Polyphenylbiphenyls and polyphenylfluorenes

Article abstract of DOI:10.1021/ja980322r

A series of highly congested polyphenylbiphenyls and polyarylfluorenes has been prepared and their X-ray structures determined. Decaphenylbiphenyl adopts a very unusual C₁-symmetric geometry (rather than the more intuitive D₂ geometr

Full text of DOI:10.1021/ja980322r

6000
J. Am. Chem. Soc. 1998, 120, 6000-6006
Polyphenylbiphenyls and Polyphenylfluorenes
Ling Tong, Heidi Lau, Douglas M. Ho, and Robert A. Pascal, Jr.*
Contribution from the Department of Chemistry, Princeton UniVersity, Princeton, New Jersey 08544
ReceiVed January 28, 1998
Abstract: A series of highly congested polyphenylbiphenyls and polyarylfluorenes has been prepared and
their X-ray structures determined. Decaphenylbiphenyl adopts a very unusual C1-symmetric geometry (rather
than the more intuitive D2 geometry) in which one of the central benzene rings is distorted into a boat
conformation. AM1 calculations confirm that the C1 geometry is the ground state but indicate that less highly
substituted biphenyls should adopt D2 geometries. The structure of 2,2′,4,4′,6,6′-hexaphenylbiphenyl supports
the latter prediction; this material has crystallographic C2 symmetry and (except for the orientation of the para
phenyl groups) approximate D2 symmetry in the solid state. Octaphenylfluorenone has been prepared in four
steps from tetraphenylcyclopentadienone. Its X-ray structure shows the fluorenone core to be twisted and
sterically shielded by the eight peripheral phenyl groups; nevertheless, phenylmagnesium bromide adds easily
to the carbonyl group of its equally hindered dimethyl derivative, 2,3,5,6,7,8-hexaphenyl-1,4-di(p-tolyl)fluo-
renone. Reduction of the resulting fluorenol with TiCl3 gives a nonaarylfluorene, 2,3,5,6,7,8,9-heptaphenyl-
1,4-di(p-tolyl)fluorene, and its X-ray structure shows distortions similar to those of octaphenylfluorenone.
Introduction
three are highly congested molecules in which the cores are
well shielded by the surrounding “picket fence” of phenyl
substituents. The crowding is most acute in 4, where the four
central ortho-phenyl groups must clash, and the molecule may
only partly relieve steric congestion by rotation about its central
carbon-carbon bond. Indeed, a CPK model of 4 is impossible
to construct, and, given the degree of steric conflict, it was
surprising to find that 4 had been prepared in 1965 in 60% yield
by heating together two commercially available starting materi-
als, tetraphenylcyclopentadienone (5) and 1,4-diphenyl-1,3-
butadiyne (6).5 The second step in this double Diels-Alder
reaction is sterically very demanding.6
Polyphenyl polycyclic aromatic hydrocarbons are remarkably
robust even when possessing highly unusual geometries. Deca-
phenylanthracene1 (1) and 9,10,11,12,13,14,15,16-octaphenyl-
dibenzo[a,c]naphthacene2 have strongly twisted acene cores, but
their many phenyl groups form protective hydrocarbon sheaths
about them which confer exceptional stability. Octaphenyl-
naphthalene1 is less distorted, but its stability and simple
synthesis permits the contruction of the albatrossenes,3,4 in which
multiple octaphenylnaphthalene subunits define large hydro-
phobic clefts. Other maximally arylated aromatic hydrocarbons,
if easily prepared, could also serve as building blocks for very
large organic structures. Among the obvious candidates are
octaphenylfluorenone (2), nonaphenylfluorene (3), and deca-
phenylbiphenyl (4).
In contrast with 4 and similar biphenyls, the “biphenyl” nuclei
of the fluorenes 2 and 3 are constrained to be more nearly planar,
although some degree of steric relief is provided by the
replacement of two of the strongly interacting ortho-phenyl
groups of 4 with the single linking atom (C-9) in a polyphenyl-
fluorene. However, unlike the polyphenylbiphenyls, of which
there are several examples, fluorenes with more than four phenyl
substituents appear to be unknown.7 Can such compounds be
unstable or unusually difficult to prepare?
Compounds 2-4 may all be considered derivatives of
biphenyl. Molecular mechanics calculations indicate that the
The peralkyl and peraryl derivatives of common hydrocarbons
and their ions are of fundamental interest to chemists: the
(1) 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.
(2) Qiao, X.; Ho, D. M.; Pascal, R. A., Jr. Angew. Chem., Int. Ed. Engl.
1997, 36, 1531-1532.
(3) Tong, L.; Ho, D. M.; Vogelaar, N. J.; Schutt, C. E.; Pascal, R. A.,
Jr. Tetrahedron Lett. 1997, 38, 7-10.
(4) Tong, L.; Ho, D. M.; Vogelaar, N. J.; Schutt, C. E.; Pascal, R. A.,
Jr. J. Am. Chem. Soc. 1997, 119, 7291-7302.
(5) Ogliaruso, M. A.; Becker, E. I. J. Org. Chem. 1965, 30, 3354-3360.
(6) Ogliaruso and Becker do not seem to have found this to be unusual;
although they recognized that decaphenylbiphenyl is a very crowded
molecule, its synthesis is reported almost without comment as “4k” in a
series of “bishexaphenyl-benzenes”!
(7) Reid, W.; Freitag, D. Chem. Ber. 1966, 99, 2675-2676.
