A combined experimental and theoretical study on the conformation of multiarmed chiral aryl ethers

Article abstract of DOI:10.1021/jo071216n

(Graph Presented) Four series of multiarmed chiral aryl ethers carrying two, three, five, or eight side-chains on a variety of aromatic core molecules (2-5) were prepared. The structure and conformation of 2 and 3 (in the solid state) were determined by the X-ray crystallographic analyses. While a pair of alternated (anti) conformers (i.e, up-down and down-up) were found in the crystal of 2, three side-arms in 3 were aligned in the same direction to give a C₃-symmetric syn-conformation. Examinations by dispersion-corrected density functional (DFT-D) calculations revealed that two out of six anti- and two out of four syn-conformers of 2 are energetically most important. Two calculated structures of anti-conformers are in good agreement with those found in the solid state by X-ray analysis. Similarly, relevant conformations of syn-3, fully alternated 4, and C₅-symmetric 5 were optimized at the DFT-D-B-LYP/TZVP level. The structure and conformation of the side-arms in 2-5 in solution were further studied by temperature dependent ¹H NMR and UV-vis spectroscopy. In addition, comparative experimental and theoretical CD spectral studies were carried out in order to elucidate the contribution of the thermodynamically less-stable minor isomers in solution. The CD spectral changes observed for 2 and 3 at varying temperatures were quite different, while the parent chiral arene 1, as well as 4 and 5, only showed an increased intensity of the negative Cotton effect for the ¹L_b band. The latter behavior is readily accounted for in terms of the conformational freezing of the chiral groups at low temperatures. The unusual CD spectral behavior observed for 2 and 3 was rationalized by the conformational alteration of the side-arms. Because of attractive van der Waals interactions between the aromatic units of the arms in nonpolar solvents, the syn-conformations become gradually more important for 2 at low temperatures, which eventually results in a weak positive Cotton effect for the ¹L_b band. This was also supported by the SCS-MP2/TZVPP single-point energy calculations for the relevant conformers of 2. For 3, the contribution of the C₃-symmetrical conformer becomes more important than the less-symmetrical isomers at low temperatures. The conformations of 2 and 3 in their excited states as well as in the oxidized states were also examined.

Full text of DOI:10.1021/jo071216n

A Combined Experimental and Theoretical Study on the
Conformation of Multiarmed Chiral Aryl Ethers
Tadashi Mori,*,†,‡ Stefan Grimme,‡ and Yoshihisa Inoue†
Department of Applied Chemistry, Graduate School of Engineering, Osaka UniVersity, 2-1 Yamada-oka,
Suita 565-0871, Japan, and Theoretische Organische Chemie, Organisch-Chemisches Institut der
UniVersita¨t Mu¨nster, Corrensstrasse 40, D-48149 Mu¨nster, Germany
tmori@chem.eng.osaka-u.ac.jp
ReceiVed June 12, 2007
Four series of multiarmed chiral aryl ethers carrying two, three, five, or eight side-chains on a variety of
aromatic core molecules (2-5) were prepared. The structure and conformation of 2 and 3 (in the solid state)
were determined by the X-ray crystallographic analyses. While a pair of alternated (anti) conformers (i.e,
up-down and down-up) were found in the crystal of 2, three side-arms in 3 were aligned in the same direction
to give a C3-symmetric syn-conformation. Examinations by dispersion-corrected density functional (DFT-D)
calculations revealed that two out of six anti- and two out of four syn-conformers of 2 are energetically most
important. Two calculated structures of anti-conformers are in good agreement with those found in the solid
state by X-ray analysis. Similarly, relevant conformations of syn-3, fully alternated 4, and C5-symmetric 5
were optimized at the DFT-D-B-LYP/TZVP level. The structure and conformation of the side-arms in 2-5 in
1
solution were further studied by temperature dependent H NMR and UV-vis spectroscopy. In addition,
comparative experimental and theoretical CD spectral studies were carried out in order to elucidate the
contribution of the thermodynamically less-stable minor isomers in solution. The CD spectral changes observed
for 2 and 3 at varying temperatures were quite different, while the parent chiral arene 1, as well as 4 and 5,
only showed an increased intensity of the negative Cotton effect for the 1Lb band. The latter behavior is readily
accounted for in terms of the conformational freezing of the chiral groups at low temperatures. The unusual
CD spectral behavior observed for 2 and 3 was rationalized by the conformational alteration of the side-arms.
Because of attractive van der Waals interactions between the aromatic units of the arms in nonpolar solvents,
the syn-conformations become gradually more important for 2 at low temperatures, which eventually results
1
in a weak positive Cotton effect for the Lb band. This was also supported by the SCS-MP2/TZVPP single-
point energy calculations for the relevant conformers of 2. For 3, the contribution of the C3-symmetrical
conformer becomes more important than the less-symmetrical isomers at low temperatures. The conformations
of 2 and 3 in their excited states as well as in the oxidized states were also examined.
Introduction
liquid crystals and hosts for organic/inorganic anions and even
solvent molecules, as well as ligands for metal ions.1 More
recently, multiarmed compounds also have been employed as
Multiply substituted molecules, or molecular podands, are
attracting growing interest as potential candidates for discotic
(1) (a) MacNicol, D. D.; Downing, G. A. ComprehensiVe Supramolecular
Chemistry; Elsevier: Oxford, 1996. (b) Brunsveld, L.; Folmer, B. J. B.;
Meijer, E. W.; Sijbesma, R. P. Chem. ReV. 2001, 101, 4071-4097.
* To whom correspondence should be addressed. Fax: +81-6-6879-7923.
† Osaka University.
‡ Universita¨t Mu¨nster.
10.1021/jo071216n CCC: $37.00 © 2007 American Chemical Society
Published on Web 08/01/2007
6998
J. Org. Chem. 2007, 72, 6998-7010

