[2.2]Paracyclophanes with N-Heterocycles as Ligands for Mono- and Dinuclear Ruthenium(II) Complexes

Article abstract of DOI:10.1002/chem.201703291

[2.2]Paracyclophane, with its unique structure, allows the design of unusual 3D structures by functionalization of this rigid and stable hydrocarbon scaffold. Therefore different mono- and homodisubstituted [2.2]paracyclophanes with pyridyl, pyrimidyl and oxazolinyl substituents were developed in order to evaluate their ability as bridging ligands for two ruthenium centres. With the successfully synthesized [2.2]paracyclophane-based N-donor functions, the cycloruthenation reaction using [RuCl₂(p-cymene)]₂ as precursor was explored. Compared to 2-phenylpyridine, the [2.2]paracyclophane derivative is clearly inferior in the cycloruthenation reaction, resulting in poor yields for the neutral complexes. By addition of KPF₆, the cationic complexes can be obtained in good yields and are formed diastereoselectively in case of a pyridyl substituent, resulting in only one diastereomer for dinuclear ruthenium complexes of bispyridyl-substituted [2.2]paracyclophanes as bridging ligands.

Full text of DOI:10.1002/chem.201703291

A Journal of
Accepted Article
Title: [2.2]Paracyclophanes with N-Heterocycles as Ligands for Mono-
and Dinuclear Ruthenium(II) Complexes
Authors: Carolin Braun, Martin Nieger, Werner R. Thiel, and Stefan
Bräse
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.201703291
Link to VoR: http://dx.doi.org/10.1002/chem.201703291
Supported by

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
[2.2]Paracyclophanes with N-Heterocycles as Ligands for Mono-
and Dinuclear Ruthenium(II) Complexes
Carolin Braun,[a] Martin Nieger,[b] Werner R. Thiel[c] and Stefan Bräse*[a,d]
Abstract: [2.2]Paracyclophane, with its unique structure, allows the
design of unusual 3D structures by functionalization of this rigid and
stable hydrocarbon scaffold. Therefore different mono- and
homodisubstituted [2.2]paracyclophanes with pyridyl, pyrimidyl and
oxazolinyl substituents were developed in order to evaluate their
ability as bridging ligands for two ruthenium centres. With the
successfully synthesized [2.2]paracyclophane-based N-donor
functions, the cycloruthenation reaction using [RuCl2(p-cymene)]2 as
precursor was explored. Compared to 2-phenylpyridine, the
[2.2]paracyclophane derivative is clearly inferior in the cyclo-
ruthenation reaction, resulting in poor yields for the neutral complexes.
By addition of KPF6, the cationic complexes can be obtained in good
yields and are formed diastereoselectively in case of a pyridyl
substituent, resulting in only one diastereomer for dinuclear ruthenium
complexes of bispyridyl-substituted [2.2]paracyclophanes as bridging
ligands.
of five-membered metallacycles, [2.2]paracyclophanes attached
to N-heterocycles that possess a sp2-nitrogen in ortho-position
are of particular interest. Therefore we chose pyridine, pyrimidine
and oxazoline rings as preferred substituents (Figure 1), which
have the advantage that synthetic routes to their monosubstituted
[2.2]paracyclophane derivatives are well established. Further-
more, the ability of the oxazoline[7] and the pyridine[8] derivative to
coordinate transition metals like palladium to yield complexes like
1
[7] and 2[9] have been already reported. Therefore the formation
of dinuclear metal complexes with an appropriate ligand system
is very likely. As a result of the particular structure of [2.2]para-
cyclophane, a greater variety of disubstituted regioisomers are
accessible than in the analogous phenyl molecules. Some of
these isomers possess planar chirality depending on the
substituents and their connectivity. Among homodisubstituted
[2.2]paracyclophanes, achiral pseudo-para derivatives are most
desirable as they are readily accessible from the respective
dibromide.[3] In case of their double cyclometallation, the metal
centres are not able to interact directly. With a para-disubstituted
derivative, the two metal centres are situated at the same phenyl
ring changing their electronic environment.
Introduction
Since the discovery of the [2.2]paracyclophane scaffold by Brown
and Farthing in 1949[1] and the pioneering work of Reich and
Cram,[2] this class of bridged aromatic compounds has been
extensively studied in terms of reactivity[3], properties[4], its use for
planar-chiral ligands[5] as well as for material science[6]. Because
of its unique 3D structure, [2.2]paracyclophanes feature trans-
annular properties which impart exceptional reactivity and make it
attractive as building block for complex molecules with an unusual
and rigid spatial structure. The [2.2]paracyclophane backbone
enables the design of bridging ligands in which the position of two
transition metals is controlled. Because of the well-known stability
[a]
C. Braun, Prof. Dr. S. Bräse
Institute of Organic Chemistry
Karlsruhe Institute of Technology (KIT)
Fritz-Haber-Weg 6, 76131 Karlsruhe, Germany
E-mail: braese@kit.edu
[b]
[c]
[d]
Dr. M. Nieger
Department of Chemistry, University of Helsinki
P.O. Box 55, 00014 University of Helsinki, Finland
Prof. Dr. Werner R. Thiel
Fachbereich Chemie, Technische Universität Kaiserslautern
Erwin-Schrödinger-Straße 54, 67663 Kaiserslautern, Germany
Prof. Dr. S. Bräse
Institute of Toxicology and Genetics
Karlsruhe Institute of Technology (KIT)
Hermann-von-Helmholtz-Platz 1, 76344 Eggenstein-Leopoldshafen,
Germany
Figure 1: Known cyclopalladated [2.2]paracyclophanes 1[7] and 2[9]
,
cycloruthenated phenylpyridine 3[10] and phenyloxazoline 4[11] as starting point
for the development of [2.2]paracyclophane-based mono- and dinuclear
ruthenium complexes.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
To explore the suitability of biscyclometallated [2.2]paracyclo-
phanes as bridging ligands with unique electronic properties, we
selected ruthenium(II) as diamagnetic transition metal centre. On
the one hand ruthenium(II) is known to form defined monomeric
complexes like 3[10] and 4[11] with the phenyl derivatives, on the
other hand, ruthenium(II)-complexes with an η6-coordinated p-
cymene ligand are usually air- and moisture stable. Therefore we
developed synthetic routes to pseudo-para- and para-disubsti-
tuted [2.2]paracyclophanes that are potentially bridging ligands
for dinuclear transition metal complexes and explored their
coordination behavior for the cyclometallation of ruthenium(II).