S0002-7863(98)00322-9 CCC: $15.00 © 1998 American Chemical Society
Published on Web 05/29/1998

Polyphenylbiphenyls and Polyphenylfluorenes
J. Am. Chem. Soc., Vol. 120, No. 24, 1998 6001
boat conformation! Why should this C1 conformation be
preferred, or is the observed structure merely due to the influence
of crystal packing forces? A search of the Cambridge Structural
Database10 showed that another extremely crowded biphenyl,
dekakis(dichloromethyl)biphenyl,11 exists in a similar C1 con-
formation, suggesting that the unusual conformational preference
of 4 is not just an artifact of crystal packing. Therefore, a more
extensive computational study of 4 and related biphenyls was
carried out in order to better define their conformational
preferences.
All of the biphenyls examined bear phenyl substituents on
the four carbons ortho to the central bond (see Table 1). AM1
calculations12,13 indicated that such biphenyls display two
distinct conformational minimasthe D2 and C1 conformations
mentioned previously. These calculations indicate that when-
ever the four ortho-phenyls are buttressed by substituents meta
to the central bond, the C1 conformation is preferred, but when
these positions are occupied by hydrogens, the D2 conformation
is best, although by only a very small margin (less than one
kcal/mol; certainly too small to be the basis for a strong
prediction). Thus decaphenylbiphenyl (4), 2,2′,3,3′,5,5′,6,6′-
octaphenylbiphenyl (7), and 3,3′,4,4′,5,5′-hexachloro-2,2′,6,6′-
tetraphenylbiphenyl (8) should exhibit C1 structures, but
2,2′,4,4′,6,6′-hexaphenylbiphenyl (9) and 2,2′,6,6′-tetraphenyl-
biphenyl (10) should exhibit D2 structures. The geometry of
the simplest member of this series, compound 10, was also
examined by means of ab initio calculations at the HF/STO-
3G and HF/3-21G levels14 (Table 1). Once again, both the D2
and C1 conformations were found to be minima, with a strong
preference for the D2 geometry for 10 at the HF/STO-3G level
and a slim, but perhaps more reliable, margin of 0.55 kcal/mol
at the HF/3-21G level. (An ab initio calculation for 4 at this
level is far beyond our current computational resources, but the
excellent agreement between the AM1 calculations and the
experimental results is reassuring.)
To test whether a D2 conformation is indeed preferred in the
less crowded biphenyls, 2,2′,4,4′,6,6′-hexaphenylbiphenyl (9)
was prepared by a literature procedure,15 and its X-ray crystal
structure was determined. Compound 9 crystallizes in the
orthorhombic space group Pnna, and the molecule lies on a
special position and possesses crystallographic C2 symmetry.
Only in the orientation of the para phenyl groups (which play
no part in the steric conflict in these molecules) does 9 deviate
from D2 symmetry (see Figure 1). All of the aromatic rings of
9 are approximately planar; there are none of the severe
distortions observed in 4.
Figure 1. X-ray crystal structures of decaphenylbiphenyl (4, above)
and 2,2′,4,4′,6,6′-hexaphenylbiphenyl (9, below). Thermal ellipsoids
have been drawn at the 50% probability level.
structures and reactions of these species are important calibration
points for theoretical work. Given our own long-standing
interest in strained polycyclic aromatic hydrocarbons, the
polyphenylbiphenyls and polyphenylfluorenes were attractive
targets for synthetic and structural studies. We report herein
the X-ray crystal structures of decaphenylbiphenyl and
2,2′,4,4′,6,6′-hexaphenylbiphenyl (which have surprisingly dif-
ferent geometries), computational studies of these and several
related polyphenylbiphenyls, and the synthesis and structures
of octaphenylfluorenone and several derivatives of nona-
phenylfluorene.
Results and Discussion
Polyphenylbiphenyls. The preferred conformation of de-
caphenylbiphenyl (4) is not obvious, but the symmetry of its
planar drawing suggests that some D2 or D2d geometry might
be best. Ogliaruso and Becker’s synthesis of 4 was repeated
without difficulty, single crystals of the chloroform solvate of
4 were obtained, and X-ray data was collected. The structure
was solved and refined in the monoclinic space group P21/c,
and it is illustrated in Figure 1. The molecular conformation
of 4 is unexpectedly asymmetric! Molecular mechanics cal-
culations (SYBYL8,9) had indicated that a D2 geometry is the
conformation of lowest energy, but the experimentally observed
C1 conformation is a distinct potential minimum of higher
energy. There are no great distortions present in the calculated
D2 conformation, but in the observed C1 structure, one of the
central benzene rings is planar and the other is distorted into a
The structures of compounds 4 and 9 are best compared in
the stereoviews in Figure 2. In compound 9, the dihedral angle
between the mean planes of the central rings is only 65.5°, while
the corresponding rings in 4 are nearly perpendicular (86.5°).
In 9, both pairs of the interacting phenyl groups ortho to the
(10) Allen, F. H.; Kennard, O.; Taylor, R. Acc. Chem. Res. 1983, 16,
146-153.
(11) Biali, S. E.; Kahr, B.; Okamoto, Y.; Aburatani, R.; Mislow, K. J.
Am. Chem. Soc. 1988, 110, 1917-1922.
(12) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P. J.
Am. Chem. Soc. 1985, 107, 3902-3909.
