Synthesis of Polyphenylene Dendrimers Related to "Cubic Graphite"

Article abstract of DOI:10.1021/ja030675c

Four large, 6-fold symmetric, polyphenylene hydrocarbons have been prepared by short syntheses that chiefly employed alkyne trimerization, palladium-catalyzed coupling, and Diels-Alder reactions. The two largest of these molecules, hexakis[4′-(pentaphenyl

Full text of DOI:10.1021/ja030675c

Published on Web 04/16/2004
Synthesis of Polyphenylene Dendrimers Related to “Cubic
Graphite”
Xianfeng Shen, Douglas M. Ho, and Robert A. Pascal, Jr.*
Contribution from the Department of Chemistry, Princeton UniVersity,
Princeton, New Jersey 08544
Received December 22, 2003; Revised Manuscript Received March 1, 2004; E-mail: snake@chemvax.princeton.edu
Abstract: Four large, 6-fold symmetric, polyphenylene hydrocarbons have been prepared by short syntheses
that chiefly employed alkyne trimerization, palladium-catalyzed coupling, and Diels-Alder reactions. The
two largest of these molecules, hexakis[4′-(pentaphenylphenyl)biphenyl-4-yl]benzene (4, C294H198) and
hexakis[4′-(2,3,4,5-tetraphenylphenyl)biphenyl-4-yl]benzene (5, C258H174) are substructures of “phenylogous
cubic graphite”, and the other two, hexakis(2′,3′,4′,5′,6′-pentaphenylbiphenyl-4-yl)benzene (26, C258H174
)
and hexakis(2′,3′,4′,5′-tetraphenylbiphenyl-4-yl)benzene (25, C222H150) are strongly pitched, six-bladed
molecular propellers. The X-ray crystal structure of compound 26 has also been determined; dendrimer 26
is at present the largest crystallographically characterized hydrocarbon.
Polyphenylene nanostructures2,5 are of great current interest,
Introduction
and substructures of phenylogous cubic graphite (2) are attractive
examples this class of compounds. We recently reported the
synthesis and X-ray structure of compound 3 (C168H112),4 a
macrocyclic substructure of 2, and in this article we describe
the synthesis of two larger, dendrimeric substructures of 2,
compounds 4 (C294H198) and 5 (C258H174), and several related,
very large, polyphenylene hydrocarbons (see Schemes 1-3).
“Cubic graphite” (1) is a hypothetical carbon allotrope first
proposed by Gibson et al. in 1946.1 Composed entirely of
benzene rings, with each benzene ring attached to six other,
different benzene rings, cubic graphite is the ultimate polyphe-
nylene structure,2 but the prospects for its synthesis are poor
because of both thermodynamic and kinetic considerations.3,4
However, the homolog, or phenylog, of cubic graphite 2,4
formed from hexaphenylbenzene subunits linked in the same
fashion as the benzene rings in cubic graphite, is a better
candidate for synthesis because the large cavities in this structure
and its larger precursors would permit the easy passage of
solvents and reagents.
Results and Discussion
The polyphenylene dendrimer 4 possesses ideal D6 symmetry,
and thus it has a central benzene ring with six large, but
identical, substituents. The trimerization of the bis(polyaryl)-
9
5798
J. AM. CHEM. SOC. 2004, 126, 5798-5805
10.1021/ja030675c CCC: $27.50 © 2004 American Chemical Society

Polyphenylene Dendrimers Related to Cubic Graphite
A R T I C L E S
Scheme 1 a
Scheme 2 a
a
(a) 4-TMSC6H4B(OH)2, (Ph3P)4Pd, K2CO3, toluene, EtOH, 50 °C, 83%.
(b) Hg[Co(CO)4]2, dioxane, 100 °C, 81%. (c) ICl, CH2Cl2, 93%. (d)
TMSCCH, (Ph3P)2PdCl2, CuI, Et3N, toluene, 60 °C, then TBAF, THF, room
temperature, 91%. (e) PhCCH, (Ph3P)2PdCl2, CuI, Et3N, toluene, 50 °C,
84%. (f) Tetracyclone, Ph2O, 280 °C, 60% for 5, 80% for 4.
a (a) PhCCH, (Ph3P)2PdCl2, CuI, piperidine, 50 °C, 62%. (b) Tetracy-
clone, Ph2O, 280 °C, 91%. (c) TMSCCH, (Ph3P)2PdCl2, CuI, Et3N, toluene,
50 °C, 96%. (d) TBAF, THF, room temperature, 94%. (e) 2-Methyl-3-
butyn-2-ol, (Ph3P)2PdCl2, CuI, Et3N, toluene, 80 °C, 76%. (f) 8, (Ph3P)4Pd,
CuI, Et3N, toluene, 50 °C, 47%. (g) 8, (Ph3P)2PdCl2, CuI, NaOH, toluene,
Bu4NI, 80 °C, 22%.
was employed along with triphenylarsine as an accelerating
ligand,7 but 13 remained the major product. The usual base,
triethylamine, was replaced with n-butylamine or Hunig’s base,
but without success. Finally, the method of Chow et al.,8 in
which a masked acetylene (the acetone adduct 11) is used to
generate low concentrations of the free alkyne 10 in situ, enabled
the synthesis of compound 12 without the formation of dimer
13. The 22% yield in this reaction was low, but attempts to
optimize the yield were abandoned when a more serious problem
was discovered: the insolubility of the alkyne 12. The vast
majority of polyphenyl aromatic compounds are readily soluble
in aromatic and halogenated organic solvents, but compound
12 proved to be quite insoluble in all common solvents.
Nevertheless, the synthesis of dendrimer 4 was attempted by
heating a suspension of 12 in refluxing dioxane in the presence
of the trimerization catalyst Hg[Co(CO)4]2,9 but only starting
material was recovered.
A new synthesis of 4 was devised in which the trimerization
of a freely soluble alkyne is carried out at an earlier stage
(Scheme 2). Bis(4-bromophenyl)acetylene10 (14) was subjected
to a double Suzuki coupling11 with 4-(trimethylsilyl)benzenebo-
ronic acid to give the alkyne 15 in 83% yield, and trimerization
of this material yielded the starlike polyphenylene 16 in 81%
yield. It was now necessary to alter six functional groups in
each of three consecutive reactions to obtain 4, but this
acetylene 12 (Scheme 1) should be a very efficient way to
prepare such a molecule, because there is no question of the
formation of regioisomers in such a reaction.
A synthesis of the precursor 12 is simple enough on paper,
and indeed only the last step was problematic in practice.
Sonogashira coupling6 of 4,4′-diiodobiphenyl (6) with 1 equiv
of phenylacetylene gave alkyne 7 in 62% yield, and its Diels-
Alder reaction with tetracyclone gave the aryl iodide 8 in 91%
yield. A second Sonogashira reaction of 8 with trimethylsil-
ylacetylene gave 9, and treatment of this material with tetrabu-
tylammonium fluoride furnished the corresponding monosub-
stituted acetylene 10 in 90% yield from 8. Unfortunately, the
third Sonogashira reaction in this sequence was not successful:
the coupling of aryl iodide 8 with alkyne 10 gave primarily the
bis(polyaryl)butadiyne 13 (the dimer of 10) instead of the desired
12.
