Synthesis, structure, and resolution of exceptionally twisted pentacenes

Article abstract of DOI:10.1021/ja065935f

9,10,11,20,21,22-hexaphenyltetrabenzo[a,c,l,n]pentacene (2) and a dimethyl derivative (2m) were prepared by the reaction of 1,3-diphenylphenanthro[9,10-c] furan with bisaryne equivalents generated from 1,2,4,5-tetrabromo-3,6- diarylbenzenes in the presenc

Full text of DOI:10.1021/ja065935f

Published on Web 12/06/2006
Synthesis, Structure, and Resolution of Exceptionally Twisted
Pentacenes
Jun Lu,† Douglas M. Ho,† Nancy J. Vogelaar,‡ Christina M. Kraml,§,|
Stefan Bernhard,† Neal Byrne,| Laura R. Kim,| and Robert A. Pascal, Jr.*,†
Contribution from the Department of Chemistry, Princeton UniVersity, Princeton, New Jersey
08544, Department of Biology, Virginia Tech, Blacksburg, Virginia 24061, Wyeth Research,
CN 8000, Princeton, New Jersey 08543, AccelaPure Corporation, Princeton, New Jersey 08544
Received August 15, 2006; E-mail: snake@chemvax.princeton.edu
Abstract: 9,10,11,20,21,22-hexaphenyltetrabenzo[a,c,l,n]pentacene (2) and a dimethyl derivative (2m) were
prepared by the reaction of 1,3-diphenylphenanthro[9,10-c]furan with bisaryne equivalents generated from
1,2,4,5-tetrabromo-3,6-diarylbenzenes in the presence of n-butyllithium, followed by deoxygenation of the
double adducts with low-valent titanium. Both are bright red solids with a strong orange fluorescence in
solution. The X-ray structures of these compounds show them to be the most highly twisted polycyclic
aromatic hydrocarbons known. Compound 2 has an end-to-end twist of 144°, and the two crystallographically
independent molecules of 2m have twists of 138° and 143°. Both molecules were resolved by
chromatography on chiral supports, and the pure enantiomers have extremely high specific rotations (for
2, [R]D ) 7400°; for 2m, 5600°), but the molecules racemize slowly at room temperature (∆Gq ) 24
rac
kcal/mol). Both the experimental geometry and the observed racemization barrier for 2 are in good agreement
with computational studies of the molecule at a variety of levels. Attempts to prepare compound 2 by reaction
of tetraphenylbenzyne with 9,10,12,13-tetraphenyl-11-oxacyclopenta[b]triphenylene (3, a twisted isoben-
zofuran) gave no adducts, and attempts to prepare tetradecaphenylpentacene by reaction of hexaphenyl-
isobenzofuran (11) with bisaryne equivalents gave only monoadducts.
Introduction
but two years ago we reported the synthesis and structure of
9,10,11,20,21,22-hexaphenyltetrabenzo[a,c,l,n]pentacene (2) as
The term “acene” calls to mind a relatively rigid, flat
molecule. However, such compounds are certainly not rigid,
inasmuch as only a few kilocalories/mole are required to impart
substantial bending or twisting distortions to naphthalene,
anthracene, or the higher acenes.1 Furthermore, there are many
examples of sterically encumbered acenes that display severe
distortions from planarity yet are perfectly stable under ordinary
conditions.1 Most such compounds are derivatives of naphtha-
lene or anthracene that bear several bulky substituents or
multiple benzannulations, but a few highly twisted naph-
thacenes,2,3 pentacenes,4,5 and even one heptacene6 have been
described. For quite a few years the most highly twisted acene
derivative was 9,10,11,12,13,14,15,16-octaphenyldibenzo-
[a,c]naphthacene (1), with an end-to-end twist of 105°,3
a preliminary communication.7 This molecule is by far the most
highly twisted acene to have been prepared, and it contains the
most twisted naphthalene (60°), anthracene (88°), naphthacene
(116°), and pentacene (144°) substructures known. In this paper
we report the full details of the successful (and some notable
but unsuccessful) syntheses of 2 and related, twisted pentacenes,
the X-ray structures of 2 and its dimethyl derivative 2m (Scheme
1), and the resolution of both compounds into pure enantiomers
with exceptional optical rotations.
† Princeton University.
‡ Virginia Tech.
§ Wyeth Research.
| AccelaPure.
(1) Review: Pascal, R. A., Jr. Chem. ReV. published online September 21,
2006 http://dx.doi.org/10.1021/cr050550l.
(2) Fagan, P. J.; Ward, M. D.; Caspar, J. V.; Calabrese, J. C.; Krusic, P. J. J.
Am. Chem. Soc. 1988, 110, 2981-2983.
(3) Qiao, X.; Ho, D. M.; Pascal, R. A., Jr. Angew. Chem., Int. Ed. Eng. 1997,
36, 1531-1532.
(4) Zhang, J.; Ho, D. M.; Pascal, R. A., Jr. Tetrahedron Lett. 1999, 40, 3859-
3862.
Results and Discussion
(5) Schuster, I. I.; Craciun, L.; Ho, D. M.; Pascal, R. A., Jr. Tetrahedron 2002,
58, 8875-8882.
Attempted Synthesis of Compound 2 via Single Aryne
Addition. The preparation of compound 1 was achieved by
(6) Duong, H. M.; Bendikov, M.; Steiger, D.; Zhang, Q.; Sonmez, G.; Yamada,
J.; Wudl, F. Org. Lett. 2003, 5, 4433-4436.
9
10.1021/ja065935f CCC: $33.50 © 2006 American Chemical Society
J. AM. CHEM. SOC. 2006, 128, 17043-17050
17043

A R T I C L E S
Scheme 1a
Lu et al.