Results and Discussion
Synthesis of pyridyl-substituted [2.2]paracyclophanes. In
order to develop [2.2]paracyclophanes as bridging ligands, we
synthesized the different mono- and homodisubstituted [2.2]para-
cyclophanes. As previously shown, pyridyl substituents are easily
accessible via Stille cross coupling reactions in only one step
(Scheme 1).[8b] This makes these substituents attractive as they
can be easily varied with respect to subsequent fine-tuning of this
ligand type. We already reported the synthesis for mono- and
pseudo-para-bispyridyl-substituted [2.2]paracyclophanes 8 and 9
from the respective bromides 5 and 6. They are accessible from
commercially available unsubstituted [2.2]paracyclophane via
single or double bromination adjusting the reaction conditions and
workup procedures.[2c-e, 12]
Scheme 2: Synthesis of para-diiodide 12 as substrate for the Stille cross
coupling reaction to yield 13.
In case of the bispyridyl-substituted 13 (Figure 2), the molecule
shows the characteristic twist between the pyridyl rings and the
[2.2]paracyclophane backbone, which was reported for the mono-
and pseudo-para-disubstituted [2.2]paracyclophanes.[8b] Note-
worthy, with 32.3° and 44.1° two different angles were observed
for the twist of the two pyridyl substituents towards the [2.2]para-
cyclophane backbone, whereas for the pseudo-para-derivative 6a
they show the same angle of 37.2°.[8b]
Scheme 1: Synthesis of 2-pyridyl-substituted [2.2]paracyclophanes 8 and 9 via
Stille cross coupling.[8b]
Figure 2: Molecular structure of bispyridyl[2.2]paracyclophane 13 (displace-
ment parameters are drawn at 50% probability level).
However, during the double bromination of unsubstituted
[2.2]paracyclophane the para-dibromide is only a side product
and its separation from the other side products of the unselective
bromination is not trivial. As a result, we required a different
approach to such para-homodisubstituted [2.2]paracyclophanes
and developed a synthetic route towards the para-diiodo-
[2.2]paracyclophane (12, Scheme 2). This para-diiodide 12 was
obtained by para-iodination of the aniline derivative 10 followed
by the Sandmeyer-type reaction of amine 11. The para-bispyridyl-
substituted [2.2]paracyclophane 13 could be prepared from the
diiodo species 12 in one step by a Stille cross coupling reaction.
We could obtain the molecular structure of the diiodide 12 as well
as the cross coupling product 13 by X-ray crystallography both
confirming the para-disubstitution (the crystal structure of 12 can
be found in the supporting information).
Synthesis of pyrimidyl-substituted [2.2]paracyclophanes.
After we had successfully synthesized the desired pyridyl-
substituted derivatives, we focused on the preparation of the
pyrimidyl-substituted [2.2]paracyclophanes.
Scheme 3: Synthesis of pyrimidyl-substituted [2.2]paracyclophanes 16 via
enaminone 15.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
The de novo synthesis of pyrimidine rings starting from enami-
nones and amidines is well established,[13] and our group already
reported the synthesis of monopyrimidyl-substituted [2.2]para-
cyclophanes 16 from the 4-acetyl[2.2]paracyclophane (14,
Scheme 3).[14] A X-ray crystal structure could be obtained from
16a that confirms the molecular structure and shows a twist of the
pyrimidyl ring against the [2.2]paracyclophane backbone of 37.8°
(see supporting information for more details). The disubstituted
derivatives should be accessible by the same synthetic route from
the respective diacetyl[2.2]paracyclophane. Unfortunately, the
synthesis of diacetyl-substituted [2.2]paracyclophanes is challen-
ging because the Friedel-Crafts acylation gives the monoacety-
lated product 14 selectively due to transannular effects between
the two phenyl rings of the [2.2]paracyclophane backbone.[15]
Another possibility to prepare the monoacetylated [2.2]paracyclo-
phane is the reaction of the carboxy[2.2]paracyclophane with
methyl lithium.[16] In case of the pseudo-para-dicarboxylic acid 17,
the pseudo-para-diacetyl[2.2]paracyclophane 19 was obtained
only with a maximum yield of 6% by reaction with methyl lithium,
probably due to the very low solubility of 17 in thf or diethyl ether.
Therefore we adjusted our reaction sequence and could success-
fully synthesize the diacetylated [2.2]paracyclophane 19 via the
Weinreb amide 18, which could be converted to 19 in good overall
yield by reaction with methyl magnesium bromide or methyl
lithium (Scheme 4). Afterwards, the synthesis of the bispyrimidyl-
substituted [2.2]paracyclophanes 21 could be completed by
reaction with dimethylfomamide dimethyl acetal (DMF-DMA) to
the enaminone 20, that is subsequently converted to the bispyri-
midyl-substituted [2.2]paracyclophanes by reaction with formami-
dine (21a) or benzamidine (21b) in moderate to good yield.
For 18, 19 and 21b the molecular structures by X-ray crystallo-
graphy confirm the pseudo-para-disubstitution of the [2.2]para-
cyclophane backbone (see supporting information for details
about the molecular structures of 18 and 19). The molecular
structure of 21b for one of the two independent molecules per unit
cell is shown in Figure 3. The observed conformations of the
bispyridyl-[8b] and bispyrimidyl-substituted [2.2]paracyclophanes
can be explained by minimization of the dipoles and packing
effects in the solid state.
Figure 3: Molecular structure of 21b (displacement parameters are drawn at
50% probability level). For clarity only one of the two independent molecules per
unit cell is shown.
Oxazolinyl-substituted [2.2]paracyclophanes. Moreover we
introduced oxazolinyl substituents into the [2.2]paracyclophane
scaffold. Monooxazolinyl-substituted [2.2]paracyclophanes 24
are well-known and could be readily prepared from the carboxy-
[2.2]paracyclophane 22 (Scheme 5).[7, 17]
Scheme 5: Synthesis of oxazolinyl-substituted [2.2]paracyclophanes 24.
