Severe energy costs of double steric interactions: Towards a molecular clamp

Article abstract of DOI:10.1002/ejoc.201000847

The factors determining the ease of rotation about carboncarbon single bonds connecting two internally rigid fragments such as phenyl, indenyl, anthracenyl and triptycyl are analysed. The internal rotation barriers in these molecules have been estimated o

Full text of DOI:10.1002/ejoc.201000847

FULL PAPER
DOI: 10.1002/ejoc.201000847
Severe Energy Costs of Double Steric Interactions: Towards a Molecular Clamp
Kirill Nikitin,*[a] Conor Fleming,[a] Helge Müller-Bunz,[a] Yannick Ortin,[a] and
Michael J. McGlinchey*[a]
Keywords: Rotational barrier / Steric hindrance / Strained molecules / Molecular machinery / Molecular devices
The factors determining the ease of rotation about carbon–
carbon single bonds connecting two internally rigid frag-
ments such as phenyl, indenyl, anthracenyl and triptycyl are
analysed. The internal rotation barriers in these molecules
have been estimated on the basis of kinetic data or variable-
temperature NMR measurements, and the crystal structures
have been analysed in terms of steric strain. Computer simu-
lation of the internal rotation indicates that the estimated
Closest Approach Distance, CAD, between sterically inter-
acting atoms of the two interconnected fragments can be a
helpful parameter for evaluating their rotational freedom, but
must be used with caution. Thus, the barrier to rotation of a
3-indenyl moiety linked to the 9-position of anthracene is
very high (∆G϶ ≈ 25 kcalmol–1) compared to that in 3-inden-
yltriptycene (∆G϶ ≈ 12 kcalmol–1) despite the fact that the
nominal CADs in both cases are very similar. Moreover, di-
meric 2-methylindenyl fragments linked by a single bond at
the 3-position undergo relatively slow rotation (∆G϶ ≈ 14–
15 kcalmol–1) owing to the simultaneous close approach of
two pairs of sterically interacting hydrogens. Although the
rotational barrier for a 2-indenyl or phenyl moiety attached
to the bridgehead atom of triptycene, or to the related di-
benzobicyclo[2.2.2]octane system, is relatively low (∆G϶ ≈ 8–
9 kcalmol–1), further extension of the bridge to dibenzobi-
cyclo[2.2.4]dioxadecane leads to an activation energy barrier
in excess of 23 kcalmol–1, attributable to an intramolecular
simultaneous “clamping” of the phenyl rings by the edges of
the aromatic rings of the dibenzobicyclo[2.2.4]dioxadecane
moiety. The X-ray crystal structures of 15 molecules, includ-
ing mono- and di-indenyl-anthracenes, racemic- and meso-
2-methylindenyl dimers, phenyl- and indenyl-triptycenes
and -barrelenes, are reported.
rotation of the paddlewheel is typically hindered by steric
repulsion between hydrogen atoms, or other molecular frag-
ments proximate to the central carbon–carbon single bond,
control of the barrier height may be achieved by structural
modification, including functional group transformation or
isomerisation. Another appealing structural class includes
flat polycyclic aromatic hydrocarbon systems that also pos-
sess tetrahedral carbon centres. Importantly, such struc-
tures, e.g. indene or fluorene, can be readily converted into
π complexed transition metal derivatives that undergo re-
versible haptotropic rearrangements involving a shuttle-like
translational movement of the metal fragment.[7]
We have recently reported the syntheses and dynamic be-
haviour of two isomers, 9-(3-indenyl)triptycene (1a) and 9-
(2-indenyl)triptycene (1b) in which the originally D3h-sym-
metric triptycene is connected to the mirror-symmetric in-
dene via an sp3–sp2 carbon–carbon single bond. Variable-
temperature NMR studies revealed that, while the rota-
tional barrier, ∆G϶, in 1a is 12 kcalmol–1, rotation in 1b is
still fast on the NMR time-scale at –80 °C, suggesting a
barrier no greater than 8 kcalmol–1.[8]
Introduction
Rotation about sp3–sp3 carbon–carbon single bonds in
non-hindered molecules is very rapid at ambient tempera-
ture.[1] Hence, rotations in alkanes are normally too fast to
be studied by dynamic NMR at room temperature since the
barriers are less than 5 kcalmol–1.[2] However, it has been
demonstrated that, in appropriately substituted systems, ro-
tation about single bonds linking two aromatic carbon
atoms can be significantly slowed down, and even stopped
on a chemically significant time-scale.[3,4] In these cases,
barriers can exceed 25 kcalmol–1 and so open up the pos-
sibility of designing controlled molecular brakes, ratchets
and other functional molecular systems.[5,6]
The key to the successful implementation of the pro-
posed molecular design is the ability to tune the rotational
barriers while maintaining the required molecular geometry
and function. Particularly interesting as molecular building
blocks are the substituted triptycenes[6a] and their structural
analogues whereby C(9), a bridgehead tetrahedral carbon
of the three-bladed paddlewheel, is linked to another rigid
moiety, as in the examples depicted in Scheme 1. Since the
The complex interplay of factors determining rotational
barriers in a wide range of molecules is a fundamental topic
in modern organic chemistry,[9] and a better understanding
of such processes would be valuable for the design of com-
ponents in molecular machinery. For example, in a very re-
cent paper we described how the low-barrier system 1b
View this journal online at
[a] School of Chemistry & Chemical Biology University College
Dublin,
Belfield, Dublin 4, Ireland
Fax: +353-1-716-1178
E-mail: kirill.nikitin@ucd.ie
michael.mcglinchey@ucd.ie
wileyonlinelibrary.com
Eur. J. Org. Chem. 2010, 5203–5216
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
5203

K. Nikitin, M. J. McGlinchey et al.
FULL PAPER
Scheme 1. Potential steric clashes and rotational barriers in indenyltriptycenes 1a (12 kcal mol–1), 1b (Ͻ8 kcal mol–1), and in the isomeric
methanophenanthrenyl-anthracene derivative 1c (Ͼ18 kcal mol–1).
could be modified by complexation to the indenyl of a π-
bonded tricarbonylchromium moiety to work as the first
haptotropically-mediated system capable of functioning as
an organometallic molecular brake.[10] However, very little
is known about the interaction of unsubstituted rigid mo-
lecular fragments although even relatively small hydrogen
atoms can cause severe steric repulsion due to spatial prox-
imity and orientation.
In this paper we continue the study of polycyclic mole-
cules incorporating multi-bladed fragments such as trip-
tycene or barrelene, and flat aromatic groups, e.g. indenyl
or phenyl. The internal rotational barrier about the single
carbon–carbon linkage between these fragments was evalu-
Figure 1. In 9-(3-indenyl)triptycene (1a) rotation of the indenyl
moiety has to overcome a steric barrier of about 12 kcal mol–1 aris-
ing from repulsion between H(23) of the indene with the triptycene
ated by variable-temperature NMR spectroscopy, or by
monitoring the kinetics of isomerisation. The key finding is
that the rate of intramolecular rotation is sensitive to very
fine distinctions in the geometries and orientation of the
interconnected moieties.
hydrogens H(1), H(8) and H(13).
Results and Discussion
It has previously been shown that in 9-(3-indenyl)trip-
tycene (1a) the proximity of the six-membered ring of the
indenyl fragment to the triptycyl paddlewheel engenders a
12 kcalmol–1 rotation barrier;[8a] this contrasts with the low
barrier to rotation observed in 9-(2-indenyl)triptycene
(1b).[8b] This drastic difference can be readily rationalised
by examining the X-ray crystal structures of both isomers,
Figure 2. In 9-(2-indenyl)triptycene (1b) the H(1), H(8) and H(13)
need not approach closer than 1.5 Å to H(18), leading to a signifi-
cantly lower rotational barrier.[8b]
shown also in space-filling style in Figures 1 and 2, respec- 0.3 Å, while in the 2-indenyl isomer 1b the analogous
tively. H(1)···H(18) CAD is 1.5 Å. Of course, it is reasonable to
The intramolecular H···H repulsion interactions can be assume that a clash of these two hydrogens in 1a can be
treated in a manner similar to the approach first applied to partially alleviated by bending the indenyl moiety away at
2,2Ј-substituted biphenyls.[9] In their systematic study Bott, the opportune moment. Indeed, even in the observed X-ray
Field, and Sternhell showed that the severity of steric inter- crystal structure of 1a, whereby the indenyl is positioned
actions can be approximated by “apparent overlap”, r*, between two blades, and which presumably represents a po-
which is defined as the projection of the van der Waals radii tential energy minimum, the indenyl moiety is bent away
in a hypothetical structure where there is no distortion of bond from the threefold axis of the triptycyl such that the angle
angles and lengths.[9b,11] In our study, the experimentally de- C(10)–C(9)–C(17) is 172°.
termined molecular structure of 1a underwent a simple
While the rotational barriers in the 9-indenyltriptycenes
computer modelling procedure involving an analogous “vir- 1a and 1b apparently are determined by repulsion between
tual rotation” of the indenyl fragment about the indenyl- hydrogen atoms in the indenyl fragments and one of the
triptycyl single bond in the 3-indenyl isomer 1a with no blades of the triptycene, it is noteworthy that in the isomeric
other geometric changes. This indicated that the nominal methano-bridged phenanthrenyl-anthracene 1c (Figure 3)
Closest Approach Distance (CAD) for H(1)···H(23) is a mere the rotational barrier exceeds 18 kcalmol–1.[8b] In this latter
5204
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5203–5216