Figure 1. Molecular structure of compound 3. Thermal ellipsoids have
been drawn at the 50% probability level, and hydrogen atoms have been
omitted for clarity.
conditions that produced it, and our best yields (ca. 30%) were
achieved only after carefully controlling the heating times for
the three stages of the process, especially the final 2-min reflux
in acetic acid (see the Experimental Section). Once isolated and
crystallized, the bright orange 3 is reasonably stable, but best
stored in a freezer. Its molecular structure is shown in Figure
1; this compound, a potential precursor of twisted acenes,
already possesses a 43° end-to-end twist.
Unfortunately, the diazotization of 4 in the presence of 3
failed to yield any of the desired epoxypentacene adduct.
Compound 3 was not completely stable to the heat (refluxing
1,2-dichloroethane) and oxidant (isoamyl nitrite) present in the
reaction mixture; indeed, it appeared that 3 was reoxidized and
opened to return diketone 6 during the reaction! In order to
prevent the exposure of 3 to oxidants, we converted compound
4 to the corresponding arenediazonium carboxylate hydrochlo-
ride by the method of Hart and Oku.11 Addition of propylene
oxide to diazonium salts of this type removes the HCl and
generates an aryne under exceptionally mild conditions.11 When
this procedure was carried out in the presence of 1,3-diphenyl-
isobenzofuran, the expected adduct was formed in 16% yield
(data not shown), but no adduct was obtained upon reaction
with the more congested isobenzofuran 3. It seemed that this
particular aryne addition was too sterically demanding, and thus
this route was abandoned.
a (a) Dibenzoylacetylene, PhBr, heat, 74%; (b) NaOH, Zn, 2-methoxy-
ethanol, reflux, 31%; (c) (i) I2, KIO3, H2SO4, 97%, (ii) aryl iodide, Cu, 230
°C, for 8, 20%, for 8m, 13%; (d) n-BuLi, hexanes, toluene, for 9, 26%, for
9m, 15%; (e) n-BuLi, TiCl3, ether, hexanes, for 2, 27%, for 2m, 17%; (f)
n-BuLi, hexanes, toluene, 28%.
diazotization of the bulky anthranilic acid 4 in the presence of
hexaphenylisobenzofuran (11, Scheme 1) followed by deoxy-
genation of the resulting epoxynaphthacene adduct.3 It was
fortunate that both of the precursors, 4 and 11, had been reported
some years before.8-10 A logical extension of this work was
the synthesis of 2 by the reaction of 4 with the as yet unknown
isobenzofuran 3, and we focused our attention on the preparation
of the latter molecule.
Synthesis of Compounds 2 and 2m via Double Aryne
Addition. The addition of 1,4-benzadiyne equivalents to
crowded cyclopentadienones has been used to prepare strongly
twisted acene derivatives,5,6 although none so crowded as 2.
The simplest way to generate these “1,4-benzadiynes” is to treat
1,2,4,5-tetrahalobenzenes with alkyllithiums,12 but this method
is incompatible with most cyclopentadienones, and thus, the
previous studies used more complex reagents to form the
required bisaryne equivalents.5,6 However, 1,4-diphenylphenan-
thro[9,10-c]furan (7, Scheme 1) is a base-insensitive Diels-
Alder diene that contains the necessary carbon framework, and
its recently reported, easy synthesis13 permitted the development
of a relatively simple synthesis of compound 2.
The synthesis of 3 proved to be challenging, but it was
ultimately accomplished by a method very similar to that
empoyed to make 11.9 The Diels-Alder reaction of dibenzoyl-
acetylene and phencyclone (5, Scheme 1) was nonproblematic,
but the reductive dehydration of diketone 6 to give isobenzofuran
3 was extremely sensitive. The product 3 was not stable to the
1,2,4,5-Tetrabromobenzene (10) was treated with iodine and
potassium iodate in sulfuric acid, and the resulting 1,2,4,5-
tetrabromo-3,6-diiodobenzene was subjected to a crossed Ull-
(7) Lu, J.; Ho, D. M.; Vogelaar, N. J.; Kraml, C. M.; Pascal, R. A., Jr. J. Am.
Chem. Soc. 2004, 126, 11168-11169.
(8) Smyth, N.; Van, Engen, D.; Pascal, R. A., Jr. J. Org. Chem. 1990, 55,
1937-1940.
(11) Hart, H.; Oku, A. J. Org. Chem. 1972, 37, 4269-4274.
(12) Hart, H.; Lai, C.; Nwokogu, G.; Shamouilian, S.; Teuerstein, A.; Zlotgorski,
C. J. Am. Chem. Soc. 1980, 102, 6649-6651.
(9) Ried, W.; Bo¨nnighausen, K. H. Liebigs Ann. Chem. 1961, 639, 61-67.
(10) Qiao, X.; Padula, M. A.; Ho, D. M.; Vogelaar, N. J.; Schutt, C. E.; Pascal,
R. A., Jr. J. Am. Chem. Soc. 1996, 118, 741-745.
(13) Marchand, A. P.; Srinivas, G.; Watson, W. H. ARKIVOC 2003 (iii), 8-15.
9
17044 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006

Exceptionally Twisted Pentacenes
A R T I C L E S
hope that it would form better crystals. The synthesis of 2m
mirrored that of 2. Ullmann coupling of 10 and 4-iodotoluene
gave 1,2,4,5-tetrabromo-3,6-di(p-tolyl)benzene (8m), and the
generation of the corresponding “bisaryne” and addition to furan
7, followed by deoxygenation of the double adduct 9m, gave
pentacene 2m without difficulty. Compound 2m is, as expected,
a bright red solid, high-melting but freely soluble, and it
crystallizes easily from several solvents. Ultimately, the crystal
structures of both compounds 2 and 2m were determined.
Having successfully prepared the polyphenyltetrabenzopen-
tacenes 2 and 2m, we used a similar strategy for an attempted
synthesis of tetradecaphenylpentacene. Compound 8 was treated
with n-butyllithium in the presence of hexphenylisobenzofuran
(11) in the hope that a double addition to this reactive diene
would occur. Unfortunately, only monoadducts, such as the
epoxyanthracene 12 (Scheme 1), were observed in these
reactions.