Scheme 4: Synthesis of pseudo-para-bispyrimidyl-substituted [2.2]paracyclo-
phanes 6b with the diacetylated [2.2]paracyclophane 18 as key intermediate.
Scheme 6: Synthesis of bisoxazolinyl-substituted [2.2]paracyclophanes 25
starting from dicarboxylic acid 17.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
The analogous reaction sequence was applied for the synthesis
of the pseudo-para-bisoxazolines, starting from the pseudo-para-
dicarboxylic acid 17 (Scheme 6). This reaction illustrates clearly
the peculiarity of planar-chiral substituted [2.2]paracyclophanes.
Through twofold reaction of the achiral acid 17[18] with the (S)-
amino alcohol, the original Ci symmetry is broken, resulting one
chiral (S,Sp,Rp,S)-diastereomer of 25a and b.
The yield of 26 is far from satisfactory, so we investigated
alternative methods to form ruthenium(II) complexes. The
literature contains a multitude of possible solutions[20] with the
most favourable appearing to be the preparation of cationic
complexes as reported by Pfeffer and Le Lagadec.[21]
Cycloruthenation of the [2.2]paracyclophane scaffold. After
completing our ligand synthesis we focused our efforts on the
synthesis of the intended ruthenium(II) complexes. Numerous
procedures exist in the literature for the synthesis of neutral
ruthenium(II) complexes of 2-phenylpyridines, such as 3, in high
yields (>90%), each using 0.5 equiv. of the dimeric ruthenium
precursor and a variable amount of a base.[10b, 19] Therefore we
tried analogous reaction conditions for the cycloruthenation of our
racemic 2-pyridyl-substituted [2.2]paracyclophane (rac)-8, but our
efforts did not result in yields higher than 9% (Scheme 7, see
supporting information for more details).
Scheme 8: Synthesis of the cationic ruthenium(II) complex 27. The relative
configuration is shown.
By addition of potassium hexafluorophosphate to the reaction
mixture, we were able to approach cycloruthenated [2.2]para-
cyclophanes in good yields (Scheme 8). The cationic ruthe-
nium(II) complex 27 is air- and moisture stable for weeks as solid
and as solution in acetonitrile. Through coordination of ruthe-
nium(II) to the ligand 8, an additional stereogenic centre is
generated, resulting in two possible diastereomers for 27. As with
the neutral complex 26, only one diastereomer was observed for
27 via 1H and 13C NMR spectroscopy, indicating again the diaste-
reoselective formation of the diastereomer with the p-cymene
ligand pointing away from the [2.2]paracyclophane backbone.
Scheme 7: Cycloruthenation of 8 with [RuCl2(p-cymene]2.
Coordination of 8 to the ruthenium(II) centre creates a new
stereogenic centre as the metal has a pseudo-tetrahedral “piano
stool” coordination geometry. In conjunction with the stereogenic
plane of [2.2]paracyclophane this could potentially lead to dia-
stereomers. The NMR spectra do not show any evidence for a
diastereomeric product mixture, indicating that a diastereo-
selective cycloruthenation took place. A plausible reason for a
single diastereomer involves the interaction of the [2.2]paracyclo-
phane framework and the p-cymene ligand. With both
coordinated to the ruthenium centre, and the [2.2]paracyclophane
in a bidentate fashion, there is no freedom in the (Rp*,S*)-dia-
stereomer for the p-cymene ligand to avoid non-bonding inter-
actions with the core of the paracyclophane (Figure 4). In contrast,
in the (Rp*,R*)-diastereomer the p-cymene is orientated away
from both decks of the paracyclophane and is therefore the more
stable product. For this discussion only the relative configuration
is stated, and all contemplations pertain inversely to the other
enantiomer.
Figure 5: Molecular structure of 27 (displacement parameters are drawn at 50%
probability level). For clarity the anion has been removed. Characteristic bond
lengths: Ru–NPy 2.046(3) Å; Ru–CPC 2.057(4) Å; Ru–NMeCN 2.077(3) Å; Ru–
Ccymene 2.185(4)–2.302(4) Å, Ru–cymene 1.726(4) Å. Characteristic bond
angles: NPy–Ru–CPC 76.95(14)°, NPy–Ru–NMeCN 85.61(13)°, NMeCN–Ru–CPC
91.04(14)°, NPy–Ru–cymene 132.1(2)°, NMeCN–Ru–cymene 127.2(2)°, CPC–Ru–
cymene 127.8(2)°. PC = [2.2]paracyclophane.
From the cationic complex 27 we could obtain single crystals that
were suitable for X-ray crystallography and confirm the stated
molecular structure of (Rp*,S*)-27 that is depicted in Figure 5. In
Figure 4: The two diastereomers of the ruthenium complex 26 and a possible
explanation for the preference for (Rp*,R*). The relative configuration is shown.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
the cationic complex 27, the twist of the pyridyl ring against the
[2.2]paracyclophane is significantly smaller compared to the free
ligand with angles of 16.4° and 37.4°, respectively. The coordi-
nation geometry around the ruthenium(II) centre shows the
Moreover, the chiral para-bispyridyl-substituted [2.2]paracyclo-
phane 13 can also undergo cycloruthenation under comparable
reaction conditions in 56% yield. The dinuclear ruthenium
complex 29 could also form a mixture of diastereomers,
expected piano stool and can be described as pseudo-tetrahedral. analogous to those discussed above. This might be further
The NPy–Ru–CPC angle [76.95(14)°] is significantly smaller than
the NPy–Ru–NMeCN and the CPC–Ru–NMeCN angle [85.61(13)° and
91.04(14)°]. These observations are most widely in agreement
with those of cycloruthenated 2-phenylpyridine,[19c, 21e] but with a
more distorted ruthenapyrrole in 27. A characteristic of 27 is the
shorter Ru–NPy bond [2.046(3) Å] compared to those in the
literature.
compounded by the fact bispyridine 13 is chiral but a racemate.
The NMR spectra suggest that a single, C2-symmetric, diastereo-
mer is formed (Scheme 10). As the racemic ligand (rac)-13 was
used, only the relative configuration is discussed.