Various conditions were examined for the ability to suppress
the undesired dimerization. The more reactive Pd2(dba)3 catalyst
(1) Gibson, J.; Holohan, M.; Riley, H. L. J. Chem. Soc. 1946, 456-461.
(2) Berresheim, A. J.; Mu¨ller, M.; Mu¨llen, K. Chem. ReV. 1999, 99, 1747-
1785.
(3) (a) O’Keeffe, M.; Adams, G. B.; Sankey, O. F. Phys. ReV. Lett. 1992, 68,
2325-2328. (b) Baughman, R. H.; Cui, C. Synth. Met. 1993, 55, 315-
320.
(7) Wagner, R. W.; Johnson, T. E.; Lindsey, J. S. J. Am. Chem. Soc. 1996,
118, 11166-11180.
(4) Shen, X.; Ho, D. M.; Pascal, R. A., Jr. Org. Lett. 2003, 5, 369-371.
(5) Watson, M. D.; Fechtenko¨tter, A.; Mu¨llen, K. Chem. ReV. 2001, 101, 1267-
1300.
(8) Chow, H.-F.; Wan, C.-W.; Low, K.-H.; Yeung, Y.-Y. J. Org. Chem. 2001,
66, 1910-1913.
(9) Dighe, S. V.; Orchin, M. Inorg. Chem. 1962, 1, 965-966.
(10) Barber, H. J.; Slack, R. J. Chem. Soc. 1944, 612-615.
(11) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457-2483.
(6) Sonogashira, K. ComprehensiVe Organic Synthesis; Trost, B. M., Fleming,
I., Eds.; Pergamon: New York, 1991; Vol. 3, pp 521-549.
9
J. AM. CHEM. SOC. VOL. 126, NO. 18, 2004 5799

A R T I C L E S
Shen et al.
Scheme 3 a
conditions (280 °C). Nevertheless, 5 was prepared in 34%
overall yield from 14.
The success of the various 6-fold functional group modifica-
tions encouraged us to try a similar strategy for the synthesis
of a more crowded set of polyphenylene dendrimers. Whereas
compound 4 is composed of six hexaphenylbenzene subunits
attached to a central hexaphenylbenzene, compound 26 (Scheme
3) is made of six hexaphenylbenzenes attached to a central
benzene ring. As such, 26 is not a substructure of phenylogous
cubic graphite, and it is much more crowded than 4. Mu¨llen
and co-workers have previously prepared the closely related
compound 25,14 which lacks six of the phenyl groups of 26,
but even so, the very fine X-ray structure of 25 shows it to be
a strongly pitched molecular propeller.14c
The synthesis of 26 began with hexakis(4-iodophenyl)-
benzene15 (20, Scheme 3). The iodines were converted to
phenylalkynyl groups via a 6-fold Sonogashira reaction with
phenylacetylene to give 24 (80% yield), and a Diels-Alder
reaction with tetracyclone gave 26 (81%). In a similar manner,
we prepared Mu¨llen’s compound 25 by elaboration of 20 to
the terminal hexayne (75%), followed by Diels-Alder reaction
with tetracyclone (85%). This synthesis is perhaps more succinct
than Mu¨llen’s, but there is no real synthetic advantage either
way.
Finally, compounds 16 and 19 were prepared by Suzuki
coupling reactions of hexaiodide 20 with 4-(trimethylsilyl)-
benzeneboronic acid and the tolaneboronic acid 22, respectively,
thus providing alternative routes to dendrimers 4 and 5.
However, the three-step synthesis of 20 from 14 proceeds in
only 48% overall yield,15 and thus the overall syntheses of 4
and 5 in this manner are not quite as efficient as those illustrated
in Scheme 2.
The two- or three-step elaborations of the D6 symmetric core
molecules 16 and 20 gave the target polyphenylene dendrimers
4, 5, 25, and 26 in good yields, despite the fact that the
modification of six functional groups was required in each of
these steps. Given this success, it is easy to imagine even shorter,
convergent syntheses of these molecules by attaching preformed
penta- or hexaphenylbenzenes to a D6 core molecule. For
example, a Suzuki reaction of 20 with 6 equiv of the (hexaphe-
nylbenzene)boronic acid 27 should yield compound 4 directly.
a
(a) 4-TMSC6H4B(OH)2, (Ph3P)4Pd, K2CO3, toluene, EtOH, 80 °C, 64%.
(b) 22, (Ph3P)4Pd, K2CO3, toluene, EtOH, 80 °C, 25%. (c) TMSCCH,
(Ph3P)2PdCl2, CuI, Et3N, toluene, 60 °C, then TBAF, THF, room temper-
ature, 75%. (d) PhCCH, (Ph3P)2PdCl2, CuI, Et3N, toluene, 50 °C, 80%. (e)
n-BuLi, THF, -78 °C, then B(OMe)3, warm to room temperature, 76%.
(f) Tetracyclone, Ph2O, 280 °C, 85% for 25, 81% for 26.
questionable tactic proved to be surprisingly efficient. Iodode-
silation12 of 16 gave 17 in 93% yield, a Sonogashira reaction
with excess phenylacetylene converted the six iodines into
phenylethynyl groups to give 19 (84%), and Diels-Alder
reaction of 19 with excess tetracyclone smoothly gave dendrimer
4 (80%). Thus, the overall yield of 4 from the simple, readily
accessible 14 was 42% in a five-step synthesis. We note that
Mu¨llen and co-workers have very recently reported a very
similar preparation of intermediates 16 and 17 during the
synthesis of entirely different targets.13
The polyphenylene dendrimer 5swhich lacks six of the
phenyl groups found in 4, but which is also a substructure of
phenylogous cubic graphiteswas prepared by this general
method as well. In that case, hexaiodide 17 was converted to
the terminal alkyne 18 by Sonogashira reaction with trimeth-
ylsilylacetylene followed by fluoride cleavage in overall 91%
yield. The final Diels-Alder reaction gave a more modest 60%
yield of 5, perhaps due to partial polymerization of the six
terminal alkyne groups under the relatively harsh reaction
Unfortunately, this reaction and similar reactions employing
hexakis(4-bromophenyl)benzene instead of the hexaiodide 20
were not successful (data not shown). The products were mainly
the polyphenylenes resulting from two, three, and four coupling
(14) (a) Iyer, V. S.; Wehmeier, M.; Brand, J. D.; Keegstra, M. A.; Mu¨llen, K.
Angew. Chem., Int. Ed. Engl. 1997, 36, 1604-1607. (b) Simpson, C. D.;
Brand, J. D.; Berresheim, A. J.; Przybilla, L.; Ra¨der, H. J.; Mu¨llen, K.
Chem.-Eur. J. 2002, 8, 1424-1429. (c) Bauer, R. E.; Enkelmann, V.;
Wiesler, U. M.; Berresheim, A. J.; Mu¨llen, K. Chem.-Eur. J. 2002, 8,
3858-3864.
(12) Felix, G.; Dunogues, J.; Pisciotti, F.; Calas, R. Angew. Chem., Int. Ed.
Engl. 1977, 16, 488-489.
(13) Wu, J.; Watson, M. D.; Mu¨llen, K. Angew. Chem., Int. Ed. 2003, 42, 5329-
5333.
(15) Hyatt, J. A. Org. Prep. Proced. Int. 1991, 23, 460-463.
