Imposing curvature on a polyarene by intramolecular palladium-catalyzed arylation reactions: A simple synthesis of dibenzo[a,g]corannulene

Article abstract of DOI:10.1021/ol005755p

(Matrix Presented) Palladium catalysis has been found to offer an effective solution-phase alternative to gas-phase flash vacuum pyrolysis as a method for converting the planar ring system of 7,10-di(2-bromophenyl)fluoranthene (4) to that of the C_28<

Full text of DOI:10.1021/ol005755p

ORGANIC
LETTERS
2000
Vol. 2, No. 10
1427-1430
Imposing Curvature on a Polyarene by
Intramolecular Palladium-Catalyzed
Arylation Reactions: A Simple
Synthesis of Dibenzo[a,g]corannulene
Helge A. Reisch, Matthew S. Bratcher, and Lawrence T. Scott*
Department of Chemistry, Merkert Chemistry Center Boston College,
Chestnut Hill, Massachusetts 02467-3860
lawrence.scott@bc.edu
Received March 5, 2000
ABSTRACT
Palladium catalysis has been found to offer an effective solution-phase alternative to gas-phase flash vacuum pyrolysis as a method for
converting the planar ring system of 7,10-di(2-bromophenyl)fluoranthene (4) to that of the C H bowl-shaped fullerene fragment dibenzo-
28 14
[a,g]corannulene (5).
The synthesis and study of bowl-shaped polycyclic aromatic
hydrocarbons (PAHs) whose networks of five- and six-
membered rings can be mapped onto those of C60 and/or
the higher fullerenes has attracted considerable attention in
recent years.1 Such geodesic polyarenes are characterized
by both concave and convex π-surfaces as well as by a high
degree of strain energy2 resulting from the pyramidalization
of interior trigonal carbon atoms. An ability to exhibit
fullerene-like chemistry at their interior carbon atoms1c,f,g sets
these novel hydrocarbons apart from their familiar planar
relatives.
separated carbon atoms in planar PAH derivatives to
construct strained fullerene fragments.3 Most subsequent
synthetic work, both in our laboratory and elsewhere, has
employed this powerful technique.1 With appropriate py-
rolysis precursors, the FVP method has been used to
synthesize a broad range of geodesic polyarenes built around
a central five-membered ring (e.g., corannulene 1), a central
alkene (e.g., the C26H12 bowl 2), or a central six-membered
ring (e.g., the chiral C30H12 bowl 3).
Some years ago, we introduced the use of flash vacuum
pyrolysis (FVP) as a strategy for stitching together distantly
(1) Several reviews have already been published: (a) Rabideau, P. W.;
Sygula, A. In AdVances in Theoretically Interesting Molecules; Thummel,
R. P., Ed.; JAI Press: Greenwich, 1995; Vol. 3, pp 1-36. (b) Rabideau, P.
W.; Sygula, A. Acc. Chem. Res. 1996, 29, 235-242. (c) Scott, L. T. Pure
Appl. Chem. 1996, 68, 291-300. (d) Mehta, G.; Rao, H. S. P. AdV. Strain
Org. Chem. 1997, 6, 139-187. (e) Mehta, G.; Rao, H. S. P. Tetrahedron
1998, 54, 13325-13370. (f) Scott, L. T.; Bronstein, H. E.; Preda, D. V.;
Ansems, R. B. M.; Bratcher, M. S.; Hagen, S. Atual. Fis.-Quim. Org. 1999,
11, 138-150. (g) Scott, L. T.; Bronstein, H. E.; Preda, D. V.; Ansems, R.
B. M.; Bratcher, M. S.; Hagen, S. Pure Appl. Chem. 1999, 71, 209-219.
(2) Kiyobayashi, T.; Nagano, Y.; Sakiyama, M.; Yamamoto, K.; Cheng,
P.-C.; Scott, L. T. J. Am. Chem. Soc. 1995, 117, 3270-1.
Despite the obvious success of this method, the FVP
strategy suffers from several drawbacks: (1) low to modest
(3) (a) Scott, L. T.; Hashemi, M. M.; Meyer, D. T.; Warren, H. B. J.
Am. Chem. Soc. 1991, 113, 7082-7084. (b) Scott, L. T.; Hashemi, M. M.;
Bratcher, M. S. J. Am. Chem. Soc. 1992, 114, 1920-1921.
10.1021/ol005755p CCC: $19.00 © 2000 American Chemical Society
Published on Web 04/26/2000