Scheme 11: Attempted cycloruthenation of pyrimidyl-substituted [2.2]para-
cyclophanes 16 to yield complexes 30.
Next, we subjected the pyrimidyl-substituted [2.2]paracyclopha-
nes to the cycloruthenation conditions (Scheme 11). For the
unsubstituted pyrimidine 16a, no conversion of the ligand to a
ruthenium complex could be observed. The phenyl-substituted
pyrimidine 16b does undergo C-H activation with [RuCl2(p-
cymene)]2 but not at the desired position. Instead reaction occurs
at the ortho-position of the phenyl ring. Based on 1H and 13C NMR
spectroscopy it appears that reaction occurs with complete
regioselectivity but gives a 1:1 mixture of the two possible
diastereomers differing by the relative stereochemistry between
the [2.2]paracyclophane and the metal centre (please see
supporting information for detailed information).
Scheme 9: Synthesis of the dinuclear ruthenium complex 28 from achiral 9.
For the synthesis of homobimetallic complexes, we subjected the
pseudo-para-bispyridyl-substituted [2.2]paracyclophane 9 to the
analogous cycloruthenation conditions. After the reaction of the
achiral 9 with [RuCl2(p-cymene)]2 the dinuclear ruthenium(II)
complex 28 was isolated in 39% yield (Scheme 9). Due to the
pseudo-tetrahedral coordinated ruthenium, two additional stereo-
genic centres are generated while the planar chirality of the
[2.2]paracyclophane stays unaffected. In principle the formation
of four configuration isomers is possible. However, the 13C NMR
spectrum of 28 shows only 25 signals for the 50 carbon atoms in
the molecule referring to a higher symmetry in the molecule and
therefore chemically equivalent phenyl rings of the [2.2]paracyclo-
phane as well as p-cymene and acetonitrile ligands. The
assumption that the p-cymene ligands in 28 point away from the
sterically demanding [2.2]paracyclophane backbone is in agree-
ment with the observed molecular structure of 27 and an achiral
meso-compound 28 with an inversion centre is obtained.
Scheme 12: Suggested structure of the obtained ruthenium(II) complex 31 after
C-H activation of the phenyl substituent.
For the formation of two isomers, two possibilities have to take
into account, with a regioselective but not diastereoselective
reaction as the first option. The alternative is a not regioselective
but diastereoselective reaction. Based on steric considerations
with regards to bulky substituents and directing groups together
Scheme 10: Synthesis of the dinuclear ruthenium complex 29 from the chiral
para-disubstituted (rac)-13. The relative configuration is shown.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
with the observed 1:1 ratio and a lack of cross peaks between the
protons of the [2.2]paracyclophane backbone and the p-cymene
ligand in a NOESY NMR experiment, a regioselective reaction for
the N1’ nitrogen atom combined with a not diastereoselective
reaction to a complex 31 seems to be the most probable option
(Scheme 12, see supporting information for more details).
two meso compounds both have an inversion centre which result
in only half the amount of signals with double intensity. In
summary the signals for the three diastereomers seem to be a
“double signal set” with a 1:1 ratio which is visualized in Figure 6.
The observed behaviour of the pyrimidyl-substituted [2.2]para-
cyclophanes 16 in the cycloruthenation reaction shows the
decreased reactivity for C-H activation of the [2.2]paracyclophane
derivatives compared to the phenyl analogues. Another evidence
is the low yield of the neutral complex 26 compared to the phenyl
analogue 3. The decreased reactivity might be explained by a
combination of steric and electronic effects of the electron-rich,
bulky [2.2]paracyclophane scaffold that impede addition of the
ruthenium and the proposed intermolecular deprotonation[10b, 22]
.
Figure 6: Verification of the observed number of signals in the NMR spectra of
32 (top) and an enlarged section of the aromatic region of the 13C NMR spectrum
of 32 (bottom).
Scheme 13: Cycloruthenation of the pseudo-para bispyrimidyl-substituted
[2.2]paracycophane 21b.
Finally, we tested the oxazolinyl-substituted [2.2]paracyclophanes
for their ability to coordinate ruthenium(II). As with 2-phenyl-
pyridine, for 2-phenyloxazoline various cycloruthenated comple-
xes are known.[11, 19c, 22-23] We subjected the tert-butyl-substituted
oxazoline 24b the cycloruthenation conditions, as only this
oxazoline that was obtained diastereomerically pure. For the
formation of the respective ruthenium complexes, a reactivity
dependent on the used diastereomer was observed. Thus we
isolated the ruthenium complex (R,Sp,S)-33b by reaction of
(Rp,S)-24b in good yield of 71% whereas with the diastereomeric
(Sp,S)-24b we obtained only a mixture of several unidentifiable
species (Scheme 14). This result can be explained by the spatial
arrangement of the tert-butyl group relative to the [2.2]paracyclo-
phane backbone and the p-cymene ligand. Reaction of (Rp,S)-
24b gives the diastereomeric complex (R,Sp,S)-33b in which the
tert-butyl substituent points towards the small acetonitrile ligand.
This is readily accommodated. The (Sp,S)-diastereomer would
result in the formation of a complex that had a highly disfavoured
non-bonding interaction between the tert-butyl group and the p-
cymene ligand. As a result, this complex does not form and a
series of side-reaction occurs instead.
However, we wanted to synthesize the respective dinuclear com-
plex of the bispyrimidyl-substituted [2.2]paracyclophane 21b
(Scheme 13). Therefore we used the analogous reaction condi-
tions and observed again the C-H-activation of the phenyl
substituents for the dinuclear product 32. Similar to the pseudo-
para bispyridyl-substituted [2.2]paracyclophane 9, the ligand 21b
itself is achiral. Analogous to 28, two new stereogenic centres are
formed via pseudo-tetrahedral coordination of two ruthenium(II)
centres while the planar-chiral configuration at the [2.2]paracyclo-
phane is retained. That enables the formation of three diastereo-
mers (see supporting information for details) that are distinguish-
able by NMR spectroscopy. Indeed, the NMR spectra of 32
displays only the double amount of signals referring to the number
of atoms with a ratio of 1:1 and is exemplarily shown for some
aromatic carbon atoms (Figure 6). If the statistical formation of
the possible stereoisomers is assumed, all of them should be
generated with the same probability. For the two chiral configu-
ration isomers one signal is obtained per atom of the molecule.