Towards a Molecular Clamp
case, an analysis of its structure indicates the simultaneous Table 1. Internal rotation barriers, ∆G϶, nominal Closest Approach
Distances, CADs, and r* values in the polycyclic systems 1–20.
approach of two pairs of hydrogen atoms with CAD values
∆G϶
H of first H of second CAD Overlap,[b] Ref.
of 0.3 and 1.2 Å leading to an estimated total steric overlap
of 3.3 Å. As noted by Sternhell,[9b] the severity of a non-
bonded interaction is not determined solely by the size of
[kcalmol–1
]
fragment
fragment
[Å]
r*
[8a]
1a
12Ϯ0.5
Ͻ8
1
1
1
9
1
8
1
8
8
7
7
8
1
1
1
14
18
1
23
18
20
12
17
12
12
17
7Ј
0.3
1.5
0.3
1.2
0.3
2.1
0.9
3.3
the interacting fragments, but also by the geometry of the 1b
c
[8b]
1c
Ͼ18[a]
system and its mode of relaxation, in particular its resis-
tance to bond bending or stretching.[11] Moreover, it was
[c]
[c]
[c]
[c]
3a
25Ϯ0.5[a]
3.3
2.4
3
suggested that one could predict rotational barriers by
1.2
using the additivity of pairwise steric contributions,[9b] but
this approach has since been questioned.[9d]
3b
13.5Ϯ0.5[a]
14.5Ϯ0.5[a]
1.15
1.25
0.9
0.9
0.9
0.8
0.7
1.05
1.0
0.7
1.65
0.85
1.6
0.9
0.95
rac-9
8Ј
meso-9 14.8Ϯ0.5[a]
16
15
18
14
14
1
3.1
[c]
[c]
[c]
[c]
10
13
14
15
Ͻ9
Ͻ9
1.7
1.35
1.4
3
Ͻ9
15.3Ϯ0.5[a]
O
[c]
[c]
[c]
16
19
20
Ͻ9
14
12
14
18
1.55
0.8
2.95
Ͻ9
4
1
8
Figure 3. Rotation of the anthracene moiety in the methano-
bridged phenanthrenyl derivative 1c has to overcome two simulta-
neous repulsion interactions, H(1)–H(20) and H(9)–H(12), such
that the barrier exceeds 18 kcal mol–1.
Ͼ23[a]
[a] Two close H···H contacts are attained simultaneously. [b] r* =
Σ(van der Waals radii) – Σ(CAD).[9b] [c] This work.
Parallels may be drawn with 2,2Ј,6,6Ј-tetrasubstituted bi-
phenyls in which rotation about the single bond linking the
aryl rings would engender two simultaneous steric interac-
tions; such systems yield configurationally stable atropiso-
mers. In contrast, analogous rotations in 2,2Ј-disubstituted
biphenyls may proceed via a transoid transition state,
whereby the bulky substituents can avoid each other.[4a,4e]
Accordingly, we chose to design, synthesise and study a
range of molecules in which two or more rigid polycyclic
systems were linked by a carbon–carbon single bond.
Herein we show that the rotational barrier can significantly
change depending on very subtle differences in the orienta-
tion and position of hydrogen atoms or other functional
groups attached to the core cyclic systems.
1
The H and 13C NMR spectra of 2a and 2b reveal that
the external rings of the anthracenyl fragment are equiva-
lent, even at low temperature; this is explicable by invoking
a simple oscillation process that generates time-averaged Cs
symmetry whereby the indenyl ring plane is orthogonal to
the anthracene plane. Moreover, even if the indenyl were
to rotate through the anthracenyl plane, thus engendering
effective C2v symmetry, such a process would still be unde-
tectable since the outer anthracene rings are already equiva-
lent. To probe the barriers in such systems, it is necessary
to prepare the 9,10-diindenyl analogues 3a and 3b, as-
suming their rotational barriers to be identical (or at least
very similar) to the undetectable processes in 2a and 2b,
respectively. The crucial point is that a 180° rotation of
either of the indenyl rings in 3a leads to interconversion of
Indenylanthracenes
Table 1 summarises the available data on the rotational the chemically distinguishable atropisomers syn-3a (effec-
barriers and nominal closest approach distances, CADs, tively C2v) and anti-3a (effectively C2h) – a process that can
and r* values in new and previously published molecules of be monitored by NMR spectroscopy.
three types: indenylanthracenes, biindenyls and three-
bladed systems (triptycenes and barrelenes).
9,10-Bis(2-indenyl)anthracene (3b) was prepared by the
palladium-catalysed coupling of 9,10-dibromoanthracene
Having established that the rotational barriers in the 9- (4) with 2-indenylboronic acid (5) (Scheme 3) analogously
indenyltriptycenes 1a and 1b depend on the site of attach- to the previously described synthesis of 2b.[8b,12] In contrast,
ment to the indenyl fragment, we wished to compare the the preparation of the 3-isomer 3a via a palladium-cata-
situation with that of their synthetic precursors 9-(3-in- lysed Heck reaction was unsuccessful. Under these condi-
denyl)anthracene (2a) and 9-(2-indenyl)anthracene (2b) tions, the formation of polycyclic indeno-dihydroace-
(Scheme 2). In 2a and 2b an sp2–sp2 carbon–carbon single anthrylenes was predominant, as had been observed in
bond links two planar aromatic fused systems in such a other Pd-catalysed reactions involving anthracenes.[8b,13]
fashion that the dihedral angles between the anthracenyl However, 9,10-bis(3-indenyl)anthracene (3a) was success-
and indenyl planes are 74° and 70°, respectively. Such a fully prepared by the Stille coupling of 9,10-dibromo-
conformation was to be expected because coplanarity of the anthracene (4) with 1-(trimethylstannyl)indene (6). The
indenyl and anthracenyl fragments is disfavoured both steri- mono-coupled product, 9-bromo-10-(3-indenyl)anthracene
cally and electronically.[8b]
(7), was also isolated after chromatographic separation; its
Eur. J. Org. Chem. 2010, 5203–5216
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5205

K. Nikitin, M. J. McGlinchey et al.
FULL PAPER
Scheme 2. 9-Indenylanthracenes 2 and 9,10-di-indenylanthracenes 3.
X-ray crystal structure is shown in Figure 4, and the di- in anti-3a is 0.3 Å, suggesting the existence of a very high
hedral angle between the indenyl and bromoanthracenyl rotational barrier, even allowing for the geometric pertur-
planes is 65°.
bations that must certainly occur. In contrast, the analo-
gous CAD in 3b, H(1)···H(12), exceeds 1.1 Å. Moreover, in
3b the syn- and anti-rotamers are disordered in the solid
state such that the methylene and methine sites in the five-
membered rings are each 50% occupied.
Scheme 3. Synthesis of the di-indenylanthracenes 3a and 3b.
Figure 4. X-ray crystal structure and spacefill view of 9-bromo-10-
(3-indenyl)anthracene (7).
Figure 5. X-ray crystal structure of 9,10-di-(3-indenyl)anthracene
(anti-3a). Rotation of the indenyl fragments is severely hindered
because of simultaneous close contacts in the pairs H(8)–H(12)
(CAD = 0.3 Å) and H(1)–H(17) (CAD = 1.2 Å).
It transpired that, as indicated by NMR, 9,10-bis(3-in-
denyl)anthracene, 3a, is a nearly equimolar mixture of syn-
and anti-atropisomers that can be separated by column
chromatography. However, the assignment of either one as
syn or anti cannot be unambiguously determined spectro-
scopically. Fortunately, one atropisomer of 3a yields X-ray
quality crystals, and the structure of anti-3a appears in Fig-
ure 5; in addition, the structure of the bis(2-indenyl) isomer
3b is shown in Figure 6. The indenyl rings adopt dihedral
angles of 81.5° (anti-3a) and 81.7° (3b) with the anthracene
plane. A computer-assisted virtual rotation of the indenyl
fragment about the single bond (whilst maintaining the re-
Figure 6. X-ray crystal structure of 9,10-di-(2-indenyl)anthracene
(3b). Rotation is markedly less hindered since the nominal closest
maining geometry) shows that the CAD for H(1)···H(17) approach distance between H(1) and H(12) is 1.15 Å.
5206 Eur. J. Org. Chem. 2010, 5203–5216
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Towards a Molecular Clamp
1
Although the H NMR spectra of the two atropisomers rac-9 the dihedral angle H(1)–C(1)–C(1Ј)–H(1Ј) is 58°, and
of 9,10-bis(3-indenyl)anthracene, 3a, are very similar, it is the corresponding angle, H(1)–C(1)–C(9)–H(9), in meso-9
possible to identify key differences in the 6.5–7.0 ppm re- is 66°. It is noteworthy that in the conformation adopted
gion, and hence to follow the interconversion of anti-3a and by rac-9 in solid state the H(1) and H(1Ј) hydrogens are
syn-3a at different temperatures. The barrier was evaluated positioned between two widely-separated methyl substitu-
as ca. 25 kcalmol–1, approximately twice the value found in ents, as shown in Scheme 4 for rac-9b.
the corresponding three-bladed system 9-(3-indenyl)trip-
tycene (1a). As noted above and depicted in Figure 1, the
triptycene 1a adopts a structure such that, although the six-
membered ring of the 3-indenyl moiety is positioned be-
tween two blades, the five-membered ring is seen to twist
out of the molecular mirror plane so as to minimise the
interaction between H(1) and H(18). This distortion pre-
sumably raises the energy of the ground state and so slightly
lowers the apparent barrier. Moreover, simulation of virtual
rotation in 1a reveals that as the benzo ring of the 3-indenyl
passes a triptycyl blade, the five-membered ring can bend
into the space between the other two blades, thus mini-
mising the energy cost of the rotation. This “duck-and-
dodge” mechanism cannot be realised in the case of the
bis(3-indenyl)anthracenes 3a, in which two unfavourable co-
planar H···H contacts are attained simultaneously. Moreover,
the observed rotational barrier of 25 kcalmol–1 is without
a doubt significantly higher than that estimated on the basis
of empirical correlation given by Sternhell.[9b]
Figure 7. X-ray crystal structures of (a) rac-9 and (b) meso-9. In
these dimers, rotation about the central C(1)–C(1Ј) or C(1)–C(9)
bond is slowed (barrier about 15 kcal mol–1) by simultaneous close
approach of H(7)–H(7Ј) and the hydrogens of the methyl groups at
C(8) and C(8Ј), in rac-9, and of H(7)–H(15) and the methyl hydro-
gens, such as H(8)–H(16), in meso-9.
1
Although the variable-temperature 500 MHz H NMR
spectra of 9,10-bis(2-indenyl)anthracene (3b) did not reveal
the presence of syn and anti atropisomers, this phenomenon
was observable in the 13C regime. Furthermore, coalescence
of each of the two pairs C(9) syn/anti and C(11) syn/anti
resonances at –26 °C indicated
a
barrier of ca.
13.5 kcalmol–1, in good agreement with estimated nominal
CAD in this molecule.
1,1Ј-Biindenyls
We recently reported that the isomeric meso- and race-
mic-2,2Ј-disubstituted-1,1Ј-biindenyls can readily be pre-
pared by oxidative dimerisation of lithiated indenes.[14]
Furthermore, it was shown that rotation about the central
carbon–carbon bond in meso-2,2Ј-diphenyl-1,1Ј-biindenyl
Scheme 4. A 180° rotation about the central bond in rac-9a leads
to a different C2 isomer, rac-9b. In contrast, the analogous rotation
(8) is slow as evidenced by its NMR spectrum in conjunc- in meso-9 merely interconverts enantiomers.
tion with DFT-level calculations.[14] Consequently, it was of
1
Interestingly, the room temperature H NMR spectra of
both rac-9 and meso-9 were broad and unresolved in both
the aromatic and aliphatic regions. However, lowering the
temperature to –15 °C led to full decoalescence of all signals
(Figure 8). At that temperature the racemic isomer of 9 is
represented by a 42:58 mixture of two slowly inter-
converting C2-symmetric rotamers, rac-9a and rac-9b
(Scheme 4), of which the latter is presumably the more
stable since it is the only one found in the solid state. As
the temperature is ramped up, the NMR signals broaden
interest to probe the rotational barriers about the central
carbon–carbon bond in the related methyl-substituted 1,1Ј-
biindenyls 9 and to correlate these activation energies with
their molecular structures.
The 2,2Ј-dimethyl-1,1Ј-biindenyls 9 were prepared by and eventually coalesce at ca. 35 °C; the rotational barrier
treatment of 1-lithio-2-methylindene with CuCl2 and, after is approximately 14.5 kcalmol–1 (Table 1). While this bar-
chromatographic separation, the racemic and meso isomers rier is perhaps higher than would have been expected on
were characterised by X-ray crystallography. As shown in the basis of relatively long (0.9 Å) crystallographically-de-
Figure 7, both molecules adopt a gauche conformation: in termined CAD in this molecule, one should note that the
Eur. J. Org. Chem. 2010, 5203–5216
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5207