Conformation of Multiarmed Chiral Aryl Ethers
a core to build dendric molecules.2 These molecules possess
unique features depending on the number of arms and the nature
of core unit. In designing such compounds, it is essential to
control the spatial arrangement of the side-arms. A number of
reports have shown that the most favored arrangement in
benzene derivatives is the one where all adjacent side-arms are
aligned in the opposite direction to give a fully alternated
conformation. For instance, X-ray crystal structures of hexakis-
(4-pyridylsulfanylmethyl)benzene and octakis(2-pyridylsulfa-
nylmethyl)biphenylene reveal that all the arms are in the anti
geometry.3 Such preorganization of hexasubstituted benzenes
has been known for a long time.4 Consequently, 1,3,5-trisub-
stituted triethylbenzenes have been extensively exploited as a
conventional tool for aligning the arms into a C3-symmetric syn-
orientation.5 Thus, a large number of successful examples of
tripodal anion receptors based on this strategy have been
reported,6 while several other examples of such aligned tripods
have employed the Kemp’s triacid7 and other cores.8 In contrast
to the tripods, control of anti/syn orientation in flexible
disubstituted hosts is rather difficult.9 Some recent successful
reports have utilized the dipole moment of the side-arms as a
means to align them in the desired orientation.10
side-arms.13 In fact, such exceptions have been reported
occasionally and also found in the crystal structures.14,15 For
some dipods, the syn-form is readily induced upon (electro)-
chemical oxidation, as demonstrated by X-ray crystallographic
analysis of the resulting radical cation.16 Such conformational
changes associated with the π-stacking of radical cation were
also reported for other systems.17 Accordingly, although the
(fully) alternated form is energetically most favored in most of
the cases, other (partially alternated or partially syn) conforma-
tions should also be considered, especially in solution in order
to understand the binding behavior in detail. It is also expected
that slight environmental changes, for example in temperature
or solvent polarity, should affect the population of the conform-
ers. Such conformational interconversion has, however, been
investigated only recently for 3-aminopyridinium-based tripodal
hosts.18 Therefore, our goal is to track the static and dynamic
behaviors of side-arm conformations in multiarmed compounds
through the investigation of their experimental and theoretical
chiroptical properties.
In this study, we synthesized new chiral multiarmed com-
pounds (2-5) and examined and discussed the spectroscopic
changes associated with the possible conformation(s) and
dynamics of the side-arms in detail, focusing on the experimental
and theoretical circular dichroism (CD)19 spectra. Scheme 1
summarizes the possible conformational changes of the relevant
polytentacled aryl ethers 2-5. We have chosen the aromatic
pendants with simple chiral (R)-2-methylpropyloxy group(s) to
observe the chiroptical signals required for the present study
(A). Obviously, the fundamental conformational variation is
based on the relative up-down/down-down (or anti/syn)
orientation of the side-arms. In addition to such anti/syn
conformation, up-down and down-up conformers should be
separately considered (in the case of chiral dipod 2), because
they are in diastereomeric relationship due to the chirality in
the side-arm (B). Note that these two conformers are essentially
mirror-imaged in structure, if one ignores the chiral group in
the side-arm. In some cases upon incorporating unsymmetrically
substituted side-arms, there is an additional pair of conformers
originating from the orientation of the aromatic plane of the
side-arm (see A in Scheme 1). To sum up, six up-down and
four down-down skeletal conformers are possible, and all of
them were considered in the following study. For simplicity,
the conformation of the terminal alkyl group was ignored
throughout this study (Vide infra). For tripod 3 and octapod 4,
all possible fully alternated conformers were considered (C and
D). A possible involvement of the partially alternated conformers
(which contain one syn or down-down orientation) was also
considered briefly. In the case of corannulene-based 5, side-
In contrast, the binding process of guest molecules with
flexible podand receptors remains more elusive.11 In some cases,
contrary to the preorientation strategy mentioned above, the
induced fit concept becomes important for flexible hosts. For a
urea-based tripodal receptor with large terminal groups, the 2-up-
1-down conformer was designated as the predominant conformer
by experiments and theory.12 The 1H NMR study of the
Enterobactin-analogues also showed the possible rotation of the
(2) Newkome, G. R.; Moorefield, C. N. Densritic Molecules: Concepts,
Syntheses, PerspectiVes; VCH: New York, 1996.
(3) McMorran, D. A.; Steel, P. J. Tetrahedron, 2003, 59, 3701-3707.
(4) MacNicol, D. D.; Hardy, A. D. U.; Wilson, D. R. Nature 1977, 266,
611-612.
(5) For reviews on the unique features of 1,3,5-trisubstituted triethyl-
benzenes in host-guest chemistry, see: Hennrich, G.; Anslyn, E. V. Chem.
Eur. J. 2002, 8, 2219-2224. (b) Best, M. D.; Tobey, S. L.; Anslyn, E. V.
Coord. Chem. ReV. 2003, 240, 3-15. (c) Wiskur, S. L.; Ait-Haddou, H.;
Lavigne, J. J.; Anslyn, E. V. Acc. Chem. Res. 2001, 34, 963-972.
(6) (a) Frontera, A.; Morey, J.; Oliver, A.; Pina, M. N.; Quinonero, D.;
Costa, A.; Ballester, P.; Deya, P. M.; Anslyn, E. V. J. Org. Chem. 2006,
71, 7185-7195. (b) Wright, A. T.; Griffin, M. J.; Zhong, Z.; McCleskey,
S. C.; Anslyn, E. V.; McDevitt, J. T. Angew. Chem., Int. Ed. 2005, 44,
6375-6378. (c) Wiskur, S. L.; Floriano, P. N.; Anslyn. E. V.; McDevitt, J.
T. Angew. Chem., Int. Ed. 2003, 42, 2070-2072. (d) Zhong, Z.; Anslyn,
E. V. J. Am. Chem. Soc. 2002, 124, 9014-9015. (e) Metzger, A.; Anslyn,
E. V. Angew. Chem., Int. Ed. 1998, 37, 649-652. (f) Metzger, A.; Lynch,
V. M.; Anslyn, E. V. Angew. Chem., Int. Ed. 1997, 36, 862-865.
(7) (a) Kemp, D. S.; Petrakis, K. S. J. Org. Chem. 1981, 46, 5140-
5143. See, also: (b) Bauer1, A.; Westka¨mper1, F.; Grimme, S.; Bach, T.
Nature, 2005, 436, 1139-1140. (c) Bach, T.; Bergmann, H.; Grosch, B.;
Harms, K. J. Am. Chem. Soc. 2002, 124, 7982-7990.
(8) (a) Fabris, F.; Pellizzaro, L.; ZOnta, C.; De Lucchi, O. Eur. J. Org.
Chem. 2007, 283-291. (b) Abe, H.; Aoyagi, Y.; Inouye, M. Org. Lett.
2005, 7, 59-61.
(9) (a) Singh, P.; Kumar, S. Tetrahedron, 2006, 62, 6379-6387. (b)
McMorran, D. A.; Hartshorn, C. M.; Steel, P. J. Polyhedron, 2004, 23,
1055-1061. (c) McMorran, D. A.; Steel, P. J. Inorg. Chem. Commun. 2003,
6, 43-47. (d) Kilway, K. V.; Siegel, J. S. Tetrahedron 2001, 57, 3615-
3627.
(13) Stack, T. D. P.; Hou, Z.; Raymond, K. N. J. Am. Chem. Soc. 1993,
115, 6466-6467.
(14) Menger, F. M.; Azov, V. A. J. Am. Chem. Soc. 2002, 124, 11159-
11166.
(15) Taha, M.; Marks, V.; Gottlieb, H. E.; Biali, S. E. J. Org. Chem.
2000, 65, 8621-8628.
(16) Sun, D. L.; Rosokha, S. V.; Lindeman, S. V.; Kochi, J. K. J. Am.
Chem. Soc. 2003, 125, 15950-15963.
(17) (a) Rathore, R.; Chebny, V. J.; Kopatz, E. J.; Guzei, I. A. Angew.
Chem., Int. Ed. 2005, 44, 2771-2774. (b) Rathore, R.; Chebny, V. J.;
Abdelwahed, S. H. J. Am. Chem. Soc. 2005, 127, 8012-8013.
(18) Wallace, K. J.; Belcher, W. J.; Turner, D. R.; Syed, K. F.; Reed, J.
W. J. Am. Chem. Soc. 2003, 125, 9699-9715.
(19) (a) Berova, N.; Nakanishi, K.; Woody, R. W. Circular Dichroism,
Principles and Applications, 2nd ed.; John Wiley & Sons: New York, 2000.
(b) Lightner, D. A.; Gurst, J. E. Organic Conformational Analysis and
Stereochemistry from Circular Dichroism Spectroscopy; John Wiley &
Sons: New York, 2000.
(10) (a) Cort, A. D.; Gasparrini, F.; Lunazzi, L.; Mandolini, L.; Mazzanti,
A.; Pasquini, C.; Peirini, M.; Rompietti, R.; Schiaffino, L. J. Org. Chem.
2005, 70, 8877-8883. (b) Betson, M. S.; Clayden, J.; Lam, H. K.; Helliwell,
M. Angew. Chem., Int. Ed. 2005, 44, 1241-1244.
(11) Steed, J. W.; Atwood, J. L. Supramolecular Chemistry; John and
Wiely & Sons: Chichester, 2000.
(12) Turner, D. R.; Paterson, M. J.; Steed, J. W. J. Org. Chem. 2006,
71, 1598-1608.
J. Org. Chem, Vol. 72, No. 18, 2007 6999

Mori et al.
SCHEME 1. Conformational Variations of 2-5
arms are attached directly to the scaffold. Thus, two possible
C5-symmetrical conformations, where the side-arms are tilted
to the opposite direction, were considered (E). These conformers
interconvert either by synchronized aryl-aryl bond rotations
or by concave-convex inversion of the corannulene moiety.
Several other combinations of partially syn (alternated) con-
formers were also possible, but we only address the simple C5
conformations in this study.
well as ORTEP drawings and crystal packing diagrams are
found in Supporting Information.
The crystalline 2 bears two independent molecules in the unit
cell, which can be ascribed to the chirality in the portal group.
Both of these molecules are in an anti orientation with respect
to the chiral arms, thus possessing C2 symmetry. Accordingly,
two conformers, i.e., up-down and down-up, are found in
crystals and they are alternately aligned in the crystalline lattice.
Although they are diastereomers, the dihedral angles between
the core and side-arm aromatic planes are almost identical:
99.7° ( 0.2° (Table S-1 in Supporting Information). The anti
orientation of 2 in the solid state is in accord with the reported
crystallographic structure of an achiral analogue, 1,2-bis(2,5-
dimethoxy-4-methyltolyl)tetramethylbenzene.16
The structure determination of 3 revealed that there are also
two independent molecules in the unit cell. Both of these enjoy
fully alternated up-down disposition with respect to the side-
arms, adopting quasi-C3 symmetry. Consequently, the chiral
donor moieties are aligned in an all-syn orientation to the central
ring. This is in accord with the reported X-ray structures of
hexaethylbenzene (D3h)20 and 1,3,5-tris(p-methoxybenzyl)tri-
ethylbenzene (C3).21 More importantly, three peripheral rings
Results and Discussion
Structural Studies of Multi-armed Chiral Aryl Ethers. (a)
X-ray Crystallographic Studies of 2 and 3. The single crystals
suitable for X-ray crystallographic analyses were obtained by
slow evaporation of the ethanol solutions of 2 or 3, containing
diethyl ether or dichloromethane, respectively. All the data
collections were performed at -60 °C, since the chiral (1-
methylpropyl) group tends to disorder at ambient temperatures
according to our previous experiences. This clearly indicates
that the conformation of the chiral group(s) is not completely
fixed at a single orientation in both solid and condensed phases,
especially at elevated temperatures. Nevertheless, the final
residues of the refinement on both structures at this temperature
were fairly good (R ) 7.4 and 7.6% for 2 and 3, respectively).
The PLUTO drawings, thus obtained for 2 and 3, are shown in
Figure 1. More details of the X-ray structural determination as
(20) Iverson, D. J.; Hunter, G.; Blount, J. F.; Damewood, J. R., Jr.;
Mislow, K. J. Am. Chem. Soc. 1981, 103, 6073-6083.
(21) Simaan, S.; Siegel, J. S.; Biali, S. E. J. Org. Chem. 2003, 68, 3699-
1701.
7000 J. Org. Chem., Vol. 72, No. 18, 2007

Conformation of Multiarmed Chiral Aryl Ethers
FIGURE 1. X-ray structures of 2 and 3 (PLUTO drawings). Pairs of independent structures were found for both 2 (left) and 3 (right) in the each
unit cell. They are diastereomeric to each other with the up-down vs down-up conformations in 2, while they differ only in tilt angles in 3.
in the latter compound are oriented in the H-perpendicular
arrangement.21 The two conformers in the crystal of 3 differ
structurally in tilt angles. Thus, their side-arms are tilted in the
opposite direction, but all are in the H-perpendicular arrange-
ment. A close inspection of the structure revealed that the tilt
angles of these perpendicularly arranged arms are entirely
different among the three moieties for both conformers, where
one of these arms is essentially situated in the almost eclipsed
arrangement. Unfortunately, no single crystals suitable for the
X-ray structural analysis were obtained for 4 and 5, despite our
repeated attempts under a variety of conditions.
molecules24 (for an overview, see ref 25). Here, we first
examined the validity of the DFT-D geometry optimization
applied to the present systems to reveal that the X-ray structures
of 2 and 3 are accurately reproduced by the calculations, despite
the apparent difference in “phase” (solid versus gas). Accord-
ingly, we employed this approach to optimize various conform-
ers of 2 and 3, and further extended the theoretical treatment to
the much larger systems (4 and 5) as follows.
1,2-Bis(arylmethyl)-3,4,5,6-tetramethylbenzene 2. All of the
conformations of 2 were fully optimized at the DFT-D-B-LYP/
TZVP level. As briefly described in the Introduction, the up-
down and the down-up conformers were separately considered
because they become different by introducing chirality in the
side-arms (Scheme 1). They are expected to give completely
different (and most probably almost mirror-imaged) CD profile
because of their diastereomeric structure, in which the relative
structures of aromatic rings are in enantiomeric relationship. In
addition, since the substituent on the benzene ring of the arm
is not symmetrical, there are additional pairs with respect to
the orientation of the aromatic plane. Thus, there are six up-
down and four down-down skeletal conformers in total. Note
that the conformation of the chiral group was taken always as
the Tg- conformation throughout this study.26 Among all the
optimized structures, two up-down (2a, 2b) and two down-
down (2g, 2h) conformations were found to be energetically
feasible (population g5%). The optimized structures of these
four conformers are shown in Figure 2, and details of the
(b) Geometrical Optimization of Multi-armed Chiral Aryl
Ethers 2-5 by Density Functional Theory. The geometries
of a variety of conformers of polytentacled donors 2-5 were
investigated using the DFT-D-B-LYP method. The DFT-D
method exploits an empirical correction by adding pairwise -C6/
R6 potentials to describe van der Waals (vdW) interactions. The
method is computationally as efficient as conventional DFT
calculations, and the calculated energies and geometries are
fairly comparable to those obtained with the improved SCS-
MP2 method.22 Thus, DFT-D is a very efficient practical
alternative for large molecules where the more accurate electron
correlated methods such as (SCS)-MP2 or CCSD(T) cannot be
applied. A number of successful applications of the DFT-D
method have been reported recently, e.g., for weakly bonded
intermolecular complexes23 and relatively large host-guest
(23) (a) Grimme, S.; Diedrich, C.; Korth, M. Angew. Chem., Int. Ed.
2006, 45, 625-629. (b) Piacenza, M.; Grimme, S. J. Am. Chem. Soc. 2005,
127, 14841-14848 (c) Piacenza, M.; Grimme, S. Chem. Phys. Chem. 2005,
6, 1554-1558.
(22) (a) Grimme, S. J. Phys. Chem. A 2005, 109, 3067-3077. (b)
Goumans, T. P. M.; Ehlers, A. W.; Lammertsma, K.; Wu¨rthwein, E.-U.;
Grimme, S. Chem. Eur. J. 2004, 10, 6468-6475. (c) Grimme, S. J. Chem.
Phys. 2003, 118, 9095-9102. (d) Gerenkamp, M.; Grimme, S. Chem. Phys.
Lett. 2004, 392, 229-235. (e) Grimme, S. J. Comput. Chem. 2003, 24,
1529-1537. (f) Piacenza, M.; Grimme, S. J. Comput. Chem. 2004, 25, 83-
99.
(24) Parac, M.; Etinski, M.; Peric, M.; Grimme, S. J. Chem. Theory
Comput. 2005, 1, 1110-1118.
(25) Grimme, S.; Antony, J.; Schwabe, T.; Mu¨ck-Lichtenfeld, C. Org.
Biomol. Chem. 2007, 741-758.
J. Org. Chem, Vol. 72, No. 18, 2007 7001

Mori et al.
FIGURE 2. The DFT-D-B-LYP/TZVP optimized structures of most stable conformers of multiarmed chiral aryl ethers 2-5.
calculation are summarized in Table 1 (see Supporting Informa-
tion for all the optimized geometries). The structures of two
up-down conformers (2a, 2b) obtained by the DFT-D method
show reasonable agreement with those obtained by the X-ray
structural analyses. Although two types of conformation were
found in (complete) 1:1 ratio in the crystals, the energy
difference calculated by the SCS-MP2 method indicates that
the up-down conformer 2a is slightly preferred by ∼0.1 kcal
mol-1 over the diastereomeric 2b. Indeed, this slight preference
turned out to be very important in understanding the experi-
mental CD spectra (Vide infra). A similar situation has already
been reported for the axial chirality in biaryls,26 where most of
the CD signals are canceled out between the conformers of
almost opposite chirality, but still the remnants sum up to give
a small but appreciable Cotton effect in the CD spectra. Two
down-down (2g, 2h) conformers were also considered since
these conformers should be involved in the interconversion
process of the side-arms. The overall calculated preference for
the down-down conformers over the up-down/down-up
conformers in the gas phase is ascribed mostly to the large vdW
attraction between the substituents for which the dispersion term
is responsible. Note that our calculations completely ignore the
solvent effect, and therefore the relative energies of the down-
down conformers are intrinsically underestimated.
1,3,5-Tris(arylmethyl)-2,4,6-triethylbenzene 3. The geom-
etry optimization of 3 was also performed at the DFT-D-B-
LYP/TZVP level employing C3 symmetry (i.e., fully alternated
form, Scheme 1). In our calculation, only one conformation (3a)
was found to be feasible (Figure 2), and the optimized structure
is in excellent agreement with one of the two structures found
in the crystal. Another conformer, in which the side-arms are
tilted to the opposite direction, did not give a local minimum
upon calculations. The latter conformer found in the crystal is,
therefore, considered to be stabilized by the packing forces in
the crystal. Two additional conformers (3b, 3c), which arise
from the different orientation of the side-arm, were optimized
by the calculations. These conformers, however, possess larger
steric repulsion between the methoxy groups and are thus higher
in energy by more than ∼15 kcal mol-1 (Table S-2 in
Supporting Information). Therefore, such conformers are practi-
cally nonexistent in solution. Similarly, other conformers with
one or two flipped side-arm(s) (partially alternated conformers,
Scheme 1) were not considered because they are also expected
to have higher relative energies. In the following, only the most
stable conformer 3a is considered in further simulations of the
CD spectra of 3.
Octakis(arylmethyl)biphenylene 4. Geometry optimization
of a molecule of this size (132 non-hydrogen and 152 hydrogen
atoms) by quantum chemical methods is challenging, even when
the very efficient DFT-D method is applied. By using a parallel
computing system and the resolution of identity approximation47
(RI) and employing C2-symmetry, we have successfully opti-
mized two fully alternated conformations of 4 at the DFT-D-
B-LYP/TZVP level within reasonable computation times. The
two calculated conformers (4a, 4b) were in a diastereomeric
relationship; the side-arms are in an alternated form and aligned
either up-down-up··· or down-up-down··· manner (Figure 2).
The relative energies at the DFT-D level were found to be quite
similar (∆E ∼ 0.3 kcal mol-1) between 4a and 4b despite having
eight arms. The SCS-MP2/TZVPP energy calculations were not
possible because of the size of molecule (Table S-2). The
partially syn (alternated) conformations (Scheme 1) found in
the crystal structure of octaethylbiphenylene15 were not feasible
in the present molecule because of the large steric hindrance of
the side-arms and therefore not considered further.
Pentaarylcorannulene 5. The geometrical optimization of
5 was also performed at the same level employing the C5
symmetry restrictions. Two conformations, in which the relative
7002 J. Org. Chem., Vol. 72, No. 18, 2007

Conformation of Multiarmed Chiral Aryl Ethers
FIGURE 3. Temperature dependence of the 1H NMR spectra of 2 and 3 (and 4 at 25 °C) in dichloromethane-d2. For other regions, see Figure S-8
in Supporting Information.
orientations of the alkoxy group are inverted, were obtained as
the most stable structures. Both structures (5a, 5b) are displayed
in Figure 2, which demonstrates that the relative orientations
of the aromatic rings in the side-arms are almost identical.
However, one of the conformers (5a) is slightly favored by
∼0.2 kcal mol-1, as determined by the SCS-MP2/TZVPP energy
calculations. The other conformers, in which the aromatic planes
of the side-arms are tilted to the opposite direction, were found
much less stable (by ∼7 kcal mol-1). Therefore, they will not
affect the overall spectra and are not considered further.
protons. The opposite tendency observed for the H3 proton (para
to H6) can be similarly explained. Thus, the deshielding effect
observed for the outside H3 of 2 at lower temperatures is quite
reasonable, as the inside H6 is shielded by the ring current of
the benzene core. The effect of deshielding was smaller for 3,
probably because the steric repulsion between the inside H6
protons, all of which are facing the center of the aromatic ring,
prevents more tight packing. Therefore, the H3 protons in 3
are anticipated to be slightly deviated from the center of the
ring (as compared with 2), and the effect of ring current is
attenuated.
1
Variable Temperature H NMR Studies of 2 and 3. To
In the above NMR experiments, a gradual freezing of the
side-arms upon cooling was confirmed for 2 and 3. This is
commonly observed behavior, but does not tell about the
possibility of the side-arm rotation to form the partially
alternated conformers. In the following, we investigated the
effect of temperature on the circular dichroism (CD) of 2-5.
By analyzing the experimental and theoretical CD spectra, we
were able to detect the possible rotation of the arm(s) leading
to the partially alternated conformations.
Measurement of UV-vis and CD Spectra of Multiarmed
Chiral Aryl Ethers 2-5. (a) Experimental CD Spectra of 2
and 3. The CD spectra of 2 were obtained in dichloromethane
solution at temperatures between 25 and -95 °C (Figure 4, top,
left). The corresponding UV-vis spectra as well as the
anisotropy (g) factors (g ) ∆ꢀ/ꢀ), both observed with the same
solutions, are also shown in Figure S-9 in Supporting Informa-
tion. The CD spectrum of 2 at 25 °C shows negative and positive
Cotton effects at the 1Lb (290 nm) and the 1La (230 nm) bands,
respectively. This pattern is quite similar to the CD spectrum
of the parent chiral benzene 1 (Figure S-13 in Supporting
Information). A slightly smaller molar ellipticity observed for
2 (having two chiral auxiliaries) is probably due to a cancelation
of the oppositely signed Cotton effects of the relevant (diaster-
eomeric) conformers. Upon cooling, a positive Cotton effect
experimentally gain information on the conformational changes
of 2 and 3 in solution, we performed 1H NMR spectral studies
at varying temperatures in dichloromethane-d2 (Figure 3). As
the temperature of the solution was decreased, the aliphatic
protons of the chiral group of 2 and 3 became broad and
structureless, e.g., at -80 °C (Figure S-8 in Supporting
Information). This implies that the peripheral alkyl group(s) are
gradually fixed at lower temperatures and the rotation of the
alkyl group(s) becomes slow in the NMR time-scale. The
changes of chemical shift in the aromatic protons, as a measure
of the side-arm dynamics, are compared in Figure 3. The H6
proton adjacent to the chiral alkoxy group (and thus most
crowded) is highly shielded in 2 and 3 compared with that in 4
(and 1, see Supporting Information for the details). As deter-
mined by X-ray structural analyses and also by the DFT-D
calculations, this proton is located just above (or below) the
central aromatic ring (see Figures 1 and 2). By lowering the
temperature, the H6 was further shielded gradually, and this
tendency was much pronounced for 3. The H6 in 3 appeared
as a broad peak at a chemical shift as low as ∼6 ppm at
-80 °C. This is rationalized by assuming that the side-arms
are more closely located to each other on the same side of the
aromatic core in 3 at lower temperatures to make the H6 protons
overcrowded and affected more severely by the neighboring
J. Org. Chem, Vol. 72, No. 18, 2007 7003

Mori et al.
FIGURE 4. Temperature dependence of the CD spectra of 2-4 (and 5 at 25 and -80 °C) in dichloromethane at temperatures varied from 25 to
-95 °C at an interval of 15 °C.
peak appeared at slightly longer wavelengths in the 1Lb transition
region (300 nm). Thus, apparently bisignate CD spectra were
obtained in this Lb region at temperatures between -50 and
-80 °C. At -95 °C, a positive Cotton effect was observed at
1Lb band (300 nm), together with a negative Cotton effect at
the1Lb band (240 nm).
Information). By lowering the temperature, a small blue shift
of the Lb band maximum was observed for both 2 and 3. In
1
1
contrast, the temperature effect on the CD spectrum of 3 was
considerably different from that of 2. As shown in Figure 4
(top, right), positive and negative Cotton effects observed for
the 1Lb (290 nm) and 1La (230 nm) bands of 3 steadily became
stronger as the temperature was decreased. The CD intensities
This temperature effect on the CD spectra of 2 is puzzling
but may be rationalized by assuming a larger contribution of
the partially alternated conformation(s) at lower temperatures
as follows. A comparison of the temperature effects on the CD
spectra of 2 and 1 (Figure 4 and Figure S13) led us to a
conclusion that the typical CD spectral changes observed for 2
should not be attributed to the side-arm conformation alone but
to the arm-arm and/or arm-core orientations. The fact that
the CD spectrum of 2 is almost identical to that of 1 at 25 °C
implies that the CD signals arising from the arm-arm and/or
arm-core orientations are almost canceled out at least at ambient
temperatures. Considering the fact that the all-syn conformer
of 3 affords a strong positive Cotton effect (Vide infra), we may
deduce that the partially alternated (i.e., syn or down-down)
conformer(s) becomes more important, leading to the sign
inversion of the CD spectra of 2, at low temperatures. Neverthe-
less, as the CD intensity of 2 is always weak relative to that of
3, contribution of the partially alternated conformers is modest
even at the lower temperatures.
1
1
of the Lb and the La bands are 20 times larger for 3 than for
2 at -95 °C. As was indicated by the VT-NMR studies (Vide
supra), the three side-arms in 3 are basically aligned in the same
direction, which gradually packed tightly as the temperature
decreases. However, it is quite rational to consider that the
conformational alteration of side-arms is also involved in 3,
especially at higher temperatures. Lowering temperature reduces
the contribution of such partially alternated conformers of 3 in
solution. These explanations are consistent with the results that
simple freezing of the chiral group in 1 only slightly enhances
(by a factor of 2) the CD intensity (Figure S-13 in Supporting
Information), but 2 and 3 display the totally different CD spectral
behavior (intensity change).
(b) Comparison of Solvent Effects on the CD Spectra of
2 and 3. The effect of solvent on the CD spectra of 2 and 3
provides additional insights into the dynamic behavior of the
side-arms of the molecules. For 2, the above-mentioned tem-
perature effect was only observed in less polar solvents such as
methylcyclohexane and dichloromethane. Interestingly, the
effect of the temperature was less significant in more polar
The temperature dependence behavior of the UV-vis spectra
of 3 was quite similar to that of 2 (Figure S-10 in Supporting
7004 J. Org. Chem., Vol. 72, No. 18, 2007

Conformation of Multiarmed Chiral Aryl Ethers
FIGURE 5. CD spectra of 2 (top) and 3 (bottom) in methylcyclohexane (left), methanol (middle), and acetonitrile (right) at two different temperatures
(0.1 mM).
methanol and acetonitrile (Figure 5, top), which is quite similar
to that found for 1. Thus, the arm-arm and/or arm-core CDs
are almost canceled out in the observed spectrum of 2 in polar
solvents at any temperatures examined, and what we observed
in the VT-CD spectra of 2 in polar solvents is a consequence
of simple freezing of the chiral auxiliaries. In less polar solvents
at low temperatures, the partially alternated (down-down)
conformers become apparent, probably because of the pro-
nounced π-π (dispersion) interaction between the arms. Such
intramolecular interactions are not feasible in polar solvents
because of the more preferred solvation.
In sharp contrast to 2, 3 exhibited a more straightforward
temperature effect in all the solvents examined, giving an
increased CD intensity at a reduced temperature (Figure 5,
bottom). The simplest explanation for this CD spectral behavior
of 3 is that the syn-(C3)-conformer becomes dominant at low
temperatures in all solvents. Accordingly, the CD intensities
increase with decreasing population to the partially alternated
conformers, which are pseudo-centrosymmetric, and thus the
CD signals are mutually canceled between the oppositely
oriented side-arms to give weaker Cotton effects.
tion). The molar circular dichroism (∆ꢀ) of the negative Cotton
effect at the 1Lb band of 4 was roughly 8 times larger than that
of 1. Since the adjacent bromomethyl groups (instead of
arylmethyl groups in this study) in polysubstituted arenes are
already conformationally fixed at room temperature (on the
NMR time-scale),26 the much larger side-arms in 4 are
completely fixed in the fully alternated form at temperatures
lower than 25 °C. Thus, the above-mentioned spectral change
can be explained by a conformational freezing of the chiral
auxiliaries by lowering the temperature. Thus, the temperature
effects in different solvents (methylcyclohexane and methanol)
are very similar to that in dichloromethane (Figure S-14 in
Supporting Information).
The UV-vis and CD spectra of 5 were also measured in
dichloromethane at 25 and -80 °C (Figure 4, bottom, right,
and Figure S-12 in Supporting Information). The (induced)
Cotton effects of 5 at around 300 nm were not apparent at the
absorption bands corresponding to the corannulene core, as was
similarly observed for the spectral changes of 4. At a low
temperature, the negative Cotton effect corresponding to the
1Lb band of the side-arms became larger. Considering the free
rotation of the side-arms as well as rapid bowl inversion of the
(c) CD Spectra of 4 and 5. The CD spectra of 4 in
dichloromethane at temperatures ranging from 25 to -95 °C
were also obtained and are shown in Figure 4 (bottom, left).
As the temperature decreases, the negative Cotton effect at the
1Lb band (290 nm) gradually increases. The (induced) Cotton
effect at around 300 nm is not apparent at any temperature in
the wavelength range corresponding to the absorption of the
biphenylene core (see also Figure S-11 in Supporting Informa-
(26) The effect of the minor conformers in respect to the terminal alkyl
group as well as overall simulated UV/CD spectra therefrom is considered
to be very small according to our previous experience. For a detailed
discussion of the effect of the conformation of the terminal chiral group,
see: Mori, T.; Inoue, Y.; Grimme, S. J. Phys. Chem. A 2007, 111, 4222-
4234.
J. Org. Chem, Vol. 72, No. 18, 2007 7005

Mori et al.
FIGURE 6. Comparison of the experimental and calculated CD spectra of 1 (left) and 2 (right). Black: experimental CD spectra in dichloromethane
at 25 °C. Red: TD-DFT-BH-LYP/TZVP calculated CD spectra. For 2, solid line: calculated spectrum averaged over all up-down conformers
(2a-f); dotted line: calculated spectrum averaged over all possible conformers (2a-j).
corannulene core in 5,27,28 this temperature effect can be ascribed
to a freezing of the chiral auxiliaries at the low temperature.
(d) Theoretical Calculation of the CD Spectrum of 2. The
TD-DFT based computation of rotatory strengths and CD spectra
has been successfully applied recently to the assignment of the
absolute configuration and conformation of chiral organic
molecules.29,30 Accordingly, we performed TD-DFT calculations
for all 10 conformers of 2 at the BH-LYP/TZVP level to
simulate the UV-vis and CD spectra. The choice of the
functional and the basis-sets is based on previous experience.26,31
The validity of our method was confirmed by the successful
reproduction of the experimental CD spectrum of 1 by theoreti-
cal calculations (Figure 6, left). While the computed UV-vis
spectra of all possible conformers of 2 are essentially the same,
the CD spectra are considerably different from each other. The
theoretical CD spectrum of 2, shown in Figure 6 (right), was
obtained as a Boltzmann population-weighted average of the
CD spectra of all conformers based on their SCS-MP2 relative
energies. Such a static approach32 nicely reproduced the pattern
and relative rotatory strength of the experimental CD spectra
of 2. The first negative and the second positive Cotton effects
were correctly predicted, although a slightly larger scaling factor
(1/20) was necessary to fit to the experimental values. Since
the dynamic equilibrium between up-down and down-up
conformers becomes evident, we further tested the effect of
partially alternated conformers (i.e., syn or down-down con-
formers). The experimental spectrum is only slightly better
reproduced when the contributions of the partially alternated
conformations are considered (Figure 6, right), but we were not
able to estimate the magnitude of contribution from such
conformers.
Theoretical computations of the CD spectra of 3-5 were also
conducted by the same TD-DFT method as well as by a
semiempirical π-electron time-dependent Pariser-Parr-Pople
(TD-PPP) scheme.33 The calculations of UV-vis and CD
spectra were performed for energetically relevant conformers
of 3-5 (i.e., 3a, 4a, 4b, 5a, and 5b). The computation of the
CD spectrum of 4 was only possible with DFT-D energy-based
averaging of the conformers since the SCS-MP2 energies were
not available. These results were then compared with the
experimental spectra.
Contrary to our expectations, all attempts to theoretically
simulate the CD spectra of 3-5 turned out to be unsuccessful
(Figure S-15 in Supporting Information). The exact reason
remains to be elucidated, but possible reasons for this failure
are as follows: (1) The dynamic behavior in solution was
neglected in our calculation, but is actually more important.32
Probably, the estimation of the conformer population by a simple
Boltzmann approach was too simple. (2) We neglected vibra-
tional Herzberg-Teller coupling effects in our calculations, but
(27) Simaan, S.; Marks, V.; Gottlieb, H. E.; Stanger, A.; Biali, S. E. J.
Org. Chem. 2003, 68, 637-640.
(28) Zavada, J.; Pankova, M.; Arnold, Z. Coll. Czech. Chem. Commun.
1976, 41, 1777-1790.
(29) (a) Seiders, T. J.; Baldridge, K. K.; Grube, G. H.; Siegel, J. K. J.
Am. Chem. Soc. 2001, 123, 517-525. (b) Scott, L. T.; Hashemi, M. M.;
Bratcher, M. S. J. Am. Chem. Soc. 1992, 114, 1920-1921. (c) Biederman,
P. U.; Pogodin, S.; Agranat, I. J. Org. Chem. 1999, 64, 3655-3662.
(30) (a) Diedrich, C.; Grimme, S. J. Phys. Chem. A 2003, 107, 2524-
2539. (b) Autschbach, J.; Ziegler, T.; van Gisbergen, S. J. A.; Baerends, E.
J. J. Chem. Phys. 2002, 116, 6930-6940.
(31) For recent examples, see: (a) McCann, D. M.; Stephens, P. J. J.
Org. Chem. 2006, 71, 6074-6098. (b) Stephens, P. J.; McCann, D. M.;
Devlin, F. J.; Smith, A. B., III. J. Nat. Prod. 2006, 69, 1055-1064. (c)
Gawronski, J. K.; Kwit, M.; Boyd, D. R.; Sharma, N. D.; Malone, J. F.;
Drake, A. F. J. Am. Chem. Soc. 2005, 127, 4308-4319. (d) Schu¨hly, W.;
Crockett Sara, L.; Fabian Walter, M. F. Chirality 2005, 17, 250-6. (e)
Forzato, C.; Furlan, G.; Nitti, P.; Pitacco, G.; Marchesan, D.; Coriani, S.;
Valentin, E. Tetrahedron: Asym. 2005, 16, 3011-3023. (f) Zuber, G.;
Goldsmith, M.-R.; Hopkins, T. D.; Beratan, D. N.; Wipf, P. Org. Lett. 2005,
7, 5269-5272. (g) Stephens, P. J.; McCann, D. M.; Cheeseman, J. R.;
Frisch, M. J. Chirality 2005, 17, S52-S64.
1
benzene’s Lb transitions are known in general to be strongly
influenced by vibronic coupling.34 (3) The current TD-DFT
method with standard hybrid functionals is not accurate enough
to describe the electronic through-space interactions (exciton
coupling) in short distances. We recently reported that the
theoretical CD spectra of some chiral cyclophanes were only
(33) For a recent treatment of the dynamic behavior on the simulation
of the chiroptical properties, see: (a) Sebek, J.; Kejik, Z.; Bour, P. J. Phys.
Chem. A 2006, 110, 4702-4711. (b) Glattli, A.; Daura, X.; Seebach, D.;
van Gunsteren, W. F. J. Am. Chem. Soc. 2002, 124, 12972-12978.
However, such treatment for the current system was practically unfeasible.
(34) (a) Vo¨gtle, F.; Hu¨nten, A.; Vogel, E.; Buschbeck, S.; Safarowsky,
O.; Recker, J.; Parham, A.-H.; Knott, M.; Mu¨ller, W. M.; Mu¨ller, U.;
Okamoto, Y.; Kubota, T.; Lindner, W.; Francotte, E.; Grimme, S. Angew.
Chem., Int. Ed. 2001, 40, 2468-2471. (b) Parac, M.; Grimme, S. Chem.
Phys. 2003, 292, 11-21.
(32) (a) Mori, T.; Inoue, Y.; Grimme, S. J. Org. Chem. 2006, 71, 9797-
9806. See also: (b) Mori, T.; Ko, Y. H.; Kim, K.; Inoue, Y. J. Org. Chem.
2006, 71, 3232-3247. (c) Mori, T.; Inoue, Y. Angew. Chem., Int. Ed. 2005,
44, 2582-2585.
7006 J. Org. Chem., Vol. 72, No. 18, 2007

Conformation of Multiarmed Chiral Aryl Ethers
TABLE 1. Computation Results for All Conformations of 2a
SCS-MP2
energy
(Hartree)
specific
rotation
specific
rotation
∆E (SCS-MP2)
%population DFT-D energy ∆E (DFT-D)
%population
(kcal mol-1
)
%population (anti only)
(Hartree)
(kcal mol-1
) %population (anti only) (aug-cc-pVDZ) (aug-SVP)
2a -1699.2443
2b -1699.2441
2c -1699.2406
2d -1699.2403
2e -1699.2314
2f -1699.2311
2g -1699.2472
2h -1699.2474
2i -1699.2450
2j -1699.2349
=0
5.8
5.1
0.6
0.5
0.0
48.1
43.0
4.9
4.0
0.0
-1702.7025
-1702.7024
-1702.7011
-1702.6995
-1702.6920
-1702.6922
-1702.7065
-1702.7061
-1702.7050
-1702.6943
=0
3.4
3.2
1.5
0.5
0.0
39.7
37.2
16.8
6.2
0.1
0.1
1528
-2491
2339
-2487
-2653
755
2339
-3540
+0.1
+2.3
+2.5
+8.1
+8.3
-1.9
-2.0
-0.5
+5.9
+0.1
+0.9
+1.9
+6.6
+6.4
-2.5
-2.2
-1.6
+5.1
-2715
811
2410
0.0
0.0
0.0
-3598
37.0
42.2
8.9
42.8
32.1
16.5
0.0
-643
-2515
1637
0.0
-1039
-1205 (-437 )b
(-44)b
a See Supporting Information for the structure of each conformer. b Averaged specific rotation for up-down/down-up conformers (2a-f).
FIGURE 7. Comparison of cyclic voltammogram (left) and Osteryoung square-wave voltammogram (right) of 2 (blue) and 3 (red) in dichloromethane
at a scan rate of 100 mV s-1
reproduced by the more sophisticated CC2 method.35 (4) The
basic chromophores in 4 (distorted biphenylene) and 5 (geodesic
corannulene) are not accurately described by the current density
functionals.
obtained incidentally, judging from the fact that the calculated
specific rotations of 1 at the BH-LYP/aug-cc-pVDZ and aug-
SVP levels are very similar to each other (-556° and -555°),
which are also ca. 13 times larger than the experimental value
(-43°). The sign of the specific rotation was, nevertheless,
always correctly predicted.
(e) Theoretical and Experimental Optical Rotations of 2.
The TD-DFT computations of optical rotations (OR) have
become a routine tool in chiroptical studies and have been
repeatedly employed for the prediction of absolute configura-
tion.36 We also performed the OR calculations for 2 with the
BH-LYP functional using the aug-cc-pVDZ or aug-SVP basis-
sets (Table 1). The calculated values for individual conformers
were Boltzmann-averaged by using the most reliable single point
SCS-MP2/TZVPP energies and compared with the experimental
values. The calculated OR with the aug-cc-pVDZ basis-set are
much larger (in absolute values) than the experimental values,
and this was not improved even when the partially alternated
(down-down) conformers were also taken into account. The
calculated OR for each conformer showed reasonable agreement
between the two basis-sets employed. However, the values
obtained after averaging show large differences (-437° versus
-44°). Such a result can be attributed primary to an incomplete
cancelation of the dynamic effects between the conformers. We
believe that the better results with the aug-SVP basis-set were
Electrochemical Studies with 2 and 3. The cyclic voltam-
mograms were obtained with dichloromethane solutions of 2
and 3 with tetra-n-butylammonium hexafluorophosphate as
electrolyte (Figure 7, left). A stepwise and reversible oxidation-
reduction curve was obtained for 2 at a scan rate of 100 mV
s-1, E1/2 being determined to be 1.02 and 1.14 V vs SCE (∆E
) 0.12 V). Due to a larger solvation effect, considerably higher
oxidation potentials (1.07 and 1.18 V) were found in a more
polar solvent (acetonitrile). These values are in good agreement
with those reported for the related achiral analogue (E1/2 ) 1.07
and 1.23 V vs SCE in dichloromethane).16 Such an apparent
stepwise oxidation of 2 with two identical donor arms indicates
that there should be strong interaction between the arms in the
ground state. The electrochemical behavior of the analogous
o-xylylene-bridged donor is known to strongly depend on
temperature, solvent, and sweep rate.16 This was also the case
with 2 (details of the effect of solvent and scan rate on the CV
behavior are found in Figure S-16 in Supporting Information).
These results are in dramatic contrast to the CV behavior of
3, where a quasi-single oxidation (3e) at much higher potential
was observed with all the scan rates examined (10-1000 mV
s-1). The oxidation potential of 3 was determined to be 1.20 V
(35) (a) Muller, D. J.; Knight, A. E. W. J. Phys. Chem. 1984, 88, 3392-
3395. (b) Berger, R.; Fischer, C.; Klessinger, M. J. Phys. Chem. A 1998,
102, 7157-7167.
(36) Mori, T.; Inoue, Y.; Grimme, S. J. Phys. Chem. A 2007, 111,
799598006.
J. Org. Chem, Vol. 72, No. 18, 2007 7007

Mori et al.
FIGURE 8. Left: Fluorescence spectra and fluorescence excitation spectra of 2 (0.01 mM, blue) and 3 (0.0067 mM, red) in methylcyclohexane
at 25 °C. Excitation at 290 nm (dotted lines). Emission monitored at 320 nm (solid lines). Right: Arrhenius-type plots for the temperature dependence
of the fluorescence maxima of 2 and 3 in methylcyclohexane.
vs SCE by a square-wave voltammogram (Figure 7, right),
which is slightly larger than the E1/2 value reported for 2,6-
dimethoxy-p-xylene and related molecules (∼1.0 V), partially
owing to the steric effect. It is to note that the peak was not
completely single but that small shoulder peaks were also found,
which indicates that appreciable, but relatively weak, interactions
between the arms are operative in the oxidation process. A
difference between the first and the second potentials, ∆E )
0.09 V, was evaluated from the Laplacian of the OSWV curve
of 3, which enabled us to estimate the stabilization energy of
the radical cation by using the relationship: -RT ln K ) -∆ΕF,
where F is the Faraday constant. Thus, the stabilization energy
of radical cation 3•+ was calculated as 2.1 kcal mol-1, which is
relatively smaller than the value for radical cation 2•+ (2.8 kcal
mol-1). The stabilization energy of radical cation 3•+ is still
significant in spite of the larger distance between the arms,
compared with that in 2•+. A large peak shift found between
the oxidation and reduction cycle (0.19 V at a scan rate of
100 mV s-1) also supports the interaction between the arms of
3 in the oxidation process.
Chemical Oxidation and the CD Spectral Changes of 2
and 3. The Meerwein salt [Et3O+SbCl6-] is known as an
effective reagent particularly for the preparation of the cation
radical salts with various aromatic aryl ethers with oxidation
potentials lower than 1.5 V vs SCE. By using this oxidant, one
can avoid the concomitant formation of byproduct(s),37 allowing
the formation of suitable crystals of pure paramagnetic aromatic
cation radicals for X-ray structure determination. Upon mixing
1.5 equiv of Et3O+SbCl6- into dichloromethane solutions of 2
or 3 under argon atmosphere, new absorption bands gradually
developed at 400-500 nm, which are assignable to a dialkoxy-
p-xylyl radical cation (Figure S-17, left, in Supporting Informa-
tion). The CD spectra of the same solutions were also measured
after oxidation for 10 h (Figure S-17, right). Relatively weak
positive Cotton effects were observed at this absorption band
among the drifting spectra. We did not further study the CD
spectral changes upon chemical oxidation, since, unfortunately,
the spectra did not appear to provide insights into the confor-
mational changes upon oxidation. Such a weak Cotton effect
observed at the D-D transition seems to be a common feature
of radical cations.38
Excited-State Properties of 2 and 3. The steady-state
fluorescence spectra of 2 and 3 were recorded in methylcyclo-
hexane at varying temperatures (Figure 8). The fluorescence
quantum yields of both 2 and 3 were determined as 0.27 at
25 °C, which are comparable to that of the parent 2,5-
dimethoxy-p-xylene (φFL ) 0.30).39 No exciplex or excimer
emission was observed upon excitation of 2 under the experi-
mental conditions examined. A small red-shift in the fluores-
cence emission observed for 3, relative to that of 2, can be
ascribed to the larger stabilization by the through-bond electronic
resonance with the additional aromatic side-arm in the excited-
state of 3. Fluorescence maxima were gradually shifted to the
blue as the temperature of the solution decreased. This suggests
that there exists quite large stabilization by reorientation of the
arms in their excited states, and such a conformational change
requires some activation energy. This assumption is in accord
with the fact that there is a dynamic conformational alteration
process of the chiral side-arms in solution in the ground state
(Vide supra) and in the excited-state as well. Thus, the plots of
the fluorescence maxima (in energy unit) against the reciprocal
temperature afforded good straight lines for both compounds.
The temperature dependence was found more pronounced for
2 than for 3, probably because of the larger stabilization in the
(37) For recent examples, see: (a) Mort, B. C.; Autschbach, J. J. Phys.
Chem. A. 2006, 110, 11381-11383. (b) Kundrat, M. D.; Autschbach, J. J.
Phys. Chem. A 2006, 110, 4115-4123. (c) Autschbach, J.; Jensen, L.;
Schatz, G. C.; Tse, Y. C. E.; Krykunov, M. J. Phys. Chem. A. 2006, 110,
2461-2473. (d) Durandin, A.; Jia, L.; Crean, C.; Kolbanovskiy, A.; Ding,
S.; Shafirovich, V.; Broyde, S.; Geacintov, N. E. Chem. Res. Toxicol., 2006,
19, 908-913. (e) Giorgio, E.; Roje, M.; Tanaka, K.; Hamersak, Z.; Sunjic,
V.; Nakanishi, K.; Rosini, C.; Berova, N. J. Org. Chem. 2005, 70, 6557-
6563. (f) Crawford, T. D.; Owens, L. S.; Tam, M. C.; Schreiner, P. R.;
Koch, H. J. Am. Chem. Soc. 2005, 127, 1368-1369. (g) Grimme, S.; Furche,
F.; Ahlrichs, R. Chem. Phys. Lett. 2002, 361, 321-328. (h) Grimme, S.
Chem. Phys. Lett. 2001, 339, 380-388. (i) Ruud, K.; Helgaker, T. Chem.
Phys. Lett. 2002, 352, 533-539. (j) Autschbach, J.; Patchkovskii, S.; Ziegler,
T.; van Gisbergen, S. J. A.; Baerends, E. J. J. Chem. Phys. 2002, 117, 581-
592. (k) Stephens, P. J.; Devlin, F. J.; Cheeseman, J. R.; Frisch, M. J. J.
Phys. Chem. A 2001, 105, 5356-5379.
(38) (a) Rathore, R.; Burns, C. L.; Deselnicu, M. I.; Denmark, S. E.;
Bui, T. Org. Synth. 2005, 82, 1-9. (b) Rathore, R.; Kumar, A. S.; Lindeman,
S. V.; Kochi, J. K. J. Org. Chem. 1998, 63, 5847-5856.
(39) (a) Mori, T.; Ko, Y. H.; Kim, K.; Inoue, Y. J. Org. Chem. 2006,
71, 3232-3247. (b) Mori, T.; Inoue, Y. Angew. Chem., Int. Ed. 2005, 44,
2582-2585. (c) Mori, T.; Inoue, Y. J. Phys. Chem. A 2005, 109, 2728-
2740. (d) Mori, T.; Izumi, H.; Inoue, Y. J. Phys. Chem. A 2004, 108, 9540-
9549.
7008 J. Org. Chem., Vol. 72, No. 18, 2007

Conformation of Multiarmed Chiral Aryl Ethers
excited-state of 2, owing to the chiral side-arms placed in the
ortho position. By the Arrhenius-type treatment, the activation
energies were evaluated as Ea ) 27.0 kcal mol-1 for 2 and
20.2 kcal mol-1 for 3, respectively. These values are much larger
than the energy for the rotation of small ethyl group (barrier of
conformational alteration: 9-12 kcal mol-1) in the polyethy-
larenes,20,40 but still in an energy range accessible at ambient
temperatures.
stabilization energy of radical cation of 2 (2.8 kcal mol-1) was
found larger than that of 3 (2.1 kcal mol-1). Both 2 and 3
exhibited strong fluorescence upon electronic excitation (φFL
) 0.27). By Arrhenius plots of the fluorescence maxima, the
activation energies of side-arm alteration were evaluated as Ea
) 27.0 kcal mol-1 and 20.2 kcal mol-1, for 2 and 3, respectively.
By using a chiral probe and experimental and theoretical CD
spectra, the dynamic behavior of the side-arms of the multiarmed
aryl ethers in solution was revealed for the first time. Thus, the
syn-conformation in bipodal donors and the partially alternated
conformation in 1,3,5-tripodal donors turned out to be more
important in solution, than were previously thought, in interpret-
ing the given spectroscopic features. The investigation of X-ray
structures alone is therefore not sufficient to understand the
conformations of multiply substituted host molecules in con-
densed phase.
Summary and Conclusions
We prepared new multiarmed chiral aryl ethers (2-5)
containing two, three, five, or eight side-arms on a series of
gradually expanded aromatic cores and investigated the con-
formation and dynamic behavior of the side-arms to find that
the CD spectra measured at varying temperatures are very
informative. While the fully side-arm-alternated conformers are
most stable in 2 and 3, the syn-conformer of 2 or the partially
alternated conformer 3 is also important in solution to interpret
all the spectroscopic observations. Scheme 1 illustrates the
conformational variations of 2-5. Important findings are
summarized as follows:
(1) X-ray crystallographic analyses: Two anti-conformers
(up-down and down-up) were found in a 1:1 ratio in
crystalline 2. The crystal structure of 3 shows that the fully
alternated up-down disposition is favored. Thus, all the chiral
side-arms in 3 are syn with respect to the central ring in the
solid state.
Experimental Section
Preparation of Multiarmed Chiral Donors. The chiral ether
1 was prepared by Mitsunobu reaction41 of 2-methyl-4-methox-
yphenol with (S)-2-butanol. Compounds 2-4, which contain two,
three, or eight chiral donor tether groups, were prepared by Ag+-
promoted aryl substitution of the corresponding benzyl bromides.27
The chiral corannulene derivative 5, which bears somewhat different
chiral aromatic side-arms with the same (R)-2-methylpropyloxy
group, was prepared by Ni-catalyzed arylation of sym-pentachlo-
rocorannulene.42 A range of core molecules (benzene, biphenylene,
and corannulene) were selected in this study in order to investigate
the effect of different scaffolds on the conformational and dynamic
behavior of the side-arms in solution. The same chiral donor (similar
one in the case of 5) was selected as side-arm, which enabled us to
study the conformational behavior in solution by experimental (and
some theoretical) circular dichroism. Details of the synthetic
procedure and full characterization details are described below or
found in the Supporting Information.
(2) DFT-D calculations: dispersion-corrected DFT calcula-
tions at the B-LYP/TZVP level yielded six anti- and four syn-
conformers of 2, only two each of which were found to be
important. The geometries of the two calculated anti-conformers
are in good agreement with the X-ray structures. The calcula-
tions with the same level of theory afforded the syn-3, fully
alternated 4, as well as C5-symmetric 5 as the stable conformers.
1,2-Bis(2-methoxy-4-methyl-5-(R)-2-methylpropyloxytolyl)tet-
ramethylbenzene (2). Yield: 16%. Mp: 115-116 °C (diethyl
1
(3) H NMR and UV-vis spectra: These spectra of poly-
1
tentacled donors were investigated in detail at different tem-
peratures to afford important information about the conformation
of the molecules (gradual freezing of the side-arms upon
cooling), although the fast isomerization of the side-arms to form
partially alternated conformers (side-arm alteration) could not
be tracked.
ether/ethanol). H NMR: 0.81 (6H, t, J ) 7.6 Hz), 1.07 (6H, d, J
) 6.4 Hz), 1.41 (2H, pseudo sept, J ) 7.6 Hz), 1.57 (2H, pseudo
sept, J ) 7.6 Hz), 2.08 (6H, s), 2.17 (6H, s), 2.25 (6H, s), 3.75
(6H, s), 3.77 (2H, m, J ) 5.6 Hz, peaks are overlapped), 3.79 (4H,
s), 6.12 (2H, s), and 6.61 (2H, s). 13C NMR: 9.7, 16.4, 16.8, 17.0,
19.4, 29.1, 29.8, 56.0, 76.2, 112.9, 116.3, 125.6, 127.4, 133.08,
133.12, 134.6, 149.9, and 151.2. MS (EI, direct): m/z ) 547 (M+
+ 1, 29%), 546 (M+, 100), 339 (16), 321 (14), 295 (25), 284 (12),
283 (27), 279 (24), 266 (18), 265 (49), 250 (19), 207 (13), 151
(42), 147 (38), 125 (26), and 121 (17). HRMS (EI): 546.3707.
C36H50O4 requires 546.3709. EA: Found: C, 78.82; H, 9.25%.
Calcd for C36H50O4: C, 79.08; H, 9.22; O, 11.70. Specific
(4) Temperature dependence of the CD spectra: In contrast,
the low-temperature CD spectra of 2 and 3 provided important
insights into such behavior. By decreasing the temperature, an
increased negative Cotton effect at the 1Lb band was only evident
for compounds 1, 4, and 5. Because of the attractive van der
Waals interactions between the arms in nonpolar solvents, the
syn-conformations become increasingly significant for 2 at low
temperatures to induce a weak positive Cotton effect at the 1Lb
band. A much larger positive Cotton effect was observed for 3
rotation: [R]25 -21.9 ( 4.2° (c 0.10, CHCl3).
D
1,3,5-Tris(2-methoxy-4-methyl-5-(R)-2-methylpropyloxytolyl)-
triethylbenzene (3). Yield: 20%. Mp: 105-106 °C (dichlo-
romethane/ethanol). 1H NMR: 0.65 (9H, t, J ) 7.6 Hz), 0.95 (9H,
d, J ) 6.0 Hz), 1.09 (9H, t, J ) 7.6 Hz), 1.22-1.45 (6H, m, J )
7.2 Hz), 2.15 (9H, s), 2.46 (6H, q, J ) 7.2 Hz), 3.62 (3H, sxt, J )
6.0 Hz), 2.84 (9H, s), 3.92 (3H, t, J ) 18.0 Hz), 4.06 (3H, t, J )
18.0 Hz), 6.09 (3H, s), and 6.64 (3H, s). 13C NMR: 9.2, 15.6, 16.4,
19.3, 23.7, 28.1, 28.5, 56.2, 76.1, 113.1, 116.6, 125.9, 127.5, 141.3,
150.1, and 151.0. MS (EI, direct): m/z ) 781 (M+ + 1, 17%), 780
1
at the Lb band due to the C3-symmetric conformer. A rapid
increase of this positive Cotton effect with decreasing temper-
ature can be explained by a larger population of partially
alternated conformers and/or slower alteration of the side-arms
at the low temperatures.
(5) Electrochemistry and fluorescence spectra: While a
stepwise and reversible oxidation-reduction (1.02 and 1.14 V
vs SCE) was found in the CV of 2, a quasi-single oxidation at
much higher potential (1.20 V) was observed for 3. The
(41) (a) Marks, V.; Gottlieb, H. E.; Biali, S. E. J. Am. Chem. Soc. 1997,
119, 9672-9679. (b) Taha, M.; Marks, V.; Gottlieb, H. E.; Biali, S. E. J.
Org. Chem. 2000, 65, 8621-8628. (c) Marks, V.; Gottlieb, H. E.; Melman,
A.; Byk, G.; Cohen, S.; Biali, S. E. J. Org. Chem. 2001, 66, 6711-6718.
(42) (a) Mitsunobu, O. Synthesis 1981, 1-28. (b) Huges, D. L. Org.
Prep. Proced. Int. 1996, 28, 127-164. (c) Dodge, J. A.; Jones, S. A. Rec.
Res. DeV. Org. Chem. 1997, 1, 273-283.
(40) Mori, T.; Takamoto, M.; Wada, T.; Inoue, Y. HelV. Chim. Acta
2001, 84, 2693-2707.
J. Org. Chem, Vol. 72, No. 18, 2007 7009

Mori et al.
(M+, 33%), 612 (19), 306 (16), 207 (23), 152 (10), 151 (100), and
121 (10). HRMS (EI): 780.5337. C51H72O6 requires 780.5329.
EA: Found: C, 78.23; H, 9.30%. Calcd for C51H72O6: C, 78.42;
O, 12.29; H, 9.29. Specific rotation: [R]25D -25.2 ( 5.8° (c 0.10,
CHCl3).
calculations) and was thus used to obtain the final Boltzmann
distribution at 298 K of the conformers. Details are reported in
Supporting Information.
All excited-state calculations have been performed at the
optimized ground-state geometries and thus correspond to vertical
transitions. The CD (and UV-vis) spectra of 1 and 2 were simulated
on the basis of time-dependent density functional theory (TD-
DFT)29 with the BH-LYP49 functional and employing the TZVP
basis set. The program module escf50 has been used in these TD-
DFT treatments. The CD spectra were simulated by overlapping
Gaussian functions for each transition where the width of the band
at 1/e height is fixed at 0.4 eV and the resulting intensities of the
combined spectra were scaled to the experimental values. As usual,
the calculated band intensities are larger than their experimental
counterparts and are thus scaled to facilitate comparison with
experiment. The somewhat smaller scaling factors found (1/5 and
1/20, for 1 and 2, respectively, in contrast to standard 1/2-1/3)
are probably due to our ignorance of the dynamic behavior in the
present theoretical treatments (see also ref 26 for more discussion).
Because of a systematic overestimation of the transition energies
compared to the experiment in the BH-LYP calculations, the spectra
were uniformly shifted by 0.5 eV. The optical rotations at the
sodium-D line wavelength of 1 and 2 were also calculated at the
TD-DFT-BH-LYP method using Dunning’s aug-cc-pVDZ51 and the
aug-SVP52 basis-sets (H, [3s2p], C/O, [4s3p2d]). The rotational
strengths from the length-gauge representation, which are known
to be numerically more robust, were used throughout. It is known
that in the absence of GIAOs, these values are origin-dependent.
Strictly speaking, the calculated optical rotations and CD intensities
can only be compared to experimental values when they are origin-
independent, but the length and velocity rotational strengths
converge to the same value in the complete basis-set limit. With
our relatively large triple-ú type AO basis-sets, the differences
between both representations were negligible (differences are mostly
less than 2-3%).
Octakis(2-methoxy-4-methyl-5-(R)-2-methylpropyloxytolyl)-
biphenylene (4). Yield: 3%. 1H NMR: 0.90 (24H, t, J ) 7.6 Hz),
1.17 (6H, d, J ) 6.0 Hz), 1.49 (8H, pseudo sept, J ) 7.6 Hz), 1.63
(8H, pseudo sept, J ) 7.8 Hz), 2.04 (24H, s), 2.19 (16H, s), 3.74
(24H, s), 3.96 (8H, sxt, J ) 6.0 Hz), 6.40 (8H, s), and 6.62 (8H,
s). 13C NMR: 9.8, 16.4, 16.5, 19.5, 29.2, 55.9, 76.2, 113.0, 115.7,
124.8, 125.9, 126.62, 137.8, 147.9, 145.0, and 151.2. MS (FD):
m/z ) 1802.16 (M+ + 1). MS (FAB): m/z ) 1802.17 (M+ + 1).
C116H152O16 requires 1801.11. Specific rotation: [R]25 -15.6 (
D
4.1° (c 0.10, CHCl3).
1,3,5,7,9-Pentakis[p-(R)-methylpropyloxyphenyl]coran-
1
nulene (5). Yield: 45%. H NMR: 1.03 (15H, t, J ) 7.5 Hz),
1.36 (15H, d, J ) 6.0 Hz), 1.68 (5H, pseudo sept, J ) 7.5 Hz),
1.82 (5H, pseudo sept, J ) 7.5 Hz), 4.38 (5H, pseudo sxt, J )
6.0 Hz), 6.97 (10H, d, J ) 8.6 Hz), 7.54 (10H, d, J ) 8.6 Hz), and
7.76 (5H, s). 13C NMR: 10.0, 19.5, 29.4, 75.2, 116.0, 125.3, 129.4,
131.3, 132.2, 135.3, 141.8, and 158.2. HRMS (FAB): 990.5231.
C70H70O5 requires 990.5223. EA: Found: C, 84.58; H, 7.08%.
Calcd for C70H70O5: C, 84.81; H, 7.12; O, 8.07. Specific rotation:
[R]25 -45.8 ( 4.5° (c 0.10, CHCl3).
D
Technical Details of the Quantum Chemical Computations.
All calculations were performed on Linux-PCs using the TURBO-
MOLE 5.8 program suite.43 All conformers were fully optimized
at the dispersion-corrected DFT-D-B-LYP level44 employing C2,
C3, or C5 symmetry constraint, where appropriate, by using an AO
basis set of valence triple-ú quality with a set of polarization
functions (denoted as TZVP;45 in standard notation: H, [3s1p], C/O,
[5s3p1d]), and numerical quadrature grid m4. The most stable Tg-
conformation was employed for the chiral alkyl group throughout
this study.26 This conformation was found also in the X-ray crystal
structures of 2 and 3 in this study (Vide infra). The resolution of
identity (RI) approximation47 was employed in all DFT-D-B-LYP
calculations, and the corresponding auxiliary basis sets were taken
from the TURBOMOLE basis-set library. Subsequent single-point
energy calculations were performed with the spin-component-scaled
(SCS)-MP2 method22 with a TZVPP basis-set that has additional
d/f and p/d functions on non-hydrogen and hydrogen atoms,
respectively. It has been shown that this simple modification of
the standard MP2 scheme, termed SCS-MP2, leads to dramatic
improvements in accuracy of calculated energies, particularly for
molecules with weak interactions where standard DFT fails.48 The
method is expected to provide the most accurate relative energies
(comparable to the computationally highly demanding CCSD(T)
Acknowledgment. T.M. thanks the Alexander von Hum-
boldt-Stiftung for the fellowship. We thank Drs. Christian Mu¨ck-
Lichtenfeld, Christian Diedrich, and Manuel Piacenza for
technical assistance and fruitful discussion of the theoretical data.
Financial support of this work by a Grant-in-Aid for Scientific
Research from the Ministry of Education, Culture, Sports,
Science, and Technology of Japan to T.M. is gratefully
acknowledged.
Supporting Information Available: Details of experimental
procedure and computation, X-ray crystallographic structure de-
termination (CCDC 650013 and 650014) as well as CIF files of 2
and 3, cyclic voltammometric studies of 2 and 3, 1H and 13C NMR
charts and theoretical and experimental UV-vis and CD spectra
of 1-5 under a variety of conditions, and Cartesian coordinates of
the optimized geometries of a variety of conformations of 1-5.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(43) Grube, G. H.; Elliott, E. L.; Steffens, R. J.; Jones, C. S.; Baldridge,
K. K.; Siegel, J. S. Org. Lett. 2003, 5, 713-716.
(44) Ahlrichs, R.; Ba¨r, M.; Baron, H.-P.; Bauernschmitt, R.; Bo¨cker, S.;
Crawford, N.; Deglmann, P.; Ehrig, M.; Eichkorn, K.; Elliott, S.; Furche,
F.; Haase, F.; Ha¨ser, M.; Horn, H.; Ha¨ttig, C.; Huber, C.; Huniar, U.;
Kattannek, M.; Ko¨hn, A.; Ko¨lmel, C.; Kollwitz, M.; May, K.; Nava, P.;
Ochsenfeld, C.; O¨ hm, H.; Patzelt, H.; Rappoport, H. D.; Rubner, O.; Scha¨fer,
A.; Schneider, U.; Sierka, M.; Treutler, O.; Unterreiner, B.; von Arnim,
M.; Weigend, F.; Weis, P.; Weiss, H. TURBOMOLE Version 5.8; Universita¨t
Karlsruhe: Karlsruhe, 2005. See also: http://www.cosmologic.de/Quan-
tumChemistry/ main_turbomole.html.
JO071216N
(45) Grimme, S. J. Comput. Chem. 2004, 25, 1463-1473.
(46) Scha¨fer, A.; Huber, C.; Ahlrichs, R. J. Chem. Phys. 1994, 100,
5829-5835.
(47) (a) Eichkorn, K.; Treutler, O.; O¨ hm, H.; Ha¨ser, M.; Ahlrichs, R.
Chem. Phys. Lett. 1995, 240, 283-289. (b) Weigend, F.; Ha¨ser, M. Theor.
Chem. Acc. 1997, 97, 331-340. (c) Grimme, S.; Waletzke, M. Phys. Chem.
Chem. Phys. 2000, 2, 2075-2081. (d) Vahtras, O.; Almlo¨f, J.; Feyereisen,
M. W. Chem. Phys. Lett. 1993, 213, 514-518. (e) Ha¨ttig, C.; Weigend, F.
J. Chem. Phys. 2000, 113, 5154 -5161.
(49) Notations from Gaussian package corresponds to BH&HLYP. See
also: (a) Maity, D. K.; Duncan, W. T.; Truong, T. N. J. Phys. Chem. A
1999, 103, 2152-2159. (b) Duncan, W. T.; Truong, T. N. J. Chem. Phys.
1995, 103, 9642-9652. (c) Liao, M.-S.; Lu, Y.; Scheiner, S. J. Comput.
Chem. 2003, 24, 623-631.
(50) Bauernschmidt, R.; Ahlrichs, R. Chem. Phys. Lett. 1996, 256, 454-
464.
(51) (a) Dunning, T. H. J. Chem. Phys. 1993, 98, 7059-7071. (b)
Kendall, R. A.; Dunning, T. H.; Harrison, R. J. J. Chem. Phys. 1992, 96,
6796-6806.
(48) (a) Grimme, S. Angew. Chem., Int. Ed. 2006, 45, 4460-4464, and
references cited therein. (b) Grimme, S.; C. Diedrich, S.; Korth, M. Angew.
Chem., Int. Ed. 2006, 45, 625-629.
(52) Hobza, P.; Sponer, J. J. Am. Chem. Soc. 2002, 124, 11802-11808.
7010 J. Org. Chem., Vol. 72, No. 18, 2007