As enantiomers are not distinguishable via NMR spectroscopy,
they result in a full signal set with double intensity. In contrast, the
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
The dinuclear complex (R,Sp,S)-34 possesses a cycloruthenated
oxazolinyl-substituted [2.2]paracyclophane but with a second
ruthenium(II) center that is η6-coordinated by the second deck of
the [2.2]paracyclophane forming a sandwich-type complex (Figure
7). [2.2]Paracyclophanes as η6-coordinating arene ligands for
ruthenium(II) are known since the initial reports particularly by
Boekelheide et al.[24] in the 1980s and have been extensively
studied.[25] Nevertheless, among the numerous examples of a
cycloruthenation with KPF6 to a cationic complex, the literature
presents only a single report for the formation of a similar tri-
cationic product.[26] This complex was formed as further reaction
of a cyclometallated 2-phenylpyridine derivative with another (η6-
benzene)Ru2+ fragment that is only occurring for cycloruthenation
products with sufficient instability. The equilibrium can be shifted
towards the tri-cation by an excess of the ruthenium(II) pre-
cursor.[26] But unfortunately, all our attempts to a targeted synthe-
sis of 34 by an excess of [RuCl2(p-cymene)]2 were unsuccessful
and gave only the cycloruthenated di-cationic mononuclear
complex, which was also the case for the reaction of the pyridyl-
substituted 8 with an excess [RuCl2(p-cymene)]2. That prompts
the conclusion that the tri-cationic dinuclear complex is a
decomposition product that was formed during the crystallization
process, which was not performed under argon atmosphere.
Hoping that only one diastereomer reacts, we subjected the inse-
parable diastereomeric mixture of 24a for the cycloruthenation.
Ideally only the (Rp,S)-24a diastereomer would react analogous
to the observed reactivity for the diastereomers of 24b in a
diastereoselective reaction, so that only one of the diastereomeric
products 33a would be formed (Scheme 15). Indeed, only the
product of the non-diastereoselective cycloruthenation could be
obtained with a ratio of 1.18:1. This is explainable by the smaller
and more flexible isopropyl group compared to the bulky tert-butyl
group that is able to avoid steric hindrance by rotation.
Scheme 14: Synthesis of the ruthenium complex 33b from diastereomeric
oxazolines 24b.
From a sample of (R,Sp,S)-33b we were able to grow single
crystals that were suitable for Xray crystallography. Interestingly
the obtained molecular structure from one of these crystals does
not show the structure of (R,Sp,S)-33b that was determined by
NMR spectroscopy and mass spectrometry, but a tri-cationic
molecule (R,Sp,S)-34 with two coordinated ruthenium(II) units.
Scheme 15: Synthesis of the ruthenium(II) complex 33a starting from oxazoline
24a as diastereomeric mixture.
Figure 7: Instead the determined structure of 33b, the measured crystal showed
a tri-cationic dinuclear ruthenium complex 34 in X-ray crystallography (displace-
ment parameters are drawn at 50% probability level). For clarity only one of the
two independent molecules per unit cell is shown and the PF6- counter ions and
solvent have been removed. Characteristic bond lengths for both independent
molecules: Ru–NOx 2.120(8) and 2.093(8) Å; Ru–CPC 2.083(9) and 2.085(10) Å;
Ru–NMeCN 2.059(8) and 2.067(9) Å; Ru–Ccymene 2.164(10)–2.279(10) and
2.167(9)–2.293(11) Å, Ru2–CPC 2.203(9)–2.353(8) and 2.199(11)–2.375(10) Å;
Ru2–Ccymene2 2.190(9)–2.243(9) und 2.202(11)–2.254(12) Å. Characteristic
bond angles: NOx–Ru–CPC 78.2(3)° and 78.2(3)°, NOx–Ru–NMeCN 85.2(3)° and
85.8(3)°, NMeCN–Ru–CPC 90.3(3)° and 91.1(3)°. The oxazolinyl unit and the
[2.2]paracyclophane are twisted by 14.7° and 19.9° against each other.
Afterwards we investigated the reaction of [RuCl2(p-cymene)]2
with the pseudo-para-bisoxazolines 25. These molecules features
both, a (Rp)- and a (Sp)-configured ring combined with an (S)-
oxazoline ring that possesses an isopropyl or a tert-butyl
substituent. In both cases, only the mononuclear ruthenium
complexes 35 was isolated in good yields (>74%, Scheme 16),
which is consistent with the above-mentioned reactivity for the two
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
diastereomeric tert-butylsubstituted monooxazolines 24b. For the
bisoxazolines 25, only the (Rp)-ring of the [2.2]paracyclophane in
combination with the (S)-configuration of the oxazoline as starting
material enables the stable coordination to the ruthenium metal.
However, this result seems surprising regarding the observed
reactivity for the [2.2]paracyclophane-based bisoxazoline 25a
with isopropyl substituents, as the respective diastereomeric
monooxazolines 24a do not show diastereospecific cyclo-
ruthenation upon reaction with [RuCl2(p-cymene)]2. Therefore,
ongoing research in our laboratories is necessary to figure out if
a diastereoselective reaction for the monooxazoline 24a or the
double cycloruthenation of 25a is obtained by further optimization
of the reaction conditions.
Experimental Section
General cycloruthenation reaction to the cationic complexes: In a
schlenk tube the [2.2]paracyclophane-based ligand (1.00 equiv.),
[RuCl2(p-cymene)]2 (1.00 equiv. or 0.50 equiv.), KOAc (1.50 equiv. or 3.00
equiv.) and KPF6 (2.00 equiv. 1.50 equiv.) in dry acetonitrile (5 mL) were
stirred at r.t. for 3 d. The solvent was removed under reduced pressure
and the residue was purified by column chromatography on silica gel
(cyclohexane/ethyl acetate 50/50 → acetonitrile) to obtain the ruthenium
complex as yellow solid.