K. Nikitin, M. J. McGlinchey et al.
FULL PAPER
interconversion of rac-9a and rac-9b requires simultaneous or 180°, giving rise to either a single mirror plane (Cs) or
close approach of both the methyls and the indenyls of the an inversion centre (Ci), respectively. At low temperature
molecule.
(–15 °C), meso-9 is represented by an individual conformer
whose NMR spectrum indicates the presence of two non-
equivalent 2-methylindenyl moieties, in accord with its un-
symmetrical structure in solid state (Figure 7, b). Rotation
about the central C(1)–C(9) bond leads to enantiomeris-
ation (see Scheme 4) in which the two methylindenyl moie-
ties exchange roles. Coalescence of the H(1) and H(9) reso-
nances at 30 °C indicates a rotational barrier of ca.
15 kcalmol–1, presumably via a mirror-symmetric transition
state. This finding is in good agreement with our earlier
data on 2,2Ј-diphenyl-1,1Ј-biindenyl (meso-8), whereby the
substituents at the 2-indenyl positions become closely proxi-
mate as the enantiomerisation proceeds. We note, however,
that DFT calculations on the diphenyl system 8 revealed
that the central carbon–carbon bond lengthens markedly in
the transition state.[14]
Substituted Triptycenes
It has long been known that the rotation barriers in 9-
substituted triptycenes are very responsive to the steric size
and orientation of the substituent;[3] for instance, the bar-
rier reaches 16 kcalmol–1 in 9-(chloromethyl)triptycene.[15]
As discussed above, the barrier to paddlewheel rotation in
indenyltriptycenes 1 is also very sensitive to the mode of
1
Figure 8. 500 MHz H NMR spectra (–15 °C, CDCl3) of (a) rac-9
as an equilibrium mixture of conformers rac-9a (42%) and rac-9b
(58%, assignments shown); (b) note that in meso-9 the H1 and H9
protons are anisochronous.
Unlike dimeric rac-9 that retains its C2 symmetry what- attachment of the indenyl moiety (Scheme 1). Accordingly,
ever the rotamer conformation, meso-9 loses all symmetry one might anticipate that variations in the structure of the
elements unless the central H–C–C–H dihedral angle is 0° paddlewheel will also challenge the rotational barrier. We
Scheme 5. Preparation of three-bladed molecules 10–15 from substituted anthracenes 2b or 2c. Reagents: (i) benzyne; (ii) tetrafluo-
robenzyne; (iii) DMAD; (iv) N-methylmaleimide.
5208
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5203–5216

Towards a Molecular Clamp
here describe the syntheses, X-ray crystal structures and dy- Moreover, there are two independent molecules of 11 in the
namic NMR behaviour of several closely related molecules, unit cell, one of which exhibits a disorder between the tetra-
10–20, (Schemes 5 and 6) in each of which a three-bladed fluorobenzo ring and one of its C6H4 partners.
paddlewheel is linked to either a phenyl or an indenyl moi-
ety.
Scheme 6. Syntheses of the diphenyl-dibenzodihydrobarrelene de-
rivative 16 and tetraphenyltetrahydrofuran (17).
Figure 10. Molecular structure of 1,2,3,4-tetrafluorotriptycene (11);
thermal ellipsoids at 50%.
Earlier reports on rotational barriers in 9-aryltriptycenes
appear to suggest that rotation is fast on the NMR time-
scale in these symmetrical molecules attributable, possibly,
to steric hindrance raising the energy of the ground state.[3]
The parent compound, 9-phenyltriptycene (10), was there-
fore prepared and characterised by X-ray crystallography
and variable-temperature NMR spectroscopy. The structure
appears as Figure 9, and reveals that the phenyl moiety is
bent off the threefold axis of the triptycene by 6° indicating
a slightly unbalanced steric repulsion. Nevertheless, as with
2-indenyltriptycene (1a), there is no evidence of slowed ro-
tation of the phenyl fragment on the NMR time-scale at
temperatures down to –60 °C, indicating a rotational bar-
rier well below 10 kcalmol–1. This is in a good agreement
with the calculated CAD of 0.7 Å between H(18)/H(22) of
the phenyl fragment and H(1)/H(8)/H(13) of the blades.
Interestingly, we note that the variable-temperature
NMR behaviour of the substituted 9-(o-tolyl)-1,2,3,4-tetra-
fluorotriptycene derivative 12 has been interpreted in terms
of an oscillation process whereby the largest barrier would
require the methyl to rotate past the fluorinated benzo
ring.[3a]
Substituted Barrelenes
Having established that paddlewheel rotation in both 9-
(2-indenyl)triptycene (1b) and 9-phenyltriptycene (10) is ra-
pid at ambient temperature, we chose to study the analo-
gous 11,12-bis(methoxycarbonyl)dibenzobarrelene deriva-
tives 13 and 14 to probe the effect of incorporating the more
sterically demanding ester substituents. These molecules
were prepared by the Diels–Alder cycloaddition of DMAD
to the corresponding substituted anthracenes, as shown in
Scheme 5.
The X-ray structures (Figure 11) reveal that in these mo-
lecules the dibenzobarrelene core almost retains its pseudo-
threefold symmetry. In the phenyl derivative, 13, the in-
terplanar angles between the two peripheral benzo rings
and those between the benzo rings and the C(11)–C(12)
double bond are 121.8°, 123.2° and 115.0°, respectively: in
the 2-indenyl system 14 the corresponding angles are
121.2°, 120.3° and 118.5°. Moreover, the phenyl substituent
in 13, and the 2-indenyl fragment in 14, deviate slightly
from coplanarity with the C(11)–C(12) double bond bear-
ing the methoxycarbonyl groups. In 13, the dihedral angle
C(12)–C(9)–C(13)–C(18) is 20°, and the angular distortion
Figure 9. Molecular structure of 9-phenyltriptycene (10) showing a
possible close contact between H(18)/H(22) and one of H(1)/H(8)/
H(13) of the triptycene.
Another way to break the threefold symmetry of a trip- of the phenyl from C(10)–C(9) axis of the paddlewheel is
tycene involves the functionalisation of one, or more, of the 4.4°; in 14, the dihedral angle C(12)–C(9)–C(13)–C(19) is
blades. Although the Diels–Alder addition of tetrafluo- 10°, and the angular distortion of the indenyl from C(10)–
robenzyne to anthracene to form 1,2,3,4-tetrafluorotrip- C(9) axis of the paddlewheel is only 3.5 degrees. To allow
tycene (11) was first reported in 1968,[16] we are unaware of access to this conformation the adjacent methoxycarbonyl
any structural characterisation of the parent compound, groups sacrifice their conjugative interaction with the
nor of any 9-substituted derivatives. The X-ray crystal double bond by rotating through ca. 75° away from the sub-
structure of 1,2,3,4-tetrafluorotriptycene is shown in Fig- stituent at C(9). Interestingly, in the parent compound, di-
ure 10, and the geometry deviates only slightly from idea- methyl dibenzobarrelene-11,12-dicarboxylate, the torsion
lity: the interplanar angle between the two benzo rings angles C(11)=C(12)–C=O and C(12)=C(11)–C=O are 116°
(123.1°) is somewhat larger that those between the tetra- and 164°, again revealing a marked twisting of the ester
fluorobenzo ring and its neighbours (120.5° and 116.4°). moieties away from the double bond.[17]
Eur. J. Org. Chem. 2010, 5203–5216
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5209