Table 1. Reaction Conditions, Product Ratios, and Isolated Yields of Products from the Reaction in Scheme 2
ratio (%)c
entry
catalyst
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
base
solventa
additive
conditionsb
4
5
6
7
yield (%)d
1
2
3
4
5
6
7
8
DBU
DBU
DBU
K2CO3
DBU
DBU
DBU
DBU
NMP
NMP
NMP
NMP
DMF
DMF
DMAc
DMF
LiI
LiI
Bu4NBr
none
Bu4NBr/PPh3
Bu4NBr/PPh3
none
130 °C/48 h
175 °C/48 h
175 °C/48 h
175 °C/48 h
130 °C/48 h
150 °C/48 h
175 °C/48 h
150 °C/72 h
100
90
90
75
95
0
0
0
0
0
0
20
0
25
0
0
4 (95)
tr
tr
tr
0
55
5
10
10
25
5
25
95
5
Pd(OAc)2
Pd[P(o-tol)3]2Cl2
Pd[P(o-tol)3]2Cl2
Pd[P(o-tol)3]2Cl2
Pd2(dba)3/P(t-Bu)3
0
0
none
70
5 (21)
6 (60)
5 (8)
5 (51)
5 (53)
5 (57)
9
10
11
12
13
14
15
16
17
18
19
Pd2(dba)3/P(t-Bu)3
Pd(PPh3)2Br2
Pd(PPh3)2Br2
palladacyclee
palladacycle
Cs2CO3
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DBU
DMF
DMF
DMF
DMF
DMAc
NMP
DMF
DMAc
DMF
DMAc
DMF
none
none
PPh3
none
none
none
none
none
none
none
none
150 °C/72 h
150 °C/72 h
150 °C/72 h
150 °C/72 h
165 °C/72 h
175 °C/12 h
150 °C/72 h
175 °C/72 h
150 °C/72 h
175 °C/72 h
150 °C/72 h
0
0
0
0
0
10
55
55
60
25
5
0
0
0
0
85
25
25
20
60
45
0
0
0
0
0
5
20
20
20
15
palladacycle
0
50
Pd2(dba)3/BINAP
Pd2(dba)3/BINAP
Pd2(dba)3
Pd2(dba)3
PdCl2(MeCN)2
100
90
0
0
4 (93)
7 (93)
7 (96)
10
100
100
100
0
0
0
a Anhydrous solvents were purged with argon for 10 min. b A mixture of 4 (0.5 mmol), catalyst (10 mol %), base (40 mmol), and solvent (10 mL) was
heated and stirred under argon. c Ratio determined by NMR integration and HPLC (reverse-phase RP-18, water-acetonitrile-CH2Cl2). tr ) traces. d Isolated
yield. e See ref 13.
yields, (2) essentially zero functional group tolerance, (3) a
limitation to relatively small scale runs, and (4) the potential
for thermal rearrangements of the molecular framework at
the high temperatures normally employed (1000-1100 °C).
In this regard, the new nonpyrolytic method reported by
Seiders and Siegel et al. for preparing such compounds, using
highly exothermic reductive couplings of benzylic bromides
with Ti(0),4 is especially noteworthy. Other workers in the
field have been quick to adopt and extend this solution-phase
alternative.5 Herein we report our own, quite different,
solution-phase alternative to the FVP method, one that is
based on palladium-catalyzed intramolecular arylation reac-
tions.6
H2O, 100 °C, 90 h],11c thereby demonstrating the feasibility
of this approach; however, we never succeeded in developing
the reaction into a practical synthetic method at that time.
Scheme 1
Aryl-aryl coupling reactions involving palladium-cata-
lyzed eliminations of HX (X ) Br, I, OTf, etc.), both intra-
and intermolecular, have enjoyed widespread popularity for
the synthesis of carbocycles,7 heterocycles,8 and various
natural products.9 The mild reaction conditions, wide func-
tional group tolerance, and infinite scaleability of this
chemistry has made it very attractive for organic synthesis.
To the best of our knowledge, however, catalytic palladium
chemistry has never been successfully applied to the synthesis
of a bowl-shaped polyarene; only one unsuccessful attempt
has been reported.10
Convinced that a preparatively useful palladium-catalyzed
variant of the double cyclization in Scheme 1 should be
possible, and impressed by the tremendous advances world-
wide in organopalladium chemistry over the past several
years, we recently resumed our activities in this area.
Numerous palladium sources were screened, with both
organic and inorganic bases, in several different dipolar
aprotic solvents, with and without additives, at temperatures
ranging from 130 to 175 °C (Table 1). The progress of the
reactions was monitored by TLC and HPLC.
Shortly after finding that dibenzo[a,g]corannulene (5)
could be prepared by FVP of 7,10-di(2-bromophenyl)-
fluoranthene (4) (Scheme 1),11,12 we began exploring pal-
ladium catalysis as a potential solution-phase alternative for
the large scale preparation of 5 from the same precursor.
Early results were encouraging, in that they gave detectable
quantities of 5 [Pd(OAc)2, K2CO3, Bu4NBr, (Ph)3P, DMF/
Gratifyingly, dibenzocorannulene (5) was indeed formed
in several of the reactions. Unfortunately, reductive de-
(4) (a) Seiders, T. J.; Baldridge, K. K.; Siegel, J. S. J. Am. Chem. Soc.
1996, 118, 2754-5. (b) Seiders, T. J.; Elliott, E. L.; Grube, G. H.; Siegel,
J. S. J. Am. Chem. Soc. 1999, 121, 7804-7813.
(5) (a) Sygula, A.; Rabideau, P. W. J. Am. Chem. Soc. 1998, 120, 12666.
(b) Rabideau, P. W.; Sygula, A. J. Am. Chem. Soc. 1999, 121, 7800.
1428
Org. Lett., Vol. 2, No. 10, 2000