Crystal Structure Determinations
The single-crystal X-ray diffraction study was carried out on an Agilant
SuperNova diffractometer with EOS detector at 173(2) K using Mo-Kα
radiation (λ = 0.71073 Å, 12), a Bruker D8 Venture diffractometer with
Photon100 detector at 123(2) K using Cu-Kα radiation (λ = 1.54178 Å, 13,
16a, 34) or using Mo-Kα radiation (λ = 0.71073 Å, 21b, 27) and a Bruker-
Nonius KappaCCD at 123(2) K using Mo-Kα radiation (λ = 0.71073 Å, 18,
19). Direct Methods (SHELXS-97)[27] or dual space methods (SHELXT for
27, 34)[28] were used for structure solution and refinement was carried out
using SHELXL-2013 or SHELXL-2014 (full-matrix least-squares on F2)[29]
.
Hydrogen atoms were localized by difference electron density
determination and refined using a riding model. An analytical absorption
correction was applied for 12, semi-empirical absorption corrections were
applied for 13, 16a, 18, 19, 21b, 27 and 34. For 13 and 16a an extinction
correction was applied. The absolute structure of 27 and the absolute
configuration of 34 were determined by refinement of Parsons’
x-parameter[30]. In 19 are 2 x 0.5 molecules and one complete molecule in
the asymmetric unit, in the complete molecule one acetyl O-atom is
disordered. In 27 and 34, respectively one anion PF6 is disordered. In 27
the anion (due to the symmetry of the space group) is localized on three
positions, one ordered, one disordered and on the third position the
refinement with the listed atoms show residual electron density due to a
heavily disordered PF6 anion which could not be refined with split atoms.
Therefore the option "SQUEEZE" of the program package PLATON[31]was
used to create a hkl file taking into account the residual electron density in
the void areas. Therefore the atoms list and unit card do not agree. In 34
the refinement with the listed atoms show residual electron density due to
three heavily disordered water molecules which could not be refined with
split atoms. The option "SQUEEZE" of the program package PLATON[31]
was used to create an hkl file taking into account the residual electron
density in the void areas. The water molecules are not included in the
atoms list and unit card (for the refinement of the disorder and SQUEEZE
of 27 and 34 see the cif-file for details).
Scheme 16: Reaction of the pseudo-para bisoxazolines 25 to the mononuclear
ruthenium complexes 35.
Conclusions
The planar-chiral [2.2]paracyclophane backbone is a versatile
hydrocarbon scaffold that enables the synthesis of unique 3D
architectures due to its unique molecular geometry. Therefore we
explored synthetic routes to pseudo-para- or para-disubstituted
[2.2]paracyclophanes with additional N-donor groups such as
pyridines, pyrimidines and oxazolines. By subjecting the obtained
N-donor-substituted [2.2]paracyclophanes to common cyclo-
ruthenation conditions under C-H activation, it can be stated that
the potential of the [2.2]paracyclophane to undergo cycloruthe-
nation is obviously inferior to a phenyl ring. Thus, only the cationic
ruthenium complexes of substituted [2.2]paracyclophanes could
be isolated in satisfactory yields. In case of a 2-pyridyl substituent
as N-donor, the cycloruthenation proceeded diastereoselectively
at the pseudo-tetrahedral coordinated ruthenium centre, pre-
ferring the configuration where the p-cymene ligand points away
from the [2.2]paracyclophane backbone. For a pyrimidine substi-
tuent as N-donor, no cycloruthenation of the [2.2]paracyclophane
backbone took place, but instead the C-H activation of a simple
aromatic substituent at the pyrimidine. For oxazoline substituents
diastereoselective cycloruthenation was observed that was in
addition diastereoselective for oxazolines with a bulky tert-butyl
substituent.
13: colourless crystals, C26H22N2, Mr = 362.45, crystal size 0.24 × 0.20 ×
0.08 mm, orthorhombic, space group Pccn (No. 56), a = 11.6872(4) Å, b =
42.5118(13) Å, c = 7.3758(2) Å, V = 3664.6(2) Å3, Z = 8, ρ = 1.314 Mg/m-
3, µ(Cu-Kα) = 0.589 mm-1, F(000) = 1536, 2max = 144.4°, 22288 reflections,
of which 3600 were independent (Rint = 0.025), 254 parameters, R1 = 0.037
(for 3362 I > 2σ(I)), wR2 = 0.098 (all data), S = 1.03, largest diff. peak /
hole = 0.275 / -0.170 e Å-3.
21b: colourless crystals, C36H28N4 ∙ CH2Cl2, Mr = 601.55, crystal size 0.16
× 0.14 × 0.02 mm, triclinic, space group P-1 (No. 2), a = 7.2898(4) Å, b =
13.8443(8) Å, c = 15.2534(9) Å, α = 73.925(3)°, ꢀ = 77.739(2)°, ꢁ =
78.526(2)°, V = 1429.29(14) Å3, Z = 2, ρ = 1.398 Mg/m-3, µ(Mo-Kα) = 0.263
mm-1, F(000) = 628, 2max = 55.2°, 50346 reflections, of which 6589 were
independent (Rint = 0.049), 388 parameters, R1 = 0.038 (for 5433 I > 2σ(I)),
wR2 = 0.098 (all data), S = 1.02, largest diff. peak / hole = 0.310 / -0.453 e
Å-3.
In summary we showed that substituted [2.2]paracyclophanes are
able to act as bridging ligands for dinuclear ruthenium complexes
if substituted with a suitable N-donor ligand. This enables the
synthesis of rigid dinuclear complexes with defined distance and
orientation between the two metal centres which are not
accessible by commonly used planar ligand structures.
This article is protected by copyright. All rights reserved.

10.1002/chem.201703291
Chemistry - A European Journal
FULL PAPER
[12]
[13]
a) G. J. Rowlands, R. J. Seacome, Beilstein J. Org. Chem. 2009,
5, No. 9; b) A. Pelter, H. Kidwell, R. A. N. C. Crump, J. Chem.
Soc., Perkin Trans. 1997, 3137-3140; c) M. Brink, Synthesis 1975,
807-808.
a) E. Bejan, H. A. Haddou, J. C. Daran, G. G. A. Balavoine,
Synthesis 1996, 1012-1018; b) F. Pezet, L. Routaboul, J.-C.
