Per-Substituted [8]Circulene and Its Non-Planar Fragments: Synthesis, Structural Analysis, and Properties

Article abstract of DOI:10.1002/chem.201600124

The syntheses, structures, and physical properties of a full series of benzannulated tetraphenylenes are reported. The palladium-catalyzed annulation of tetraiodo-substituted 2,3,6,7,10,11,14,15-octamethyltetraphenylene with insufficient di(4-anisyl)ethyne yielded a mixture of per-substituted [8]circulene and its non-planar fragments, including mono-, para-di-, ortho-di-, and triannulated products. Their structures were unambiguously verified by X-ray crystallography. Successive benzannulations significantly affect the molecular geometries, dynamic behaviors, and physical properties of the compounds. In this series of compounds, [8]circulene is the most strained one, as reflected by the significant deplanarization of the phenanthrene moieties (ca. 63° in the bay region) and the fact that it has the highest strain energy (120.6 kcal mol^?1). The dynamic behaviors of these compounds were examined both experimentally and theoretically. The ring flipping of per-substituted [8]circulene is confirmed to proceed through pseudorotation with a barrier of around 21 kcal mol^?1, whereas its non-planar fragments require much more energy for the ring inversion. The photophysical and electrochemical properties of the investigated compounds depend strongly on the extent of efficient π conjugation. The successive benzannulations red-shift both the absorption and the emission bands, and reduce the first oxidation potential.

Full text of DOI:10.1002/chem.201600124

DOI: 10.1002/chem.201600124
Full Paper
&
Circulenes
Per-Substituted [8]Circulene and Its Non-Planar Fragments:
Synthesis, Structural Analysis, and Properties
[a]
[a]
[a]
[b]
[a]
Chieh-Ning Feng, Wen-Ching Hsu, Jen-Yi Li, Ming-Yu Kuo,* and Yao-Ting Wu*
Dedicated to Professor Yoshito Tobe on the occasion of his 65th birthday
Abstract: The syntheses, structures, and physical properties
of a full series of benzannulated tetraphenylenes are report-
ed. The palladium-catalyzed annulation of tetraiodo-substi-
tuted 2,3,6,7,10,11,14,15-octamethyltetraphenylene with in-
sufficient di(4-anisyl)ethyne yielded a mixture of per-substi-
tuted [8]circulene and its non-planar fragments, including
mono-, para-di-, ortho-di-, and triannulated products. Their
structures were unambiguously verified by X-ray crystallog-
raphy. Successive benzannulations significantly affect the
molecular geometries, dynamic behaviors, and physical
properties of the compounds. In this series of compounds,
638 in the bay region) and the fact that it has the highest
strain energy (120.6 kcalmol ). The dynamic behaviors of
À1
these compounds were examined both experimentally and
theoretically. The ring flipping of per-substituted [8]circulene
is confirmed to proceed through pseudorotation with a barri-
À1
er of around 21 kcalmol , whereas its non-planar fragments
require much more energy for the ring inversion. The photo-
physical and electrochemical properties of the investigated
compounds depend strongly on the extent of efficient p
conjugation. The successive benzannulations red-shift both
the absorption and the emission bands, and reduce the first
oxidation potential.
[
8]circulene is the most strained one, as reflected by the sig-
nificant deplanarization of the phenanthrene moieties (ca.
Introduction
lene is the most curved aromatic bowl with a POAV (p-orbital
[
8]
axis vector) pyramidalization angle of up to 178.
2
The introduction of non-six-membered rings into sp -hybrid-
Both [7]circulene and [8]circulene are saddle-shaped, but the
inclusion of a larger ring creates a more crowed environment.
Accordingly, the latter compound has a higher strain than the
ized carbon hexagonal p networks forms non-planar polycylic
arenes. [n]Circulene is the best candidate to explain systemati-
cally the structural variation because it contains a central n-
sided polygon that is completely surrounded and fused by
[
9]
former one. Replacement of the benzenoids in non-planar
[n]circulenes with small rings may form planar analogues, rep-
[
1]
[10]
benzenoids. Three members of the circulene family—[5]circu-
resentative examples of which include tetraoxo[8]circulene,
[2]
[3]
[11]
lene (or corannulene), [6]circulene (or coronene), and [7]cir-
octathio[8]circulene,
tetracyclopenta[def,jkl,pqr,vwx]tetraphe-
[4]
[12]
[13]
culene (or pleiadannulene) —have so far been prepared, and
their structures have been identified conclusively. [5]Circulene
and [7]circulene deviate from planarity as they form a bowl
and a saddle shape, respectively. Although studies of the
parent compounds [4]circulene and [8]circulene remain theo-
retical, this geometric rule applies to the recently reported tet-
nylene (TCT),
and tetrabenzotetraaza[8]circulene (TBAC).
Circulenes and their derivatives described above have various
structures that endow them with many interesting physical
properties and chemical reactivity, making them worthy of fun-
damental study and enabling them to be (potentially) applied
in organic materials.
[5]
[6]
rabenzo[4]circulene (quadrannulene) and the per-substituted
Most studies of the per-substituted [8]circulene (1) and tet-
[
6]
[7]
[7]
[
8]circulene (1) /tetrabenzo[8]circulene, which exhibit bowl-
rabenzo[8]circulene have focused on their synthesis and
and saddle-shaped structures, respectively. Tetrabenzo[4]circu-
structural analysis; their properties are barely known. One of
the most interesting features of [8]circulene is the mechanism
of its saddle-to-saddle inversion. Based on theoretical investi-
gations, pseudorotation is the most energetically favorable
route, and the height of its barrier has been found to be only
[
a] C.-N. Feng, W.-C. Hsu, J.-Y. Li, Prof. Y.-T. Wu
Department of Chemistry, National Cheng Kung University
No.1 Ta-Hsueh Road, 70101 Tainan (Taiwan)
E-mail: ytwuchem@mail.ncku.edu.tw
À1 [7a]
around 2.9 kcalmol , which is too small to be confirmed ex-
perimentally. Studies of a full series of benzannulated tetraphe-
nylenes may provide more information about the chemistry of
[
b] Prof. M.-Y. Kuo
Department of Applied Chemistry, National Chi Nan University
No. 1 University Rd., 54561 Puli, Nantou (Taiwan)
[
8]circulene. Therefore, this investigation elucidates the synthe-
sis, structural analysis, and physical properties of per-substitut-
ed [8]circulene and its non-planar fragments.
Supporting information for this article can be found under
http://dx.doi.org/10.1002/chem.201600124.
Chem. Eur. J. 2016, 22, 9198 – 9208
9198
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
revealing that they were statically in a non-planar environment
at room temperature on the NMR timescale.