9
5800 J. AM. CHEM. SOC. VOL. 126, NO. 18, 2004

Polyphenylene Dendrimers Related to Cubic Graphite
A R T I C L E S
events, and each of these appeared to be mixtures of regioiso-
mers. These results stand in contrast to the ostensibly similar
Suzuki coupling reactions used to make 16 and 19 from 20,
which gave reasonable yields (64% and 25%, respectively) of
6-fold reacted products. Even better was Mu¨llen’s recent
synthesis of 16 by 6-fold coupling of 4-(trimethylsilyl)ben-
zeneboronic acid and hexakis(4-bromophenyl)benzene (92%).13
Indeed, it has been our experience, in this and related projects,
that palladium-catalyzed coupling reactions involving two large
reactants generally proceed poorly, even if both are freely
soluble and there is no apparent steric encumbrance of the
reactive sites, but reactions involving one large reactant and
one small one tend to be highly efficient. Of course, this
observation may be due to steric hindrance of a slightly different
sort: even though the reactive sites are exposed, the large
pendant groups of the two reactants may collide with each other
or with large phosphine ligands surrounding the palladium in
some critical transition state. In any event, it may be wise to
avoid the coupling of two large molecules where possible when
planning syntheses of highly branched polyphenylenes.
fact, combined with the presence of large voids filled with
disordered dichloromethane molecules, makes a truly satisfac-
tory refinement of the structure impossible (see the Experimental
Section for details), but this X-ray structure determination does
reveal the composition and conformation of the molecule. A
search of the Cambridge Structural Database18 indicates that
compound 26 is the largest hydrocarbon to date to have its
structure determined by X-ray crystallography, and it is one of
only four crystallographically characterized hydrocarbons con-
taining more than 200 carbon atoms.14c,19
The molecular structure of 26 is illustrated in Figure 1. The
molecule is a relatively crowded, six-bladed molecular propeller.
Compound 25, which lacks six of the interior phenyls in 26, is
less crowded, and its X-ray structure14c shows more variably
tilted polyphenylbenzene “blades” than 26, but neither molecule
shows any large distortions due to steric conflicts. This is
consistent with the 13C NMR spectra of both molecules, which
show resonances that are few and sharp enough to suggest that
all of the phenyl groups undergo rapid rotation on the NMR
time scale. The packing of one layer of the molecules of 26 is
also shown in Figure 1. The crystals contain large channels
parallel to the c axis of the crystal, and these are filled with
severely disordered solvent molecules. Additional solvent
molecules are located in cavities above and below the central
ring of the dendrimer; overall, this is a highly porous, but also
fragile, crystal.
The characterization of large polyphenylenes requires special
care. The most useful simple method is mass spectrometry,
which, when combined with the knowledge of the structures of
the precursors, gives the best evidence of successful reactions.
All of the compounds reported here give strong molecular ions
in FAB or MALDI mass spectra. Less useful are NMR data.
1
Although the molecules are usually soluble, their H NMR
The structures of the larger dendrimers 4 and 5 would be
interesting to obtain, because they might be very different than
those of 25 and 26. The six polyphenylbenzene blades in 4 and
5 are further removed from the core, and they would have the
opportunity to flatten out to a much greater degree than those
in 25 and 26, perhaps giving discoidal structures approximately
40 Å in diameter. However, it seems unlikely that these
structures will be obtained: in addition to the parent hydrocar-
bons 4 and 5, extensive crystallization experiments with a variety
of halogenated and oxygenated derivatives (a dodecabromo
derivative, 4b, is described in the Experimental Section) have
thus far failed to yield satisfactory crystals.
spectra are not very informative, and even the 13C NMR spectra
of the larger compounds can suffer from overlapping resonances,
which are inevitable for molecules composed entirely of benzene
rings. However, the 13C NMR spectra do give an indication of
the degree of symmetry present in the molecules, and thus
provide an important check on the mass spectral data. Com-
pounds 4, 5, 25, and 26 all have relatively simple 13C NMR
spectra consistent with their high symmetry (and all are
reproduced in the Supporting Information).
Because of the limitations of spectroscopic methods and to
determine the conformations of our large polyphenylenes, a
special effort has been made to obtain X-ray crystal structures
of these molecules. Unfortunately, crystallization experiments
with the new dendrimers have failed to yield, with one
exception, single crystals large enough for X-ray analysis. This
was surprising, because the crystal structure of the known
compound 25 has been reported by Mu¨llen and co-workers,14c
and we have previously reported the X-ray structures of several
polyphenylene hydrocarbons containing more than 100 carbon
atoms.4,16,17
The exception is compound 26, which gives long needles
from many solvent combinations, provided that dichloromethane
is one of the components. These crystals are trigonal, and, after
data collection, the initial solution of the structure was rapid.
Unfortunately, it soon became apparent that the compound
crystallizes in the polar space group R3m with Z ) 3 (hexagonal
setting), which implies that each molecule of 26 resides on a
site with crystallographic C3V symmetry. However, compound
26 itself possesses only C3 symmetry, and thus the entire
structure is disordered across the vertical mirror planes. This
Conclusion
The successful synthesis of the polyphenylene 4 in only five
steps and 42% yield from the readily accessible dibromotolane
14, as well as the preparation of dendrimers 5, 25, and 26 by
syntheses of comparable brevity and yield, suggest that even
more complicated polyphenylene nanostructures should be
accessible. The high solubility of these compounds, already
containing nearly 300 carbon atoms, encourages us to believe
that studies of larger structures can still be accomplished by
using conventional chemical methods for synthesis and purifica-
tion, although common spectroscopic methods will be increas-
ingly strained for the characterization of such molecules.
Therefore, having now prepared and characterized both mac-
rocyclic (3)4 and dendrimeric (4, 5) fragments of “phenylogous
cubic graphite”, we hope to synthesize even more elaborate
(18) Allen, F. H.; Kennard, O.; Taylor, R. Acc. Chem. Res. 1983, 16, 146-
153.
(19) (a) Mu¨ller, P.; Uson, I.; Hensel, V.; Schlu¨ter, A. D.; Sheldrick, G. M. HelV.
Chim. Acta 2001, 84, 778-785. (b) Nielsen, M. B.; Schreiber, M.; Baek,
Y. G.; Seiler, P.; Lecomte, S.; Boudon, C.; Tykwinski, R. R.; Gisselbrecht,
J.-P.; Gramlich, V.; Skinner, P. J.; Bosshard, C.; Gunter, P.; Gross, M.;
Diederich, F. Chem.-Eur. J. 2001, 7, 3263-3280.
(16) Tong, L.; Ho, D. M.; Vogelaar, N. J.; Schutt, C. E.; Pascal, R. A., Jr. J.
Am. Chem. Soc. 1997, 119, 7291-7302.
(17) Pascal, R. A., Jr.; Hayashi, N.; Ho, D. M. Tetrahedron 2001, 57, 3549-
3555.
9
J. AM. CHEM. SOC. VOL. 126, NO. 18, 2004 5801

A R T I C L E S
Shen et al.