27: yellow crystals, C33H35N2Ru ∙ PF6, Mr = 705.67, crystal size 0.22 × 0.18
× 0.12 mm, trigonal, space group P31c (No. 159), a = 17.0065(6) Å, c =
18.2539(6) Å, V = 4572.1(4) Å3, Z = 6, ρ = 1.538 Mg/m-3, µ(Mo-Kα) = 0.630
mm-1, F(000) = 2160, 2max = 55.2°, 91047 reflections, of which 6947 were
independent (Rint = 0.033), 371 parameters, 1 restraint, R1 = 0.028 (for
6709 I > 2σ(I)), wR2 = 0.068 (all data), S = 1.07, largest diff. peak / hole =
0.847 / -0.819 e Å-3, x = -0.028(5).
̈
Daran, I. Sasaki, H. Aıt-Haddou, G. G. A. Balavoine, Tetrahedron
2000, 56, 8489-8493; c) K. Muller, A. Schubert, T. Jozak, A.
Ahrens-Botzong, V. Schünemann, W. R. Thiel, ChemCatChem
2011, 3, 887-892; d) L. Taghizadeh Ghoochany, C. Kerner, S.
Farsadpour, F. Menges, Y. Sun, G. Niedner-Schatteburg, W. R.
Thiel, Eur. J. Inorg. Chem. 2013, 4305-4317.
34: yellow crystals, C45H57N2ORu2 ∙ 3 PF6 ∙ CHCl3, Mr = 1398.34, crystal
size 0.18 × 0.10 × 0.02 mm, triclinic, space group P1 (No. 1), a =
10.6537(3) Å, b = 15.7694(4) Å, c = 16.4628(4) Å, α = 87.271(1)°, ꢀ =
85.346(1)°, ꢁ = 81.364(1)°, V = 2723.66(12) Å3, Z = 2, ρ = 1.705 Mg/m-3,
µ(Cu-Kα) = 7.574 mm-1, F(000) = 1404, 2max = 144.8°, 31716 reflections,
of which 16299 were independent (Rint = 0.039), 1344 parameters, 3409
restraints, R1 = 0.055 (for 15521 I > 2σ(I)), wR2 = 0.147 (all data), S = 1.03,
largest diff. peak / hole = 2.333 (near Ru atoms) / -1.030 (near disordered
P-atom) e Å-3, x = -0.034(9).
[14]
[15]
M. Busch, M. Cayir, M. Nieger, W. R. Thiel, S. Bräse, Eur. J. Org.
Chem. 2013, 6108-6123.
a) E. A. Truesdale, D. J. Cram, J. Org. Chem. 1980, 45, 3974-
3981; b) J. E. Gready, T. W. Hambley, K. Kakiuchi, K. Kobiro, S.
Sternhell, C. W. Tansey, Y. Tobe, J. Am. Chem. Soc. 1990, 112,
7537-7540.
V. I. Rozenberg, N. V. Dubrovina, E. V. Vorontsov, E. V. Sergeeva,
Y. N. Belokon', Tetrahedron: Asymmetry 1999, 10, 511-517.
A. Marchand, A. Maxwell, B. Mootoo, A. Pelter, A. Reid,
Tetrahedron 2000, 56, 7331-7338.
[16]
[17]
[18]
[19]
H. Allgeier, M. G. Siegel, R. C. Helgeson, E. Schmidt, D. J. Cram,
J. Am. Chem. Soc. 1975, 97, 3782-3789.
CCDC 1558680 (12), 1558681 (13), 1558682 (16a), 1558683 (18),
1558684 (19), 1558685 (21b), 1558686 (27), and 1558687 (34) contain
the supplementary crystallographic data for this paper. These data can be
obtained free of charge from The Cambridge Crystallographic Data Centre
via www.ccdc.cam.ac.uk/data_request/cif.
a) Q. Yu, L. Hu, Y. Wang, S. Zheng, J. Huang, Angew. Chem. Int.
Ed. 2015, 54, 15284-15288; Angew. Chem. 2015, 127, 15499-
15503; b) A. J. Paterson, S. St John-Campbell, M. F. Mahon, N. J.
Press, C. G. Frost, Chem. Commun. 2015, 51, 12807-12810; c) B.
Li, C. Darcel, T. Roisnel, P. H. Dixneuf, J. Organomet. Chem.
2015, 793, 200-209; d) N. Hofmann, L. Ackermann, J. Am. Chem.
Soc. 2013, 135, 5877-5884.
[20]
[21]
a) H. C. L. Abbenhuis, M. Pfeffer, J. P. Sutter, A. de Cian, J.
Fischer, H. L. Ji, J. H. Nelson, Organometallics 1993, 12, 4464-
4472; b) S. Attar, J. H. Nelson, J. Fischer, A. de Cian, J.-P. Sutter,
M. Pfeffer, Organometallics 1995, 14, 4559-4569; c) M. Pfeffer, J.-
P. Sutter, E. P. Urriolabeitia, Inorg. Chim. Acta 1996, 249, 63-67.
a) S. Fernandez, M. Pfeffer, V. Ritleng, C. Sirlin, Organometallics
1999, 18, 2390-2394; b) B. Boff, M. Ali, L. Alexandrova, N. Á.
Espinosa-Jalapa, R. O. Saavedra-Díaz, R. Le Lagadec, M. Pfeffer,
Organometallics 2013, 32, 5092-5097; c) A. D. Ryabov, V. S.
Sukharev, L. Alexandrova, R. Le Lagadec, M. Pfeffer, Inorg.
Chem. 2001, 40, 6529-6532; d) R. Le Lagadec, L. Rubio, L.
Alexandrova, R. A. Toscano, E. V. Ivanova, R. Meškys, V.
Laurinavičius, M. Pfeffer, A. D. Ryabov, J. Organomet. Chem.
2004, 689, 4820-4832; e) V. Martínez Cornejo, J. Olvera Mancilla,
S. López Morales, J. A. Oviedo Fortino, S. Hernández-Ortega, L.
Alexandrova, R. Le Lagadec, J. Organomet. Chem. 2015, 799–
800, 299-310.