(13) Semiempirical and ab initio molecular orbital calculations were
performed by using the SPARTAN program package (version 4.1; Wave-
function, Inc., Irvine, California, USA), and its built-in default thresholds
for wave function and gradient convergence were employed. Frequency
calculations were performed on the AM1-optimized equilibrium geometries
to verify that these were true potential minima.
(8) Molecular mechanics calculations were performed by using the
SYBYL9 force field implemented in the SPARTAN program package
(versions 3.0 and 4.1; Wavefunction, Inc., Irvine, California, U.S.A.).
(9) Clark, M.; Cramer, R. D., III.; Van Opdenbosch, N. J. Comput. Chem.
1989, 10, 982-1012.
(14) 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.
(15) Fujioka, Y.; Ozasa, S.; Sato, K.; Ibuki, E. Chem. Pharm. Bull. 1985,
33, 22-29.

6002 J. Am. Chem. Soc., Vol. 120, No. 24, 1998
Tong et al.
Table 1. Computational Data for Various Conformations of Polysubstituted Biphenyls
computational
levela
∆Hf (AM1, kcal/mol)
or E (HF, aub)
difference
central bond
length (Å)
substituents
symm
(C1 - D2, kcal/mol)c
X ) Y ) Ph (4)
AM1
AM1
AM1
AM1
AM1
AM1
AM1
AM1
AM1
AM1
HF/STO-3G
HF/STO-3G
HF/3-21G
HF/3-21G
C1
D2
C1
D2
C1
D2
C1
D2
C1
D2
C1
D2
C1
D2
354.67
356.67
288.39
289.59
144.90
147.57
223.83
223.24
168.38
-2.00
-1.20
-2.67
0.59
1.476
1.474
1.476
1.475
1.476
1.474
1.472
1.470
1.473
1.472
1.526
1.520
1.506
1.501
X ) Ph, Y ) H (7)
X ) Y ) Cl (8)
X ) H, Y ) Ph (9)
X ) Y ) H (10)
0.72
167.66
-1361.64463
-1361.65205
-1370.73715
-1370.73803
4.66
0.55
a See note 13 for some computational details. b 1 au ) 627.503 kcal/mol. c Negative values favor the C1 conformation.
experimental and computational data agree, and observed
preference must arise from small steric conflicts in the D2
structure exacerbated by the buttressing groups in the positions
meta to the central bond. It is noteworthy that the experimental
central bond length in 4 [1.501(4) Å] is a bit shorter than that
in the less crowded 9 [1.515(4) Å], which perhaps compensates
in part for the distortion of the benzene ring in 4. However, no
significant difference between the central bond lengths of the
pairs of D2 and C1 conformations of the various biphenyls is
observed in the computational studies (Table 1).
A more extensive search of the Cambridge Structural
Database found three additional examples of biphenyls with
geometries similar to those of 4: 2,2′,4,4′,6,6′-hexabromobi-
phenyl,16 2,4,6-trinitro-2′,4′,6′-tris(N-pyrrolidinyl)biphenyl,17 and
4,4′,6,6′-tetranitro-2,2′-diphenic acid.18 The first two possess
only small deviations from ideal D2 and C2 conformations,
respectively, but the third is quite similar to 4, although less
highly distorted. AM1 calculations show discrete C2 and C1
conformations for this molecule, with the observed C1 structure
disfaVored by 2.6 kcal/mol. However, the dimethyl ester of
4,4′,6,6′-tetranitro-2,2′-diphenic acid19 as well as 6,6′-dinitro-
2,2′-diphenic acid,20 display undistorted, more nearly symmetric
structures, so the observed preference for a higher energy C1
conformation by solid 4,4′,6,6′-tetranitro-2,2′-diphenic acid is
likely to be the result of packing forces.
Polyphenylfluorenes. Given that a high-temperature Diels-
Alder reaction sufficed for the synthesis of 4, we chose similar
reactions for key steps in the synthesis of octaphenylfluorenone
(2, Scheme 1). Cycloaddition of tetraphenylcyclopentadienone
(5) and 4-pentenoic acid in refluxing xylenes gave a highly
crystalline adduct, which was then aromatized with bromine to
give the (tetraphenylphenyl)propanoic acid 11 in 44% overall
Figure 2. Stereoviews of the X-ray structures of compounds 4 (above)
and 9 (below).
(16) Field, L. D.; Skelton, B. W.; Sternhell, S.; White, A. H. Aust. J.
Chem. 1985, 38, 391-399.
(17) Effenberger, F.; Agster, W.; Fischer, P.; Jogun, K. H.; Stezkowski,
J. J.; Daltrozzo, E.; Kollmansberger-von Nell, G. J. Org. Chem. 1983, 48,
4649-4658.
central bond are stacked face-to-face, but in 4, one pair of ortho
phenyls are face-to-face and the other pair adopts an edge-to-
face orientation. This, indeed, is the characteristic difference
between the D2 and C1 conformations in all of the calculated
biphenyls as well. However, it remains difficult to understand
why the D2 conformation of 4 is not the preferred geometry:
there is no single interaction in the calculated D2 structure of 4
which is the obvious source of its destabilization. Nevertheless,
(18) Popova, E. G.; Chetkina, L. A.; Bel’skii, V. K.; Andrievskii, A.
M.; Poplovskii, A. N.; Dyumaev, K. M. Zh. Struct. Khim. 1987, 28, 129-
132.
(19) Popova, E. G.; Chetkina, L. A.; Sobolev, A. N. Zh. Struct. Khim.
1991, 32, 130-133.
(20) Popova, E. G.; Chetkina, L. A.; Belk’skii, V. K.; Andrievskii, A.