4-Iodo-4′-(phenylethnyl)biphenyl (7). 4,4′-Diiodobiphenyl (6, 10.0
g, 24.6 mmol), CuI (0.23 g, 1.2 mmol), and (Ph3P)2PdCl2 (0.43 g, 0.62
mmol) were stirred in piperidine (120 mL) at 50 °C under argon for
20 min. Phenylacetylene (2.29 g, 22.4 mmol) was added to the reaction
mixture over the next 20 min, and stirring was continued for 3 h at 50
°C. Saturated NH4Cl was added, and the mixture was extracted twice
with toluene. The combined extracts were washed with water, dried
over Na2SO4, and concentrated to dryness. This residue was chromato-
graphed on a silica gel column (4:1 hexanes-benzene), and the fraction
with Rf 0.29 (silica gel TLC; 4:1 hexanes-benzene) was collected and
concentrated to give compound 7 as a white powder (5.81 g, 15.3 mmol,
1
62%), mp 218-219 °C. H NMR (300 MHz, CDCl3): δ 7.32-7.38
(m, 5 H), 7.53-7.62 (m, 6 H), 7.78 (m, 2 H). 13C NMR (75 MHz,
CDCl3): δ 89.3, 90.6, 93.7, 122.9, 123.4, 127.0, 128.6, 129.0, 131.8,
132.4, 138.2, 140.0, 140.1 (13 of 14 expected resonances). MS (EI)
m/z: 380 (M+, 100), 252 (M - HI, 57). Exact mass, 380.0050; calcd
for C20H13I, 380.0062.
4-Iodo-4′-(pentaphenylphenyl)biphenyl (8). Compound 7 (1.15 g,
3.02 mmol), tetracyclone (1.20 g, 3.13 mmol), and diphenyl ether (3
mL) were mixed in a screw-capped Pyrex tube, and the tube was placed
in a metal bath heated at 280 °C for 2 h. After cooling, the mixture
was dissolved in CH2Cl2, and then methanol was added until no more
precipitate appeared. Suction filtration gave compound 8 as an off-
1
white powder (2.02 g, 2.75 mmol, 91%), mp 351-353 °C. H NMR
(300 MHz, CDCl3): δ 6.81-6.88 (m, 25 H), 6.89 and 7.08 (AA′BB′
system, 4 H), 7.16 and 7.64 (AA′BB′ system, 4 H). 13C NMR (75 MHz,
CDCl3): δ 92.7, 125.1, 125.4, 125.5, 126.8, 126.9, 128.8, 131.59,
131.61, 131.62, 132.2, 136.4, 137.8, 139.9, 140.50, 140.52, 140.53,
140.6, 140.70, 140.72, 140.75, 140.76 (22 of 24 expected resonances).
MS (EI) m/z: 736 (M+, 100). Exact mass, 736.1632; calcd for C48H33I,
736.1627.
4-Ethynyl-4′-(pentaphenylphenyl)biphenyl (10). Compound 8 (500
mg, 0.679 mmol), CuI (6.5 mg, 0.034 mmol), and (Ph3P)2PdCl2 (11.9
mg, 0.017 mmol) were stirred in toluene (15 mL) and triethylamine (5
mL) at 50 °C under argon for 20 min. Trimethylsilylacetylene (133
mg, 1.36 mmol) was added over 5 min, and stirring was continued for
3 h at 50 °C. Saturated NH4Cl was added, and the mixture was extracted
twice with toluene. The combined extracts were washed with water,
dried over Na2SO4, and concentrated to dryness. The residue was
chromatographed on a silica gel column (2:1 hexanes-CH2Cl2), and
the fraction with Rf 0.43 (silica gel TLC; 2:1 hexanes-CH2Cl2) was
collected and concentrated to give 4-(pentaphenylphenyl)-4′-(trimeth-
ylsilylethynyl)biphenyl (9) as a white powder (461 mg, 0.65 mmol,
1
96%). H NMR (400 MHz, CDCl3): δ 0.26 (s, 9 H), 6.84-6.88 (m,
25 H), 6.91 and 7.12 (AA′BB′ system, 4 H), 7.37 and 7.43 (AA′BB′
system, 4 H). 13C NMR (100 MHz, CDCl3): δ 0.22, 94.8, 105.3, 121.7,
125.3, 125.4, 125.5, 126.6, 126.8, 126.9, 131.59, 131.60, 131.62, 132.1,
132.4, 136.7, 139.9, 140.45, 140.53, 140.64, 140.68, 140.72, 140.76,
140.77, 140.9 (25 of 27 expected resonances).
Figure 1. Top: Molecular structure of compound 26. Hydrogen atoms
have been omitted. One symmetry-independent third of the molecule is
drawn with 40% probability thermal ellipsoids, and another third shows
the crystallographic numbering scheme. Bottom: Unit cell and one layer
of the packing of compound 26 viewed down the c axis. For clarity, only
one of the two disordered molecules at each site is illustrated.
The silylated alkyne 9 (461 mg, 0.652 mmol) was dissolved in THF
(15 mL) at room temperature. Tetrabutylammonium fluoride (1.0 M
in THF, 1.3 mL, 1.3 mmol) was added, and the solution was stirred
for 2 h. Water was added, and the mixture was extracted twice with
ethyl acetate. The combined extracts were washed with water, dried
substructures of this material containing multiple macrocycles
in a three-dimensional network.
Experimental Section
over Na2SO4, and concentrated to dryness. The residue was chromato-
graphed on a silica gel column (2:1 hexanes-CH2Cl2), and the fraction
with Rf 0.34 (silica gel TLC; 2:1 hexanes-CH2Cl2) was collected and
concentrated to give compound 10 as a white powder (389 mg, 0.613
Bis(4-bromophenyl)acetylene (14, mp 183-185 °C [ref 10, 182-
184 °C]), 1-bromo-4-(phenylethynyl)benzene (21, mp 84-85 °C [ref
20, 83-84 °C]), hexakis(4-iodophenyl)benzene (20, mp > 400 °C [ref
15, > 330 °C]), 3,4-bis(bromophenyl)-2,5-diphenylcyclopentadienone
(mp 246-247 °C [ref 21, 249.5-250 °C]), and Hg[Co(CO)4]2 were
prepared by literature procedures.
1
mmol, 94%), mp 308-311 °C. H NMR (300 MHz, CDCl3): δ 3.09
(s, 1 H), 6.81-6.88 (m, 25 H), 6.90 and 7.12 (AA′BB′ system, 4 H),
7.39 and 7.46 (AA′BB′ system, 4 H). 13C NMR (75 MHz, CDCl3): δ
77.8, 83.9, 120.7, 125.3, 125.4, 125.5, 126.76, 126.81, 126.9, 131.60,
131.62, 131.63, 132.2, 132.5, 136.6, 139.9, 140.5, 140.58, 140.64,
140.70, 140.72, 140.76, 140.77, 141.4 (24 of 26 expected resonances).
MS (EI) m/z: 634 (M+, 100). Exact mass, 634.2682; calcd for C50H34,
634.2660.
9
(20) Kabalka, G. W.; Wang, L.; Pagni, R. M. Tetrahedron 2001, 57, 8017-
8028.
(21) Coan, S. B.; Trucker, D. E.; Becker, E. I. J. Am. Chem. Soc. 1955, 77,
60-66.