I. Fabre, N. von Wolff, G. Le Duc, E. Ferrer Flegeau, C. Bruneau,
P. H. Dixneuf, A. Jutand, Chem. Eur. J. 2013, 19, 7595-7604.
a) Y. Boutadla, D. L. Davies, R. C. Jones, K. Singh, Chem. Eur. J.
2011, 17, 3438-3448; b) W.-G. Jia, T. Zhang, D. Xie, Q.-T. Xu, S.
Ling, Q. Zhang, Dalton Trans. 2016, 45, 14230-14237.
a) E. D. Laganis, R. G. Finke, V. Boekelheide, Tetrahedron Lett.
1980, 21, 4405-4408; b) R. T. Swann, V. Boekelheide,
Tetrahedron Lett. 1984, 25, 899-900; c) R. T. Swann, A. W.
Hanson, V. Boekelheide, J. Am. Chem. Soc. 1984, 106, 818-819;
d) R. T. Swann, A. W. Hanson, V. Boekelheide, J. Am. Chem. Soc.
1986, 108, 3324-3334.
a) M. R. J. Elsegood, D. A. Tocher, J. Organomet. Chem. 1988,
356, C29-C31; b) M. R. J. Elsegood, D. A. Tocher, Inorg. Chim.
Acta 1989, 161, 147-149; c) M. R. J. Elsegood, J. W. Steed, D. A.
Tocher, J. Chem Soc., Dalton Trans. 1992, 1797-1801; d) P. J.
Dyson, B. F. G. Johnson, C. M. Martin, A. J. Blake, D. Braga, F.
Grepioni, E. Parisini, Organometallics 1994, 13, 2113-2117; e) B.
Gollas, B. Speiser, J. Sieglen, J. Strähle, Organometallics 1996,
15, 260-271; f) B. Gollas, B. Speiser, I. Zagos, C. Maichle-
Mössmer, J. Organomet. Chem. 2000, 602, 75-90; g) R. Bhalla, C.
J. Boxwell, S. B. Duckett, P. J. Dyson, D. G. Humphrey, J. W.
Steed, P. Suman, Organometallics 2002, 21, 924-928.
J.-P. Djukic, L. Fetzer, A. Czysz, W. Iali, C. Sirlin, M. Pfeffer,
Organometallics 2010, 29, 1675-1679.
Acknowledgements
We thank the SFB/TR 88 for financial support and N. B. Heine for
his help with the synthesis of 12. C. B. gratefully acknowledges
the Evonik Stiftung for financial support.
Keywords: Cyclophanes • Ruthenium • Chirality •
Diastereoselectivity • Metallacycle
[1]
[2]
C. J. Brown, A. C. Farthing, Nature 1949, 164, 915-916.
a) H. J. Reich, D. J. Cram, J. Am. Chem. Soc. 1967, 89, 3078-
3080; b) H. J. Reich, D. J. Cram, J. Am. Chem. Soc. 1968, 90,
1365-1367; c) H. J. Reich, D. J. Cram, J. Am. Chem. Soc. 1969,
91, 3534-3543; d) H. J. Reich, D. J. Cram, J. Am. Chem. Soc.
1969, 91, 3527-3533; e) H. J. Reich, D. J. Cram, J. Am. Chem.
Soc. 1969, 91, 3517-3526; f) H. J. Reich, D. J. Cram, J. Am.
Chem. Soc. 1969, 91, 3505-3516.
[22]
[23]
[24]
[3]
[4]
[5]
O. R. P. David, Tetrahedron 2012, 68, 8977-8993.
D. J. Cram, J. M. Cram, Acc. Chem. Res. 1971, 4, 204-213.
a) S. E. Gibson, J. D. Knight, Org. Biomol. Chem. 2003, 1, 1256-
1269; b) S. Bräse, S. Dahmen, S. Höfener, F. Lauterwasser, M.
Kreis, R. E. Ziegert, Synlett 2004, 2647-2669; c) G. J. Rowlands,
Org. Biomol. Chem. 2008, 6, 1527-1534; d) J. Paradies, Synthesis
2011, 3749-3766; e) G. J. Rowlands, Isr. J. Chem. 2012, 52, 60-
75.
[25]
[6]
a) H. Hopf, Angew. Chem. Int. Ed. 2008, 47, 9808-9812; Angew.
Chem. 2008, 120, 9954-9958; b) E. Elacqua, L. R. MacGillivray,
Eur. J. Org. Chem. 2010, 6883-6894.
[7]
[8]
C. Bolm, K. Wenz, G. Raabe, J. Organomet. Chem. 2002, 662, 23-
33.
a) J. R. Fulton, J. E. Glover, L. Kamara, G. J. Rowlands, Chem.
Commun. 2011, 47, 433-435; b) C. Braun, E. Spuling, N. B. Heine,
M. Cakici, M. Nieger, S. Bräse, Adv. Synth. Catal. 2016, 358,
1664-1670.
[26]
[9]
J. E. Glover, P. G. Plieger, G. J. Rowlands, Aust. J. Chem. 2014,
67, 374-380.
[27]
[28]
[29]
[30]
G. M. Sheldrick, Acta Crystallogr. A 2008, 64, 112-122.
G. M. Sheldrick, Acta Crystallogr. A 2015, 71, 3-8.
G. M. Sheldrick, Acta Crystallogr. C 2015, 71, 3-8.
S. Parsons, H. D. Flack, T. Wagner, Acta Crystallogr. B 2013, 69,
249-259.
A. Spek, Acta Crystallogr. D 2009, 65, 148-155.
[10]
a) J.-P. Djukic, A. Berger, M. Duquenne, M. Pfeffer, A. de Cian, N.
Kyritsakas-Gruber, J. Vachon, J. Lacour, Organometallics 2004,
23, 5757-5767; b) E. Ferrer Flegeau, C. Bruneau, P. H. Dixneuf, A.
Jutand, J. Am. Chem. Soc. 2011, 133, 10161-10170.
D. L. Davies, O. Al-Duaij, J. Fawcett, K. Singh, Organometallics
2010, 29, 1413-1420.
[11]
[31]
This article is protected by copyright. All rights reserved.