M.; Poplavskii, A. N.; Dyumaev, K. M. Dokl. Akad. Nauk SSSR 1989, 304,
127.

Polyphenylbiphenyls and Polyphenylfluorenes
J. Am. Chem. Soc., Vol. 120, No. 24, 1998 6003
Scheme 1
The mean planes of all the phenyl groups are roughly perpen-
dicular to the mean plane of the fluorene, so the carbonyl group
is quite sheltered. Compound 2 proved to be profoundly
insoluble, a surprising property, given that all of our other
polyphenyl aromatics1-4,21 are quite soluble in common organic
solvents. However, the X-ray structure shows the crystals to
be especially close-packed (Figure 3), with many interlocking
phenyl-phenyl interactions.
Because of the low solubility, further transformations of this
material proved impossible. To continue the synthesis of a
nonaarylfluorene, the dimethyl derivative 14 was prepared by
Diels-Alder reaction of 13 and 2,3-diphenyl-1,4-di(p-tolyl)-
cyclopentadienone22 (39% yield). This material is quite soluble;
presumably the methyls break up the close packing observed
in the crystals of 2. Addition of phenylmagnesium bromide to
14 gave nonaarylfluorenol 15 (59% yield); the Grignard reagent
has little difficulty in adding to the hindered carbonyl group.
Finally, Ti(III) reduction of 15 gave the nonaarylfluorene 17
(61% yield). Interestingly, a standard procedure for reduction
of 9-phenyl-9-fluorenols is to treat them with hydrogen halides
in ethanol.23 We verified that HCl in ethanol reduces 9-phenyl-
9-fluorenol to the 9-phenylfluorene, as reported,23 but treatment
of 15 under the same conditions gave only the ethyl ether 16.
Crystals of the nonaarylfluorene 17 were obtained from
CH2Cl2-acetone, its structure was solved and refined in the
space group P1h, and it is illustrated in Figure 4. The newly
introduced C(9)-phenyl group is easily accommodated. Indeed,
the fluorene nucleus is less distorted than in 2, with a C(4)-
C(4a)-C(4b)-C(5) torsion angle of only 19.3°, since the C(9)-
phenyl need not lie directly between the C(1)-tolyl and C(8)-
phenyl groups. However, there remain several close contacts
between phenyl substituents: the distances between the ipso
carbons C(29) and C(36), C(10) and C(60), and C(54) and C(60)
are 3.09, 3.12, and 3.06 Å, respectively. The central “biphenyl”
bonds in both 2 [C(6)-C(6′)] and 17 [C(4a)-C(4b)] are of
normal length, 1.501(9) Å and 1.510(5) Å, respectively.
yield. Surprisingly, the seemingly simpler reaction of 5 with
4-pentynoic acid to give 11 in one step gave yields no higher
than 21% in several attempts. Cyclization of 11 to the indanone
12 was best accomplished by converting 11 to the acid chloride
followed by an intramolecular Friedel-Crafts acylation (28%
yield). Again, a one-step cyclization of 11 with hot polyphos-
phoric acid was much inferior (6% yield). NBS bromination
of 12 followed by an elimination gave indenone 13 as a brilliant
yellow solid (32% yield). The final Diels-Alder reaction of
13 and 5 was carried out in refluxing nitrobenzene to promote
decarbonylation and dehydrogenation of the initial adduct. The
yield of this reaction was low (11%), but single crystals of
octaphenylfluorenone (2), suitable for X-ray analysis, formed
upon cooling the reaction mixture, and additional 2 was
precipitated by addition of methanol.
X-ray diffraction data were collected by using one of the
crystals of 2 that formed upon cooling its synthetic reaction
mixture. The structure was solved and refined without difficulty
in the space group P2/c, and it is illustrated in Figure 3.
Compound 2 lies on a special position and possesses crystal-
lographic C2 symmetry. The structure is generally similar to
that of octaphenylcarbazole,21 but the fluorene core of 2 exhibits
a somewhat greater twisting distortion from planarity. The twist
is imparted by the very close interaction of the C(5) and C(5′)
phenyl groups; the ipso carbons of these phenyls [C(26) and
C(26′)] are only 3.12 Å apart, well within their van der Waal
radii, and the torsional angle C(5)-C(6)-C(6′)-C(5′) is 33.2°.
Finally, the nonaphenylfluorenyl cation might be an unusually
stable species due to steric shielding and delocalization of the
charge throughout the fluorene nucleus, and this cation is a very
likely intermediate in the transformations of 15 into 16 and 17.
However, attempts to prepare the cation by dissolving fluorenol
15 in D2SO4 yielded a deep blue solution which gave only a
very highly broadened 13C NMR spectrum (unlike the sharp
spectrum of the triphenylmethyl cation24 recorded under the
same conditions). Further attempts to obtain a crystalline
nonaarylfluorenyl cation for X-ray analysis by treatment of 15
with acetic anhydride and perchloric acid25 were also unsuc-
cessful. In retrospect, it may be seen that the cation is more
highly strained than the alcohol 15, the ether 16, or the
hydrocarbon 17, because sp3 to sp2 rehybridization at C(9) to
form the cation will require the attached phenyl group to slide
between the C(1) and C(8) phenyls, forcing them apart (see the
space-filling view of 17 in Figure 4). By the same argument,
the addition of any nucleophile to the cation to reform an sp3
carbon will be favorable, perhaps enough so to prevent its easy
isolation.