9
5802 J. AM. CHEM. SOC. VOL. 126, NO. 18, 2004

Polyphenylene Dendrimers Related to Cubic Graphite
A R T I C L E S
4-(3-Hydroxy-3-methyl-1-butynyl)-4′-(pentaphenylphenyl)biphe-
nyl (11). Compound 8 (500 mg, 0.679 mmol), CuI (12.9 mg, 0.068
mmol), and (Ph3P)2PdCl2 (23.7 mg, 0.034 mmol) were stirred in toluene
(15 mL) and triethylamine (8 mL) at 80 °C under argon for 20 min.
2-Methyl-3-butyn-2-ol (0.20 mL, 2.06 mmol) was added over 5 min,
and stirring was continued for 3 h at 80 °C. Saturated NH4Cl was added,
and the mixture was extracted twice with ethyl acetate. The combined
extracts were washed with water, dried over Na2SO4, and concentrated
to dryness. The residue was chromatographed on a silica gel column
(5:1 hexanes-ethyl acetate), and the fraction with Rf 0.12 (silica gel
TLC; 5:1 hexanes-ethyl acetate) was collected and concentrated to
give compound 11 as a light orange powder (358 mg, 0.517 mmol,
76%), mp 326-329 °C. 1H NMR (300 MHz, CDCl3): δ 1.61 (s, 6 H),
6.82-6.87 (m, 25 H), 6.88 and 7.11 (AA′BB′ system, 4 H), 7.37 (s, 4
H). 13C NMR (75 MHz, CDCl3): δ 31.7, 65.9, 82.3, 94.3, 121.3, 125.2,
125.4, 125.5, 126.7, 126.8, 126.9, 131.60, 131.62, 131.63, 132.0, 132.1,
136.7, 139.9, 140.4, 140.5, 140.6, 140.68, 140.74, 140.76, 140.78 (25
of 28 expected resonances). MS (EI) m/z: 692 (M+, 100), 674 (M -
H2O, 88), 634 (M - CH3COCH3, 45). Exact mass, 692.3087; calcd
for C53H40O, 692.3079.
Hexakis[4′-(trimethylsilyl)biphenyl-4-yl]benzene (16) (by Tri-
merization of 15). A mixture of compound 15 (210 mg, 0.442 mmol),
Hg[Co(CO)4]2 (17 mg, 0.031 mmol), and dioxane (25 mL) was sealed
in a screw-capped Pyrex tube under argon, and the mixture was stirred
at 100 °C for 2 days. After cooling, the mixture was filtered through
a short Celite pad to remove the catalyst residue. The filtrate was
concentrated to dryness, and the residue was dissolved in CH2Cl2. This
solution was diluted with methanol to precipitate the desired product
16 as an off-white powder (170 mg, 0.119 mmol, 81%), mp > 400 °C.
1H NMR (400 MHz, CDCl3): δ 0.24 (s, 54 H), 6.95 and 7.16 (AA′BB′
system, 24 H), 7.41 and 7.46 (AA′BB′ system, 24 H). 13C NMR (100
MHz, CDCl3): δ -1.0, 125.5, 126.3, 132.2, 133.8, 137.7, 138.9, 140.0,
140.5, 141.2 (10 of 10 expected resonances). FAB MS m/z: 1422 (M+,
77), 978 (100).
Hexakis(4′-iodobiphenyl-4-yl)benzene (17). Compound 16 (0.67
g, 0.47 mmol) was dissolved in CH2Cl2 (40 mL). A solution of ICl in
CH2Cl2 (3.3 mL of a 1.0 M solution, 3.3 mmol) was added, and the
reaction was stirred under argon for 12 h. Methanol was added to
precipitate the desired product 17 as a white powder (0.76 g, 0.44 mmol,
1
93%), mp > 500 °C. H NMR (400 MHz, CDCl3): δ 6.93 and 7.10
(AA′BB′ system, 24 H), 7.14 and 7.63 (AA′BB′ system, 24 H). 13C
NMR (100 MHz, CDCl3): δ 92.9, 125.4, 128.8, 132.2, 137.9, 140.1,
140.3, 140.4 (eight of nine expected resonances). FAB MS m/z: 1746
(M+, 100).
1,4-Bis[4′-(pentaphenylphenyl)biphenyl-4-yl]-1,3-butadiyne (13).
Aryl iodide 8 (42.4 mg, 57.6 mmol), CuI (1.1 mg, 5.8 µmol), and
(Ph3P)4Pd (3.4 mg, 2.9 µmol) were stirred in toluene (4 mL) and
triethylamine (1 mL) at 50 °C under argon for 20 min. Alkyne 10 (36.5
mg, 57.5 mmol) in toluene (1 mL) was added in a dropwise fashion,
and the reaction mixture was stirred at 50 °C for 18 h. Saturated NH4-
Cl was added, and the mixture was extracted twice with toluene. The
combined extracts were washed with water, dried over Na2SO4, and
concentrated to dryness. The residue was chromatographed on a silica
gel column (1:1 hexanes-CH2Cl2), and the fraction with Rf 0.19 (silica
gel TLC; solvent, 1:1 hexanes-CH2Cl2) was collected and concentrated
to give compound 13 as a white powder (34 mg, 27 mmol, 47%), mp
Hexakis(4′-ethynylbiphenyl-4-yl)benzene (18). Compound 17 (200
mg, 0.115 mmol), CuI (5 mg, 0.026 mmol), and (Ph3P)2PdCl2 (24 mg,
0.034 mmol) were stirred in toluene (30 mL) and triethylamine (7 mL)
at 60 °C under argon for 20 min. Trimethylsilylacetylene (132 mg,
1.34 mmol) was added over 5 min, and the reaction mixture was stirred
at 60 °C for 18 h. Saturated NH4Cl was added, and the mixture was
extracted twice with toluene. The combined extracts were washed with
water, dried over Na2SO4, and concentrated to dryness. The solid residue
was dissolved in THF (20 mL), and tetrabutylammonium fluoride (1.0
M in THF, 1.1 mL, 1.1 mmol) was added. The solution was stirred for
2 h at room temperature. Water was added, and the mixture was
extracted twice with ethyl acetate. The combined extracts were washed
with water, dried over Na2SO4, and concentrated to dryness. The residue
was chromatographed on a silica gel column (3:1 hexanes-ethyl
acetate) to give compound 18 as an off-white powder (119 mg, 0.18
1
385-387 °C. H NMR (300 MHz, CDCl3): δ 6.82-6.88 (m, 50 H),
6.90 and 7.11 (AA′BB′ system, 8 H), 7.39 and 7.47 (AA′BB′ system,
8 H). FAB MS m/z: 1268 (M + H [13C1], 100). Because of the very
low solubility of 13, a 13C NMR spectrum was not recorded.
Bis[4′-(pentaphenylphenyl)biphenyl-4-yl]acetylene (12). Aryl io-
dide 8 (50 mg, 0.068 mmol), alkyne 11 (47 mg, 0.068 mmol), CuI
(1.3 mg, 6.8 µmol), (Ph3P)4PdCl2 (4.8 mg, 6.8 µmol), and tetrabutyl-
ammonium iodide (2.5 mg, 6.8 µmol) were heated in a mixture of
toluene (6 mL) and aqueous NaOH (5 M, 0.1 mL, 0.5 mmol) in a
screw-capped tube at 80 °C for 24 h. Saturated NH4Cl was added, and
the mixture was extracted twice with toluene. The combined extracts
were washed with water, dried over Na2SO4, and concentrated to
dryness. The residue was chromatographed on a silica gel column (3:2
hexanes-toluene), and the fraction with Rf 0.15 (silica gel TLC; solvent,
1:1 hexanes-toluene) was collected and concentrated to give compound
12 as an off-white powder (18 mg, 0.014 mmol, 22%), mp > 400 °C.
1H NMR (500 MHz, CDCl3): δ 6.80-6.88 (m, 50 H), 6.88 and 7.16
(AA′BB′ system, 8 H), 7.51 and 7.56 (AA′BB′ system, 8 H). MS
(MALDI) m/z: 1242 (M+, 100). Because of the very low solubility of
12, a 13C NMR spectrum was not recorded.
1
mmol, 91%), mp > 400 °C. H NMR (300 MHz, CDCl3): δ 3.08 (s,
6 H), 6.96 and 7.16 (AA′BB′ system, 24 H), 7.30 and 7.43 (AA′BB′
system, 24 H). 13C NMR (75 MHz, CDCl3): δ 77.9, 83.8, 120.8, 125.6,
126.8, 132.1, 132.6, 137.0, 140.2, 140.4, 141.1 (11 of 11 expected
resonances). FAB MS m/z: 1134 (M+, 100).
Hexakis[4′-(phenylethynyl)biphenyl-4-yl]benzene (19). Compound
17 (300 mg, 0.172 mmol), CuI (6 mg, 0.032 mmol), and (Ph3P)2PdCl2
(36 mg, 0.051 mmol) were stirred in toluene (45 mL) and triethylamine
(10 mL) at 50 °C under argon for 20 min. Phenylacetylene (214 mg,
2.09 mmol) was added over 5 min, and the reaction mixture was stirred
at 50 °C for 18 h. Saturated NH4Cl was added, and the mixture was
extracted twice with toluene. The combined extracts were washed with
water, dried over Na2SO4, and concentrated to dryness. The residue
was chromatographed on a silica gel column. A byproduct, 1,4-
diphenyl-1,3-butadiyne, was eluted with solvent 2:1 hexanes-toluene,
and compound 19 was then eluted with 1:1 hexanes-toluene. Evapora-
tion of the solvent gave 19 as a light yellow powder (230 mg, 0.144
mmol, 84%), mp > 400 °C. 1H NMR (300 MHz, CDCl3): δ 6.99 and
7.21 (AA′BB′ system, 24 H), 7.31-7.34 (m, 18 H), 7.43 and 7.49
(AA′BB′ system, 24 H), 7.50-7.54 (m, 12 H). 13C NMR (75 MHz,
CDCl3): δ 89.6, 90.2, 122.0, 123.5, 125.6, 126.8, 128.4, 128.5, 131.8,
132.1, 132.2, 137.1, 140.2, 140.5, 140.6 (15 of 15 expected resonances).
FAB MS m/z: 1592 (M + H [13C1], 100).
Bis[4′-(trimethylsilyl)biphenyl-4-yl]acetylene (15). Bis(4-bro-
mophenyl)acetylene (14, 500 mg, 1.49 mmol) and 4-(trimethylsilyl)-
benzeneboronic acid (722 mg, 3.72 mmol) were dissolved in toluene
(30 mL). Ethanol (4 mL) and 2 M K2CO3 (10 mL) were added, and
the mixture was flushed with argon. (Ph3P)4Pd (170 mg, 0.15 mmol)
was added, and the mixture was heated at 50 °C under argon for 24 h.
Addition of ethyl acetate (10 mL) to the reaction mixture induced the
crystallization of the desired product 15, which was collected as off-
1
white plates (590 mg, 1.24 mmol, 83%), mp 269-271 °C. H NMR
(400 MHz, CDCl3): δ 0.31 (s, 18 H), 7.62 (s, 16 H). 13C NMR (100
MHz, CDCl3): δ -0.9, 90.3, 122.5, 126.5, 127.2, 132.3, 134.1, 140.0,
140.9, 141.1 (10 of 10 expected resonances). MS (EI) m/z: 474 (M+,
100), 459 (M - CH3, 61). Exact mass, 474.2193; calcd for C32H34Si2,
474.2199.
Hexakis[4′-(2,3,4,5-tetraphenylphenyl)biphenyl-4-yl]benzene (5).
Compound 18 (34 mg, 0.030 mmol), tetracyclone (103 mg, 0.27 mmol),
and diphenyl ether (1 mL) were mixed in a screw-capped Pyrex tube,
and the tube was placed in a metal bath heated at 280 °C for 3 h. After
9
J. AM. CHEM. SOC. VOL. 126, NO. 18, 2004 5803

A R T I C L E S
Shen et al.
cooling, CH2Cl2 (3 mL) was added, the solution was then added
dropwise into methanol (30 mL), and the resulting precipitate was
collected. This material was chromatographed on a silica gel column
(5:1 hexanes-ethyl acetate) to give compound 5 as a pale yellow solid
(60 mg, 0.018 mmol, 60%), mp > 500 °C. 1H NMR (300 MHz,
CDCl3): δ 6.42-7.00 (m, 174 H). 13C NMR (75 MHz, CDCl3): δ
125.3, 125.5, 125.8, 126.1, 126.4, 126.8, 127.1, 127.2, 127.8, 130.1,
130.4, 131.7, 137.1, 138.6, 139.3, 139.4, 139.8, 140.2, 140.4, 140.5,
141.0, 141.9, 142.0 (23 of 31 expected resonances). MS (MALDI)
m/z: 3273 (M+ [13C3], 100).
Hexakis[4′-(phenylethynyl)biphenyl-4-yl]benzene (19) (by Ary-
lation of 20). Compound 20 (200 mg, 0.155 mmol) and compound 22
(300 mg, 1.50 mmol) were dissolved in toluene (50 mL). Ethanol (4
mL) and 2 M K2CO3 (10 mL) were added, and the mixture was flushed
with argon. (Ph3P)4Pd (18 mg, 0.016 mmol) was added, and the mixture
was heated at 80 °C under argon for 24 h. Water was added, and the
mixture was extracted twice with ethyl acetate. The combined extracts
were washed with water, dried over Na2SO4, and concentrated to
dryness. The residue was chromatographed on a silica gel column.
Impurities were eluted with 2:1 hexanes-toluene, and compound 19
was then eluted with solvent 1:1 hexanes-toluene. Evaporation of the
solvent gave 19 as a light yellow powder (62 mg, 0.039 mmol, 25%),
mp > 400 °C. (Spectral data are provided above.)
Hexakis[4′-(pentaphenylphenyl)biphenyl-4-yl]benzene (4). Com-
pound 19 (40 mg, 0.025 mmol), tetracyclone (87 mg, 0.23 mmol), and
diphenyl ether (1 mL) were mixed in a screw-capped Pyrex tube, and
the tube was placed in a metal bath heated at 280 °C for 3 h. After
cooling, CH2Cl2 (3 mL) was added, the solution was then added
dropwise into methanol (30 mL), and the resulting precipitate was
collected. This material was chromatographed on a silica gel column
(1:1 hexanes-CH2Cl2) to give compound 4 as a pale yellow solid (75
Hexakis(4-ethynylphenyl)benzene (23). Compound 20 (300 mg,
0.233 mmol), CuI (7 mg, 0.037 mmol), and (Ph3P)2PdCl2 (24 mg, 0.034
mmol) were stirred in toluene (20 mL) and triethylamine (5 mL) at 50
°C under argon for 20 min. Trimethylsilylacetylene (271 mg, 2.76
mmol) was added over 5 min, and the reaction mixture was stirred at
50 °C for 18 h. Saturated NH4Cl was added, and the mixture was
extracted twice with ethyl acetate. The combined extracts were washed
with water, dried over Na2SO4, and concentrated to dryness. The solid
residue was dissolved in THF (30 mL), and tetrabutylammonium
fluoride (1.0 M in THF, 2.1 mL, 2.1 mmol) was added. The solution
was stirred for 2 h at room temperature. Water was added, and the
mixture was extracted twice with ethyl acetate. The combined extracts
were washed with water, dried over Na2SO4, and concentrated to
dryness. The residue was chromatographed on a silica gel column (1:1
hexanes-CH2Cl2) to give compound 23 as a white powder (119 mg,
1
mg, 0.020 mmol, 80%), mp > 500 °C. H NMR (300 MHz, CDCl3):
δ 6.62-6.96 (m, 198 H). 13C NMR (75 MHz, CDCl3): δ 125.2, 125.4,
126.8, 126.9, 131.6, 131.8, 137.47, 137.53, 139.3, 139.4, 140.0, 140.3,
140.50, 140.56, 140.61, 140.8 (16 of 25 expected resonances). MS
(MALDI) m/z: 3731 (M+ [13C5], 100).
Hexakis{4′-[3,4-bis(4-bromophenyl)-2,5,6-triphenylphenyl]biphe-
nyl-4-yl}benzene (4b). Compound 19 (30 mg, 0.019 mmol), 3,4-bis-
(4-bromophenyl)-2,5-diphenylcyclopentadienone (93 mg, 0.17 mmol),
and diphenyl ether (1 mL) were mixed in a screw-capped Pyrex tube,
and the tube was placed in a metal bath heated at 280 °C for 3 h.
Workup and purification as described for 4 gave compound 4b as a
pale yellow solid (50 mg, 0.011 mmol, 57%), mp > 500 °C. 1H NMR
(300 MHz, CDCl3): δ 6.60-7.20 (m, 186 H). 13C NMR (75 MHz,
CDCl3): δ 119.9, 125.2, 125.5, 125.7, 126.9, 127.08, 127.13, 130.3,
131.5, 131.6, 131.8, 133.1, 137.4, 137.7, 139.0, 139.1, 139.2, 140.26,
140.32, 140.49, 140.54, 140.59, 140.7, 141.1 (24 of 35 expected
resonances). MS (MALDI) m/z: 4678 (M+ [13C479Br681Br6], 100).
Hexakis[4′-(trimethylsilyl)biphenyl-4-yl]benzene (16) (by Ary-
lation of 20). Compound 20 (1.00 g, 0.78 mmol) and 4-(trimethylsilyl)-
benzeneboronic acid (1.80 g, 9.3 mmol) were dissolved in toluene (100
mL). Ethanol (6 mL) and 2 M K2CO3 (20 mL) were added, and the
mixture was flushed with argon. (Ph3P)4Pd (90 mg, 0.078 mmol) was
added, and the mixture was heated at 80 °C under argon for 24 h.
Water was added, and the mixture was extracted twice with ethyl
acetate. The combined extracts were washed with water, dried over
Na2SO4, and concentrated to dryness. The residue was chromatographed
on a silica gel column. A byproduct, 4,4′-bis(trimethylsilyl)biphenyl,
was eluted with hexanes, and compound 16 was then eluted with ethyl
acetate. Evaporation of the solvent gave 16 as an off-white powder
(0.71 g, 0.50 mmol, 64%), mp > 400 °C. (Spectral data are provided
above.)
1
0.175 mmol, 75%), mp > 400 °C [ref 13, >300 °C]. H NMR (300
MHz, CDCl3): δ 2.98 (s, 6 H), 6.72 and 7.03 (AA′BB′ system, 24 H).
13C NMR (75 MHz, CDCl3): δ 77.5, 83.8, 119.7, 131.17, 131.24, 140.0,
140.5 (seven of seven expected resonances). MS (EI) m/z: 678 (M+,
100). Exact mass, 678.2350; calcd for C54H30, 678.2347.
Hexakis[4-(phenylethynyl)phenyl]benzene (24). Hexakis(4-io-
dophenyl)benzene (20, 300 mg, 0.233 mmol), CuI (7 mg, 0.037 mmol),
and (Ph3P)2PdCl2 (24 mg, 0.034 mmol) were stirred in toluene (20 mL)
and triethylamine (5 mL) at 50 °C under argon for 20 min. Phenyl-
acetylene (288 mg, 2.82 mmol) was added over 5 min, and the reaction
mixture was stirred at 50 °C for 18 h. Saturated NH4Cl was added,
and the mixture was extracted twice with ethyl acetate. The combined
extracts were washed with water, dried over Na2SO4, and concentrated
to dryness. The residue was chromatographed on a silica gel column
(1:1 hexanes-CH2Cl2) to give compound 24 as a light yellow powder
(211 mg, 0.186 mmol, 80%), mp > 400 °C. 1H NMR (300 MHz,
CDCl3): δ 6.82 and 7.12 (AA′BB′ system, 24 H), 7.27-7.32 (m, 18
H), 7.42-7.49 (m, 12 H). 13C NMR (75 MHz, CDCl3): δ 89.66, 89.72,
120.7, 123.4, 128.4, 128.5, 130.7, 131.4, 131.7, 140.20, 140.21 (11 of
11 expected resonances). FAB MS m/z: 1136 (M + H [13C1], 100).
Hexakis(2′,3′,4′,5′-tetraphenylbiphenyl-4-yl)benzene (25). Com-
pound 23 (115 mg, 0.169 mmol), tetracyclone (585 mg, 1.52 mmol),
and diphenyl ether (3 mL) were mixed in a screw-capped Pyrex tube,
and the tube was placed in a metal bath heated at 280 °C for 3 h. After
cooling, CH2Cl2 (5 mL) was added, the solution was then added
dropwise into methanol (50 mL), and the resulting precipitate was
collected. The residue was chromatographed on a silica gel column
(3:2 hexanes-CH2Cl2) to give compound 25 as an off-white solid (406
4-(Phenylethynyl)benzeneboronic Acid (22). Compound 21 (2.00
g, 7.78 mmol) was dissolved in freshly distilled THF (30 mL) and
cooled to -78 °C under argon. To this solution was added n-
butyllithium (1.6 M in pentane, 5.3 mL, 8.5 mmol), and after 30 min,
trimethyl borate (1.21 g, 11.67 mmol) was added over 15 min. The
reaction was allowed to warm to room temperature over 3 h, and it
was quenched by adding 2 N HCl (50 mL). The resulting mixture was
extracted twice with ethyl acetate, and the combined extracts were
washed with water, dried over Na2SO4, and concentrated to dryness.
Crystallization of the crude product from toluene gave pure compound
22 as a white powder (1.32 g, 5.94 mmol, 76%), mp 204-206 °C. 1H
NMR (400 MHz, DMSO-d6): δ 7.42-7.45 (m, 3 H), 7.51 and 7.82
(AA′BB′ system, 4 H), 7.55-7.58 (m, 2 H), 8.19 (s, 2 H). 13C NMR
(100 MHz, DMSO-d6): δ 89.6, 90.1, 122.3, 123.7, 128.8, 128.9, 130.3,
131.4, 134.3 (nine of 10 expected resonances). MS (EI) m/z: 222 (M+,
10), 204 (M - H2O, 8), 178 (M - B(OH)2, 100). Exact mass, 222.0845;
calcd for C14H11BO2, 222.0852.
1
mg, 0.144 mmol, 85%), mp > 400 °C [ref 13, > 300 °C]. H NMR
(300 MHz, CDCl3): δ 6.40 and 6.66 (AA′BB′ system, 24 H), 6.72-
6.98 (m, 90 H), 7.14-7.20 (m, 30 H), 7.49 (s, 6 H). 13C NMR (75
MHz, CDCl3): δ 125.5, 125.7, 126.4, 126.8, 127.0, 127.1, 127.7, 128.6,
130.2, 131.5, 131.7, 131.8, 131.9, 138.3, 138.6, 139.2, 139.3, 140.1,
140.2, 140.3, 140.6, 140.7, 140.8, 141.86, 141.9 (25 of 27 expected
resonances). FAB MS m/z: 2819 (M + H [13C3], 100).
Hexakis(2′,3′,4′,5′,6′-pentaphenylbiphenyl-4-yl)benzene (26). Com-
pound 24 (70 mg, 0.062 mmol), tetracyclone (213 mg, 0.55 mmol),
and diphenyl ether (1 mL) were mixed in a screw-capped Pyrex tube,
9
5804 J. AM. CHEM. SOC. VOL. 126, NO. 18, 2004

Polyphenylene Dendrimers Related to Cubic Graphite
A R T I C L E S
and the tube was placed in a metal bath heated at 280 °C for 3 h. After
cooling, CH2Cl2 (3 mL) was added, the solution was then added
dropwise into methanol (30 mL), and the resulting precipitate was
collected. The crude product was chromatographed on a silica gel
column (5:1 hexanes-ethyl acetate) to give compound 26 as a pale
However, 26 itself possesses only C3 symmetry, and thus the entire
structure is disordered across the vertical mirror planes. For this reason,
it was necessary to employ ideal rigid bodies for the benzene rings,
restraints on many inter-ring distances, and restraints on the anisotropic
displacement coefficients of all non-hydrogen atoms. A riding model
was employed for the hydrogen atoms [C-H ) 0.93 Å, U(H) )
1.2U(C)], and several molecules of CH2Cl2 were included as rigid
bodies. Refinement of this discrete atom model converged to R(F) )
0.192 with no peaks in the ∆F map greater than 0.50 e/Å3, but it was
clear that the solvent was very poorly described. For this reason, the
SQUEEZE/BYPASS procedure24 implemented in PLATON25 was
employed to account for the solvent electron density. With only the
title molecule model included in the instruction file for PLATON, the
SQUEEZE option found a total electron count of 1724 e in a volume
of 7947 Å3 for the solvent regions in the unit cell (33% of the unit cell
volume). This electron count corresponds to 41.0 molecules of CH2-
Cl2 (42 e each), and thus the composition of the crystal has been
formulated as C258H174‚14CH2Cl2 (because Z ) 3, the number of CH2-
Cl2 molecules in the cell would be 42), but it seems likely that even
more solvent is present.
1
yellow solid (163 mg, 0.050 mmol, 81%), mp > 500 °C. H NMR
(300 MHz, CD2Cl2): δ 5.59 and 6.24 (AA′BB′ system, 24 H), 6.69-
7.02 (m, 150 H). 13C NMR (75 MHz, CD2Cl2): δ 125.7, 125.8, 127.0,
127.2, 130.7, 131.5, 131.8, 131.9, 132.2, 132.9, 136.7, 137.6, 140.7,
140.8, 141.1, 141.4, 141.59, 141.60, 141.62 (19 of 21 expected
resonances). FAB MS m/z: 3275 (M + H [13C3], 100).
X-ray Crystal Structure of 26. Compound 26 crystallized upon
the evaporation of a solution in CH2Cl2-CHCl3-benzene at room
temperature. A colorless needle was cut to dimensions of 0.12 mm ×
0.15 mm × 0.55 mm, sealed in a glass capillary (because of rapid
solvent loss by exposed crystals), and transferred to a Nonius Kappa-
CCD diffractometer. Frames of data (1400) were collected using Mo
KR radiation (λ ) 0.71073 Å) at 298 (2) K with an ω oscillation range
of 0.5°/frame and an exposure time of 480 s/deg. A total of 62 392
reflections were collected (θmax ) 18.84°), and these were reduced to
8378 unique reflections, of which 5416 had I > 2σ(I), by using the
program DENZO-SMN.22 For subsequent refinement in the space group
R3m (see below), the 8378 reflections were merged to 4402 unique
reflections by using Siemens SHELXTL.23 Postrefinement of the unit
cell parameters gave a ) b ) 48.293 (1), c ) 11.866 (1), and V )
23,965 (2) Å3. Axial photographs, systematic absences, and the observed
mean |E‚E-1| value of 0.658 were consistent with those expected for
the noncentric, trigonal space groups R3 (No. 146), R32 (No. 155),
and R3m (No. 160).
The SQUEEZE-processed data were used for all subsequent cycles
of refinement. The refinement converged to R(F) ) 0.1321, wR(F2) )
0.3431, and S ) 1.204 for 2777 reflections with I > 2σ(I), and R(F) )
0.1545, wR(F2) ) 0.3751, and S ) 1.033 for 4402 unique reflections,
601 parameters, and 389 restraints. The maximum ∆/σ in the final cycle
of least squares was less than 0.001, and the residual peaks on the
final ∆F map ranged from -0.33 to 0.33 e/Å3.
Acknowledgment. This work was supported by National
Science Foundation Grants CHE-0077990 and CHE-0314873,
which are gratefully acknowledged.
Supporting Information Available: 13C NMR spectra of
compounds 4, 4b, 5, 7-11, 15-19, and 22-26 (PDF) and X-ray
structural information for compound 26 (CIF). This material is
available free of charge via the Internet at http://pubs.acs.org.
JA030675C
The structure was solved by direct methods in the space group R3,
and the initial model was refined by full-matrix least-squares on F2
using SHELXTL. In such a model, each molecule of 26 resides on a
special position with crystallographic C3 symmetry, and, in the
hexagonal setting for R3, Z ) 3. Unfortunately, it soon became apparent
that the true space group was R3m with Z ) 3, which implies that
each molecule of 26 lies on a site with crystallographic C3V symmetry.
(22) Otwinowski, Z.; Minor, W. Methods Enzymol. 1997, 276, 307-326.
(23) Sheldrick, G. M. SHELXTL, version 5.04; Siemens Analytical X-ray
Instruments: Madison, WI, 1996.
(24) Van der Sluis, P.; Spek, A. L. Acta Crystallogr., Sect. A 1990, 46, 194-
201.
(25) Spek, A. L. Acta Crystallogr., Sect. A 1990, 46, C34.
9
J. AM. CHEM. SOC. VOL. 126, NO. 18, 2004 5805