Structural analysis
X-ray-quality crystals of OMTP and compounds 3, 4, 5, 6, and
1
d were obtained by slow evaporation of a CHCl /MeOH sol-
3
[
15]
vent mixture. Table 1 presents the relevant crystallographic
data. The remarkable features of the structures in this family
include the effect of successive annulations on molecular
shape, deformation of the central eight-membered ring, and
[
16]
the extent of the deplanarization of the benzenoids.
The
central eight-membered ring is tub-shaped, and a change in
its conformation can be evaluated by measuring the bending
angle a (Table 2). The distortion of a six-membered ring is
identified by the torsion angle (q ) of either C1-C2-C3-C4 or
1
C1-C2-C4-C5 (end-to-end twist). Additionally, all of the annulat-
Results and Discussion
Synthesis
The per-substituted [8]circulenes 1a–c have been
generated in moderate yields by palladium-catalyzed
benzannulations of tetraiodo-substituted tetrapheny-
[6]
lenes with the corresponding diarylethynes. Howev-
er, the non-planar fragments of compound 1 cannot
be selectively prepared by using this synthetic proto-
col owing to the following problems: The iodina-
tion of 2,3,6,7,10,11,14,15-octamethyltetraphenylene
(
OMTP) must be performed by using excess iodine,
so the number and positions of iodine atoms cannot
be manipulated. Moreover, multi-iodinated OMTP
and its annulated products, which were formed by
annulations
with
bis(3,5-dimethylphenyl)ethyne,
cannot be separated by column chromatography or
crystallization. Fortunately, the per-substituted [8]cir-
culene 1d and its non-planar fragments, including
the monoannulated compound 3, the para-diannulat-
[
14]
ed compound 4, the ortho-diannulated compound
, and the triannulated compound 6, were obtained
5
by the palladium-catalyzed benzannulation of com-
pound 2 and its regioisomers with insufficient di(4-
anisyl)ethyne (two equivalents), as depicted in
Scheme 1. All of these compounds were successfully
separated by column chromatography because they
had a significantly different polarity. Palladium-cata-
lyzed reductive deiodination removed some iodine
substituents, which did not undergo annulation Scheme 1. Synthesis of per-substituted [8]circulene and its non-planar fragments.
owing to the shortness of the alkyne. However, com-
pounds 3–5 had to be treated with excess of n-butyl-
lithium to yield their pure forms (see the Supporting Informa-
ed products contain a substructure of 9,10-dianisyl-1,2,7,8-tet-
ramethylphenanthrene, whose out-of-plane distortion is quan-
1
13
tion). As revealed by the H and C NMR spectra recorded at
room temperature, the anisyl hydrogen atoms and the carbon
atoms in all of the aforementioned products yielded more sig-
nals than would have been obtained from a planar structure,
tified by gauging the bay torsion angle (q ) of C4-C4a-C4b-C5
2
(Table 2). In a crowded environment, the non-bonded distance
(d ) between the C4 and C5 atoms in the phenanthrenyl bay
4,5
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9199
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
15]
Table 1. Crystallographic data.
1
d
3
4
5
6·3CHCl
3
OMTP
formula
[gmol
solvent
C
96
H
80
O
8
C
48
H
44
O
2
C
64
H
56
O
4
C
64
H
56
O
4
C
83
H
71Cl
9
O
6
32 32
C H
À1
M
W
]
1361.60
CHCl /MeOH
652.83
CHCl /MeOH
889.08
CHCl /MeOH
889.08
CHCl /MeOH
1483.44
CHCl /MeOH
416.57
CHCl /MeOH
3
3
3
3
3
3
T [K]
crystal system
space group
100(2)
orthorhombic
Pnma
100(2)
monoclinic
P2 /n
1
100(2)
triclinic
P1¯
100(2)
triclinic
P1¯
100(2)
triclinic
P1¯
100(2)
monoclinic
P2/n
Z
4
4
2
2
2
2
unit cell dimensions
a [Š]
b [Š]
c [Š]
a [8]
b [8]
g [8]
16.645(9)
26.247(15)
22.438(13)
90
90
90
13.4878(7)
12.0884(6)
22.8930(12)
90
106.027(2)
90
13.8013(9)
13.9855(9)
14.1886(9)
74.264(4)
67.850(4)
68.149(4)
2326.9(3)
35305
12.2707(4)
13.4545(5)
17.1381(5)
100.1680(10)
106.2700(10)
96.2290(10)
2635.84(15)
40173
11.4359(7)
17.2220(10)
19.6013(12)
77.0850(10)
87.2990(10)
74.5930(10)
3627.3(4)
56814
10.6508(15)
9.3570(14)
11.8669(17)
90
94.643(4)
90
3
V [Š ]
9803(10)
3587.5(3)
28984
1178.8(3)
9251
reflections collected 72188
goodness-of-fit on
F
0.921
1.045
1.007
1.087
1.041
1.045
2
final R indices
R
1
=0.0984,
R
1
=0.0758,
R
1
=0.0872,
R
1
=0.0666,
R
1
=0.1002,
R
1
=0.0378,
[
I>2s(I
0
)]
2
wR =0.2173
2
wR =0.2005
2
wR =0.2359
2
wR =0.1982
2
wR =0.2837
2
wR =0.1003
Table 2. Selected structural data of compounds 1a, 1d, 3–6, and OMTP.
1
a
1d
3
4
5
6
OMTP
six-membered-ring twist angle q
bay torsion angle q [8]
non-conduct distance d4,5 [Š]
bending angle a [8]
1
[8] 32.2, 29.2 31.9, 29.8 25.6, 13.4
25.4, 15.1
33.2
2.921
27.1, 31.6, 15.9
52.6
3.262
32.1, 30.9, 29.9, 16.4 <2
2
61.9
3.325
47.0
39.2
62.9
3.343
47.2
35.2
35.7
3.014
47.6, 63.9
3.127, 3.443
–
–
52.3, 46.7, 45.6, 52.6 48.4, 47.6, 45,3, 49.3 41.8, 56.8, 50.0, 49.7 43.2, 51.2, 46.3, 48.1 50.7
22.1 19.2 36.8, 23.4 38.1, 27.7, 6.8
peri-torsion angle q
3
[8]
–
region can be smaller than the sum of the van der Waals radii
[17]
of the two carbon atoms (i.e., 3.40 Š).
Taking the bond
lengths, the aromaticity of a six-membered ring is quantized
by using the HOMA (harmonic oscillator model of aromaticity)
[
18]
index. HOMA values of 1 and 0 specify aromaticity and non-
aromaticity, respectively.