Scheme 2
bromination of the 2-bromophenyl groups, either just on one
side, giving 6, or on both sides, giving 7 (Scheme 2), often
prevailed over the desired cyclizations.
to reductive debromination and formation of 7 as the major
product (entry 7).
With Pd[P(t-Bu)3]2, a catalyst known to be effective for
palladium-catalyzed Heck and Suzuki couplings involving
normally recalcitrant aryl chlorides,14,15 25% of the reaction
took the desired course to produce 5 when DBU was used
as the base (entry 8). The yield of 5 dropped significantly,
however, when the inorganic base Cs2CO3 was used (entry
9). The major product obtained using Pd[P(t-Bu)3]2 was
always the monoclosed hydrocarbon 6.
The catalytic system Pd2(dba)3 and BINAP showed almost
no reactivity; most of the starting material was recovered
even after 72 h at 150-170 °C (entries 15 and 16). The use
of phoshine-free Pd2(dba)3 and PdCl2(MeCN)2 led only to
reductive debromination, yielding 7 nearly quantitatively
(entries 17-19).
While the factors responsible for the success of the
cyclizations listed in entries 10-12 remain uncertain, it is
now clear that palladium-mediated dehydrobrominations can
be used for intramolecular aryl-aryl couplings that generate
strained geodesic polyarenes from relatively simple PAH
derivatives. In the present case, dibromide 4 was prepared
in just two steps from commercially available starting
materials (Schemes 3 and 4).
It is clear from the results listed above that variations in
the reaction conditions have a profound influence on the
product distribution. The best yields of dibenzocorannulene
(5) that we have achieved so far have come from the
treatment of dibromide 4 with excess DBU and 10 mol %
of either Pd(PPh3)2Br2 or Herrmann’s palladacycle13 in
dimethylformamide (DMF) at 150 °C (entries 10-12). The
isolated yield of 5 under these conditions (50-60%) exceeds
that obtained by the FVP method (35-40%) by a significant
margin and gives hope that this chemistry could prove useful
for the synthesis of other geodesic polyarenes.
It is interesting to note that the addition of triphenylphos-
phine had no influence on the product ratio in this catalytic
reaction (compare entries 10 and 11) and that higher
temperatures in either N-methylpyrrolidinone (NMP) or
dimethylacetamide (DMAc) led to significantly more reduc-
tive debromination, with a corresponding decrease in the
amount of dibenzocorannulene (5) formed (compare entries
12-14).
Notwithstanding our encouraging early results with
Pd(OAc)2 (vide supra), all subsequent experiments using this
catalyst were completely unsuccessful. At temperatures lower
than 150 °C in NMP, only starting material was recovered
(entry 1). At high temperature, reductive debromination
leading to diphenylfluoranthene 7 was the only reaction
observed (entries 2-4). Neither LiI nor Bu4NBr as additives
promoted the desired cyclizations, and the use of K2CO3 as
base simply accelerated the formation of 7 (entry 4).
Only marginally greater success was achieved with Pd[P(o-
tol)3]2Cl2. With this catalyst, dibenzocorannulene (5) was
formed in low yield at 150 °C in DMF (entry 6). At lower
temperatures, however, only starting material was recovered
(entry 5), whereas increasing the temperature to 175 °C led
Scheme 3
(6) Aspects of this work have been reported in preliminary form: Scott,
L. T.; Reisch, H. A. Abstracts of Papers, 219th National Meeting of the
American Chemical Society, San Francisco, CA.; American Chemical
Society: Washington, DC; March, 2000; ORGN-515.
(7) (a) Rice, J. E.; Cai, Z.-W. J. Org. Chem. 1993, 58, 1415. (b) Dyker,
G. J. Org. Chem. 1993, 58, 6426. (c) Dyker, G. Chem. Ber. 1994, 127,
739. (d) Rice, J. E.; Cai, Z.-W.; He, Z.-M.; LaVoie, E. J. Org. Chem. 1995,
60, 8101. (e) Gonza´lez, J. J.; Garcia, N.; Go´mez-Lor, B.; Echavarren, A. J.
Org. Chem. 1997, 62, 1286. (f) De Frutos, O.; Gomez-Lor, B.; Granier, T.;
Monge, M. A.; Gutierrez-Puebla, E.; Echavarren, A. M. Angew. Chem.,
Int. Ed. 1999, 38, 204-207. (g) Gomez-Lor, B.; de Frutos, O.; Echavarren,
A. M. Chem. Commun. 1999, 2431-2432.
(8) (a) Ames, D. E.; Bull, D. Tetrahedron 1982, 38, 383. (b) Ames, D.
E.; Opalko, A. Tetrahedron 1984, 40, 1919-1925. (c) Ames, D. E.; Opalko,
A. Synthesis 1984, 234. (d) Rawal, V. H.; Hennings, D. D.; Iwasa, S. Angew.
Chem., Int. Ed. Engl. 1997, 36, 2-3.
Taking advantage of our previously reported one-pot
synthesis of 7,10-disubstituted fluoranthenes,3 1,3-di(2-
bromophenyl)acetone (8) was heated with acenaphthene-
Org. Lett., Vol. 2, No. 10, 2000
1429

The 1,3-di(2-bromophenyl)acetone (8) required for this
synthesis is easily prepared by carbonylation of 2-bromo-
benzyl bromide with diironnonacarbonyl16 (71% yield)
(Scheme 4).
Scheme 4
By this three-step synthesis, dibenzo[a,g]corannulene (5)
can now be prepared without pyrolysis in quantities sufficient
to use as the starting point for syntheses of larger fullerene
fragments.
quinone (9) in the presence of a base and norbornadiene. A
double Knoevenagel condensation reaction presumably gen-
erates cyclopentadienone 10, which then undergoes a Diels-
Alder reaction with the norbornadiene to give adduct 11.
Carbon monoxide and cyclopentadiene are lost spontaneously
under the reaction conditions, giving 4 in 40% yield (Scheme
3). The entire assembly of the disubstituted fluoranthene
takes place in a single reaction vessel without isolation of
intermediates.
Acknowledgment. We thank the National Science Foun-
dation for financial support of this work.
Supporting Information Available: Characterization
data for compounds 4-6. This material is available free of
charge via the Internet at http://pubs.acs.org.
OL005755P
(12) A second FVP synthesis of 5 has recently been reported by Mehta,
G.; Srirama Sarma, P. V. V. Chem. Commun. 2000, 19-20.
(13) (a) Herrmann, W. A.; Brossmer, C.; Oefele, K.; Reisinger, C.-P.;
Priermeier, T.; Beller, M.; Fischer, H. Angew. Chem., Int. Ed. Engl. 1995,
34, 1844-7. (b) Beller, M.; Fischer, H.; Herrmann, W. A.; Oefele, K.;
Brossmer, C. Angew. Chem., Int. Ed. Engl. 1995, 34, 1848-1849.
(14) Fu, G. C.; Littke, A. F. J. Org. Chem. 1999, 64, 10.
(15) In the Pd2(dba)3/P(t-Bu)3 system, Pd[P(t-Bu)3]2 is the only species
observable by 31P NMR spectroscopy, although recent studies indicate that
Pd[P(t-Bu)3]1 is most likely the catalytically active species: Fu, G. C.; Littke,
A. F.; Dai, C. Book of Abstracts, 219th National Meeting of the American
Chemical Society, San Francisco, March 26-30, 2000; American Chemical
Society: Washington, DC; 2000, ORGN-067.
(9) (a) Brinkmann, G.; Jansen, J. R.; Rink, H. P. Angew. Chem., Int. Ed.
Engl. 1986, 25, 913. (b) Deshpande, P. P.; Martin, O. R. Tetrahedron Lett.
1990, 31, 6313-6316. (c) Matsumoto, T.; Suzuki, K.; Takashiro, E.; Hosoya,
T. J. Am. Chem. Soc. 1994, 116, 1004. (d) Grigg, R.; Brown, D.; Sridharan,
V.; Tambyrajah, V. Tetrahedron Lett. 1995, 36, 8137-8140.
(10) Pogodin, S.; Biedermann, P. U.; Agranat, I. J. Org. Chem. 1997,
62, 2285-2287.
(11) (a) Bratcher, M. S.; Scott, L. T. Abstracts of Papers, 207th National
Meeting of the American Chemical Society, San Diego, CA.; American
Chemical Society: Washington, DC; March, 1994; ORGN 420. (b) Scott,
L. T.; Bratcher, M. S. 8th International Symposium on Novel Aromatic
Compounds, Braunschweig, Germany, July, 1995, abstr. no. 7. (c) Bratcher,
M. S. Ph.D. Dissertation, Boston College, 1996. (d) See also ref 1c.
(16) Bennett, M. A.; Neumann, H. Aust. J. Chem. 1980, 33, 1251-1259.
1430
Org. Lett., Vol. 2, No. 10, 2000