(22) Mehr, L.; Becker, E. I.; Spoerri, P. E. J. Am. Chem. Soc. 1955, 77,
986-989.
(23) Kohler, E. P.; Blanchard, L. W., Jr. J. Am. Chem. Soc. 1935, 57,
367-371.
(24) Olah, G. A.; Baker, E. B.; Comisarow, M. B. J. Am. Chem. Soc.
1964, 86, 1265.
(21) Qiao, X.; Ho, D. M.; Pascal, R. A., Jr. J. Org. Chem. 1996, 61,
6748-6750.
(25) Chance, J. M.; Geiger, J. H.; Okamoto, Y.; Aburatani, R.; Mislow,
K. J. Am. Chem. Soc. 1990, 112, 3540-3547.

6004 J. Am. Chem. Soc., Vol. 120, No. 24, 1998
Tong et al.
Figure 3. X-ray structure of octaphenylfluorenone (2). Thermal ellipsoids have been drawn at the 50% probability level. The close packing of 2
is illustrated (right side).
Figure 4. X-ray structure of 2,3,5,6,7,8-heptaphenyl-1,4-di(p-tolyl)fluorene (17). Thermal ellipsoids have been drawn at the 50% probability level.
346-348 °C (lit.15 345 °C, lit.23 348 °C). Crystals suitable for X-ray
analysis were obtained by the slow evaporation of a solution in benzene-
Conclusion
The unique conformational dichotomy observed for the
polyphenyl-biphenyls once again demonstrates that surprises
await the unwary even for molecules having simple, “ordinary-
looking” structures on paper, and that careful searches for all
possible conformations must be made in computational studies
if one is to accurately predict molecular structures. The
polyphenylfluorenes held no such surprises, but their large size
(C60-C70), well-defined conformations, short and versatile
syntheses, and great stability indicate that they, like octaphe-
nylnaphthalene, will be suitable subunits for the assembly of
very large organic structures.
ethanol.
X-ray Crystallographic Analysis of Decaphenylbiphenyl (4).
Formula C72H50‚CHCl3; triclinic, space group P1h, a ) 11.073 (1) Å, b
) 12.096 (1) Å, c ) 23.303 (3) Å, R ) 88.649 (8)°, â ) 88.657 (10)°,
γ ) 64.251 (8)°, V ) 2810.2 (6) Å3, Z ) 2, Dcalcd ) 1.223 g/cm3.
Intensity measurements were made at 298 K with 3° e 2θ e 50° on
a Siemens P4 diffractometer using Mo KR radiation (λ ) 0.71073 Å)
and a crystal with dimensions of 0.05 mm × 0.30 mm × 0.50 mm. A
total of 10 184 reflections were measured of which 9645 were unique
(Rint ) 0.036). The structure was solved by direct methods (SHELX-
TL26) and refined by full-matrix least-squares on F2 (SHELXL-9327).
All non-hydrogen atoms were refined with anisotropic displacement
coefficients, and hydrogen atoms were included with a riding model
and isotropic displacement coefficients [U(H) ) 1.2U(C)]. The
chloroform in the lattice was disordered, and it was included in the
refinement with a two-site model with distance and similarity restraints
and a site occupancy parameter. The refinement converged to R(F) )
0.050, wR(F2) ) 0.083, and S ) 1.04 for 3413 data with I > 2σ(I),
Experimental Section
Decaphenylbiphenyl (4,2′,3′,5′,6′,2′′,3′′,5′′,6′′-octaphenyl-p-quater-
phenyl) was prepared by the method of Ogliaruso and Becker;5 mp
225-227 °C (lit.5 222-224 °C). Crystals suitable for X-ray analysis
were obtained by the slow evaporation of a solution in chloroform-
ethanol.
(26) Sheldrick, G. M. SHELXTL, Version 4.2. Siemens Analytical X-ray
Instruments, Madison, WI, 1991.
(27) Sheldrick, G. M. SHELXL-93. Program for the Refinement of Crystal
Structures; University of Gottingen, Germany, 1993.
2,2′,4,4′,6,6′-Hexaphenylbiphenyl (9, 3′,5′,2′′,6′′-tetraphenyl-p-
quaterphenyl) was prepared by the method of Fujioka et al.;15 mp

Polyphenylbiphenyls and Polyphenylfluorenes
J. Am. Chem. Soc., Vol. 120, No. 24, 1998 6005
and R(F) ) 0.170, wR(F2) ) 0.108, and S ) 0.74 for 9645 data, 723
variables, and 24 restraints. Full details are provided in the Supporting
Information.
Octaphenylfluorenone (2). A solution of compounds 13 (26 mg,
0.060 mmol) and 5 (50 mg, 0.13 mmol) in nitrobenzene (0.5 mL) was
heated to 210 °C for 40 h. The reaction mixture was allowed to cool
slowly to room temperature, methanol (2 mL) was added, and the
reaction mixture was left in the refrigerator overnight. The resulting
bright yellow crystals of compound 2 (5 mg, 11%) were collected by
filtration. These crystals are remarkably insoluble, but the larger ones
were suitable for X-ray analysis. Mp > 400 °C; MS, m/z 788 (M+,
100), 711 (M - C6H5, 30); exact mass 788.3049, calcd for C61H40O
788.3079.
X-ray Crystallographic Analysis of Octaphenylfluorenone (2).