K. Nikitin, M. J. McGlinchey et al.
FULL PAPER
Figure 12. X-ray crystal structure of 2,2,5,5-tetraphenyltetra-
hydrofuran (17): bird’s eye view and edge-on view (space-filling)
emphasising the C2 character of the molecule.
only rather slight steric strain associated with the phenyl-
dibenzocyclooctane interaction, as is apparent from the X-
ray crystal structure of 16 (Figure 13). There is a substantial
CAD of 0.85 Å between H(1)/H(8) and H(14)/H(18), and
no double steric repulsion is involved.
Figure 11. Molecular structures of dimethyl 9-phenyldibenzobarre-
lene-11,12-dicarboxylate, 13, and dimethyl 9-(2-indenyl)dibenzo-
barrelene-11,12-dicarboxylate, 14.
The simplicity of the NMR spectra of both 13 and 14
indicates that rapid rotation occurs even at –70 °C, signify-
ing in each case a low barrier about the C(9)–(C13) linkage.
It has previously been shown that an out-of-plane carbonyl
Figure 13. X-Ray crystal structure of 9,10-diphenyl-11,12-dihydro-
dibenzobarrelene (16). In this molecule, phenyl rotation is fast on
group poses no significant barrier to rotation in 2-substi- the NMR time-scale, since the H(1)/H(14) CAD is relatively long,
tuted biphenyls.[9c] Similarly, in the case of 13 or 14, since
and no double interactions are involved.
the esters are twisted out of the plane of the double bond
We note that as the interplanar angle between peripheral
they do not significantly hinder rotation of the 9-indenyl or
benzo rings gradually approaches planarity, the configura-
9-phenyl groups. These low barriers are consistent with
tion of the transition state on the rotation energy profile
would resemble that of a substituted anthracene, i.e. double
steric repulsions would be engendered. In the triptycenes
and barrelenes discussed above, the interplanar angle be-
CADs in excess of 1 Å (Table 1) in the absence of double
steric interactions – behaviour similar to that seen in the
previously discussed triptycenes 1a, 1b and 10.
tween peripheral benzo rings is approximately 120°. How-
ever, in the dihydrobarrelene, 16, the anthracene framework
is somewhat closer to planarity such that this value is now
Substituted Dihydrobarrelenes
Further studies have been carried out using 9,10-di- 134°, while the angle between the benzo rings and the eth-
phenyl-11,12-dihydrodibenzobarrelene (9,10-diphenyl-9,10- ano-bridge is 113°Ϯ2°. The corresponding angles in the N-
ethanoanthracene) (16) and also the Diels–Alder adduct 15 methylmaleimide–9-phenylanthracene Diels–Alder adduct
prepared by addition of N-methylmaleimide to 9-phenylan- 15 are 131.2° and 114.4°, respectively.
thracene. Molecule 16 was prepared by an elegant intramo-
In the N-methylmaleimide/9-phenylanthracene Diels–
lecular double cyclisation[18] of 1,1,4,4-tetraphenylbutane- Alder adduct 15 (Figure 14) the incorporation of the pyrrol-
1,4-diol in hexafluorophosphoric acid medium (Scheme 6). idone fragment renders the two aromatic blades non-equiv-
Interestingly, cyclisation using H2SO4 in acetic acid yields alent, and so slowed rotation of the phenyl substituent
pure 2,2,5,5-tetraphenyltetrahydrofuran (17)[19] in which the would split the degeneracy of its ortho and meta positions.
1
bulky phenyl substituents engender a C2-symmetric struc- In the room temperature 500 MHz H NMR spectrum of
ture (Figure 12) such that pairs of phenyls adopt dihedral 15 the signals assignable to the peripheral benzo rings are
angles of 77°Ϯ1° and 47°Ϯ1° relative to the plane contain- clearly resolved whereas the ortho and meta positions of the
ing the oxygen and its adjacent carbons.
phenyl ring are broadened. However, at –15 °C the ortho
The NMR spectrum of the diphenyl-ethanoanthracene protons (at δ = 8.10 and 7.41) and meta protons (at δ =
16 at room temperature revealed that the ortho and also the 7.67 and 7.53) have clearly decoalesced (Figure 15), indicat-
meta positions of both phenyls were equivalent, as were the ing that rotation has slowed on the NMR timescale. Subse-
blades of dibenzobicyclooctane moiety. Variable-tempera- quent observation of coalescence for the ortho and meta
ture measurements confirmed that no significant slowed ro- resonances at 70 °C and 40 °C, respectively, yielded a sub-
tation was evident at –60 °C. This finding is consistent with stantial rotational barrier of 15.5Ϯ0.5 kcalmol–1 for this
5210
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5203–5216

Towards a Molecular Clamp
process. It is unlikely that this barrier is associated with the was correct. The diol 18 was readily converted into its di-
H(1)···H(14) CAD of 0.7 Å. However, a bending of the methyl ether 19 (Figure 16, b). The variable-temperature 1H
phenyl through 7.6° away from C(9)–C(10) axis of the bicy- NMR spectra of 19 indicate that the phenyl groups are ap-
clooctane core implies a steric repulsion between the car- parently free to rotate at –60 °C despite the fact that this
bonyl oxygen, O(1), and the phenyl ortho hydrogens H(14)/ rotation would involve simultaneous approach of the H(4)/
H(18) for which the CAD is 1.65 Å. Bearing in mind the H(12) and H(5)/H(16) pairs. The virtual rotation analysis
substantially larger atomic radius of the carbonyl oxygen revealed that no significant H···H repulsive interactions are
compared to hydrogen, it is more likely that this is the inter- present in this molecule since the CAD is very large (1.6 Å).
action that increases the rotational barrier. Interestingly,
NMR spectroscopic data on 15 at elevated temperatures
suggest that the Diels–Alder addition of N-methylmale-
imide to 9-phenylanthracene is reversible; this parallels the
very recent report by Lehn that reversible Diels–Alder reac-
tions of cyanoethylenes with 9,10-dimethylanthracene can
be used in fluorescent optical switches.[20]
Scheme 7. Reagents and conditions: (i) methanol, H2SO4, reflux,
3 h, 100%; (ii) ethylene glycol, toluene, TSA, 60 °C, 24 h, 52%.
Figure 14. The X-ray crystal structure of 11,12-(N-methylpyrrolid-
ino)-9-phenyldibenzobarrelene (15) suggests that close approach of
H(18) and carbonyl O(1) creates the observed substantial phenyl
rotation barrier of 15.5 kcal/mol.
Figure 16. X-Ray crystal structures (a) of the trans-diol 18, and (b)
the dimethyl diether 19.
When the trans-diol 18 was heated with ethylene glycol
in toluene in the presence of acid, the novel bicyclic ether
20 was isolated in 52% yield, and its structure appears as
Figure 17. Since the initial configuration of the diol 18 is
trans, and both ether linkages in 20 are necessarily on the
same side of the original anthracene plane, the reaction evi-
dently proceeds with inversion of configuration at one of
the benzylic centres. The peripheral benzo rings in 20 are
no longer coplanar as in the diol 18, but the dihedral angle
is 166°, substantially larger than the value of 134° observed
in the 9,10-ethanoanthracene 16. Clearly, extension of the
bridge from a two-atom chain in 16 to a four-atom chain
in 20 flattens the central ring system, pushes the phenyls at
the bridgehead positions into the narrow space between the
ortho hydrogen atoms of the peripheral rings and severely
Figure 15. 500-MHz 1H NMR spectrum of the Diels-Alder adduct
15 in the aromatic region.
1,6-Diphenyl-7,8,9,10-dibenzo-2,5-dioxabicyclo[4.2.2]decane
In light of the already discussed correlation between the limits their freedom.
rotational barriers in paddlewheel-type systems and the
Virtual rotation simulation for 20 shows that the shortest
room for manoeuvre available to substituents at C(9), it was H···H distances between the phenyl ortho-hydrogen atoms,
decided to attempt to construct a molecular system in H(14)/H(18), and the paddlewheel blade hydrogen atoms
which movement of the phenyl substituents was constrained H(1)/H(8) are attained simultaneously with a CAD of ca.
by modifying the length of the bridge between C(9) and 0.9 Å. Unlike the situation in triptycenes 1 and 10, or the
C(10). To this end, treatment of anthraquinone with phen- three-bladed barrelene derivatives, 10, 11, 13–16, the phenyl
ylmagnesium bromide furnished 9,10-diphenyldihydro- moiety cannot escape strong lateral repulsion by bending
anthracene-9,10-diol 18 (Scheme 7, Figure 16). The struc- away and slipping between the blades. Consequently, the
ture (Figure 16, a) confirms that the previously pos- phenyls are locked in an orientation perpendicular to the
tulated[21] trans configuration of the two hydroxy groups surface of the dihydroanthracene moiety.
Eur. J. Org. Chem. 2010, 5203–5216
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5211

K. Nikitin, M. J. McGlinchey et al.
FULL PAPER
As emphasised by Hunter,[22] solvent effects can also be im-
portant parameters, and so in this work all barriers were
measured in similar systems (CD2Cl2, CDCl3 or C2D2Cl4,
depending on the temperature required for peak coalesc-
ence). The structural parameters of the molecules have been
fully established by X-ray crystallography augmented by a
computational “virtual rotation” process, and the rotational
barriers have been estimated from variable-temperature
NMR and kinetic data. The goal was to probe the combi-
nation of geometric molecular parameters that led to the
sometimes surprisingly high, or low, rotational barriers in
polycyclic systems.
Comparison of the rotational barriers for the 3-indenyl-
anthracenes 2a and 3a with that for the apparently closely
related 3-indenyltriptycene (1a) revealed that they can be
very different despite the similarity in their nominal H···H
Closest Approach Distances (CADs) obtained from the vir-
tual rotation of molecular fragments. The simultaneous de-
velopment of two strong steric repulsions between two
planar aromatic fused systems renders the rotation about
the single C–C bond in 3-indenylanthacenes very slow at
room temperature. In the paddlewheel-shaped 3-indenyl-
triptycene this seemingly equally strong repulsion can be
alleviated by an energetically more favourable “duck and
dodge” motion of the indenyl moiety. The idea that not
only the strength of an individual steric interaction, but also
the number of simultaneously attained interactions, and the
ability of the molecular fragments to adapt to the changing
environment and so determine the energy cost of the rota-
tion, helps to rationalise a wide range of new observations
in this area.
Thus, the rotational barriers in 2,2Ј-disubstituted-1,1Ј-bi-
indenyls are, similarly to 3-indenylanthracenes, governed by
strong H···H interactions in the transition state, and can be
interpreted in terms of simultaneous steric repulsions and
the CADs between hydrogens of the aromatic rings and the
substituents. Moreover, one can conclude that the rota-
tional barriers in the 9-aryl-11,12-bis(methoxycarbonyl)di-
benzobarrelenes 13 and 14 are no higher than in structur-
ally related 9-substituted triptycenes (8–9 kcal/mol) because
in each case the adjacent methoxycarbonyl moiety rotates
out of the plane of the dibenzobarrelene blade. So, in the
absence of multiple steric repulsion interactions, rotation in
these systems is reasonably facile. However, this relatively
low rotational barrier can be significantly increased by in-
troducing a rigidly oriented heteroatom pointing in the di-
rection of the rotating phenyl as in the imide 15.
Figure 17. As shown in the X-ray crystal structure of 20, rotation
of the phenyls has to overcome simultaneous repulsion of H(1)/H14
and H(8)/H(18) with CAD 0.9 Å; the space-filling representation of
20 shows the phenyls constrained by the “calipers” of the dibenzo-
bicyclodioxadecane core.
1
The 500 MHz H NMR spectrum of 20 in the aromatic
region is shown in Figure 18, and clearly illustrates that all
five aromatic protons of the phenyl ring are non-equivalent.
The chemical shift separation of about 1.4 ppm reveals very
different magnetic environments for the phenyl protons,
H(14) and H(18), presumably attributable to the combined
effects of the two closely located aromatic ring systems of
the paddlewheel. An NMR study over the temperature
range 30–90 °C showed no peak broadening nor 2D EXSY
exchange between H(14) and H(18); hence, one can only
calculate a minimum rotational barrier on the basis of the
NMR chemical shift difference between two non-coalescing
signals, such as H(15) and H(17), that would be equilibrated
if rotation were to occur. These data yield a minimum value
for the rotational barrier of 23 kcalmol–1, but it is undoubt-
edly substantially higher.
In this context, the very striking difference (at least
10 kcalmol–1) in the rotational barriers between dihy-
droanthracene derivatives 19 and 20 can be clearly under-
stood in terms of CADs and multiple steric interactions. In
the di-ether 19 a possible double steric repulsion is not real-
ised owing to relatively long CAD; in the bridged di-ether,
20, the nearly-planar conformation of the benzo blades
leads to substantially shorter CADs. Consequently, the ro-
Figure 18. 500 MHz 1H NMR spectrum of the cyclic ether 20 in
the aromatic region.
Conclusions
In this paper the factors determining the ease of thermal tation of the phenyls is severely restricted because of simul-
rotation about a carbon–carbon single bond intercon- taneous steric interactions with both blades. The combina-
necting two rigid molecular fragments have been studied. tion of steric effects that have been exemplified herein may
5212
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5203–5216

Towards a Molecular Clamp
9,10-Bis(2-indenyl)anthracene (3b): To a stirred suspension of 4
(0.27 g, 0.8 mmol) in a mixture of ethanol (6 mL) and toluene
(12 mL), the boronic acid, 5 (0.32 g, 2 mmol), sodium carbonate
(0.42 g, 4 mmol), dichloro-bis(diphenylphosphanylferrocene)palla-
dium (8 mg, 0.01 mmol) were added. The reaction mixture was
stirred for 80 h at 80 °C after which time it was filtered, the solid
was extracted with copious amounts of chloroform, the organic
layer was concentrated to give 3b as a yellowish solid (0.27 g, 83%),
m.p. Ͼ340 °C. 1H NMR (500 MHz, CDCl3, numbering as in
Scheme 1): δ = 8.00 (d, J = 6 Hz, 4 H, H1, H4, H5, H8), 7.59 (d,
J = 6.8 Hz, 2 H, H13, H20), 7.57 (d, J = 6.8 Hz, 2 H, H16, H23),
7.42 (d, J = 6.8 Hz, 4 H, H2, H3, H6, H7), 7.39 (t, J = 6 Hz, 2 H,
H14, H21), 7.30 (t, J = 6.8 Hz, 2 H, H15, H22), 7.08 (s, 2 H, H12,
H19), 3.94 (s, 4 H, 2ϫ H17, 2ϫ H24) ppm. 13C NMR (125 MHz,
CDCl3): δ = 145.3 (C16a), 145.1 (C11), 143.8 (C12a), 133.5 (C12),
133.4 (C9), 129.7 (C8a), 126.7 (C14), 126.6 (C1), 125.3 (C2), 124.8
(C15), 123.7 (C16), 121.1 (C13), 44.5 (C17) ppm. C32H22 (406.53):
calcd. C 94.55, H 5.45; found C 94.43, H, 5.50.
have subtle implications for the design of molecular ma-
chines, and provide continued motivation for studies in the
area of mechano-stereochemistry.[23]
Experimental Section
General: All reactions were carried out under a nitrogen atmo-
sphere unless otherwise stated. Column chromatography separa-
tions were carried out on a Buchi Sepacor machine with UV ab-
sorbance detector using silica gel particle size 40–63 mm. NMR
spectra were acquired on Varian VNMRS 400 and 600 or Inova
500 MHz spectrometers at 25 °C unless otherwise stated. Assign-
ments were based on standard 1H-1H and 1H-13C two-dimensional
techniques, and NOE measurements. Rotational barriers were ob-
tained by standard peak coalescence measurements.[24] Melting
points were determined on a Gallenkamp instrument in air and are
uncorrected. Elemental analyses were carried out by the Microana-
lytical Laboratory at University College Dublin. Compounds 1a–
c,[8] 2a,[8a] 2b,[8b] 10,[25]11[16] 16,[18] 17,[19] 18 and 19[21] were pre-
pared as described elsewhere.
2,2Ј-Dimethyl-1,1Ј-biindenyl (racemic-9 and meso-9): To a stirred
solution of 2-methylindene (0.8 mL, 6 mmol) in diethyl ether
(6 mL) a solution of nBuLi (6.6 mmol) in hexanes was added and
the reaction mixture was stirred for 1 h resulting in the formation
of a pale yellow suspension which was added slowly to a cooled
(–20 °C) suspension of cupric chloride (1 g, 7 mmol) in THF
(10 mL). The resulting mixture was warmed to 5 °C and was stirred
for one day, after which time it was quenched with methanol
(5 mL), extracted with dichloromethane and separated (2–5%
dichloromethane in cyclohexane) to give meso-9 (310 mg, 40%),
mixed fraction (152 mg, 19%), and rac-9 (248 mg, 32%); meso-9: a
9,10-Bis(3-indenyl)anthracenes (3a) and 9-Bromo-10-(3-indenyl)-
anthracene (7): To a stirred suspension of the dibromide 4 (0.135 g,
0.4 mmol) in 1,4-dioxane (2 mL), dichloro-bis(tri-o-tolylphos-
phane)palladium (3 mg, 0.004 mmol) and indenyltrimethyltin
(0.34 g, 1.2 mmol) were added. The reaction mixture was stirred for
40 h at 120 °C (sealed tube) after which time it was concentrated,
extracted with copious amounts of dichloromethane, washed suc-
cessively with aqueous hydrochloric acid, ammonia, hydrochloric
acid, sodium hydrogen carbonate and separated to afford 3a
(102 mg, 63%) as a syn/anti mixture, and 9-bromo-10-(3-indenyl)-
anthracene (7) (18 mg, 12%) as a yellow solid. 1H NMR (500 MHz,
CDCl3): δ = 8.60 (d, J = 8 Hz, 2 H), 7.90 (d, J = 8 Hz, 2 H), 7.63
(d, J = 7.5 Hz, 1 H), 7.55 (t, J = 7.5 Hz, 2 H), 7.36 (t, J = 7.5 Hz,
2 H), 7.28 (t, J = 7.5 Hz, 1 H), 7.12 (t, J = 7.5 Hz, 1 H), 6.72 (s, 1
H), 6.65 (d, J = 7.5 Hz, 1 H), 3.84 (s, 2 H) ppm. 13C NMR
(125 MHz, CDCl3): δ = 146.5, 143.4, 141.8, 135.4, 131.7, 131.0,
130.4, 128.0, 127.1, 126.5, 125.6, 123.0, 120.8, 38.8 ppm. C23H15Br
(371.28): calcd. C 74.41, H 4.07, Br 21.52; found C 74.20, H 4.22,
Br 21.16. Careful separation of 3a (2% dichloromethane in cyclo-
hexane) yielded syn-3a (30 mg) and anti-3a (60 mg): anti-3a is a
1
yellow solid; m.p. 95–96 °C. H NMR (500 MHz, CDCl3, –15 °C,
numbering as in Figure 7b): δ = 7.64 (d, J = 7 Hz, 1 H, H7), 7.36
(t, J = 7 Hz, 1 H, H5), 7.29 (t, J = 7 Hz, 1 H, H6), 7.29 (d, J =
7 Hz, 1 H, H4), 7.22 (d, J = 7 Hz, 1 H, H12), 7.13 (t, J = 7 Hz, 1
H, H13), 6.71 (t, J = 7 Hz, 1 H, H14), 6.60 (s, 1 H, H11), 6.28 (s,
1 H, H3), 5.88 (d, J = 7 Hz, 1 H, H15), 4.06 (s, 1 H, H9), 3.93 (s,
1 H, H1), 2.39 (s, 3 H, 3ϫ H16), 1.59 (s, 3 H, 3ϫ H8) ppm. 13C
NMR (125 MHz, CDCl3): δ = 148.2 (C2), 147.1 (C10), 146.9 (C7a),
145.1 (C3a), 143.4 (C15a), 144.4 (C11a), 127.3 (C3), 128.1 (C11),
120.0 (C4), 126.9 (C5), 124.1 (C6), 122.6 (C7), 122.4 (C15), 123.6
(C14), 126.7 (C13), 119.9 (C12), 80.5 (C1), 53.7 (C9), 16.1 (C8),
16.0 (C16) ppm. C20H18 (258.36): calcd. C 92.98, H 7.02; found C
92.93, H 7.07; racemic-9: a yellow solid, m.p. 106–107 °C. 1H NMR
ofhe predominant (58%) conformer rac-9b (500 MHz, CDCl3,
–15 °C, numbering as in Figure 7a): δ = 7.12 (d, J = 7 Hz, 2 H,
H4, H4Ј), 7.03 (t, J = 7 Hz, 2 H, H5, H5Ј), 6.94 (d, J = 7 Hz, 2 H,
H7, H7Ј), 6.81 (t, J = 7 Hz, 2 H, H6, H6Ј), 6.68 (s, 2 H, 2 H, H3,
H3Ј), 3.81 (s, 2 H, 2 H, H1, H1Ј), 2.41 (s, 6 H, 2ϫ CH3) ppm. 13C
NMR (125 MHz, CDCl3): δ = 147.3 (C2), 143.2 (C7a), 144.7 (C3a),
128.7 (C3), 119.7 (C4), 126.6 (C5), 123.3 (C6), 122.2 (C7), 51.1
(C1), 16.0 (C8) ppm. C20H18 (258.36): calcd. C 92.98, H 7.02; found
C 92.93, H 7.00.
1
yellow solid; m.p. 295–300 °C (dec.). H NMR (500 MHz, CDCl3,
numbering as in Scheme 1): δ = 7.96 (d, J = 7 Hz, 4 H, H1, H4,
H5, H8), 7.67 (d, J = 7.5 Hz, 2 H, H14, H21), 7.31 (d, J = 7 Hz,
4 H, H2, H3, H6, H7), 7.28 (t, J = 7.3 Hz, 2 H, H15, H22), 7.15
(t, J = 7.3 Hz, 2 H, H16, H23), 6.80 (d, J = 7.5 Hz, 2 H, H17,
H24), 6.80 (s, 2 H, H12, H19), 3.87 (s, 4 H, 2ϫ H13, 2ϫ H20)
ppm. 13C NMR (125 MHz, CDCl3): δ = 146.8 (C13b), 143.7
(C13a), 142.4 (C9), 135.0 (C12), 131.1 (C11), 130.0 (C4a), 126.9
(C1), 126.4 (C16), 125.1 (C2), 125.0 (C15), 123.8 (C14), 121.2
(C17), 44.5 (C13) ppm. C32H22 (406.53): calcd. C 94.55, H 5.45; 9-Phenyl-11,12-bis(methoxycarbonyl)dibenzobarrelene (13): To
a
found C 94.16, H 5.94. syn-3a is a yellow solid; m.p. 295–300 °C
(dec.). H NMR (500 MHz, CDCl3): δ = 7.96 (d, J = 7 Hz, 4 H,
suspension of 2c (203 mg, 0.8 mmol) in 1,4-dioxane (2 mL) di-
methyl acetylenedicarboxylate (0.48 mL, 4 mmol) was added and
1
H1, H4, H5, H8), 7.67 (d, J = 7.5 Hz, 2 H, H14, H21), 7.30 (t, J the mixture was heated at 120 °C for 2 days after which time it
= 7 Hz, 4 H, H2, H3, H6, H7), 7.28 (t, J = 7.3 Hz, 2 H, H15, H22), was concentrated under ca. 1 mbar (to remove excess DMAD) and
7.17 (t, J = 7.3 Hz, 2 H, H16, H23), 6.85 (d, J = 7.5 Hz, 2 H, H17,
H24), 6.75 (s, 2 H, H12, H19), 3.85 (s, 4 H, 2ϫ H13, 2ϫ H20)
separated to give crude 13 (0.25 g, 79%) containing approx. 25%
of the isomeric 1,4 adduct. Pure 13 was isolated by fractional crys-
ppm. 13C NMR (125 MHz, CDCl3): δ = 146.9 (C13b), 143.7 tallisation from acetone-hexane as a white solid: m.p. 158–159 °C.
(C13a), 142.4 (C9), 135.1 (C12), 131.2 (C11), 130.0 (C4a), 126.9
(C1), 126.4 (C16), 125.2 (C2), 125.1 (C15), 123.9 (C14), 121.1
1H NMR (400 MHz, CDCl3, numbering as in Figure 11): δ = 7.75
(d, J = 7.5 Hz, 2 H, H14, H18), 7.51 (t, J = 7.5 Hz, 2 H, H15,
(C17), 39.0 (C13) ppm. C32H22·0.5C6H12 (448.61): calcd. C 93.71, H17), 7.46 (t, J = 7.5 Hz, 1 H, H16), 7.42 (d, J = 7.5 Hz, 2 H, H1,
H 6.29; found C 93.80, H 6.07 (from cyclohexane). H8), 7.10 (d, J = 7.5 Hz, 2 H, H4, H5), 7.03 (t, J = 7.5 Hz, 2 H,
Eur. J. Org. Chem. 2010, 5203–5216
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5213

K. Nikitin, M. J. McGlinchey et al.
FULL PAPER
H2, H7), 6.95 (t, J = 7.5 Hz, 2 H, H3, H6), 5.68 (s, 1 H, H10), 3.65 J = 7 Hz, 2 H, H3, H6), 7.14 (t, J = 7 Hz, 1 H, H2), 7.05 (t, J =
(s, 3 H, 3ϫ H20), 3.77 (s, 3 H, 3ϫ H19) ppm. 13C NMR (100 MHz, 7 Hz, 1 H, H7), 6.44 (d, J = 7 Hz, 1 H, H8), 4.86 (d, J = 3 Hz, 1
CDCl3): δ = 168.1 (C12a), 164.0 (C11a), 154.3 (C11), 144.6 (C12), H, H10), 3.90 (d, J = 8.2 Hz, 1 H, H12), 3.30 (dd, J = 3, J =
146.0 (C4a) 145.3 (C8a), 134.9 (C13), 130.4 (C14, C18), 128.3 (C15, 8.2 Hz, 1 H, H11), 2.40 (s, 3 H, 3ϫ H19) ppm. 13C NMR
C17), 127.6 (C16), 125.4 (C3), 124.8 (C2, C4), 123.7 (C1), 62.8 (125 MHz, CDCl3): δ = 177.0 (C11a), 175.0 (C12a), 144.8 (C8a),
(C9), 52.4 (C19), 52.1 (C20), 51.4 (C10) ppm. MS (ES) 397 (31%, 140.7 (C8b), 140.2 (C4b), 138.0 (C4a), 135.3 (C13), 132.4 (C14),
[M + H]+). C26H20O4 (396.44): calcd. C 78.77, H 5.09; found C
78.77, H 5.10.
129.8 (C18), 128.4 (C17), 127.6 (C16), 127.4 (C15), 127.2 (C3),
127.1 (C6), 126.7 (C2), 126.2 (C7), 125.4 (C8), 125.0 (C1,4), 123.5
(C5), 56.3 (C9), 49.1 (C11), 47.0 (C12), 46.2 (C10), 24.2 (C19) ppm.
C25H19NO2 (365.43): calcd. C 82.17, H 5.24, N 3.83; found C
81.87, H 5.25, N 3.85 (from chloroform).
Preparation of 9-(2-Indenyl)-11,12-bis(methoxycarbonyl)dibenzo-
barrelene (14): Analogously to 13, 14 was prepared in a 21% yield
as a white solid, m.p. 122 °C. 1H NMR (400 MHz, CDCl3, number-
ing as in Figure 11): δ = 7.59 (d, J = 7.5 Hz, 1 H, H18), 7.53 (d, J
Preparation of 1,6-Diphenyl-7,8,9,10-dibenzo-2,5-dioxabicyclo[4.2.2]-
= 7.5 Hz, 1 H, H15), 7.47 (d, J = 7.5 Hz, 2 H, H4, H5), 7.42 (d, J decane (20): To a suspension of the diol 18 (50 mg, 0.15 mmol) in
= 7.5 Hz, 2 H, H1, H8), 7.37 (t, J = 7.5 Hz, 1 H, H16), 7.30 (t, J toluene (6 mL) ethyleneglycol (0.15 mL) and p-toluenesulfonic acid
= 7.5 Hz, 1 H, H17), 7.09 (t, J = 7.5 Hz, 2 H, H2, H7), 7.03 (t, J (10 mg) were added and the mixture was heated at 60 °C for 3 h
= 7.5 Hz, 2 H, H3, H6), 5.71 (s, 1 H, H10), 3.94 (s, 2 H, 2ϫ H19), after which time it was separated (20% ethyl acetate in cyclohex-
3.73 (s, 3 H, 3ϫ H21), 3.72 (s, 3 H, 3ϫ H20) ppm. 13C NMR
ane) to give 20 (30 mg, 52%) as a white solid, m.p. 229–231 °C. 1H
(100 MHz, CDCl3): δ = 168.6 (C12a), 164.0 (C11a), 155.0 (C11), NMR (500 MHz, CDCl3, numbering as in Figure 17): δ = 8.22 (d,
141.6 (C12), 146.3 (C4a, C8b), 145.4 (C4b, C8a), 144.2 (C14a), J = 7.5 Hz, 2 H, H18, H24), 7.55 (t, J = 7.5 Hz, 2 H, H17, H23),
143.5 (C18a), 136.5 (C14), 127.0 (C16), 126.0 (C2, C7), 125.6 (C17),
125.4 (C3, C6), 124.4 (C4, C5), 124.3 (C1, C8), 123.9 (C18), 121.6
7.39 (t, J = 7.5 Hz, 2 H, H16, H22), 7.28 (t, J = 7.5 Hz, 2 H, H15,
H21), 7.20 (dd, J = 9 Hz, 4 H, H2, H3, H6, H7), 6.93 (d, J =
(C15), 61.3 (C9), 52.8 (C20, C21), 51.3 (C10) ppm. MS (ES) 433 7.5 Hz, 2 H, H14, H20), 6.85 (d, J = 9 Hz, 4 H, H1, H4, H5, H8),
(100%, [M – H]–). C29H22O4 (434.49): calcd. C 80.17, H 5.10; found
C 80.11; H 4.99.
3.47 (s, 4 H, 2ϫ H11, 2ϫ H12) ppm. 13C NMR (125 MHz): δ =
145.1 (C13, C19), 141.7 (C4a, C4b, C8a, C8b), 129.6 (C14, C20),
129.2 (C1, C4, C5, C8), 128.5 (C2, C3, C6, C7), 127.6 (C15, C21),
127.5 (C17, C23), 127.2 (C18, C24), 127.1 (C16, C22), 79.2 (C9,
C10), 67.5 (C11, C12) ppm. C28H24O2 (392.50): calcd. C 85.68, H
6.16; found C 86.00, H 5.95.
Preparation of 9-Phenyl-11,12-(N-methylpyrrolidiono)dibenzobarrel-
ene (15): To a suspension of 2c (25 mg, 0.1 mmol) in 1,4-dioxane
(0.4 mL) N-methylmaleimide (25 mg, 0.22 mmol) was added and
the mixture was heated at 120 °C for 1 day after which time it was
cooled to give 15 (30 mg, 82%) as a white solid, m.p. 280–282 °C. X-Ray Measurements for 7, anti-3a, 3b, rac-9, meso-9, 10, 11, 13–
1H NMR (500 MHz, CDCl3, –15 °C, numbering as in Figure 14):
20: Crystallographic data were collected using a Bruker SMART
δ = 8.10 (d, J = 8 Hz, 1 H, H18), 7.67 (m, 1 H, H17), 7.53 (m, 2 APEX CCD area detector diffractometer equipped with a Bruker
H, H15, H16), 7.45 (d, J = 7 Hz, 1 H, H5), 7.41 (m, 1 H, H14), SMART 1K CCD area detector and a rotating anode, using graph-
7.37 (d, J = 7 Hz, 1 H, H4), 7.29 (d, J = 7 Hz, 1 H, H1), 7.20 (t,
ite-monochromated Mo-Kα radiation (λ = 0.71073 Å), and are
Table 2. Crystallographic data for anti-3a, 3b, 7, rac-9 and meso-9.
Compound
anti-3a
3b
7
rac-9
meso-9
Empirical formula
Formula weight
Crystal system
Space group
a /Å
b /Å
c /Å
α /deg
C32H22
406.50
C32H22
406.50
C23H15Br
371.26
C20H18
258.34
C20H18
258.34
triclinic
monoclinic
P21/c (#14)
12.2245(16)
8.0153(11)
11.7765(16)
90
113.182(3)
90
1060.7(2)
2
1.273
100(2)
0.072
428
monoclinic
P21/n (#14)
6.7890(13)
9.1933(18)
17.284(3)
90
101.006(5)
90
1058.9(4)
2
1.275
100(2)
0.072
428
triclinic
monoclinic
C2/c (#15)
19.672(2)
5.4423(6)
15.2330(18)
90
122.087(2)
90
1381.7(3)
4
1.242
100(2)
0.070
552
¯
¯
P1 (#2)
P1 (#2)
6.8093(4)
10.8190(7)
11.9887(8)
104.193(1)
105.234(1)
101.668(1)
791.87(9)
2
1.557
100(2)
2.595
376
8.3004(5)
10.1830(7)
18.7058(12)
96.752(1)
99.324(1)
111.291(1)
1426.69(16)
4
1.203
100(2)
0.068
552
β /deg
γ /deg
Volume /Å3
Z
Density (σcalc /gcm–3)
Temperature /K
Absorption coefficient (µ /mm–1)
F(000)
θ range /°
Index ranges
1.81–23.29
–13 Յ h Յ 13
–8 Յ k Յ 8
–13 Յ l Յ 13
6899
2.40–23.28
–7 Յ h Յ 7
–10 Յ k Յ 10
–19 Յ l Յ 19
6869
1.86–28.32
–9 Յ h Յ 9
–14 Յ k Յ 14
–15 Յ l Յ 15
16306
2.44–27.99
–25 Յ h Յ 25
–7 Յ k Յ 7
–20 Յ l Յ 20
6624
2.19–26.74
–10 Յ h Յ 10
–12 Յ k Յ 12
–23 Յ l Յ 23
26052
Reflections measured
Reflections used (Rint
)
1523 (0.0302)
1523/0/145
0.0385, 0.1015
0.0515, 0.1086
1.067
1527 (0.0401)
1527/0/145
0.0500, 0.1050
0.0614, 0.1095
1.146
3924 (0.0411)
3924/0/277
0.0319, 0.0772
0.0367, 0.0792
1.044
1678 (0.0239)
1678/0/127
0.0441, 0.1147
0.0490, 0.1194
1.061
6065 (0.0267)
6065/0/505
0.0400, 0.0967
0.0457, 0.1003
1.025
Data/restraints/parameters
Final R values IϾ2σ(I): R1, wR2
R values (all data):R1, wR2
Goodness-of-fit on F2
Largest diff peak and hole /eÅ–3
0.226, –0.132
0.342, –0.168
0.600, –0.287
0.429, –0.217
0.287, –0.175
5214
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5203–5216

Towards a Molecular Clamp
listed in Table 2, Table 3 and Table 4. A full sphere of the reciprocal
space was scanned by phi-omega scans. A semi-empirical absorp-
tion correction, based on redundant reflections, was performed by
the program SADABS.[26] The structures were solved by direct
methods and refined by full-matrix least-squares on F2 for all data
using the program library SHELXTL.[27,28] Hydrogen atom treat-
ment varied from compound to compound, depending on the crys-
tal quality. In 7, meso-9, rac-9, 14, and 17–21 all hydrogen atoms
were located in the difference Fourier map and allowed to refine
freely. In 3a, 3b, 10, 11, 15 and 16 all hydrogen atoms were added
Table 3. Crystallographic data for 10, 11, 13, 14 and 15.
Compound
10
11
13
14
15
Empirical formula
Formula weight
Crystal system
Space group
a /Å
b /Å
c /Å
C26H18
330.4
C20H10F4
326.28
C26H20O4
396.42
C29H22O4
434.47
C25H19NO2
365.41
orthorhombic
Pbca (#61)
8.379(2)
19.994(5)
20.615(5)
90
orthorhombic
P212121 (#19)
10.7580(12)
12.5118(14)
21.704(2)
90
90
90
2921.3(6)
8
1.484
triclinic
triclinic
monoclinic
P21 (#4)
12.9175(12)
7.8430(8)
17.9211(17)
90
91.152(2)
90
1815.3(3)
4
¯
¯
P1 (#2)
P1 (#2)
8.0579(12)
10.5738(16)
11.9703(19)
79.394(3)
84.371(3)
74.707(3)
965.7(3)
8.1407(10)
15.5856(19)
18.628(2)
66.062(3)
83.831(3)
86.388(3)
2147.2(4)
4
α /deg
β /deg
90
90
γ /deg
Volume /Å3
Z
3445.1(15)
8
1.274
2
Density (σcalc /gcm–3)
Temperature /K
1.363
100(2)
0.091
1.344
100(2)
0.089
1.337
293(2)
0.085
100(2)
100(2)
0.121
Absorption coefficient (µ /mm–1) 0.072
F(000)
θ range /°
Index ranges
1392
1328
416
912
768
1.98–23.35
–9 Յ h Յ 8
–22 Յ k Յ 22
–21 Յ l Յ 22
14419
1.88–26.37
–13 Յ h Յ 13
–15 Յ k Յ 15
–27 Յ l Յ 27
25884
1.73–26.44
–10 Յ h Յ 10
–13 Յ k Յ 13
–14 Յ l Յ 14
17283
1.43–23.28
–9 Յ h Յ 9
–17 Յ k Յ 17
–20 Յ l Յ 20
14609
1.14–26.43
–16 Յ h Յ 16
–9 Յ k Յ 9
–22 Յ l Յ 22
16191
Reflections measured
Reflections used (Rint
Data/restraints/parameters
Final R values IϾ2σ(I): R1, wR2 0.0593, 0.1505
)
2454 (0.0588)
2454/0/235
3353 (0.0300)
3353/40/518
0.0569, 0.1477
0.0601, 0.1511
1.081
3957 (0.0305)
3957/0/351
0.0355, 0.0863
0.0416, 0.0902
1.038
6107 (0.0286)
6107/0/599
0.0610, 0.1613
0.0694, 0.1663
1.065
4000 (0.0292)
4000/1/507
0.0363, 0.0885
0.0405, 0.0907
1.031
R values (all data):R1, wR2
0.0757, 0.1662
1.023
Goodness-of-fit on F2
Largest diff peak and hole /eÅ–3 0.482, –0.252
0.654, –0.215
0.313, –0.203
0.400, –0.251
0.160, –0.224
Table 4. Crystallographic data for 16, 17, 18, 19 and 20.
Compound
16
17
18
19
20
Empirical formula
Formula weight
Crystal system
Space group
a /Å
b /Å
c /Å
C28H22
358.46
C28H24O
376.47
C26H20O2
364.42
C28H24O2
392.47
C28H22O2
390.46
monoclinic
P21/c (#14)
20.3035(14)
18.5851(13)
9.8377(7)
90
90.085(2)
90
3712.2(5)
8
triclinic
triclinic
triclinic
monoclinic
C2/c (#15)
22.0418(16)
8.5713(6)
21.1386(15)
90
92.319(1)
90
3990.4(5)
8
¯
¯
¯
P1 (#2)
P1 (#2)
P1 (#2)
9.6559(8)
10.5285(8)
10.6096(9)
88.107(2)
72.505(1)
74.525(1)
990.10(14)
2
6.7449(10)
8.0370(11)
8.9216(13)
102.967(3)
96.351(3)
106.587(3)
443.71(11)
1
7.0964(5)
8.9046(7)
8.9274(7)
104.086(2)
95.407(2)
113.174(1)
491.46(6)
1
α /deg
β /deg
γ /deg
volume /Å3
Z
Density (σcalc /gcm–3)
Temperature /K
1.283
100(2)
1.263
100(2)
0.075
1.364
100(2)
0.085
1.326
100(2)
0.082
1.300
100(2)
0.080
Absorption coefficient (µ /mm–1) 0.072
F(000)
θ range /°
Index ranges
1520
400
192
208
1648
1.10–23.32
–22 Յ h Յ 22
–20 Յ k Յ 20
–8 Յ l Յ 10
18062
2.01–26.00
–11 Յ h Յ 11
–12 Յ k Յ 12
–13 Յ l Յ 13
17084
2.38–30.47
–9 Յ h Յ 9
–11 Յ k Յ 11
–12 Յ l Յ 12
10310
2.41–31.98
–10 Յ h Յ 10
–13 Յ k Յ 12
–13 Յ l Յ 13
12003
1.85–28.31
–29 Յ h Յ 29
–11 Յ k Յ 11
–27 Յ l Յ 28
19710
Reflections measured
Reflections used (Rint
Data/restraints/parameters
Final R values IϾ2σ(I): R1, wR2 0.0337, 0.0822
)
5355 (0.0268)
5355/0/506
3886 (0.0207)
3886/0/358
0.0375, 0.0942
0.0400, 0.0961
1.033
2661 (0.0258)
2661/0/167
0.0472, 0.1295
0.0538, 0.1353
1.037
3203 (0.0242)
3203/0/184
0.0479, 0.1310
0.0547, 0.1369
1.036
4958 (0.0254)
4928/0/362
0.0445, 0.1145
0.0514, 0.1194
1.027
R values (all data):R1, wR2
0.0413, 0.0862
1.035
Goodness-of-fit on F2
Largest diff peak and hole /eÅ–3 0.196, –0.161
0.351, –0.220
0.541, –0.209
0.553, –0.163
0.424, –0.200
Eur. J. Org. Chem. 2010, 5203–5216
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
5215

K. Nikitin, M. J. McGlinchey et al.
FULL PAPER
Am. Chem. Soc. 2000, 122, 6935–6949; c) L. Grill, K.-H. Rie-
der, F. Moresco, G. Rapenne, S. Stojkovic, X. Bouju, C.
Joachim, Nat. Nanotechnol. 2007, 2, 95–98.
at calculated positions and refined using a riding model. Their iso-
tropic temperature factors were fixed to 1.2 times the equivalent
isotropic displacement parameters of the carbon atom to which the
H-atom is attached.
[7]
a) T. A. Albright, P. Hofmann, R. Hoffmann, C. P. Lillya, P. A.
Dobosh, J. Am. Chem. Soc. 1983, 105, 3396–3411; b) Y. F. Op-
runenko, Russ. Chem. Rev. 2000, 69, 683–704; c) S. Brydges, N.
Reginato, L. P. Cuffe, C. M. Seward, M. J. McGlinchey, C. R.
Chim. 2005, 8, 1497–1505; d) E. Kirillov, S. Kahlal, T. Roisnel,
T. Georgelin, J.-Y. Saillard, J. F. Carpentier, Organometallics
2008, 27, 387–393; e) G. Yamamoto, Pure Appl. Chem. 1990,
62, 569–574.
a) L. E. Harrington, L. S. Cahill, M. J. McGlinchey, Organo-
metallics 2004, 23, 2884–2891; b) K. Nikitin, H. Müller-Bunz,
Y. Ortin, M. J. McGlinchey, Org. Biomol. Chem. 2007, 5, 1952–
1960.
a) H. Förster, F. Vögtle, Angew. Chem. 1977, 89, 443–455; An-
gew. Chem. Int. Ed. Engl. 1977, 16, 429–441; b) G. Bott, L. D.
Field, S. Sternhell, J. Am. Chem. Soc. 1980, 102, 5618–5626; c)
L. Lunazzi, A. Mazzanti, M. Minzoni, J. Org. Chem. 2006,
71, 9297–9301; d) R. Ruzziconi, S. Spizzichino, L. Lunazzi, A.
Mazzanti, M. Schlosser, Chem. Eur. J. 2009, 15, 2645–2652.
CCDC-756592 (for 7), -756591 (for anti-3a), -756588 (for 3b),
-756593 (for rac-9), -756590 (for meso-9), -756595 (for 10), -756586
(for 11), -756596 (for 13), -756587 (for 14), -756599 (for 15),
-756600 (for 16), -756598 (for 17), -756589 (for 18), -756597 (for
19), -756594 (for 20) contain the supplementary X-ray crystallo-
graphic 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.
[8]
[9]
Acknowledgments
We thank Science Foundation Ireland and University College Dub-
lin for generous financial support, and Dr. J. Muldoon for the 2D
EXSY data on molecule 20.
[10]
[11]
K. Nikitin, H. Müller-Bunz, Y. Ortin, M. J. McGlinchey,
Chem. Eur. J. 2009, 15, 1836–1843.
As emphasised by Sternhell the geometric parameters, such
as CAD or r* have been introduced merely as a convenient
measure of the severity of non-bonded interactions and give
no insight into the actual details of geometry of the transition
state.
D.-W. Lee, J. Yun, Bull. Korean Chem. Soc. 2004, 25, 29–30.
H. Dang, M. A. Garcia-Garibay, J. Am. Chem. Soc. 2001, 123,
355–356.
K. Nikitin, H. Müller-Bunz, Y. Ortin, W. Risse, M. J. McGlin-
chey, Eur. J. Org. Chem. 2008, 3079–3084.
N. M. Sergeyev, K. F. Abdulla, V. R. Skvarchenko, J. Chem.
Soc., Chem. Commun. 1972, 368–370.
J. P. N. Brewer, I. F. Eckhard, H. Heaney, B. A. Marples, J.
Chem. Soc. C 1968, 664–676.
J. Trotter, F. C. Wireko, Acta Crystallogr., Sect. C 1991, 47,
1275–1277.
L. R. C. Barclay, R. A. Chapman, Can. J. Chem. 1965, 43,
1754–1760.
W. J. Wasserman, M. C. Kloetzel, J. Am. Chem. Soc. 1953, 75,
3036–3038.
P. Reutenauer, P. J. Boul, J.-M. Lehn, Eur. J. Org. Chem. 2009,
1691–1697.
C. Pinazzi, Ann. Chim. 1962, 7, 397–408.
J. L. Cook, C. A. Hunter, C. M. R. Low, A. Perez-Velasco, J. G.
Vinter, Angew. Chem. 2007, 119, 3780–3783; Angew. Chem. Int.
Ed. 2007, 46, 3706–3709.
[1] a) V. Pophristic, L. Goodman, Nature 2001, 411, 565; b) M.
Oki, Top. Stereochem. 1983, 14, 1–81; c) S. Brydges, L. E. Har-
rington, M. J. McGlinchey, Coord. Chem. Rev. 2002, 233/234,
75–105.
[2] a) M. J. Hafezi, F. J. Sharif, THEOCHEM 2008, 814, 43–49;
b) J. R. Durig, D. A. C. Compton, J. Phys. Chem. 1979, 83,
265–268.
[9b]
[12]
[13]
[3] a) M. Nakamura, M. Oki, Bull. Chem. Soc. Jpn. 1975, 48,
2106–2111; b) M. Oki, I. Fujino, D. Kawaguchi, K. Chuda, Y.
Moritaka, Y. Yamamoto, S. Tsuda, T. Akinaga, M. Aki, H.
Kojima, N. Morita, M. Sakurai, S. Toyota, Y. Tanaka, T. Tan-
uma, G. Yamamoto, Bull. Chem. Soc. Jpn. 1997, 70, 457–469;
c) G. Yamamoto, M. Suzuki, M. Oki, Angew. Chem. 1981, 93,
580–581; Angew. Chem. Int. Ed. Engl. 1981, 20, 607–608; d) D.
Zehm, W. Fudickar, M. Hans, U. Schilde, A. Kelling, T. Linker,
Chem. Eur. J. 2008, 14, 11429–11441; e) A. Caspar, S. Alten-
burger-Combrisson, F. Gobert, Org. Magn. Reson. 1978, 11,
603–606; f) P. J. Marriott, Y.-H. Lai, J. Chromatogr. 1988, 447,
29–41.
[4] a) K. M. Mislow, Introduction to Stereochemistry, W. A. Benja-
min, Inc., 1965, pp. 78–79; b) G. H. Christie, J. J. Kenner, J.
Chem. Soc. 1922, 121, 614–620; c) R. Adams, H. C. Yuan,
Chem. Rev. 1933, 12, 261–338; d) G. Yamamoto, T. Nemoto,
Y. Ohashi, Bull. Chem. Soc. Jpn. 1992, 65, 1957–1966; e) A.
Mazzanti, L. Lunazzi, M. Minzoni, J. E. Anderson, J. Org.
Chem. 2006, 71, 5474–5481; f) G. Bringmann, A. J. P. Morti-
mer, P. A. Keller, M. J. Gresser, J. Garner, M. Breuning, Angew.
Chem. 2005, 117, 5518–5563; Angew. Chem. Int. Ed. 2005, 44,
5384–5427.
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[5] a) T. R. Kelly, M. C. Bowyer, K. V. Bhaskar, D. Bebbington,
A. Garcia, F. Lang, M. H. Kim, M. P. Jette, J. Am. Chem. Soc.
1994, 116, 3657–3658; b) T. R. Kelly, J. P. Sestelo, I. N. Tellitu,
J. Org. Chem. 1998, 63, 3655–3665; c) T. R. Kelly, Acc. Chem.
Res. 2001, 34, 514–522; d) M. Llunell, P. Alemany, J. M. Bofill,
ChemPhysChem 2008, 9, 1117–1119; e) T. F. Magnera, J. Michl,
Top. Curr. Chem. 2005, 262, 63–97, and references cited therein;
f) I. Yoon, D. Bentez, Y.-L. Zhao, O. Miljanic, S.-Y. Kim, E.
Tkatchouk, K. C.-F. Leung, S. I. Khan, W. A. Goddard, J. F.
Stoddart, Chem. Eur. J. 2009, 15, 1115–1122; g) R. D. Ras-
berry, K. D. Shimizu, Org. Biomol. Chem. 2009, 7, 3899–3905;
h) V. Serreli, C.-F. Lee, E. R. Kay, D. A. Leigh, Nature 2007,
445, 523–527; i) J. Hernandez, E. R. Kay, D. A. Leigh, Science
2004, 306, 1532–1537; j) G. S. Kottas, L. I. Clarke, D. Horinek,
J. Michl, Chem. Rev. 2005, 105, 1281–1376.
[23]
D. B. Li, W. F. Paxton, R. H. Baughman, T. J. Huang, J. F.
Stoddart, P. S. Weiss, MRS Bull. 2009, 34, 671–681, and refer-
ences cited therein.
H. S. Gutowsky, C. H. Holm, J. Chem. Phys. 1956, 25, 1228–
1234.
B. H. Klandermann, T. R. Criswell, J. Org. Chem. 1969, 34,
3426–3430.
G. M. Sheldrick, SADABS, Bruker AXS Inc., Madison, WI
53711, 2000.
G. M. Sheldrick, SHELXS-97, University of Göttingen, Ger-
many, 1997.
G. M. Sheldrick, SHELXL-97-2, University of Göttingen,
Germany, 1997.
[24]
[25]
[26]
[27]
[28]
[6] a) H. Iwamura, K. Mislow, Acc. Chem. Res. 1988, 21, 175–182;
b) T. R. Kelly, R. A. Silva, H. De Silva, S. Jasmin, Y. Zhao, J.
Received: June 13, 2010
Published Online: August 16, 2010
5216
www.eurjoc.org
© 2010 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2010, 5203–5216