The per-substituted [8]circulene 1d was determined to have
the orthorhombic space group Pnma (Figure 1 and Table 1), as
[
6]
does compound 1a. Both compounds contain only one con-
former with the o-xylyl moieties arrayed in the curve-up/curve-
down positions. The absence of the conformer 1’ was caused
by its poor thermodynamic stability (see below). As presented
in Table 2, compounds 1a and 1d have highly similar struc-
tures, including the twist angles of the six-membered rings
(
q1 =308), the bay torsion angle of the 9,10-diaryl-1,2,7,8-tetra-
methylphenanthrene (q =638), the non-bonded distance
2
(
(
d4,5 =3.33 Š), the bending angle of the eight-membered ring
a=478), and most of the bond lengths and angles. A signifi-
cant difference between compounds 1a and 1d is that the
Figure 1. Structure and selected structural data of compound 1d.
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9200

Full Paper
latter has longer bonds a [1.459(3) vs. 1.447(1) Š], b [1.463(3)
vs. 1.447(1) Š], and e [1.365(3) vs. 1.352(1) Š]. These results
arise from the crystal packing and/or the poor crystal quality
of compound 1d, but not from the electronic effect of the me-
thoxy group, because the bond length of the sigma bond be-
tween the backbone and the anisyl substituent is not unusual.
Hence, changing in the aryl substituents did not strongly alter
the structure. All six-membered rings exhibit significant bond
alternations in which the bonds a and b are much longer than
the bonds c, d, and e, and their aromaticity is moderate to
weak, according to the HOMA analysis. Therefore, compound
[19]
1
is more like a [8]radialene than a 1,3,5,7-cyclooctatetraene-
based arene.
Figure 3. Structure and selected structural data of compound 4.
The saddle-shaped OMTP crystallizes in the monoclinic
group P2/n (Table 1). Its central eight-membered ring was de-
termined to be tub-shaped with a bending angle a of around
[
20]
5
18, which exceeds that of 1,3,5,7-cyclooctatetraene (438). All
structures and aromaticities to those in compound 3 (Table 2).
The substituents affect the conformation of the backbone. The
main differences between compound 4 and its parent com-
six-membered rings in OMTP are almost planar (q <28), and
they can be treated as pure aromatic units with a HOMA value
of 0.97. Monoannulation of OMTP generated the phenan-
1
[
23]
pound, dibenzo[def,pqr]tetraphenylene,
are that the latter
threne-based cyclophane 3 with C symmetry (Figure 2). The
has a smaller dihedral angle between the two phenathrenyl
frameworks (68.68) and some shorter perimeter bonds al-
though the degrees of deplanarization of their phenanthrene
2
substructures are similar (q =34.6 vs. 33.28 in compound 4).
2
The ortho-diannulated compound 5 was determined to ex-
hibit the triclinic space group P 1¯ , and a saddle-shaped struc-
ture with C symmetry (Figure 4). The mean bending angle
s
Figure 2. Structure and selected structural data of compound 3.
eight-membered ring is quasi-tub-shaped with some twist with
bending angles of around 46 and 528. The twist angles q of
1
the rings C and D were found to be 25.6 and 13.48, respective-
ly, whereas ring E is nearly planar—which is consistent with
the results of the HOMA analysis that the rings C and D/E have
weak and high aromaticity, respectively. The extent of distor-
tion of the phenanthrene framework in compound 3 (q =
2
3
5.78) is very close to those in 4,5-dimethylphenanthrene
[
21]
[22]
(
33.08) and 5,6,7,8-tetraphenyl[5]phenacene (31.18).
Figure 4. Structure and selected structural data of compound 5.
a of the eight-membered ring is 49.68, and the angle is largest
(
56.88) next to ring K. The rings H (q =27.18) and I (31.68) are
1
much more twisted than the rings J (15.98) and K (ꢀ08). Con-
sequently, the rings H (0.36) and I (0.39) have lower aromaticity
than the rings J (0.76) and K (0.98), based on HOMA analysis.
The bay torsion angle (q ) of a phenanthrene moiety in com-
2
The para-diannulated compound 4 was found to have the
triclinic space group P 1¯ and a structure with D2 symmetry
Figure 3). The two phenanthrene moieties in compound 4 are
twisted with a dihedral angle of 77.98 and have highly similar
pound 5 was obtained to be 558, which exceeds those in com-
pounds 3 and 4 (ꢀ358). Notably, the [5]helicene framework in
compound 5 does not adopt a helical shape, but its conforma-
tion is like the structure of the transition state for the racemi-
(
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9201
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[
24]
zation of the parent [5]helicene. The two extremities (i.e.,
the OMTP counterpart. This structural characteristic arises from
[
28]
rings J and J’) are oriented face-to-face.
The triannulated compound 6 crystallizes with three mole-
cules of chloroform in the triclinic space group P 1¯ , and has
the high barrier to the ring flipping of tetraphenylene and/
or the fact that the conformer observed herein is the most
stable one (see below). In all annulated products described
above, the methyl groups at the C1- and C8-positions in the
9,10-diaryl-1,2,7,8-tetramethylphenanthrene moiety align in
such a way that one twists upward and the other distorts
downward from two aryl substituents, as shown in Table 2. The
steric repulsion between an aryl ring and its neighboring
a saddle-shaped structure with C2 symmetry (Figure 5). The
average bending angle of the eight-membered ring in com-
pound 6 (a=about 478) is slightly less than that in compound
5
. The high twist angles q of the rings L (32.18), M (30.98), and
1
methyl group affects the peri-torsion angle q that involves the
3
methyl carbon atom, the C1 and C10 atoms, and the indicated
aryl carbon atom. A high peri-torsion angle was found in a com-
pound with continued phenanthrene subunits (Table 2), indi-
cating the accumulation of strain on the annulation. Notably,
all of the studied compounds herein lack significant intermo-
lecular interactions, presumably owing to the alignments of
aryl groups perpendicular to their backbones. The geometric
calculations for compounds 1b, 1d, and 3–6 by using density
functional theory (DFT) at the wB97X-D/6-31G** level of
[
29]
[30]
theory with the Gaussian 09 program are highly consistent
with the X-ray structural data (see the Supporting Information).
The parent and substituted compounds exhibit similar shapes,
but some of them have significantly different bond lengths
and deplanarizations of their arene moieties. The parent series
has less twisted benzenoids and phenanthrene moieties than
the substituted derivatives.
Figure 5. Structure and selected structural data of compound 6.