Formula C61H40O; monoclinic, space group P21/c; a ) 15.031 (2) Å,
b ) 11.437 (2) Å, c ) 12.326 (1) Å, â ) 96.702 (9)°, V ) 2104.4 (5)
Å3, Z ) 2, Dcalcd ) 1.245 g/cm3. Intensity measurements were made
at 223 K with 4° e 2θ e 50° on a Siemens P4 diffractometer using
Mo KR radiation (λ ) 0.71073 Å) and a crystal with dimensions of
0.10 mm × 0.18 mm × 0.25 mm. A total of 3868 reflections were
measured of which 3720 were unique (Rint ) 0.065). The structure
was solved by direct methods (SHELXTL) and refined by full-matrix
least-squares on F2 (SHELXL-93). All non-hydrogen atoms were
refined with anisotropic displacement coefficients, and hydrogen atoms
were included with a riding model and isotropic displacement coef-
ficients [U(H) ) 1.2U(C)]. The refinement converged to R(F) ) 0.078,
wR(F2) ) 0.197, and S ) 1.29 for 1651 data with I > 2σ(I), and R(F)
) 0.164, wR(F2) ) 0.308, and S ) 0.96 for 3713 data and 281 variables
(7 reflections were suppressed). Full details are provided in the
Supporting Information.
2,3,5,6,7,8-Hexaphenyl-1,4-di(p-tolyl)fluorenone (14). A solution
of compound 13 (204 mg, 0.470 mmol) and 2,3-diphenyl-1,4-di(p-
tolyl)cyclopentadienone22 (387 mg, 0.94 mmol) in nitrobenzene (2 mL)
was heated to 210 °C for 48 h. After cooling, methanol (8 mL) was
added slowly, and the resulting solution was left in the refrigerator
overnight to yield bright yellow crystals of compound 14 (150 mg,
0.184 mmol, 39%); mp > 350 °C. 1H NMR (CDCl3) δ 2.09 (s, 3 H),
2.18 (s, 3 H), 6.14 (d, J ) 7 Hz, 2 H), 6.26 (d, J ) 7 Hz, 2 H), 6.39
(d, J ) 8 Hz, 2 H), 6.44 (m, 4 H), 6.63 (m, 6 H), 6.70-6.81 (m, 14
H), 6.84 (d, J ) 8 Hz, 2 H), 6.91 (d, J ) 8 Hz, 2 H), 7.04 (s, 4 H);
MS, m/z 816 (M+, 100), 739 (M - C6H5, 22); exact mass 816.3399,
calcd for C63H44O 816.3392.
X-ray Crystallographic Analysis of 2,2′,4,4′,6,6′-Hexaphenyl-
biphenyl (9). Formula C48H34; orthorhombic, space group Pnna; a )
20.912 (2) Å, b ) 16.579 (1) Å, c ) 9.737 (1) Å, V ) 3375.7 (4) Å3,
Z ) 4, Dcalcd ) 1.202 g/cm3. Intensity measurements were made at
230 K with 4° e 2θ e 55° on a Siemens P4 diffractometer using Mo
KR radiation (λ ) 0.71073 Å) and a crystal with dimensions of 0.08
mm × 0.35 mm × 0.38 mm. A total of 4881 reflections were measured
of which 3889 were unique (Rint ) 0.026). The structure was solved
by direct methods (SHELXTL) and refined by full-matrix least-squares
on F2 (SHELXL-93). The positional and thermal parameters for all
atoms were refined; the hydrogens were refined isotropically, and the
carbons were refined anisotropically. The refinement converged to R(F)
) 0.052, wR(F2) ) 0.112, and S ) 1.09 for 1691 data with I > 2σ(I),
and R(F) ) 0.131, wR(F2) ) 0.137, and S ) 0.83 for 3888 data and
289 variables (one reflection was suppressed). Full details are provided
in the Supporting Information.
3-(2,3,4,5-Tetraphenylphenyl)propanoic acid (11). 4-Pentenoic
acid (2.65 g, 26.5 mmol) and tetraphenylcyclopentadienone (5, 11.2 g,
29 mmol) were heated in refluxing xylenes (30 mL) for 2 days. The
product was chromatographed on a column of silica gel eluted
successively with toluene, 1:2 toluene-ethyl acetate, and 20:1 toluene-
methanol. The fractions containing 3-[(2,3,4,5-tetraphenyl-1,6-dihy-
dro)phenyl]propanoic acid were combined, concentrated to dryness, and
recrystallized from ethanol. This dihydro acid (5.61 g, 12.3 mmol)
was heated in benzene (150 mL) in a 1 L round-bottom flask. After
the compound had completely dissolved, the solution was cooled to
room temperature. A solution of bromine (1.97 g, 12.3 mmol) in
benzene (50 mL) was added dropwise. The resulting solution was
heated at reflux overnight. Cooling and concentration gave compound
11 (5.26 g, 11.6 mmol, 44%), mp 145-150 °C. 1H NMR (CDCl3) δ
2.55 (t, J ) 7 Hz, 2 H), 2.89 (t, J ) 7 Hz, 2 H), 6.74-7.19 (m, 20 H),
7.42 (s, 1 H); MS m/z 454 (M+, 47), 436 (M - H2O, 63), 393 (22),
358 (18), 315 (38), 157 (27), 91 (100); exact mass 454.1911, calcd for
C33H26O2 454.1933. Compound 11 was also prepared by heating
4-pentynoic acid and 5 in xylenes, but the yield was only 21%.
4,5,6,7-Tetraphenyl-1-indanone (12). Compound 11 (2.78 g, 6.12
mmol), thionyl chloride (15 mL), and 1,1,1-trichloroethane (100 mL)
were heated at reflux for 4 h. Unreacted thionyl chloride was largely
removed by distillation of the reaction mixture to one-third of its original
volume. After cooling, 1,1,1-trichloroethane (65 mL) and AlCl3 (1.63
g, 12.2 mmol) were added, and the mixture was stirred at room-
temperature overnight. The solution was heated at reflux for 1 h,
allowed to cool, and poured into a mixture of 1 N HCl (100 mL) and
ice (200 g). After stirring for 1 h, the organic phase was separated
and washed 5 times with water. The organic layer was dried over
Na2SO4 and concentrated to dryness. This material was chromato-
graphed on a silica gel column eluted successively with toluene and
99:1 toluene-ethyl acetate. Concentration of the appropriate fractions
gave compound 12 (0.760 g, 1.74 mmol, 28%), mp 171-173 °C. 1H
NMR (CDCl3) δ 2.66 (t, J ) 6 Hz, 2 H), 2.93 (t, J ) 6 Hz, 2 H),
6.72-7.20 (m, 20 H); MS m/z 436 (M+, 4), 372 (10), 129 (28), 105
(100); exact mass 436.1834, calcd for C33H24O 436.1827. The direct
cyclization of 11 to 12 was also accomplished by heating 11 in
polyphosphoric acid at 190 °C for 2 days, but the yield was only 6%.
9-Hydroxy-2,3,5,6,7,8,9-heptaphenyl-1,4-di(p-tolyl)fluorene (15).
A solution of compound 9 (50 mg, 0.061 mmol) in toluene (15 mL)
was chilled in an ice bath. A solution of phenylmagnesium bromide
(3 mL, 3 M in ether, 9 mmol) was added, and the resulting mixture
was heated at reflux overnight. The reaction mixture was washed with
water, dried over Na2SO4, and concentrated to dryness. The residue
was fractionated by preparative TLC (silica gel GF; 3:1 toluene-
hexanes) to yield compound 15 (32 mg, 0.036 mmol, 59%); mp 312.5-
313.5 °C. 1H NMR (CDCl3) δ 2.12 (s, 3 H), 2.14 (s, 3 H), 2.74 (br s,
1 H), 5.74 (d, J ) 8 Hz, 1 H), 5.86 (d, J ) 8 Hz, 1 H), 6.29 (m, 3 H),
6.37-6.54 (m, 10 H), 6.60-6.88 (m, 24 H), 7.02 (t, J ) 7 Hz, 1 H),
7.08 (m, 2 H), 7.24 (d, J ) 8 Hz, 1 H); MS, m/z 894 (M+, 69), 876 (M
- H2O, 100), 817 (M - C6H5, 33), 799 (M - H2O - C6H5, 31).
9-Ethoxy-2,3,5,6,7,8,9-heptaphenyl-1,4-di(p-tolyl)fluorene (16).
Compound 15 (32 mg, 0.036 mmol) in ethanol (5 mL) was heated to
boiling in a Pyrex screw-capped tube. Concentrated HCl (2 mL) was
added, the tube was sealed, and it was heated at 110 °C overnight.
After cooling, CH2Cl2 was added, and the mixture was washed with
dilute NaOH, dried over Na2SO4, and concentrated to dryness. The
residue was fractionated by preparative TLC (silica gel GF; 1:9
toluene-hexanes) to yield compound 16 (25 mg, 0.027 mmol, 76%);
mp 305.5-307.5 °C. 1H NMR (CDCl3) δ 1.28 (t, J ) 7 Hz, 3H), 2.07
(s, 3H), 2.14 (s, 3 H), 3.67 (m, 2 H), 5.60 (d, J ) 8 Hz, 1 H), 5.73 (d,
J ) 8 Hz, 1 H), 6.18 (m, 3 H), 6.29 (m, 2 H), 6.39 (m, 3 H), 6.47 (m,
4 H), 6.52 (d, J ) 7 Hz, 2 H), 6.59 (m, 2 H), 6.68 (m, 9 H), 6.75 (m,
11 H), 6.94 (m, 6 H), 7.12 (t, J ) 7 Hz, 1 H); MS, m/z 922 (M+, 100),
877 (M - OEt, 86), 799 (M - EtOH - C6H5, 54); exact mass 922.4179,
calcd for C71H54O 922.4177.
4,5,6,7-Tetraphenylindenone (13). Compound 12 (0.76 g, 1.74
mmol), N-bromosuccinimide (0.34 g, 1.9 mmol), and carbon tetrachlo-
ride (50 mL) were heated at reflux for 2 h while under illumination by
a 250 W tungsten lamp. Triethylamine (5 mL) was then added, and
heating was continued overnight. After cooling, the solution was
washed three times with 1 N HCl; the organic layer was dried over
Na2SO4 and concentrated to give a light brown solid. This material
was chromatographed on a silica gel column eluted with 1:2 hexanes-
toluene, and compound 13 was recovered as a bright yellow solid (0.24
g, 0.55 mmol, 32%), mp 229-230 °C. 1H NMR (CDCl3) δ 5.86 (d,
J ) 6 Hz, 1 H), 6.71-7.23 (m, 20 H), 7.44 (d, J ) 6 Hz, 1 H); MS,
m/z 434 (M+, 100), 357 (M - C6H5, 30); exact mass 434.1692, calcd
for C33H22O 434.1671.
2,3,5,6,7,8,9-Heptaphenyl-1,4-di(p-tolyl)fluorene (17). Compound
15 (25 mg, 0.028 mmol) was dissolved in THF (1 mL), and TiCl3 (0.5
g) was dissolved in ethanol (2 mL) with brief heating. The solutions
were combined in a Pyrex screw-capped tube, and this was heated at

6006 J. Am. Chem. Soc., Vol. 120, No. 24, 1998
Tong et al.
110 °C for 4 h. After the mixture cooled, water (1 mL) was added,
and the mixture was poured into toluene and 1 N NaOH. The organic
layer was separated, dried over Na2SO4, and concentrated to give pure
compound 17 (15 mg, 0.017 mmol, 61%); mp 293.5-295 °C. 1H NMR
(CDCl3) δ 2.10 (s, 3 H), 2.11 (s, 3 H), 5.16 (s, 1 H), 5.78 (d, J ) 8 Hz,
1 H), 5.91 (d, J ) 8 Hz, 1 H), 5.97 (br s, 2 H), 6.15 (d, J ) 7 Hz, 1
H), 6.30 (m, 3 H), 6.36 (d, J ) 8 Hz, 1 H), 6.41 (m, 4 H), 6.47 (t, J
) 7 Hz, 1 H), 6.61 (m, 8 H), 6.71 (m, 10 H), 6.75 (m, 6 H), 6.86 (t,
J ) 7 Hz, 1 H), 6.96 (d, J ) 7 Hz, 1 H), 7.18 (m, 2 H), 7.30 (d, J )
8 Hz, 1 H); 13C NMR (CDCl3) δ 21.3, 21.4, 55.1, 124.7, 124.9, 125.0,
125.1, 125.6, 126.2, 126.4, 126.6, 126.7, 126.8, 127.0, 127.3, 127.85,
127.89, 128.0, 128.5, 129.2, 129.5, 130.2, 130.5, 130.7, 131.0, 131.5,
131.7, 131.8, 133.7, 135.0, 136.50, 136.53, 136.7, 137.7, 138.0, 138.2,
138.8, 139.0, 139.8, 140.5, 140.58, 140.64, 140.7, 140.8, 140.9, 141.0,
141.3, 148.4, 148.6 (49 of 51 expected resonances if there is free aryl
group rotation); MS, m/z 878 (M+, 100), 801 (M - C6H5, 22), 787 (M
- C6H4CH3, 5), 723 (M - C6H5 - C6H6, 11), 709 (M - C6H4CH3 -
C6H6, 8); exact mass, 878.3911, calcd for C69H50 878.3912. Crystals
suitable for X-ray analysis were obtained by the slow evaporation of a
solution in CH2Cl2-acetone.
with discrete atoms. Therefore, the SQUEEZE/BYPASS28 procedure
implemented in PLATON-9629 was used to account for the solvent
electron density. A total electron count of 52.7 e in a total volume of
354.7 Å3 was found in a single potential solvent area, consistent with
the presence of 1.6 acetones (32 e) per unit cell (52.7/32 ) 1.6) and
thus a formulation of 1:0.8 17:C3H6O ratio. NMR analysis indicated
the presence of 0.5 molecules of acetone per molecule of 17. The
SQUEEZE-processed data was used for all subsequent cycles of
refinement. All non-hydrogen atoms were refined with anisotropic
displacement coefficients, and hydrogen atoms were included with a
riding model and isotropic displacement coefficients [U(H) ) 1.2U(C)
or 1.5U(Cmethyl)]. The refinement converged to R(F) ) 0.051, wR(F2)
) 0.099, and S ) 1.08 for 2316 data with I > 2σ(I), and R(F) ) 0.159,
wR(F2) ) 0.129, and S ) 0.68 for 7109 data and 624 variables (three
reflections were suppressed). Full details are provided in the Supporting
Information.
Acknowledgment. This work was supported by NSF Grant
CHE-9707958, which is gratefully acknowledged.
X-ray Crystallographic Analysis of 2,3,5,6,7,8,9-Heptaphenyl-
1,4-di(p-tolyl)fluorene (17). Formula C69H50‚0.5C3H6O; triclinic, space
group P1h, a ) 13.771 (1) Å, b ) 14.170 (1) Å, c ) 14.233 (2) Å, R
) 83.891 (9)°, â ) 83.790 (10)°, γ ) 81.373 (7)°, V ) 2718.5 (5) Å3,
Z ) 2, Dcalcd ) 1.109 g/cm3. Intensity measurements were made at
298 K with 3° e 2θ e 45° on a Siemens P4 diffractometer using Mo
KR radiation (λ ) 0.71073 Å) and a crystal with dimensions of 0.20
mm × 0.20 mm × 0.25 mm. A total of 7473 reflections were measured
of which 7112 were unique (Rint ) 0.032). The structure was solved
by direct methods (SHELXTL) and refined by full-matrix least-squares
on F2 (SHELXL-93). The presence of a disordered solvent molecule
in the lattice was evident, but this electron density was not well modeled
Supporting Information Available: Crystal structure re-
ports for compounds 2, 4, 9, and 17, which include full
experimental details, tables of atomic coordinates, bond dis-
tances, bond angles, and thermal parameters, and selected figures
(93 pages, print/PDF). See any current masthead page for
ordering information and Web access instructions.
JA980322R
(28) Van der Sluis, P.; Spek, A. L. Acta Crystallogr., Sect. A 1990, 46,
194-201.
(29) Spek, A. L. Acta Crystallogr., Sect. A 1990, 46, C34.