Figure 2. Molecular structure of compound 9. Thermal ellipsoids have
been drawn at the 50% probability level, and hydrogen atoms have been
omitted for clarity.
Molecular and Crystal Structures of Compounds 2 and
2m. Compound 2 is easily crystallized from benzene-ethanol
or benzene-2-propanol, but as mentioned above, the crystals
tend to be multiple. One of these was cut to yield a small, more
nearly single crystal, and a preliminary structure of 2 was
determined from a weak X-ray data set collected on a
conventional diffractometer. A much better data set was obtained
later by using a small crystal fragment and a synchrotron X-ray
source, and the resulting determination is presented here. The
molecular structure of 2 is illustrated in Figure 3, and a
stereoview of the molecule is plotted in Figure 4.
mann coupling reaction with iodobenzene and copper powder
at 230 °C. The yield of the desired 1,2,4,5-tetrabromo-3,6-
diphenylbenzene (8) was a modest 20%, but the use of this
solvent-free, classical, high-temperature reaction was neces-
sitated by the low solubility of the hexahalobenzene intermedi-
ate. Fortunately, the purification of 8 was straightforward.
Treatment of compound 8 with n-butyllithium in the presence
of furan 7 gave the double adduct 9 in 26% yield.
Compound 9 crystallized from CH2Cl2-DMSO, its X-ray
structure was determined, and the molecular structure of this
adduct is shown in Figure 2. The molecule has crystallographic
Ci symmetry and approximate C2h symmetry; thus, diepoxide
9 has a trans geometry, with oxygen atoms on opposite faces
of the pentacene nucleus. The corresponding cis isomer, which
may have been present in small quantities in the reaction
mixture, but was not isolated, possesses C2V symmetry. Ab initio
calculations at the HF/3-21G level indicate that the trans
diastereomer is only 0.8 kcal/mol more stable than the cis.
The final step is the deoxygenation of 9 to give pentacene 2.
Similar deoxygenations had been problematic in past twisted
acene syntheses, with yields as low as 3% in the case of
decaphenylanthracene.3,10 However, the deoxygenation of 9 by
low-valent titanium (TiCl3/n-butyllithium14,15) proceeded rela-
tively smoothly: 90% pure 2 was obtained in 51% yield, and
one recrystallization from CH2Cl2-CH3OH gave homogeneous
2 in 27% yield.16
Compound 2 is a bright red solid with a strong orange
fluorescence in solution. It is high-melting (mp >470 °C) but
freely soluble in common organic solvents. It crystallizes readily
from a variety of solvents to give blood-red needles or plates,
but these crystals tend to be multiple, and it was some time
before a satisfactory X-ray structure could be obtained (see
below). For this reason, we also synthesized a simple methyl
derivative of 2, the 10,21-di(p-tolyl) compound 2m, with the
Compound 2 crystallizes in the common monoclinic space
group P21/c with Z ) 4, and the molecule lies on a general
position. As expected, the molecule adopts an extraordinarily
twisted conformation with approximate D2 symmetry. The end-
to-end twist of the pentacene nucleus is 143.6°, and thus,
compound 2 is by far the most highly twisted polycyclic
aromatic hydrocarbon (PAH) yet prepared.1 The twist is evenly
distributed, with the five pentacene rings (left-to-right in Figure
3) contributing 30°, 30°, 27°, 29°, and 27° of the twist. The
degree of “compression” observed in the intramolecular non-
bonded contacts is similarly uniform. The four C-C contacts
between ipso carbons of the phenyl substituents (e.g., C(23)-
C(29)] average 2.94 Å, and the four C-H contacts between
ipso carbons and benzo hydrogens [e.g., C(23)-H(8)] average
2.38 Å.17 Both types of contact distances are approximately 0.5
Å shorter than the sum of the van der Waal radii of the
contacting atoms, but these nonbonded contacts, though short,
are far from the closest such C-C and C-H contacts known,
2.61 and 1.98 Å, respectively.18
Although highly distorted from planarity, crystalline 2 is
completely stable in air at room temperature. Solutions of 2
are stable indefinitely in the dark, but in the presence of bright
room light and air, dissolved 2 oxidizes slowly (t1/2 ∼ 36 h) to
give products of unknown structure. Mass spectra of the mixture
of decomposition products show the addition of up to three
oxygen atoms. Compound 2 is thus much less reactive than
9,11,20,22-tetraphenyltetrabenzo[a,c,l,n]pentacene, which, lack-
(14) Hart, H.; Nwokogu, G. J. Org. Chem. 1981, 46, 1251-1255.
(15) Hart, H.; Lai, C.; Nwokogu, G. C.; Shamouilian, S. Tetrahedron 1987, 43,
5203-5224.
(16) The successful deoxygenation of 9 led us to try the same method for
deoxygenation of 9,10-epoxy-9,10-dihydrodecaphenylanthracene.10 Results
similar to those observed for 9 would have given a 10-fold increase in the
overall yield for the synthesis of decaphenylanthracene. Unfortunately, the
extremely hindered anthracene oxide was untouched by the titanium reagent.
(17) For the estimation of nonbonded contact distances, the C-H bond distances
were “improved” to 1.083 Å, the standard value observed in neutron
diffraction experiments (Allen, F. H.; Kennard, O.; Watson, D. G.;
Brammer, L.; Orpen, A. G.; Taylor, R. J. Chem. Soc., Perkin Trans. 1987,
2, S1-S19.).
(18) Review: Pascal, R. A., Jr. Eur. J. Org. Chem. 2004, 3763-3771.
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 17045

A R T I C L E S
Lu et al.
Figure 5. Molecular structure of compound 2m; both of the crystallo-
graphically independent molecules are shown. Thermal ellipsoids have been
drawn at the 50% probability level, and hydrogen atoms have been omitted
for clarity.
Figure 3. Molecular structure of compound 2. In the top view, thermal
ellipsoids have been drawn at the 50% probability level, and all but the
four sterically encumbered hydrogen atoms have been omitted for clarity.
Figure 4. Stereoview of compound 2.
Figure 6. Packing of compound 2m along the ac diagonal of the unit cell.
The disordered molecules of ethyl acetate, located in the channel, have been
omitted.
ing the two phenyl groups on the central pentacene ring, rapidly
forms the 10,22-endoperoxide in ambient room light.5
molecules that form it, is illustrated in Figure 6. The cross
section of the channel is irregular, but its dimensions are roughly
5 × 10 Å.
Compound 2m crystallized as red prisms from ethyl acetate-
ethanol in the chiral space group P21 (Z ) 4). For a moment
we were able to hope that a spontaneous resolution of compound
2m had occurred, but the structure solution revealed that both
enantiomers were present in the crystal as crystallographically
independent molecules, even though the crystal structure itself
is chiral. The two independent molecules of 2m are illustrated
in Figure 5. With the exception of the additional methyl groups,
both are virtual clones of compound 2. Both adopt conforma-
tions with approximate D2 symmetry, and the end-to-end twists
of the two molecules of 2m are 137.7° and 142.8°, only slightly
less than the twist observed in 2. However, while the crystals
of 2 are those of a pure hydrocarbon, compound 2m crystallized
as an ethyl acetate solvate, and the disordered ethyl acetate
molecules are located in chiral channels along the ac diagonal
of the unit cell. One of these channels, and the stacks of 2m
Computational Studies of Compound 2. The experimental
structure of compound 2 is well-reproduced by a variety of
computational methods. The end-to-end twists of the pentacene
in the D2-symmetric ground state of 2 are calculated to be 143°
at the AM1 level, 144° at the HF/3-21G level, and 149° at the
B3LYP/6-31G(d) level.19,20 A second, C2h-symmetric (and thus
achiral) conformation of compound 2 is also a potential
minimum at these levels, calculated to lie 12.6 kcal/mol (AM1),
17.6 kcal/mol (HF/3-21G), and 19.7 kcal/mol (B3LYP/ 6-31G-
(19) After our initial publication of the synthesis of 2,7 Houk and coworkers
carried out an extensive computational study of its electronic structure and
properties.20 Our own computational studies of 2 are chiefly meant to clarify
the molecule’s conformational properties.
(20) Norton, J. E.; Houk, K. N. J. Am. Chem. Soc. 2005, 127, 4162-4163.
9
17046 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006

Exceptionally Twisted Pentacenes
A R T I C L E S
Figure 7. Calculated structures (AM1) of the C2h (left) and D2 (right) conformations of compound 2 as well as the C1-symmetric transition state for their
interconversion (center).
(d)) above the D2 ground state. Many twisted acenes with D2
(or C2) ground states possess higher energy conformations with
determined by two methods: the half-life for the decay of the
specific rotation at 25 °C was 9.3 h, and the half-life for the
loss of enantiomeric excess at 27 °C, as judged by chiral HPLC
C2h (or Ci or Cs) symmetry, and when the difference in energy
is small (<4 kcal/mol), the higher energy conformation is
sometimes observed in the solid-state due to more favorable
crystal packing.1,5,6,21 In the case of 2, where the energy
difference is much greater, the experimental observation of the
analysis, was 6.2 h. Each of these values yields a ∆Gq
)
rac
23.8 kcal/mol (via the Eyring equation and assuming a transmis-
sion coefficient of 1).
The 7400° specific rotation of 2 is comparable to those of
the helicenes: for hexahelicene, [R]24D ) 3640°; for [9]helicene,
[R]25 ) 7500°; and for [13]helicene, [R]25 ) 8840°.22
C2h conformation is extremely unlikely. However, this same C2h
conformation is a likely intermediate in any racemization
pathway for compound 2. Although the calculated energy of
the C2h conformation is not high enough to predict that 2 should
be resolvable at room temperature, even a modest activation
energy for the interconversion of the D2 and C2h conformations
would confer configurational stability to compound 2.
D
D
However, the racemization barriers for the helicenes are
substantially greater than that for 2: for hexahelicene, ∆Gq
rac
) 36.2 kcal/mol, and for [9]helicene, ∆Gqrac ) 43.5 kcal/mol.22
In order to confer similar configurational stability to twisted
acenes analogous to 2, it is probable that the length of the
polyphenyl core must be increased or that additional/bulkier
substituents must be employed.
The resolution of 2 into pure enantiomers with (presumably)
very high optical rotations was of great interest to us. None of
our previously reported twisted PAHs could be resolved because
their barriers to racemization (∆Gqrac) are too low. However,
long before the actual synthesis of 2, we had computationally
examined its racemization at the AM1 level of theory. The
enantiomeric D2 ground states were found to interconvert via
the C2h intermediate through enantiomeric C1-symmetric transi-
tion states. Thus, the racemization of 2 (at the AM1 level) entails
Compound 2m proved to be much more difficult to resolve
than 2, despite its structural similarity. One can only guess that
the extra methyl groups block some critical interaction with the
chiral chromatographic media. The best resolution was achieved
by using a Chiralcel OJ-H column eluted with 4:1 ethanol:
methanol. The (-)-enantiomer eluted first and, by taking the
leading part of the peak, a pure sample was obtained. This
displayed [R]25 ) -5600° ( 200° (c ) 0.002 04). The more
q
q
the sequence (+)-D2 f (+)-C1 f C2h f (-)-C1 f (-)-D2.
One-half of this sequence is illustrated in Figure 7. Most
importantly, the C1 transition state was calculated to lie 23.0
kcal/mol above the ground state (zero-point energy corrections
for the ground and transition states are included in this value).
This calculated barrier corresponds to a half-life of a few hours
at room temperatureslong enough to permit at least a partial
resolution of 2.
D
slowly eluting component gave [R]25D ) +2300° ( 200° (c )
0.001 80), but this was estimated to be a 72:28 mixture of the
(+)- and (-)-enantiomers (44% ee) by analytical chromatog-
raphy; thus, the observed rotation corresponds to [R]25
)
D
+5200° for the pure (+)-enantiomer. In addition, the racem-
ization barrier (∆Gqrac) for compound 2m was found to be 24
kcal/mol based upon the decay of its specific rotation.
Resolution, Specific Rotation, and Racemization of 2 and
2m. Compound 2 was easily resolved by preparative HPLC on
a Chiralcel OD column eluted with ethanol (it is fortunate that
these twisted hydrocarbons are readily soluble even in polar
solvents); thus, the computational prediction of configurational
stability was at least qualitatively correct. The first peak to elute
Conclusion
Compounds 2 and 2m are extremely distorted from planarity,
but due to steric protection of the pentacene nuclei by the
peripheral phenyl groups, these molecules are quite stable. There
is every reason to believe that even longer, more highly twisted,
polyphenyl acenes would also be stable, because the pitch of
the helical cores does not increase dramatically with increased
length, and good π-orbital overlap in the acenes could be
maintained. Twisted acenes with a greater helical pitch, and
correspondingly poorer π-orbital overlap than observed in 2,
exhibited [R]25 ) +7440° ( 150° (c ) 0.006 69), and the
D
second [R]25D ) -7420° ( 150° (c ) 0.003 63); both specific
rotations were measured within a few minutes of elution from
the HPLC column. The barrier to racemization for 2 was
(21) L’Esperance, R. P.; Van Engen, D.; Dayal, R.; Pascal, R. A., Jr. J. Org.
Chem. 1991, 56, 688-694.
(22) Meurer, K. P.; Vogtle, F. Top. Curr. Chem. 1985, 127, 1-76.
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 17047

A R T I C L E S
Lu et al.
the mixture was poured into ice water (300 mL), and sodium bisulfite
powder was added with stirring until the mixture turned yellow. The
precipitate was collected by suction filtration, washed with water, and
dried under vacuum overnight to give a gray solid, 1,2,4,5-tetrabromo-
3,6-diiodobenzene (9.52 g, 14.7 mmol, 97%): mp 328-331 °C. MS
(EI) m/z 646 (M+, 100). A portion of this material (2.00 g, 3.10 mmol),
iodobenzene (1.5 mL, 2.7 g, 13 mmol), and Cu powder (4.7 g) were
thoroughly mixed in a Pyrex tube and placed in a metal bath at 230
°C. After heating for 15 min, the tube was removed from the bath and
toluene (10 mL) was added immediately. The excess copper was filtered
away, and concentration of the filtrate gave crude 8. This material was
chromatographed on a silica gel column (solvent, hexanes), and the
material with Rf 0.30 (TLC, hexanes) was collected. Recrystallization
from CH2Cl2-MeOH gave pure compound 8 (334 mg, 0.612 mmol,
would also be interesting, but the synthesis of such molecules
will require the incorporation of bulkier peripheral substituents.
The obvious course is to incorporate tertiary butyl groups, but
our own experimental and computational studies of poly-tert-
butylnaphthalenes suggest that poly-tert-butylacenes would be
relatively unstable, both to the loss of tert-butyl groups and to
isomerization to Dewar isomers.23 Other very bulky groups are
likely to give rise to similar problems. However, any polysub-
stituted acenes possessing greater length or greater pitch than 2
should be configurationally stable at room temperature. They
would probably have even higher specific rotations than 2, and
their configurational stability would make it simpler to study
other extreme chiroptical properties that these exceptional
molecules might possess.
1
19.7%): mp 262-265 °C; H NMR (CDCl3) δ 7.21 (m, 4 H), 7.49
(m, 6 H); 13C NMR (CDCl3) δ 127.3, 128.7, 128.8, 128.9, 144.1, 146.0
(6 of 6 expected resonances); MS (EI) m/z 546 (M+, 36), 386 (M -
Br2, 36), 226 (M - Br4, 100); exact mass 545.7468, calcd for C18H10-
79Br281Br2 545.7475.
Experimental Section
2,3-Dibenzoyl-1,4-diphenyltriphenylene (6). Phencyclone24,25 (5,
800 mg, 2.09 mmol), dibenzoylacetylene26 (490 mg, 2.09 mmol), and
bromobenzene (3.5 mL) were mixed in a screw-capped tube. The tube
was placed in a metal bath at 80 °C, and it was heated to 186 °C over
1 h. After 30 min further of heating, the reaction mixture was cooled,
and the solvent was removed. The residue was chromatographed on
silica gel (solvent, toluene), and the material with Rf 0.72 on TLC
(solvent, toluene) proved to be compound 6 (906 mg, 1.54 mmol,
74%): mp 256-259 °C; 1H NMR (CDCl3) δ 6.87 (br, 4 H), 7.11 (m,
8 H), 7.29 (td, J ) 7 Hz, 1 Hz, 4 H), 7.47 (m, 6 H), 7.73 (dd, J ) 8
Hz, 1 Hz, 4 H), 8.48 (dd, J ) 8 Hz, 1 Hz, 2 H); 13C NMR (CDCl3) δ
123.7, 126.1, 127.7, 127.9, 129.0, 129.5, 130.1, 130.2, 131.8, 132.1,
132.2, 132.7, 135.9, 138.3, 139.0, 140.7, 199.9 (17 of 18 expected
resonances); MS (EI) m/z 588 (M+, 100), 511 (M - C6H5, 40), 376
(84); exact mass 588.2084, calcd for C44H28O2 588.2090.
1,2,4,5-Tetrabromo-3,6-di(p-tolyl)benzene (8m). Compound 8m
was prepared in 13% yield from 1,2,4,5-tetrabromobenzene and
4-iodotoluene by using the procedure described above: mp 258-261
°C; 1H NMR (CDCl3) δ 2.45 (s, 6 H), 7.09 and 7.31 (AA′BB′ system,
4 H); 13C NMR (CDCl3) δ 21.7, 127.4, 128.5, 129.5, 138.4, 141.4,
145.9 (7 of 7 expected resonances); MS (EI) m/z 574 (M+, 100), 494
(24), 414 (42), 335 (38); exact mass 573.7785, calcd for C20H1479Br2-
81Br2 573.7788. This material was judged to be roughly 85% pure from
the NMR analysis but was used without further purification.
9,10,11,20,21,22-Hexaphenyltetrabenzo[a,c,l,n]pentacene 9,22:
11,20-diepoxide (9). 1,3-Diphenylphenanthro[9,10-c]furan13 (7, 150 mg,
0.405 mmol) and compound 8 (100 mg, 0.183 mmol) were dissolved
in dry toluene (5 mL) and cooled to -78 °C with stirring under Ar.
n-Butyllithium (1.6 M in hexanes, 0.5 mL, 0.8 mmol) was diluted in
dry hexanes (1.5 mL) and added dropwise to the cold solution. The
reaction mixture was allowed to warm to room temperature and it was
stirred overnight. The reacting mixture was then cooled again to -78
°C, and another portion of 1.6 M n-butyllithium (1.0 mL, 1.6 mmol)
was added dropwise. After warming to room temperature and stirring
overnight again, the reaction was quenched with methanol (1.0 mL).
Ether (10 mL) was added, and the resulting suspension was washed
twice with saturated NaCl. The aqueous phase was back-extracted twice
with CH2Cl2, and the combined extracts were dried over Na2SO4.
Evaporation of the solvent left a white solid, 9 (34 mg). The ether
layer was dried over Na2SO4, concentrated, and fractionated by
preparative TLC (solvent, 1:2 hexanes-benzene) to give another portion
of compound 9 (12 mg, Rf 0.32). Thus, a total of 46 mg of 9 was
9,10,12,13-Tetraphenyl-11-oxacyclopenta[b]triphenylene (3). Com-
mercial zinc powder (6 g) was mixed with 2% aqueous HCl (15 mL)
and stirred for 1 min. The zinc powder was collected by suction filtration
and washed three times with water, twice with EtOH, and once with
ether. The product was dried over phosphorus pentoxide under vacuum
overnight. Compound 6 (300 mg, 0.51 mmol) was mixed with
2-methoxyethanol (8 mL) and heated to reflux. Freshly ground NaOH
powder (0.1 g) was added, and the solution gradually became deep
brown after heating for 1 h with stirring. The activated zinc powder
(0.6 g) was then added, and heating was continued for 15 min. The
hot solution was filtered by suction filtration into acetic acid (10 mL),
and the mixture was heated to reflux for 2 min. The solvent was
removed, and purification by preparative TLC (solvent, 3:1 hexanes-
benzene) gave compound 3 as a bright orange solid (89 mg, 0.16 mmol,
31%): mp 277-280 °C; 1H NMR (CDCl3) δ 6.85 (td, J ) 8 Hz, 1 Hz,
2 H), 7.06 (m, 14 H), 7.20 (m, 6 H), 7.29 (d, J ) 8 Hz, 4 H), 8.05 (dd,
J ) 8 Hz, 1 Hz, 2 H); 13C NMR (CDCl3) δ 121.8, 123.7, 126.1, 126.9,
127.0, 127.6, 127.7, 128.8, 128.9, 129.3, 131.8, 131.9, 132.4, 133.3,
139.4, 146.3 (16 of 18 expected resonances); MS (EI) m/z 572 (M+,
100), 467 (41), 389 (66); exact mass 572.2119, calcd for C44H28O
572.2141. Single crystals for X-ray analysis were obtained from
CHCl3-MeOH.
1
obtained (0.048 mmol, 26%): mp >470 °C; H NMR (DMSO-d6) δ
6.91 (t, J ) 8 Hz, 8 H), 7.02 (t, J ) 8 Hz, 4 H), 7.10 (t, J ) 7 Hz, 2
H), 7.15 (d, J ) 7 Hz, 8 H), 7.23 (m, 8 H), 7.89 (m, 8 H), 8.20 (d, J
) 8 Hz, 4 H), 9.18 (d, J ) 8 Hz, 4 H); MS (FAB) m/z 967 (M + H,
100), 861 (32), 756 (32). Single crystals for X-ray diffraction were
obtained from CH2Cl2-DMSO.
9,11,20,22-Tetraphenyl-10,21-di(p-tolyl)tetrabenzo[a,c,l,n]penta-
cene 9,22:11,20-diepoxide (9m). Compound 9m was prepared in 15%
yield from compounds 7 and 8m by using the procedure described
above: 1H NMR (CDCl3) δ 2.18 (s, 6 H), 6.8-7.9 (m, 36 H), 8.29 (d,
J ) 8 Hz, 4 H), 8.99 (d, J ) 8 Hz, 4 H); MS (FAB) m/z 995 (M + H,
100).
1,2,4,5-Tetrabromo-3,6-diphenylbenzene (8). Iodine (21.3 g, 84
mmol) and potassium iodate (2.55 g, 12 mmol) were stirred with
concentrated sulfuric acid (30 mL) for 30 min. 1,2,4,5-Tetrabromoben-
zene (10, 6.00 g, 15.2 mmol) in concentrated sulfuric acid (35 mL)
was added in one portion. After stirring for 5 days at room temperature,
9,10,11,20,21,22-Hexaphenyltetrabenzo[a,c,l,n]pentacene (2). TiCl3
(1.5 g, 9.7 mmol) was mixed with ether (25 mL) and cooled to -78
°C with stirring under Ar. n-Butyllithium (1.6 M in hexanes, 20 mL,
32 mmol) was added dropwise. The mixture was stirred at -78 °C for
1 h, and then it was allowed to warm to room temperature over 1 h,
yielding a dark green solution. A suspension of compound 9 (77 mg,
0.080 mmol) in ether (25 mL) was added dropwise. After stirring
overnight, the reaction mixture was washed twice with saturated NaCl.
(23) Zhang, J.; Ho, D. M.; Pascal, R. A., Jr. J. Am. Chem. Soc. 2001, 123,
10919-10926.
(24) Dilthey, W.; Henkels, S.; Schaefer, A. Ber. Dtsch. Chem. Ges. 1938, 71,
974-979.
(25) Pascal, R. A., Jr.; Van Engen, D.; Kahr, B.; McMillan, W. D. J. Org. Chem.
1988, 53, 1687-1689.
(26) Lutz, R. E.; Smithey, W. R., Jr. J. Org. Chem. 1951, 16, 51-56.
9
17048 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006

Exceptionally Twisted Pentacenes
A R T I C L E S
Table 1. Crystallographic Data for Compounds 2, 2m, 3, and 9
2
2m
‚
2EtOAc
3
9‚3.25CH2Cl2‚0.75DMSO
chemical formula
formula weight
space group
a, Å
b, Å
C74H46
935.11
C76H50‚2.5C4H8O2
1159.40
P21 (no. 4)
14.8577(3)
29.2657(6)
15.1524(5)
105.778(1)
6340.3(3)
4
C44H28O
572.66
C74H46O2‚3.25CH2Cl2‚0.75C2H6SO
1301.71
P21/c (no. 14)
12.6834(2)
18.7000(3)
16.3911(2)
126.091(1)
3141.53(9)
2
P21/c (no. 14)
15.5738(5)
17.2429(8)
18.2439(8)
90.816(2)
4898.7(4)
4
1.268
0.155
100(2)
P21/c (no. 14)
13.4096(3)
13.0716(2)
21.8350(5)
127.643(1)
3030.6(1)
4
1.255
0.073
200(2)
c, Å
â, deg
V, Å3
Z
F
calcd, g/cm3
1.215
0.074
200(2)
1.376
0.372
200(2)
µ, mm-1
T, K
λ, Å
0.97791
31.79
0.71073
22.51
0.71073
22.55
0.71073
27.50
θ
max, deg
reflections
total
unique
83784
6110
5433
0.0635
0.1847
1.075
0.0683
0.1937
1.056
74302
16570
13774
0.0630
0.1460
1.092
0.0767
0.1531
1.038
22268
3990
3363
0.0552
0.1111
1.169
0.0721
0.1325
1.124
42871
7211
5328
0.0727
0.1937
1.051
0.0964
0.2231
1.032
observed [I > 2σ(I)]
R(F) (obs data)a
wR(F2) (obs data)a
S (obs data)a
R(F) (all data)a
wR(F2) (all data)a
S (all data)a
2
2
2
a R(F) ) ∑||Fo| - |Fc||/∑|Fo|; wR(F2) ) [∑w(Fo2 - Fc )2/∑w(Fo )2]1/2; S ) goodness-of-fit on F2 ) [∑w(Fo2 - Fc )2/(n - p)]1/2, where n is the number
of reflections and p is the number of parameters refined.
The red ethereal solution was dried over Na2SO4 and concentrated.
The crude product was fractionated by preparative TLC (solvent, 2:1
hexanes-benzene), and the bright red band of Rf 0.51 was collected to
give compound 2 (38 mg) containing ca. 10% of an unknown impurity.
Recrystallization from CH2Cl2-MeOH gave pure compound 2 (20 mg,
0.021 mmol, 27%): mp >470 °C; 1H NMR (CDCl3) δ 6.41 (t, J ) 7.5
Hz, 4 H), 6.52 (t, J ) 7.5 Hz, 2 H), 6.56 (dd, J ) 8 Hz, 1 Hz, 4 H),
6.62-6.71 (m, 12H), 6.75 (m, 8 H), 6.86 (tt, J ) 7.5 Hz, 1 Hz, 4 H),
6.92 (td, J ) 7.5 Hz, 1 Hz, 4 H), 7.23 (td, J ) 7.5 Hz, 1 Hz, 4 H), 8.07
(dd, J ) 8 Hz, 1 Hz, 4 H); 13C NMR (CDCl3) δ 123.7, 125.6, 126.0,
126.5, 126.9, 128.0, 128.1, 128.3, 130.3, 132.4, 132.9, 133.2, 133.3,
133.4, 134.4, 134.5, 135.3, 140.2, 141.4 (19 of 22 expected resonances);
MS (FAB) m/z 934 (M+, 100); UV (CHCl3) λmax (log ꢀ) 376 (4.84,
sh), 393 (5.10), 482 (3.88, sh), 506 (3.97), 538 (3.86); luminescence
(CHCl3) λmax ) 589 nm, Φ ) 7.4%, [Ru(bipy)3](PF6)2 reference. Single
crystals for X-ray diffraction were obtained from benzene-EtOH.
9,11,20,22-Tetraphenyl-10,21-di(p-tolyl)tetrabenzo[a,c,l,n]penta-
cene (2m). TiCl3 (0.6 g, 3.9 mmol) was mixed with ether (10 mL) and
cooled to -78 °C with stirring under Ar. n-Butyllithium (1.6 M in
hexanes, 8 mL, 13 mmol) was added dropwise. The mixture was stirred
at -78 °C for 1 h, and then it was allowed to warm to room temperature
over 1 h, yielding a dark green solution. A suspension of compound
9m (23 mg, 0.023 mmol) in ether (8 mL) was added dropwise. After
stirring overnight, the reaction mixture was washed twice with saturated
NaCl. The red ethereal solution was dried over Na2SO4 and concen-
trated. The crude product was fractionated by preparative TLC (solvent,
1:1 hexanes-benzene), and the bright red band of Rf 0.75 was collected
to give compound 2m (8 mg, ca. 90% pure). Recrystallization from
CH2Cl2-MeOH gave pure compound 2m (4 mg, 0.004 mmol, 17%):
mp 417-422 °C (dec); 1H NMR (CDCl3) δ 1.92 (s, 6 H), 6.16 (d, J )
7 Hz, 4 H), 6.45 (dd, J ) 8 Hz, 2 Hz, 4 H), 6.59 (d, J ) 8 Hz, 4 H),
6.67 (m, 8 H), 6.75 (m, 8 H), 6.88 (m, 8 H), 7.22 (m, 4 H), 8.06 (d, J
) 8 Hz, 4 H); 13C NMR (CDCl3) δ 21.0, 123.7, 125.5, 126.2, 126.7,
127.4, 127.8, 127.9, 130.4, 132.5, 132.8, 133.2, 133.3, 133.5, 134.4,
134.7, 135.1, 135.2, 137.4, 141.6 (20 of 23 expected resonances); MS
(FAB) m/z 962 (M+, 100); UV (EtOH) λmax (log ꢀ) 246 (4.7), 256 (4.8),
374 (4.9, sh), 386 (5.0), 478 (3.9, sh), 502 (4.0), 532 (3.9). Single
crystals for X-ray diffraction were obtained from EtOAc-EtOH.
9,10-Epoxy-9,10-dihydro-1,2,3,4,5,8,9,10-octaphenylanthracene
(12). Hexaphenylisobenzofuran9,10 (11, 150 mg, 0.26 mmol) and
compound 8 (70 mg, 0.13 mmol) were dissolved in toluene (8 mL)
and cooled to -78 °C with stirring under Ar. n-Butyllithium (1.6 M in
hexanes, 0.5 mL, 0.8 mmol) was diluted in dry hexanes (1.5 mL) and
added dropwise to the cold solution. The reaction mixture was allowed
to warm to room temperature and it was stirred overnight. The reacting
mixture was then cooled again to -78 °C, and another portion of 1.6
M n-butyllithium (1.0 mL, 1.6 mmol) was added dropwise. After
warming to room temperature and stirring overnight again, the reaction
was quenched with methanol (0.5 mL). After removing the solvent,
the crude product was chromatographed on a silica gel column (solvent,
1:1 hexanes-benzene) to give crude compound 12 (60 mg). This
material was recrystallized from CHCl3-MeOH to give pure 12 (29
1
mg, 0.036 mmol, 28%): mp 237-239 °C (dec); H NMR (CDCl3) δ
6.5-7.3 (m, 42 H); 13C NMR (CDCl3) δ 92.2, 125.2, 126.1, 126.2,
126.3, 126.4, 126.9, 127.1, 127.4, 127.8, 129.7, 129.8, 130.2, 131.2,
131.3, 131.4, 132.2, 134.8, 135.5, 136.1, 139.0, 140.4, 140.8, 141.6,
147.6, 148.4 (26 resonances observed; 23 resonances are expected if
all phenyl rotations are fast, but 31 are expected if all phenyl rotations
are slow); MS (FAB) m/z 803 (M + H, 100), 698 (48).
General X-ray Crystallographic Procedures. X-ray data for
compounds 2m, 3, and 9 were collected at 200 K using graphite-
monochromated Mo KR radiation (λ ) 0.71073 Å) on a Nonius
KappaCCD diffractometer. The data for compound 2 were collected
at 100 K using radiation with λ ) 0.977 91 Å at the National
Synchrotron Light Source (beamline X-29). All diffraction data were
processed by using the program DENZO.27 All structures were solved
by direct methods using Siemens SHELXTL,28 and all were refined
by full-matrix least-squares on F2 using SHELXTL. All nonhydrogen
atoms were refined anisotropically, and hydrogens were included with
a riding model. The structure of compound 2m contained substantial
amounts of highly disordered solvent of crystallization; this was treated
by using the SQUEEZE/BYPASS procedure29 implemented in PLA-
TON.30 Specific crystal, reflection, and refinement data are contained
in Table 1.
(27) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307-326.
(28) Sheldrick, G. M. SHELXTL, Version 5; Siemens Analytical X-ray Instru-
ments: Madison, WI, 1996.
(29) Van der Sluis, P.; Spek, A. L. Acta Crystallogr., Sect. A 1990, 46, 194-
201.
(30) Spek, A. L. J. Appl. Crystallogr. 2003, 36, 7-13.
9
J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006 17049

A R T I C L E S
Lu et al.
Computational Studies. All semiempirical (AM131), ab initio [HF/
3-21G32], and hybrid density functional [B3LYP/6-31G(d)33,34] calcula-
tions were performed by using Gaussian 98;35 the built-in default
thresholds for wave function and gradient convergence were employed.
Transition states for conformational interconversions were located by
using the QST3 function in Gaussian 98. When comparing experimental
and calculated structures, the function OFIT in Siemens SHELXTL
was used to determine the best fit of the experimental and calculated
geometries and the deviations of the atomic positions.
Acknowledgment. This work was supported by National
Science Foundation Grants CHE-0314873 and CHE-0614879,
which are gratefully acknowledged.
Supporting Information Available: NMR spectra of com-
pounds 2, 2m, 3, 6, 8, 8m, 9, and 12; racemization data for 2;
complete ref 35; four crystallographic information files (CIF)
for compounds 2, 2m, 3, and 9; an ASCII text file containing
the coordinates of the calculated structures of 2 at various levels
of theory. This material is available free of charge via the
Internet at http://pubs.acs.org.
(31) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J. Am.
Chem. Soc. 1985, 107, 3902-3909.
(32) Hehre, W. J.; Radom, L.; Schleyer, P. v. R.; Pople, J. A. Ab Initio Molecular
Orbital Theory; John Wiley & Sons: New York, 1986; pp 63-100.
(33) Becke, A. D. J. Chem. Phys. 1993, 98, 5648-5652.
(34) Lee, C.; Yang, W.; Parr, R. G. Phys. ReV. B 1988, 37, 785-789.
(35) Frisch, M. J.; et al. Gaussian 98; Gaussian, Inc., Pittsburgh, PA, 1998.
JA065935F
9
17050 J. AM. CHEM. SOC. VOL. 128, NO. 51, 2006