N (29.98) reveal significant deformation from planarity, and all
of them exhibit moderate aromaticity. The less twisted ring O
Analysis of the strain energy
(
16.48) has high aromaticity. The maximum bay torsion angle
The strain energies of compound 1d and its fragments were
calculated at the wB97X-D/6-31G** level of theory based on
two homodesmotic reactions presented in Figure 6. Biphenyl,
4-methoxybiphenyl, and phenanthrene are assumed to be
strain-free. Successive annulations gradually introduce strain,
and this finding is consistent with the results of the above-pre-
sented structural analyses, which revealed severe stepwise in-
creases in the deplanarization of the 9,10-dianisyl-1,2,7,8-tetra-
methylphenanthrenyl moiety. Accordingly, compound 1d, the
most strained member of this series of compounds, acquires
of the phenanthrene framework was observed at the position
M-L-M’, and the value (648) is very close to that of compound
1
helix, and the two extremities, that is the rings O and O’, are
almost perpendicular to each other with a dihedral angle of
(628). The [7]helicene moiety in compound 6 is a highly bent
7
9.28. For comparison, the maximum twist angle of the benze-
noids (q ), the torsion angle of the phenanthrene moiety (q ),
1
2
and the dihedral angle between two extremities in the parent
[25]
[
7]helicene are 24.4, 24.7, and 32.08, respectively. Two ana-
[26]
À1
logues of closed [7]helicene, that is, compounds 7 and the
an extra 120.6 kcalmol of strain energy by the benzannula-
[
27]
metallocene 8, have been prepared. The first is planar, and
the structure of the other in the solid state has not been eluci-
dated.
tions of OMTP, and a large strain is introduced in four steps.
Each new annulation needs energy in the range of 26.2–
À1
33.4 kcalmol . The triannulated compound 6 is obtained from
Changes to the structural features upon successive benzan-
nulations can be summarized as follows. The structure alters
from a saddle (OMTP) through a non-saddle-shaped (com-
pounds 3 and 4) to a saddle (compounds 5, 6, and 1) again.
Annulations slightly reduce the bending angle a of the eight-
membered ring from 51 (OMTP) to 478 (compound 1). The bay
torsion angle q2 of a phenanthrene moiety increases from
around 35 (for compounds 3 and 4) to about 638 (for com-
pounds 1 and 6). The newly formed six-membered ring, which
contains two 4-anisyl substituents, is always located at
a bridged position to connect a curve-up o-xylyl moiety with
a curve-down o-xylyl ring, as maintaining the saddle shape of
the monoannulated compound 3 through either the para-di-
annulated compound 4 or the ortho-diannulated compound 5.
À1
The first step in the former route requires 26.2 kcalmol , but
À1
the second step is the most endothermic one (33.4 kcalmol )
in the entire annulation processes. The fourth annulation to
À1
form compound 1d introduces 31.0 kcalmol of additional
strain. The results of the theoretical studies of the strain
energy agree with the experimental findings herein that all an-
nulated products were non-selectively formed with compara-
ble yields (Scheme 1). It is not unexpected that annulations of
the parent series introduce much less strain energy at this
[
31]
level of theory.
9202
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
À1
Figure 6. Extra energy cost (in kcalmol ) for each annulation calculated at the wB97X-D/6-31G** level of theory.
Dynamic behavior
in the Supporting Information. Notably, peaks of the aryl hy-
drogen atoms of the m-xylyl substituent became highly broad
at 443 K, and the coalescence of their signals could not be ac-
curately identified.
The inversion mechanisms of the saddle-shaped circulenes
have not yet been conclusively established. The barriers to the
inversion of the parent compounds [7]circulene and [8]circu-
lene through pseudorotation were calculated theoretically to
As was stated above, the 4-anisyl substituent in all annulat-
ed products, such as in compounds 1d and 3–6, is statically in
a non-planar environment at room temperature. The barrier of
the ring flipping could be estimated by measuring the coales-
cence temperature of the corresponding anisyl signals in the
[
32]
À1[7a]
be only 0.01 and 2.9 kcalmol
, respectively, whereas the
ring flipping of [8]circulene through a planar transition state
À1
requires 125 kcalmol , which significantly exceeds that of
À1
[
7]circulene (8.5 kcalmol ) through the same inversion mecha-
1
1
nism. Therefore, pseudorotation is the energetically favorable
route. However, the barrier to the ring inversion of 1,2-dihy-
H NMR spectrum. The 2D H NMR techniques COSY and
NOESY provided useful information concerning the establish-
ment of the relationships among the four hydrogen atoms
and a methoxy group within an anisyl ring, but not the
accurate position of the 4-anisyl ring in compound 5 or 6.
Given these relationships, the dynamic behaviors of solutions
dro[7]circulene (9) has been determined to be 12.2 kcal
À1 [4b]
mol .
If the inversion of [7]circulene proceeds through
a planar transition state, then its barrier should be smaller than
that of the inversion of compound 9. This statement is based
[
33]
on the fact that 4,5-dimethyl-9,10-dihydrophenanthrene has
a higher inversion barrier than 4,5-dimethylphenanthrene (23.4
of compounds 3–6 in [D ]o-dichlorobenzene were examined
4
by using VT NMR techniques. Compounds 3 and 4 exhibited
very similar coalescence temperatures of approximately 340 K,
and their barriers were determined to be 16.4 and 17.2 kcal
À1
vs. 16.4 kcalmol ).
The pseudorotation of the parent compounds [7]circulene
and [8]circulene cannot be examined experimentally because
their structures are highly symmetric and/or the barriers are
very low. However, measurements of the per-substituted [8]cir-
culene 1 may be possible. The coalescence of the m-xylyl sig-
À1
mol , respectively. The spectra of compounds 5 and 6 at high
temperature became complex because they have more anisyl
groups and some of their signals overlap the residual peaks of
the solvent. From dynamic NMR simulations, the barriers for
compounds 5 and 6 were estimated to be approximately 19
1
nals in the variable-temperature (VT) H NMR experiments that
À1
were performed with a solution of compound 1b in degassed
and 20 kcalmol , respectively (see the Supporting Informa-
[
D ]o-dichlorobenzene revealed that the peaks of two methyl
4
tion).
groups in a xylyl substituent merged into a broad band at
All annulated products investigated herein, contain a sub-
structure of non-planar 9,10-diaryl-1,2,7,8-tetramethylphenan-
threne, but the rotational behavior of the symmetric aryl moi-
eties cannot be reflected in the coalescence of symmetry relat-
4
43 K, which is assumed to be the coalescence temperature,
À1
yielding a barrier value of 20.7 kcalmol . Dynamic NMR simu-
[
34]
À1
lations gave an identical result (20.7 kcalmol ), as presented
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9203
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
ed diastereotopic signals. A VT H NMR experiment was con-
dorotation was examined (Figure 8). In this calculation, all aryl
substituents in compound 1 were simplified as phenyl groups
because investigations on compound 1b met a convergence
problem. The ring-flipping process converts the saddle 1e
(Ar=Ph) to the saddle 1e’. Because the conformer 1e has four
identical peri clashes between a phenyl moiety and its neigh-
boring methyl group, the entire process of the pseudorotation
resembles the moving of a running wave around the molecular
periphery, and generates a dynamic symmetry of D , causing
ducted with a solution of 9,10-di(3-tolyl)-1,2,3,6,7,8-hexame-
thylphenanthrene (10) in [D ]o-dichlorobenzene, in order to
4
understand the rotational behavior of the aryl moieties in the
annulated products studied herein. The methyl signals of two
m-tolyl groups remained separated below 318 K. Unexpectedly,
8h
the merging of the symmetry related diastereotopic signals of
the pendant aryl group. Unlike the parent [8]circulene, com-
pound 1 has a high peri strain, which allows the ring inversion
to be experimentally quantified. The energy difference be-
tween the saddles 1e and 1e’ was estimated to be 16.3 kcal
À1
the measured barrier to the m-tolyl rotation in compound 10
mol . The pseudo-intrinsic reaction coordinate (pseudo-IRC)
À1
(
16.3 kcalmol ) is smaller than these of 9,10-di(3-tolyl)phenan-
method provided more information concerning the intercon-
version between the conformations 1e and 1e’. This process
proceeds through the twisted saddle-shaped transition state
11 and the intermediate 12, and their potential energies are
À1
threne (20.3 kcalmol ) and 9,10-bis[(3-(bromomethyl)phenyl]-
À1 [35]
phenanthrene (21.5 kcalmol ).
In this investigation, the
methyl groups at the C-1-/8- and C-2-/7-positions in the phe-
[
36]
À1
nanthyl framework should contribute to peri interactions
15.3 and 14.0 kcalmol , respectively, higher than that of con-
[37]
and the buttressing effects, respectively, as increasing steric
hindrance. The dynamic behaviors of a highly crowded phen-
anthrene are examined in another study by the authors.
The inversion route of 1,2-dihydro[7]circulene (9) was theo-
retically studied by using the quadratic synchronous transit
former 1e. The conformer 1e’ is unstable, and spontaneously
converts to the stable conformer 1e’’, which is an “enantio-
mer” of structure 1e, through pseudorotation. Therefore, the
ring-inversion process can be reformulated as follows. The in-
terconversion between two stable saddle forms 1e and 1e’’
proceeds through the transition state 1e’ with a barrier of
[
38]
(
QST3) method. The optimized structure of compound 9 is
À1
a twisted saddle with C symmetry (Figure 7). The inversion be-
16.3 kcalmol , which is smaller than the experimental value of
2
À1
2
0.7 kcalmol measured for compound 1b (Ar=m-xylyl). The
latter is expected to have a higher inversion barrier than the
former because the methyl groups in the m-xylyl substituent
provide additional steric influence. The outcome of the insta-
bility of structure 1’ is consistent with the experimental ab-
sence of this conformer. Furthermore, when a solution of com-
pound 1b in [D ]o-dichlorobenzene was heated to 463 K, and
4
1
then cooled to room temperature, its H NMR spectrum was
[
6]
identical to that of an unheated sample. Although the con-
formers 1e and 1e’ have very similar structures, a difference
between the two conformers lies in the relative positions of
the methyl groups at the C1- and C8-positions in their 9,10-di-
phenyl-1,2,7,8-tetramethylphenanthrene subunits. In structure
1
e, one twists up and the other distorts down from two aryl
substituents in an anti manner, whereas both methyl groups in
structure 1e’ point in the same direction. Accordingly, confor-
mer 1e’ is more strained with a high peri-torsion angle (q =
3
4
7.8 vs. 36.98 in structure 1e).
The ring flipping of 4,5-dimethylphenanthrene and diben-
Figure 7. Energy profile for the ring-inversion process of 1,2-dihydro[7]circu-
lene (9).
zo[a,c]cyclooctatetraene has been investigated to proceed
[
24a]
through a planar
and an energetically favorable non-planar
[
39]
À1 [28f]
tween two twisted saddle-shaped structures involves pseudo-
transition state with barriers of 16.1 and 28.5 kcalmol ,
respectively. Although compound 3 contains similar substruc-
tures to both of the aforementioned compounds, its ring inver-
sion requires much more energy, owing to the bay-region
clashes. Issues similar to those associated with the ring inver-
sion of compound 4 are expected to arise. Therefore, the coa-
lescence of the p-anisyl signals of compounds 3 and 4 is
caused by symmetry degeneracy, but not the ring flipping (see
below).
rotation through a transition state with a C -symmetric saddle-
s
shaped geometry, which is highly similar to [7]circulene. The
barrier calculated at the B3LYP/6-31G** level of theory is
À1
1
2.9 kcalmol , which is very close to the experimental value
À1 [4b]
of 12.2 kcalmol . Hence, the saddle-to-saddle inversion of
[
7]circulene should proceed through pseudorotation.
At the wB97X-D/6-31G** level of theory, the saddle-to-
saddle inversion of the per-substituted [8]circulene 1 by pseu-
9204
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
À1
Figure 8. Energy profile for the ring-inversion process of conformer 1e (Ar=Ph). The relative energies are reported in [kcalmol ].
À1
As was stated above, the [5]helicene framework in com-
46.1 kcalmol . Therefore, the coalescence of the p-anisyl sig-
nals identified herein is not caused by the ring flipping.
pound 5 adopts a shape like the transition state of the parent
[40]
compound [5]helicene in racemization. The ring inversion of
The closed helicene has a lower barrier to racemization than
the corresponding regular helicene. This finding is supported
by the experimental results concerning 1,16-dehydro[6]heli-
compound 5 was proposed to proceed through the twisting
of the ring K, generating the C -symmetric transition state 13
2
À1 [41]
À1 [42]
(
Figure 9), in which the dihedral angle between the rings K and
cene (14) (17 kcalmol ) and [6]helicene (36.2 kcalmol ).
The ring flipping of compound 6 was assumed to proceed
through the saddle-shaped transition state 15 with C symme-
try (Figure 9), as confirmed by the pseudo-IRC method. The ge-
ometry of the transition state 15 is like that of structure 1’, and
the [7]helicene subunit of the former adopts a conformation
similar to the transition state for the racemization of the
H is approximately 458. The pseudo-IRC method confirmed the
inversion mechanism. Although the [5]helicene moiety in struc-
ture 13 is “more like” a regular helix, it is more strained than
compound 5, as evidenced by the fact that the bay torsion
angle at the I-H-I’ position (63.58) and the peri-torsion angle
50.28) both exceed those in the ground state (52.6 and 36.88,
respectively). At the wB97X-D/6-31G** level of theory, the
energy barrier to the formation of the transition state 13 is
s
(
[
24]
parent compound [7]helicene. However, the ring inversion of
compound 6 requires more energy than that of compound
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9205
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
À1
pseudorotation with a barrier of around 21 kcalmol was veri-
fied both experimentally and theoretically. However, ring inver-
sions of compounds 5 and 6 require more energy (over 35 kcal
À1
mol ) than that of compound 1, according to theoretical in-
vestigations. A surprising fact is that the barriers to the coales-
cence of the p-anisyl signals for compounds 3–6 are very close
À1
to each other (17–21 kcalmol ), and similar to that of com-
pound 1. A major difference in their ring-flip processes is that
pseudorotation of compound 1 mainly proceeds at the molec-
ular periphery, whereas other examples (i.e., compounds 3–6)
exhibit a significant change to the backbone that involves
a bay-region clash. Therefore, the coalescence of the p-anisyl
signals was caused by symmetry degeneracy, which is created
by a pendant aryl group rather than by the deformation of the
backbone because the former is energetically favorable. One
possibility is that the bending and the rotation of an aryl sub-
stituent overcome the peri strain, as a kind of “entrance ticket”
to pseudorotation of compound 1. This statement is supported
by the fact that the free rotation of the m-tolyl group in com-
pound 10 presents a similar dynamic behavior.
Photophysical properties
The photophysical properties of all annulated products were
investigated (see Table 3 and the Supporting Information), and
they were found to be influenced by the efficient p conjuga-
tion of their backbones. The monoannulated compound 3 and
the para-diannulated compound 4 are both colorless, the
ortho-diannulated compound 5 has a visible color (yellow), the
triannulated compound 6 is bright red, and the per-substituted
[8]circulene 1d is dark red. The longest wavelength of the ab-
sorption peaks (lmax) of compounds 3, 4, 5, 6, and 1d are l=
342, 342, 441, 540 and 543 nm, respectively. Generally, the ab-
sorption bands are significantly red-shifted by successive annu-
lations. The similarity between the photoabsorption properties
of compounds 3 and 4 arises from the fact that interactions
between any two aryl moieties in the backbone with a large
dihedral angle (>758) are inefficient, as established by the X-
ray structural analysis. The value of lmax for compound 6 is
Figure 9. Structures of the transition states 13 and 15.
1
because the rings O and O’ firstly suffer a bay-region clash,
and the non-bonded distance between the two indicated hy-
drogen atoms in structure 15 was found to be 1.82 Š, which is
less than the sum of the van der Waals radii of two hydrogen
[
17]
atoms (i.e., 2.40 Š). Indeed, the calculated barrier (35.9 kcal
À1
mol ) at the wB97X-D/6-31G** level significantly exceeds that
of the ring inversion of compound 1 through pseudorotation,
and that for the coalescence of the p-anisyl signals (20.4 kcal
À1
mol ).
The dynamic behaviors of the studied compounds are sum-
marized as follows. The ring flipping of compound 1 through
[
a]
Table 3. Photophysical and electrochemical properties.
ox
red
E
g
P
l
abs [nm]
l
em [nm]
Stokes shift [nm]
Q.Y.
E
[V]
E
1/2 [V]
HOMO [eV]
LUMO [eV]
E
[eV]
g
E [eV]
[
b]
sh
sh
sh
1
1
1
3
4
5
6
a
387, 413 , 475 , 502, 545
589
591
594
437
431
521
589
–
–
–
–
À0.06, 0.26
À0.05, 0.25
À0.08, 0.25
0.49
0.49
0.17, 0.43
0.05, 0.40
À1.93
À4.74
À4.75
À4.72
À5.29
À5.29
À4.97
À4.85
À2.88
1.86
1.95
1.85
–
–
–
2.05
2.06
2.11
3.25
3.22
2.50
2.15
[
b]
sh
sh
sh
b
d
391 , 411, 470 , 505, 550
À2.00
À2.80
sh
sh
389, 475 , 505, 543
51
95
89
56
49
0.14
0.17
0.34
0.09
0.19
À1.92
À2.87
342
342
–
–
–
–
–
–
–
–
sh
sh
341, 441, 465
386, 506, 540
sh
–
À5
[
(
a] The photophysical properties were measured at a concentration of 10 m in CH
for compounds 3 and 4), rhodamine 110 (for compound 5), and rhodamine 101 (for compounds 6 and 1d) were used as standards for the determination
of the fluorescence quantum yields (Q. Y.). E1/2 and E values are half-wave reductive and irreversible oxidative potentials (vs. the ferrocene/ferrocenium
Fc/Fc ) couple), respectively, measured in a background electrolyte solution of 0.1m nBu
2 2
Cl , and the absorption bands over l=330 nm are listed. Coumarin
red
ox
+
À1
(
4 6 2 2
NPF in CH Cl with a scan rate of 50 mVs . For an irreversible
oxidative wave, the potential was determined by its onset value. The HOMO and LUMO energy levels were calculated according to the equations HOMO=
onset
onset
onset
onset
À(4.8+Eox
) and LUMO=À(4.8+Ered
), where Eox
and Ered
are the onset potentials of the first oxidative and reductive redox wave, respectively.
E
P
The E
g
value is the electrochemical energy gap derived from LUMOÀHOMO, and the E
g
value corresponds to the optical energy gap derived from the
P
onset
lowest energy absorption onset in the absorption spectra (E
g
=1240/labs ). [b] The data of photophysical properties were taken from reference [6].
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9206
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
very close to that of compound 1d. The optical HOMO–LUMO
redshifts of both the absorption and the emission bands, and
a decrease in the first oxidation potential.
P
gaps (E s) of compounds 3 and 4 (around 3.25 eV) and 1d
g
(
2.11 eV) were found to be the largest and the smallest, respec-
tively, in this compound series. Aryl substituents do not partici-
pate significantly in the p system of the backbone in com-
pound 1, as evidenced by the similarity among the photoab-
sorption properties of compounds 1a, b, and d.
All benzannulated products are fluorescent, and the emis-
sion maxima for compounds 3, 4, 5, 6, and 1d were deter-
mined to be at l=437, 431, 521, 589 and 594 nm, respectively.
The emission maxima of compounds 3 and 4 as well as 1 and
Acknowledgements
This work was supported by the Ministry of Science and Tech-
nology of Taiwan (101-2628M-006-002-MY3). We thank Prof.
Sue-Lein Wang and Pei-Lin Chen (National Tsing Hua University,
Taiwan) for the X-ray structure analyses.
Keywords: circulenes
· density functional calculations ·
6
are comparable (see Table 3 and the Supporting Informa-
palladium · pseudorotation · tetraphenylenes
tion). Successive annulations typically cause progressive red
shifts. The Stokes shifts of compounds 3, 4, 5, 6, and 1d were
found to be around 95, 89, 56, 49 and 51 nm, respectively, and
this value decreases as the annulations increase (Table 3), indi-
cating the high structural rigidity of compounds 6 and 1d in
[
[
1] The compound class circulene was first defined by Wynberg, see:
a) J. H. Dopper, H. Wynberg, Tetrahedron Lett. 1972, 13, 763. For reviews
for circulenes, see: b) J. S. Siegel, T. J. Seiders, Chem. in Britain 1995, 31,
313; c) C.-S. Chang, Y.-T. Wu in Science of Synthesis, Vol. 45b (Eds.: J.
Siegel, Y. Tobe), Thieme, Stuttgart, 2009, p.955; d) C.-N. Feng, Y.-C.
Hsieh, Y.-T. Wu, Chem. Rec. 2015, 15, 266.
2] a) W. E. Barth, R. G. Lawton, J. Am. Chem. Soc. 1966, 88, 380. For X-ray
crystallographic studies, see: b) M. A. Petrukhina, K. W. Mack, J. Andreini,
L. T. Scott, J. Org. Chem. 2005, 70, 5713; c) J. C. Hanson, C. E. Nordman,
Acta Crystallogr. Sect. B 1976, 32, 1147. For reviews on corannulene-
based buckybowls, see: d) Y.-T. Wu, J. Siegel, Top. Curr. Chem. 2014, 349,
63; e) Fragments of Fullerenes and Carbon Nanotubes: Designed Synthesis
Unusual Reactions, and Coordination Chemistry (Eds.: M. A. Petrukhina,
L. T. Scott), Wiley, New York, 2012; f) A. Sygula, Eur. J. Org. Chem. 2011,
[
43]
this series.
Electrochemical properties
Table 3 presents the redox properties of the annulated prod-
ucts, obtained by using cyclic voltammetry. The oxidation
waves of all compounds are irreversible, revealing the low
thermodynamic stability of these cationic species. The first oxi-
dation potentials of the studied compounds are very sensitive
to the aromatic p system in the backbone, and their values
1611; g) Y.-T. Wu, J. S. Siegel, Chem. Rev. 2006, 106, 4843.
[3] R. Scholl, K. Meyer, Ber. Dtsch. Chem. Ges. A 1932, 65, 902.
[4] a) K. Yamamoto, T. Harada, M. Nakazaki, T. Naka, Y. Kai, S. Harada, N.
Kasai, J. Am. Chem. Soc. 1983, 105, 7171; b) K. Yamamoto, H. Sonobe, H.
Matsubara, M. Sato, S. Okamoto, K. Kitaura, Angew. Chem. Int. Ed. Engl.
generally decline with successive benzannulations. Compounds
+
3
and 4 [0.49 V vs. ferrocene/ferrocenium (Fc/Fc )] and com-
1996, 35, 69; Angew. Chem. 1996, 108, 69.
pound 1d (À0.08 V) have the highest and lowest oxidation po-
tentials, respectively. A low oxidation potential reflects the in-
stability of the compound under aerobic conditions. Indeed,
solutions of compound 1 at room temperature are stable for
only up to two weeks. Unlike the other compounds, only com-
pound 1 yields a reduction signal. Compounds 1a, 1b, and 1d
[5] Bharat, R. Bhola, T. Bally, A. Valente, M. K. Cyra n´ ski, Ł. Dobrzycki, S. M.
Spain, P. Rempała, M. R. Chin, B. T. King, Angew. Chem. Int. Ed. 2010, 49,
399; Angew. Chem. 2010, 122, 409.
[6] C.-N. Feng, M.-Y. Kuo, Y.-T. Wu, Angew. Chem. Int. Ed. 2013, 52, 7791;
Angew. Chem. 2013, 125, 7945.
[7] a) Y. Sakamoto, T. Suzuki, J. Am. Chem. Soc. 2013, 135, 14074; b) R. W.
Miller, A. K. Duncan, S. T. Schneebeli, D. L. Gray, A. C. Whalley, Chem. Eur.
J. 2014, 20, 3705.
exhibit very similar redox properties, and their electrochemical
[8] a) R. C. Haddon, L. T. Scott, Pure Appl. Chem. 1986, 58, 137; b) R. C.
E
energy gaps (E s) were determined to be approximately
g
Haddon, Acc. Chem. Res. 1988, 21, 243.
1
.9 eV.
[9] H. Christoph, J. Grunenberg, H. Hopf, I. Dix, P. G. Jones, M. Scholtissek,
G. Maier, Chem. Eur. J. 2008, 14, 5604.
[
[
10] H. Erdtman, H.-E. Hçgberg, Tetrahedron Lett. 1970, 11, 3389.
11] K. Y. Chernichenko, V. V. Sumerin, R. V. Shpanchenko, E. S. Balenkova,
V. G. Nenajdenko, Angew. Chem. Int. Ed. 2006, 45, 7367; Angew. Chem.
Conclusion
2
006, 118, 7527.
The chemistry and fundamental properties of per-substituted
[
12] S. Nobusue, H. Miyoshi, A. Shimizu, I. Hisaki, K. Fukuda, M. Nakano, Y.
Tobe, Angew. Chem. Int. Ed. 2015, 54, 2090; Angew. Chem. 2015, 127,
2118.
13] F. Chen, Y. S. Hong, S. Shimizu, D. Kim, T. Tanaka, A. Osuka, Angew.
Chem. Int. Ed. 2015, 54, 10639; Angew. Chem. 2015, 127, 10785.
14] For the synthesis of dibenzo[def,pqr]tetraphenylene, the parent com-
pound of compound 4, see: a) B. Thulin, O. Wennerstrçm, Acta Chem.
Scand. B 1976, 30, 369; b) D. N. Leach, J. A. Reiss, J. Org. Chem. 1978,
[
8]circulene and its non-planar fragments have been compre-
hensively investigated. A full series of benzannulated tetraphe-
nylenes was synthesized, and their structures were unambigu-
ously verified by X-ray crystallography. Successive benzannula-
tions influence the molecular geometries, dynamic behaviors,
and physical properties of the compounds. The strain energy
of a compound is related to the number of benzannulations,
and [8]circulene is the most strained compound in this series.
The ring flipping of per-substituted [8]circulene was confirmed
to proceed by pseudorotation with a barrier of around 21 kcal
[
[
4
3, 2484.
15] X-ray crystallographic data for the compounds: CCDC 1446132 (1d),
446128 (3), 1446129 (4), 1446130 (5), 1446131 (6), and 1446127
OMTP) contain the supplementary crystallographic data for this paper.
[
[
1
(
These data can be obtained free of charge from The Cambridge Crystal-
lographic Data Centre.
À1
mol , whereas the ring inversion of its non-planar fragments
16] For the correlation between the strain and the deplanarization of ben-
zenoids, see: a) P. J. Smith, J. F. Liebman in Strained Aromatic Molecules
Strained Hydrocarbons: Beyond the van’t Hoff and Le Bel Hypothesis (Ed.:
H. Dodziuk), Wiley-VCH, Weinheim. 2009, p.147; b) A. Greenberg, J. F.
requires much more energy. The successive benzannulations
increase the efficient p conjugation of the molecule backbone,
and reduce the HOMO–LUMO energy gap, as reflected by the
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9207
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Liebman, Strained Organic Molecules, Academic Press, New York, 1978,
p.139; c) R. A. Pascal, Jr., Chem. Rev. 2006, 106, 4809.
17] L. Pauling, The Nature of the Chemical Bond, 3rd ed., Cornell University
Press, Ithaca, 1960.
18] a) J. Kruszewski, T. M. Krygowski, Tetrahedron Lett. 1972, 13, 3839. For
a review on HOMA, see:b) T. M. Krygowski, M. K. Cyra n´ ski, Chem. Rev.
R. Kobayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S.
Iyengar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox,
J. B. Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Strat-
mann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L.
Martin, K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dan-
nenberg, S. Dapprich, A. D. Daniels, O. Farkas, J. B. Foresman, J. V. Ortiz,
J. Cioslowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
[
[
2001, 101, 1385.
[
19] a) H. Hopf, G. Maas, Angew. Chem. Int. Ed. Engl. 1992, 31, 931; Angew.
Chem. 1992, 104, 953; b) G. Maas, H. Hopf in The Chemistry of Functional
Groups, Vol. I (Eds.: S. Patai, Z. Rappoport), Wiley, New York, 1997,
p.927.
[31] Decaphenylcorannulene is more strained than corannulene by 53.2 kcal
À1
mol , see: Q. Zhang, K. Kawasumi, Y. Segawa, K. Itami, L. T. Scott, J. Am.
Chem. Soc. 2012, 134, 15664.
[32] M. Shen, I. S. Ignatyev, Y. Xie, H. F. Schaefer, III, J. Phys. Chem. 1993, 97,
3212.
[33] a) R. Cosmo, S. Sternhell, Aust. J. Chem. 1987, 40, 35; b) O. Yamamoto,
H. Nakanishi, Tetrahedron 1973, 29, 781.
[34] Dynamic NMR simulations were performed with the DNMR3 method
implemented into the SpinWorks program package: K. Marat, Spin-
Works, version 2.5.4, 2006.
[35] a) Y.-H. Lai, J. Chem. Soc. Perkin Trans. 2 1986, 1667; b) Y. H. Lai, S. M.
Lee, J. Org. Chem. 1988, 53, 4472.
[
[
[
20] M. Trætteberg, Acta Chem. Scand. 1966, 20, 1724.
21] R. Cosmo, T. W. Hambley, S. Sternhell, J. Org. Chem. 1987, 52, 3119.
22] Z. Gu, G. B. Boursalian, V. Gandon, R. Padilla, H. Shen, T. V. Timofeeva, P.
Tongwa, K. P. C. Vollhardt, A. A. Yakovenko, Angew. Chem. Int. Ed. 2011,
5
0, 9413; Angew. Chem. 2011, 123, 9585.
[
[
23] H. Irngartinger, W. R. K. Reibel, Acta Crystallogr. B 1981, 37, 1768.
24] a) S. Grimme, S. D. Peyerimhoff, Chem. Phys. 1996, 204, 411; b) R. H.
Janke, G. Haufe, E.-U. Würthwein, J. H. Borkent, J. Am. Chem. Soc. 1996,
1
18, 6031.
[36] a) F. Cozzi, M. Cinquini, R. Annuziata, J. S. Siegel, J. Am. Chem. Soc. 1993,
115, 5330; b) H. O. House, J. A. Hrabie, D. VanDerveer, J. Org. Chem.
1986, 51, 921; c) R. L. Clough, J. D. Roberts, J. Am. Chem. Soc. 1976, 98,
1018.
[37] R. N. Armstrong, H. L. Ammon, J. N. Darnow, J. Am. Chem. Soc. 1987,
109, 2077.
[
[
25] M. J. Fuchter, M. Weimar, X. Yang, D. K. Judge, A. J. P. White, Tetrahedron
Lett. 2012, 53, 1108.
26] A. Rajca, M. Miyasaka, S. Xiao, P. J. Boratynski, M. Pink, S. Rajca, J. Org.
Chem. 2009, 74, 9105.
27] A. Sudhakar, T. J. Katz, J. Am. Chem. Soc. 1986, 108, 179.
28] For experimental studies, see: a) H. Huang, T. Stewart, M. Gutmann, T.
Ohhara, N. Niimura, Y.-X. Li, J.-F. Wen, R. Bau, H. N. C. Wong, J. Org.
Chem. 2009, 74, 359; b) P. Rashidi-Ranjbar, Y.-M. Man, J. Sandstrom,
H. N. C. Wong, J. Org. Chem. 1989, 54, 4888; c) A. Rosdahl, J. Sandstrçm,
Tetrahedron Lett. 1972, 13, 4187; d) G. H. Senkler, Jr. , D. Gust, P. X. Ricco-
bono, K. Mislow, J. Am. Chem. Soc. 1972, 94, 8626; e) D. Gust, G. H. Sen-
kler, K. Mislow, J. Chem. Soc. Chem. Commun. 1972, 1345a. Theoretical
[
[
[38] C. Gonzalez, H. B. Schlegel, J. Phys. Chem. 1990, 94, 5523.
[39] H. Scherübl, U. Fritzsche, A. Mannschreck, Chem. Ber. 1984, 117, 336.
[40] The energy barrier to the racemization of the parent [5]helicene has
À1
been determined to be 24.1 kcalmol , see: C. Goedicke, H. Stegemey-
er, Tetrahedron Lett. 1970, 11, 937.
[41] K. Yamamoto, T. Harada, Y. Okamoto, H. Chikamatsu, M. Nakazaki, Y. Kai,
T. Nakao, M. Tanaka, S. Harada, N. Kasai, J. Am. Chem. Soc. 1988, 110,
3578.
À1
studies suggested that the inversion barrier is around 80 kcalmol . For
details, see: f) S. M. Bachrach, J. Org. Chem. 2009, 74, 3609.
[42] R. H. Martin, M. J. Marchant, Tetrahedron 1974, 30, 347.
[43] E. V. Anslyn, D. A. Dougherty, Modern Physical Organic Chemistry, Univer-
sity Science Press, Sausalito, 2006, p.946.
[
[
29] J. D. Chai, M. Head-Gordon, Phys. Chem. Chem. Phys. 2008, 10, 6615.
30] Gaussian 09, Revision A.01, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, T. Keith,
Received: January 12, 2016
Revised: March 24, 2016
Published online on May 31, 2016
Chem. Eur. J. 2016, 22, 9198 – 9208
www.chemeurj.org
9208
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim