may look like an un-
necessary detour given the availability of 1,8-dihydroxyanthra-
cene,[10] which could in principle be used directly. In our ap-
Scheme 1. Synthesis sequence to rotor-shaped monomer 1 through the not isolated partially dehydrogenated intermediates 6–2H and 6–4H. DIPEA=N,N’-
diisopropylethylamine; DDQ=2,3-dichloro-5,6-dicyanobenzoquinone.
Chem. Eur. J. 2014, 20, 6934 – 6938
www.chemeurj.org
6935
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
proach, however, difficulties with the sensitivity of
the latter in basic media were avoided, because such
conditions are required for the nucleophilic displace-
ment of cyanuric chloride. The conversion of 2 to 3
was inspired by the conditions described by Miao
and co-workers, which converted 5,12-tetracenoqui-
none into 5,12-dihydrotetracene using LiAlH4 and
AlCl3.[11] According to this protocol, compound 3 was
obtained on the 20 g scale demonstrating its efficien-
cy compared to reported procedures employing ex-
cessive quantities of Zn in aqueous sodium hydrox-
ide.[12] Reaction of compound 3 with cyanuric chlo-
ride in a 2.5:1 stoichiometry at 08C gave intermediate
4 (yield 97%), underlining the known selectivity of
the trichloride. The next step was carried out at room
temperature and required more diluted conditions
(3.3 mmolmLꢀ1) to prevent formation of open-chain
and cyclic oligomers. This way, compound 5 was ob-
tained from a stoichiometric mixture of 3 and 4 on
the 5 g scale after simple precipitation. Temperature-
dependent NMR spectroscopy of intermediate 5 re-
vealed its conformational flexibility (see the Support-
ing Information). Regarding the conversion from 5 to
6, we wondered whether there was an intrinsic
reason why such transformations had not been de-
scribed in the literature to date, despite the large
body of reactions performed with cyanuric chloride
and its derivate. Gratifyingly, this conversion turned
out to be uncritical and was achieved under dilute
conditions (1.7 mmolmLꢀ1) upon heating at reflux in
THF by using potassium carbonate as base. After the
first day of reaction, precursor 6 started to precipitate
as fine needles, and after completion (TLC), 6 was ob-
tained on the 3 g scale by centrifugation of the con-
Figure 2. ORTEP plots of compounds 6 (a,b) and 1 (c–f). a),b) Side and top views show-
ing the slightly kinked blades and the C3h symmetry. c),d) Side and top views showing
planar nature of the blades and the C3h symmetry. e) Side view of the dense-layered
packing of the double-decker. f) Packing within a layer showing edge-to-face stacking.
Solvent molecules are disordered and could therefore not be assigned.
centrated reaction mixture (the NMR spectra of 6 are described
in the Supporting Information).
troscopy, and the intermediates [6–2H] and [6–4H] were con-
firmed by HR MALDI mass spectrometry. After 2 h, the reaction
was completed. 1H NMR spectroscopy analysis of 1 was not
possible due to the insolubility of the latter also in potent deu-
terated solvents. Upon cooling to room temperature, com-
pound 1 started to precipitate from the reaction mixture. After
removing the solvent under reduced pressure and subsequent
washing, compound 1 was obtained in an overall yield from 3
of 61%. Finally, it was noted that apart from the dehydrogena-
tion reaction, the entire sequence was carried out under ambi-
ent air, and that at no point column chromatographic purifica-
tions were required. All compounds could be isolated very
easily, usually by washing and subsequent centrifugation,
which renders the sequence scalable.
Single crystals of 6 were obtained from 1,2-dichloroethane
(DCE) and n-hexane by slow vapor diffusion. Compound 6 has
C3h symmetry (Figure 2a and b), which made us confident to
achieve the same conformation for target 1. The adoption of
this C3h symmetric conformation, as opposed to one, in which
two of the three blades stack, can be explained by the other-
wise large sterical interactions of the axial hydrogen atoms in
the 9 and 10 position. Compound 6 is packed in layers. The
crystals contain disordered DCE solvent molecules dispersed
between the layers, but they show also some molecules in-
cluded into the cavities of the layers. The distance between
the two triazine entities is only 5 ꢂ, which may be attractive
for capturing small ions.[13] The final step of the sequence, the
dehydrogenation of 6, was carried out with 2,3-dichloro-5,6-di-
cyano-1,4-benzoquinone (DDQ). Because the two intermedi-
ates [6–2H] and [6–4H], as well as the target compound 1,
itself were expected to be poorly soluble, this step was per-
formed under dilute conditions (0.5 mgmLꢀ1 of 6) in tetra-
chloroethane (TCE) while heating at reflux with 4.5 equivalents
of DDQ per equivalent of 6. The reaction was monitored by 13C
cross-polarization magic-angle spinning (CP MAS) NMR spec-
Needle-shaped single crystals of compound 1 were grown
from hot benzonitrile or 2-cyanopyridine. The resulting crystal
structure shows that the double-deckers adopt the desired C3h
symmetry, and that the monomers are arranged in layers, in
which the triazine units of the monomers stack densely above
each other, always rotated by 608. This hexagonal composite
unit of two monomers beautifully exemplifies the Piedfort con-
cept.[14] The anthracene blades of neighboring compounds are
stacked edge-to-face (etf). Although the etf stacking does not
Chem. Eur. J. 2014, 20, 6934 – 6938
www.chemeurj.org
6936
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
allow the desired lateral polymerization leading to 2DP, we
noted that for a structurally related monomer it was recently
shown in a rather comprehensive crystallization study[7b] that
etf can be rendered into ftf, if the appropriate solvent and the
optimum crystallization conditions are chosen. While this
paper was under review, crystallization of compound 1 into
layered ftf-stacked crystals was also successful.[15]
neglect attractive long-range dispersion effects. They can be
efficiently included through a dispersion correction for atom
pairs at sufficiently large distance. Because of the interactions
between the three anthracene moieties, we expected that dis-
persion interactions play an important role and thus employed
Grimme’s dispersion corrections D and D3.[16] Detailed results
obtained for different computational set-ups are reported in
the Supporting Information. We found that 1B is energetically
favored over 1A by about ꢀ3 to ꢀ4 kcalmolꢀ1, if we take dis-
persion interaction energies into account. By contrast, without
dispersion interactions, 1B is disfavored over 1A by around
1 kcalmolꢀ1. Attractive dispersion interactions also affect the
molecular structures. Thus, in 1B, the dispersion-affected ftf-
stacking blades are by 1.7 ꢂ closer together than in the struc-
ture for which dispersion was not taken into account in the
structure optimization. Comparing dispersion corrected, as well
as dispersion-free results, structures 1A and B are more similar
in energy (DEel(1A!1B) ranges from ꢀ10 to 1 kcalmolꢀ1) than
As was stated before, a computational analysis was initiated
in the beginning of our synthetic efforts to predict the confor-
mational stability of double-decker 1. The intrinsic bent shape
of its oxygen bond can, in principle, lead to various conformers
of 1. We focused on three (Figure 3a): the C3h symmetric rotor-
structures 1A and
C
(DEel(1A!1C) ranges from 14 to
19 kcalmolꢀ1). Because structure 1C has a significantly higher
electronic energy than structure 1A, it is not likely that 1C will
be formed from 1A.
To draw conclusions about transition probabilities from one
minimum to the other, it is necessary to analyze transition bar-
riers between the minima. The calculations revealed that for
the transformation from structure 1A to 1B, at least
19 kcalmolꢀ1 have to be provided to reach the first transition
state, TS-1, and after a shallow energy minimum (1AB),
a second transition state, TS-2, of about the same electronic
energy as TS-1, was found, which connects 1AB with structure
1B. In Figure 3b, we show the rearrangement energy path cal-
culated with BP86, BP86-D3 and TPSS-D3.[17] The electronic-
energy profiles are very similar, and the addition of the D3 dis-
persion correction to the BP86 functional does not change the
electronic energy profile significantly. Due to the double-
decker nature of 1, it is reasonable that the transition from 1A
to 1B occurs in two steps and requires two transition states.
This implies that the C-O-C angles change one after the other
with an energetically shallow intermediate, in which the C-O-C
angles are of the same size, but the upper and the lower part
of the blades are twisted in different directions (1AB). The elec-
tronic-energy difference between the minimum 1AB and the
two transition states is very small (about 1 kcalmolꢀ1). By care-
fully probing our results on technical artifacts (such as optimi-
zation thresholds or numerical grid sizes), we confirmed the
existence of a small dip in the potential energy path from 1A
to 1B (see the Supporting Information).
Figure 3. Results of quantum chemical calculations. a) Quantum chemically
optimized conformations of compound 1. b) Electronic-energy transition
path from structure 1A to 1B calculated with density functionals BP86,
BP86-D3, and TPSS-D3 (all structures were fully optimized with a given func-
tional; see the Supporting Information) and with the def2-TZVP basis set.
The BP86/def2-TZVP-optimized structures are depicted. For TS-1 and TS-2,
the mode with imaginary frequency is shown. Color code: C gray; H white;
N blue; O red.
shaped form (1A), a non-C3 symmetric form (1B), which is ac-
cessible from 1A through a flipping motion of one of the
three blades, hereby twisting the two overlaying oxygen link-
ers to the opposite side, and another C3 symmetric screw-
shaped form (1C), in which the three oxygen bonds connect-
ing one triazine core face in opposite direction compared to
the other ones. To investigate the stability of these conformers,
as well as potential rearrangement pathways, density function-
al theory (DFT) calculations were performed. A detailed assess-
ment of our computation methodology is provided in the Sup-
porting Information.
The calculated energy barrier theoretically permits transfor-
mations from isolated 1A to isolated 1B at elevated tempera-
tures. However, under experimental conditions, also solvent–
solute interactions have to be taken into account. The solubility
of 1 was not known in advance. DMSO is known for its general
strong solvation and was therefore chosen as an example sol-
vent for BOMD simulations of explicitly solvated structures 1A
and 1B to predict whether these interactions would play a role.
The analysis of BOMD snapshot structures of the solvated
conformers indicates that conformer 1A is slightly more stable
First the structures 1A, B, and C were optimized. It is impor-
tant to note that standard DFT calculations usually completely
Chem. Eur. J. 2014, 20, 6934 – 6938
www.chemeurj.org
6937
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
compared to 1B. This result is in contrast to the theoretical re-
sults obtained for the isolated conformers. To understand why
the energy of structure 1A in DMSO is lower than the one of
structure 1B in DMSO, the inter- and intramolecular contribu-
tions to the electronic dispersion-free and dispersion-affected
interaction energies were analyzed. We observed that on aver-
age, the dispersion interaction energy between structure 1A
and all solvating DMSO molecules in a given number of snap-
shots (intermolecular contribution) is by 5.4 kcalmolꢀ1 more
exothermic than the dispersion interaction energy between
structure 1B and its DMSO solvent shell. The BP86 dispersion-
free intrinsic interaction energy of 1A with its DMSO solvent
shell is, on average, by 4.1 kcalmolꢀ1 larger than for 1B: the in-
trinsic interaction energy was calculated from two single-point
DFT calculations for the solute and the solvent-shell, respec-
tively, subtracted from the total energy of a snapshot (see the
Supporting Information for further details).
Acknowledgements
We thank Priv. Doz. Dr. J. Sakamoto for helpful discussions and
Dr. Nils Trapp, Dr. Michael Wçrle, Dr. W. B. Schweizer, and Mi-
chael Solar for X-ray structure analysis. M. Slusarczyk is thanked
for the help with synthesis, and Dr. K. Feldman and Prof. P.
Smith for access to their optical-microscopy equipment. We
would like to thank Dr. R. Verel for help with the acquisition of
the solid-state NMR spectra. Rolf Hꢃfliger (Laboratory of Organ-
ic Chemistry, ETH Zꢀrich) for high-resolution mass determina-
tion. M.B. gratefully acknowledges support by a fellowship of
the Fonds der Chemischen Industrie.
Keywords: Born–Oppenheimer
cyclophanes · density functional calculations · polymers
molecular
dynamics
·
[1] F. C. Schaefer, J. T. Thurston, J. R. Dudley, J. Am. Chem. Soc. 1951, 73,
2990–2992.
[2] a) M. B. Steffensen, E. Hollink, F. Kuschel, M. Bauer, E. E. Simanek, J.
Polym. Sci. Part A 2006, 44, 3411–3433; b) M. B. Steffensen, E. E. Sima-
nek, Angew. Chem. 2004, 116, 5290–5292; Angew. Chem. Int. Ed. 2004,
43, 5178–5180.
Because in 1B two anthracene blades are stacked ftf to each
other with no DMSO molecules between them, less solvent–
shell–solute interactions occur than for 1A (detailed informa-
tion can be found in the Supporting Information). From this
analysis, we therefore deduce a preference for the C3h symmet-
ric conformation 1A in solution. However, the formation of 1B
is not ruled out. It should be noted that the fact that X-ray
crystal-structure analysis of 1, revealed solely the C3h symmetric
conformation 1A might be due to possible unfavorable pack-
ing effects of conformation 1B in the crystal, or the fact that
only small amounts of 1B are formed upon heating.
[3] G. Blotny, Tetrahedron 2006, 62, 9507–9522.
[4] a) M. X. Wang, H. B. Yang, J. Am. Chem. Soc. 2004, 126, 15412–15422;
b) M. M. Naseer, D. Wang, L. Zhao, Z. Huang, M. X. Wang, J. Org. Chem.
2011, 76, 1804–1813.
[5] Modern Cyclophane Chemistry, Gleiter, R.; Hopf, H.; (eds). WILEY VCH,
Weinheim, 2004, DOI: 10.1002/3527603964.
[6] J. Sakamoto, J. van Heijst, O. Lukin, A. D. Schlꢀter, Angew. Chem. 2009,
121, 1048–1089; Angew. Chem. Int. Ed. 2009, 48, 1030–1069.
[7] a) P. Kissel, R. Erni, W. B. Schweizer, M. D. Rossell, B. T. King, T. Bauer, S.
Gçtzinger, A. D. Schlꢀter, J. Sakamoto, Nat. Chem. 2012, 4, 287–291;
b) R. Bhola, P. Payamyar, D. J. Murray, B. Kumar, A. J. Teator, M. U.
Schmidt, S. M. Hammer, A. Saha, J. Sakamoto, A. D. Schlꢀter, B. T. King, J.
Am. Chem. Soc. 2013, 135, 14134–14140.
Conclusion
We could show that monomer 1 can be easily obtained from
commercially available starting materials in five steps, taking
advantage of the selectivity difference for the nucleophilic sub-
stitution of the chlorine atoms present in cyanuric chloride. As
part of this synthetic strategy, a new efficient protocol for the
synthesis of 9,10-tetrahydro anthracene was established. The
crystal structure of the rotor-shaped precursor 6 was deter-
mined revealing its C3h symmetric nature. To understand possi-
ble conformations of monomer 1, a theoretical analysis was
conducted, which revealed two stable conformations similar in
energy. One, in which the three blades all face “clockwise tan-
gentially”, hereby creating a C3h symmetry (1A); and one, in
which two blades stack ftf (1B). Furthermore, a transition path
connecting the most stable conformers, 1A and 1B, was calcu-
lated. When solvent–solute interaction energies were taken
into account, a significant stabilization of conformation 1A
over the ftf-stacked conformation 1B could be determined.
Monomer 1 could be crystallized, and its crystal structure
showed in fact the C3h symmetry desired for a potential mono-
mer for two-dimensional polymerization. Although the packing
of monomer 1 has not yet allowed its polymerization within
the single crystal, the fact that it assumes a layered packing
shows promise that by carefully choosing the solvent and the
crystallization conditions, one will eventually be able to ach-
ieve this goal.[15]
[8] P. Payamyar, K. Kaja, C. Ruiz-Vargas, A. Stemmer, D. J. Murray, C. J. John-
son, B. T. King, F. Schiffmann, J. VandeVondele, A. Renn, S. Gçtzinger, P.
Ceroni, A. Schꢀtz, L. T. Lee, Z. Zheng, J. Sakamoto, A. D. Schlꢀter, Adv.
Mater. 2014, 26, 2052–2058.
[9] H. Bouas-Laurent, A. Castellan, J. P. Desvergne, R. Lapouyade, Chem.
Soc. Rev. 2001, 30, 248–263.
[10] L. Federer, P. Kissel, J. Sakamoto, A. D. Schlꢀter, F. Weibel, Synlett 2008,
12, 1793–1796.
[11] Z. Liang, W. Zhao, S. Wang, Q. Tang, S. Lam, Q. Miao, Org. Lett. 2008, 10,
2007–2010.
[12] C. Zhang, C. Lu, J. Zhu, C. Wang, G. Lu, C. Wang, D. Wu, F. Liu, Y. Cui,
Chem. Mater. 2008, 20, 4628–4641.
[13] a) C. Zuo, J. M. Quan, Y. Wu, Org. Lett. 2007, 9, 4219–4222; b) A. Fron-
tera, P. Gamez, M. Mascal, T. J. Moiibroek, Angew. Chem. 2011, 123,
9736–9756; Angew. Chem. Int. Ed. 2011, 50, 9564–9583.
[14] A. S. Jessiman, D. D. MacNicol, P. R. Mallinson, I. J. Vallence, J. Chem. Soc.
Chem. Commun. 1990, 1619–1621.
[15] M. Kory, M. Wçrle, T. Weber, P. Payamyar, J. Dschemuchadse, N. Trapp,
A. D. Schlꢀter, submitted.
[16] a) S. J. Grimme, Comput. Chem. 2006, 27, 1787–1799; b) S. Grimme, J.
Antony, S. Ehrlich, H. A. Krieg, J. Chem. Phys. 2010, 132, 154104.
[17] a) A. D. Becke, Phys. Rev. A 1988, 38, 3098–3100; b) J. P. Perdew, Phys.
Rev. B 1986, 33, 8822–8824; c) P. A. M. Dirac, Proc. Royal. Soc. (London)
1929, 123, 714–733; d) J. C. Slater, Phys. Rev. 1951, 81, 385–390; e) J. P.
Perdew, Y. Wang, Phys. Rev. B 1992, 45, 13244–13249; f) J. P. Perdew,
V. N. Staroverov, G. E. Scuseria, Phys. Rev. Lett. 2003, 91, 146401.
Received: January 28, 2014
Published online on April 15, 2014
Chem. Eur. J. 2014, 20, 6934 – 6938
www.chemeurj.org
6938
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim