Rhodium-catalyzed asymmetric cyclodimerization of oxabenzonorbornadienes and azabenzonorbornadienes: Scope and limitations

Article abstract of DOI:10.1021/jo7012884

(Chemical Equation Presented) Cationic rhodium(I)-catalyzed cyclodimerization of oxabenzonorbornadienes produced naphtho[1,2-b]-furan ring systems in a single step with excellent yields and excellent enantioselectivities. The effect of various Rh(I) catalysts, Ag(I) salts, solvents, and phosphine ligands on the yield and enantioselectivity of the reaction was investigated, and the scope and limitations of this reaction with various oxabicyclic alkenes were studied. Similar results were obtained with the azabenzonorbornadiene analogues, providing the corresponding cyclodimerization products in excellent yields and excellent enantioselectivities.

Full text of DOI:10.1021/jo7012884

Rhodium-Catalyzed Asymmetric Cyclodimerization of
Oxabenzonorbornadienes and Azabenzonorbornadienes: Scope and
Limitations
Anna Allen, Paul Le Marquand, Ryan Burton, Karine Villeneuve, and William Tam*
Guelph-Waterloo Centre for Graduate Work in Chemistry and Biochemistry, Department of Chemistry,
UniVersity of Guelph, Guelph, Ontario, Canada N1G 2W1
wtam@uoguelph.ca
ReceiVed June 19, 2007
Cationic rhodium(I)-catalyzed cyclodimerization of oxabenzonorbornadienes produced naphtho[1,2-b]-
furan ring systems in a single step with excellent yields and excellent enantioselectivities. The effect of
various Rh(I) catalysts, Ag(I) salts, solvents, and phosphine ligands on the yield and enantioselectivity
of the reaction was investigated, and the scope and limitations of this reaction with various oxabicyclic
alkenes were studied. Similar results were obtained with the azabenzonorbornadiene analogues, providing
the corresponding cyclodimerization products in excellent yields and excellent enantioselectivities.
Introduction
1a and found that, depending on the reaction conditions, several
products (2-6) could be obtained (Scheme 1). For example,
when oxabenzonorbornadiene 1a is treated with an alkyne in
the presence of the ruthenium catalyst, Cp*Ru(COD)Cl, a [2
+ 2] cycloaddition is observed and cyclobutene cycloadduct 2
is formed.3 When oxabenzonorbornadiene 1a is treated with the
secondary propargylic alcohol 7 in the presence of the neutral
Ru catalyst, Cp*Ru(COD)Cl, in MeOH or using a cationic Ru
catalyst (e.g., [CpRu(CH3CN)3]PF6), isochromene 3 is formed.4
On the other hand, if the same reaction between 1a and 7 is
carried out with Cp*Ru(COD)Cl in THF, cyclopropane 4 is
produced.5 More recently, we have observed that in the absence
of an alkyne, Cp*Ru(COD)Cl catalyzes the isomerization of
1a to the corresponding naphthalene oxide 5 or naphthol 6.6
Oxabicyclic alkenes are valuable synthetic intermediates as
they can serve as a general template to create highly substituted
ring systems. For instance, asymmetric ring opening of these
alkenes allows for the formation of several stereocenters in a
single step.1 They are also useful building blocks in molecular
architecture.2 We have recently examined different aspects of
ruthenium-catalyzed reactions involving oxabenzonorbornadiene
(1) For selected examples, see: (a) Lautens, M.; Fagnou, K.; Heibert,
S. Acc. Chem. Res. 2003, 36, 48. (b) Lautens, M.; Hiebert, S.; Renaud,
J.-L. J. Am. Chem. Soc. 2001, 123, 6834. (c) Lautens, M.; Hiebert, S. J.
Am. Chem. Soc. 2004, 126, 1437. (d) Lautens, M.; Fagnou, K.; Yang, D. J.
Am. Chem. Soc. 2003, 125, 14884. (e) Wu, M.-S.; Jeganmohan, M.; Cheng,
C.-H. J. Org. Chem. 2005, 70, 9545. (f) Nakamura, M.; Matsuo, K.; Inoue,
T.; Nakamura, E. Org. Lett. 2003, 5, 1373. (g) Zhang, W.; Wang, L.-X.;
Shi, W.-J.; Zhou, Q.-L. J. Org. Chem. 2005, 70, 3734. (h) Cabrera, S.;
Arraya´s, R. G.; Alonso, I.; Carretero, J. C. J. Am. Chem. Soc. Chem. 2005,
127, 17938. (i) Li, M.; Yan, X.-X.; Zhu, X.-Z.; Cao, B.-X.; Sun, J.; Hou,
X.-L. Org. Lett. 2004, 6, 2833 and references therein.
(2) For selected examples of molecular architecture, see: (a) Girreser,
U.; Giuffrida, D.; Kohnke, F. H.; Mathias, J. P.; Philp, D.; Stoddart, J. F.
Pure Appl. Chem. 1993, 65, 119. (b) Warrener, R. N.; Margetic, D.; Sun,
G.; Amarasekara, A. S.; Foley, P.; Butler, D. N.; Russell, R. A. Tetrahedron
Lett. 1999, 40, 4111. (c) Warrener, R. N.; Butler, D. N.; Margetic, D.;
Pfeffer, F. M.; Russell, R. A. Tetrahedron Lett. 2000, 41, 4671. (d) Warrener,
R. N. Eur. J. Org. Chem. 2000, 3363. (e) Warrener, R. N.; Margetic, D.;
Foley, P.; Butler, D. N.; Winling, A.; Beales, K. A.; Russell, R. A.
Tetrahedron 2001, 57, 571. (f) Dalphond, J.; Bazzi, H. S.; Kahrim, K.;
Sleiman, H. F. Macromol. Chem. Phys. 2002, 203, 1988. (g) Lautens, M.;
Colucci, J. T.; Hiebert, S.; Smith, N. D.; Bouchain, G. Org. Lett. 2002, 4,
1879.
(3) For our recent studies of Ru-catalyzed [2 + 2] cycloadditions of
bicyclic alkenes and alkynes, see: (a) Jordan, R. W.; Tam, W. Org. Lett.
2000, 2, 3031. (b) Jordan, R. W.; Tam, W. Org. Lett. 2001, 3, 2367. (c)
Jordan, R. W.; Tam, W. Tetrahedron Lett. 2002, 43, 6051. (d) Villeneuve,
K.; Jordan, R. W.; Tam, W. Synlett 2003, 2123. (e) Villeneuve, K.; Tam,
K. Angew. Chem., Int. Ed. 2004, 43, 610. (f) Villeneuve, K.; Riddell, N.
G.; Jordan, R. W.; Tsui, G.; Tam, W. Org. Lett. 2004, 6, 4543. (g) Jordan,
R. W.; Khoury, P. R.; Goddard, J. D.; Tam, W. J. Org. Chem. 2004, 69,
8467. (h) Riddell, N. G.; Villeneuve, K.; Tam, W. Org. Lett. 2005, 7, 3681.
(i) Riddell, N. G.; Tam, W. J. Org. Chem. 2006, 71, 1943. (j) Liu, P.;
Jordan, R. W.; Goddard, J. D.; Tam, W. J. Org. Chem. 2006, 71, 3793. (k)
Jordan, R. W.; Villeneuve, K.; Tam, W. J. Org. Chem. 2006, 71, 5830.
(4) Villeneuve, K.; Tam, W. Eur. J. Org. Chem. 2006, 5499.
(5) Villeneuve, K.; Tam, W. Organometallics 2006, 25, 843.
(6) Villeneuve, K.; Tam, W. J. Am. Chem. Soc. 2006, 128, 3514.
10.1021/jo7012884 CCC: $37.00 © 2007 American Chemical Society
Published on Web 09/20/2007
J. Org. Chem. 2007, 72, 7849-7857
7849

Allen et al.
SCHEME 1. Ru-Catalyzed Reactions of
Oxabenzonorbornadiene 1a
FIGURE 1. Examples of natural products that contain the naphtho-
[1,2-b]furan ring system.
ring system is present in a number of biologically important
natural products (Figure 1),10 and multiple step syntheses are
usually required to generate such a ring system.11 More
importantly, use of a chiral ligand would possibly produce a
chiral naphtho[1,2-b]furan ring system, making this novel
reaction an especially attractive synthetic method. In fact, when
(R)-BINAP was employed instead of (()-BINAP, cyclodimer-
ization product 8a was formed in 96% ee, vide infra.
SCHEME 2. Rh-Catalyzed Cyclodimerization of
Oxabenzonorbornadiene 1a
In recent years, there have been several previous examples
of the construction of fused ring systems through dimerization
reactions.12 In 2002, Junjappa and co-workers developed an
expedient route to the 1H-cyclopenta[c]carbazole framework
through a cascade of reactions that included the dimerization
of the simple indole precursors (eq 1).12a Around the same time,
Isaccs and co-workers devised a synthetic route to methylene
bridged glycoluril dimers, using a variety of glycoluril deriva-
Since the naphthalene oxide 5 produced contains two stereo-
genic centers, nucleophilic ring opening of this epoxide would
produce a useful synthetic intermediate.6 An attempt to inves-
tigate the asymmetric version of this reaction by the addition
of chiral ligands to various Ru catalysts resulted in either no
reaction or no asymmetric induction. Since certain rhodium
catalysts are known to generate naphthol 6 from oxabenzonor-
bornadiene 1a,1a we decided to investigate the use of Rh
catalysts instead of Ru catalysts. When oxabenzonorbornadiene
1a was treated with [RhCl(COD)]2 (2 mol %), AgBF4 (4 mol
%), and (()-BINAP (4 mol %) in 1,2-dichloroethane (DCE) at
60 °C, cyclodimerization product (()-8a was formed in 95%
isolated yield in 15 min as a single diastereomer, instead of the
expected naphthalene epoxide (Scheme 2). Concurrently, the
same transformation was discovered by Hayashi and co-workers,
using a similar catalytic system (1 mol % [RhCl(R)-BINAP]2
and 2 mol % NaBArF4 in DCE at 40 °C, where ArF ) 3,5-bis-
(trifluoromethyl)phenyl).7 The structure and relative stereo-
chemistry of (()-8a was established by NMR experiments (1D
GOESY and 2D COSY)8 and confirmed by X-ray crystal
structure analysis.9 The absolute stereochemistry was determined
through comparison to Hayashi’s published results. This unex-
pected transformation provides a novel and very efficient method
for the construction of the naphtho[1,2-b]furan ring system
(Scheme 2, highlighted in red). This type of naphtho[1,2-b]furan
(10) For examples of natural products containing a naphtho[1,2-b]furan
ring system, see: (a) Otani, T.; Sugimoto, Y.; Aoyagi, Y.; Igarashi, Y.;
Furumai, T.; Saito, N.; Yamada, Y.; Asao, T.; Oki, T. J. Antibiot. 2000,
53, 337. (b) Corey, E. J. J. Am. Chem. Soc. 1955, 77, 1044. (c) Marco, J.
A.; Sanz, J. F.; Falco, E.; Jakupovic, J.; Lex, J. Tetrahedron 1990, 46, 7941.
(d) Hoeksama, H.; Kruger, W. C. J. Antibiot. 1976, 29, 704. (e) Uchiyama,
N.; Kiuchi, F.; Ito, M.; Honda, G.; Takeda, Y.; Khodzhimatov, O. K.;
Ashurmetov, O. A. J. Nat. Prod. 2003, 66, 128. (f) Tomoda, H.; Tabata,
N.; Masuma, R.; Si, S.-Y.; Omura, S. J. Antibiot. 1998, 51, 618. (g) Butler,
M. S.; Capon, R. J. Aust. J. Chem. 1993, 46, 1363. (h) Thomas, R. Pure
Appl. Chem. 1973, 34, 515. (i) da Costa, G. M.; de Lemos, T. L. G.; Pessoa,
O. D. L.; Monte, F. J. Q.; Braz-filho, R. J. Nat. Prod. 1999, 62, 1044. (j)
Gaspar-Marques, C.; Simoes, M. F.; Rodriguez, B. J. Nat. Prod. 2005, 68,
1408.
(11) (a) Clive, D. L. J.; Fletcher, S. P. Chem. Commun. 2003, 2464. (b)
Kawamata, T.; Nagashima, K.; Nakai, R.; Tsuji, T. Synth. Commun. 1996,
26, 139. (c) Kraus, G. A.; Chen, L. Synlett 1991, 89. (d) Emrich, D. E.;
Larock, R. C. J. Organomet. Chem. 2004, 689, 3756. (e) Marino, J. P.;
Laborde, E.; Paley, R. S. J. Am. Chem. Soc. 1988, 110, 966. (f) Maertens,
F.; Toppet, S.; Compernolle, F.; Hoornaert, G. J. Eur. J. Org. Chem. 2004,
2707. (g) Clive, D. L. J.; Fletcher, S. P.; Liu, D. J. Org. Chem. 2004, 69,
3282. (h) Brimble, M. A.; Elliott, R. J.; Turner, P. Tetrahedron 1998, 54,
14053. (i) Takeya, T.; Doi, H.; Ogata, T.; Okamoto, I.; Kotani, E.
Tetrahedron 2004, 60, 9049. (j) Noland, W. E.; Kedrowski, B. L. J. Org.
Chem. 1999, 64, 596.
(12) For examples of the creation of fused ring systems through
dimerization reactions, see: (a) Venkatesh, C.; Ila, H.; Junjappa, H.; Mathur,
S.; Huch, V. J. Org. Chem. 2002, 67, 9477. (b) Wu, A.; Chakraborty, A.;
Will, D.; Lagona, J.; Damkaci, F.; Ofori, M. A.; Chiles, J. K.; Fettinger, J.
C.; Isaacs, L. J. Org. Chem. 2002, 67, 5817. (c) Ohkita, M.; Sano, K.;
Suzuki, T.; Tsuji, T. Tetrahedron Lett. 2001, 42, 7295. (d) Abakumov, G.
A.; Druzhkov, N. O.; Kurskii, Y. A.; Abakumova, L. G.; Shavyrim, A. S.;
Fukin, G. K.; Poddel’skii, A. I.; Cherkasov, V. K.; Okhlopkova, L. S. Russ.
Chem. Bull., Int. Ed. 2005, 54, 2571. (e) Klys, A.; Czardybon, W.;
Warkentin, J.; Werstuik, N. H. Can. J. Chem. 2004, 82, 1769. (f) Abdou,
W. M.; Kamel, A. A. Tetrahedron 2000, 56, 7573.
(7) (a) This reaction was discovered by the Ph.D. student Karine
Villeneuve on May 9, 2006, and the results were presented in her Ph.D.
thesis which was successfully defended on October 20, 2006: Villeneuve,
K. Ph.D. thesis, University of Guelph, 2006. (b) During the preparation of
this manuscript, a paper closely related to this study (using a slightly different
catalyst precursor) appeared: Nishimura, T.; Kawamoto, T.; Sasaki, K.;
Tsurumaki, E.; Hayashi, T. J. Am. Chem. Soc. 2007, 129, 1492.
(8) GOESY: Gradient enhanced nuclear Overhauser enhancement
spectroscopy. See: (a) Stonehouse, J.; Adell, P.; Keeler, J.; Shaka, A. J. J.
Am. Chem. Soc. 1994, 116, 6037. (b) Stott, K.; Stonehouse, J.; Keeler, J.;
Hwang, T.-L.; Shaka, A. J. J. Am. Chem. Soc. 1995, 117, 4199. (c) Dixon,
A. M.; Widmalm, G.; Bull, T. E. J. Magn. Reson. 2000, 147, 266.
(9) Lough, A. J.; Allen, A.; Tam, W. Acta Crystallogr. 2007, E63,
o1462-o1463.
7850 J. Org. Chem., Vol. 72, No. 21, 2007

Rhodium-Catalyzed Asymmetric Cyclodimerization
tives (eq 2).12b Tsuji and co-workers achieved the dimerization
of naphtho[2,3-c]tropone through photochemical treatment of
the naphthocyclobutene valence isomer (eq 3).12c These are just
a few examples of the dimerization strategies available to create
fused ring systems.
TABLE 1. Optimization of Rhodium Catalysts
entry
Rh(I)
time (h)
yieldd (%)
1
2
[RhCl(COD)]2
[RhCl(CO)2]2
[RhCl(PPh3)3]
[Rh(COD)2][BF4]
[Rh(COD)2][OTf]
[RhCl(CH2)CH2)2]2
[RhCl(NBD)]2
0.25
0.25
18
0.25
0.25
0.25
1
95
79
83
93
93
92
91
0c
3a
4a,b
5a,b
6
7
8
[RhCl(C8H14)2]2
0.25
a 4 mol % of Rh catalyst was used. b No AgBF4 was used. c Obtained
43% of 9. d Isolated yield after column chromatography.
SCHEME 3. Generation of Oxabenzonorbornenol 9 with
Chlorobis(cyclooctene)rhodium(I) Dimer
TABLE 2. Optimization of Silver Salts
entry
AgX
time (h)
yieldb (%)
1
2
3
4
AgBF4
AgNO3
AgSbF6
AgOTf
0.25
8
0.25
0.25
95
72a
88
80
Results and Discussion
Rhodium-Catalyzed Cyclodimerization of Oxabenzonor-
bornadiene 1a. To explore the scope and limitation of this
reaction, we have studied the effect of various Rh(I) catalysts,
Ag(I) salts, solvents, and phosphine ligands. Control experiments
were initially conducted, and it was found that all components
are necessary for the success of this reaction. Omission of a
Rh(I) complex leads to no reaction, while use of a neutral
Rh(I) complex, rather than cationic, or exclusion of the phos-
phine ligand results in decomposition of the starting material.
Results by Hayashi also confirm the necessity of the cationic
Rh(I) species.7
a Obtained 12% of 1-naphthol 10. b Isolated yield after column chro-
matography.
SCHEME 4. Generation of 1-Naphthol Side Product
After screening through a variety of Rh(I) complexes, it was
found that all but one of the chosen catalysts, in their cationic
form, promoted the cyclodimerization in good to excellent yields
(Table 1). In the case of the chlorobis(cyclooctene)rhodium(I)
dimer (entry 8), instead of obtaining the cyclodimerization
product 8a, the reaction provided an alternate isomerized dimer,
alcohol 9, in 43% yield (Scheme 3). It should also be noted
that chlorodicarbonylrhodium(I) dimer and Wilkinson’s catalyst
(entries 2-3) both resulted in slightly lower yields of 79 and
83%, respectively. With the neutral Rh(I) complexes, the catonic
forms of the catalysts were generated by the addition of the
Ag(I) salt (entries 1-3 and 6-8). Through the evaluation of
the Rh(I) catalysts, it can be seen that the best yields were
obtained for those catalysts bearing at least one 1,5-cycloocta-
diene (COD) ligand (entries 1, 4, 5). In most of the cases, all
starting oxabenzonorbornadiene 1a was consumed within 15
min. The chloronorbornadienerhodium(I) dimer was slightly
more sluggish than the majority, requiring a reaction time of 1
h (entry 7), while Wilkinson’s catalyst was significantly slower
requiring 18 h for the complete consumption of 1a (entry 3).
To test the effect of the counterion of the cationic rhodium
catalyst, four common Ag(I) salts were evaluated. All four
resulted in the cyclodimerization product 8a in good to excellent
yields (Table 2). The silver tetrafluoroborate salt, used in the
original catalytic system, provided a superior yield compared
to the other three salts. It should also be noted that in the case
of silver nitrate, the reaction also resulted in the isomerization
of 1a to yield 12% of 1-naphthol 10 (Scheme 4). The reaction
with silver nitrate was also found to be much slower than with
the other three silver salts, requiring 8 h instead of 15 min to
go to completion. Slightly lower yields were obtained with silver
J. Org. Chem, Vol. 72, No. 21, 2007 7851

Allen et al.
TABLE 3. Optimization of Solvents
alkylphosphine ligand evaluated, tricyclohexylphosphine (entry
5), showed only decomposition of starting materials.
Through the studies described above, it was concluded that
although various conditions were found to promote the cyclo-
dimerization of oxabenzonorbornadiene 1a, the optimal condi-
tions for the racemic cyclodimerization are [RhCl(COD)]2 (2
mol %), AgBF4 (4 mol %), (()-BINAP (4 mol %), in
1,2-dichloroethane (0.5 M) at 60 °C.
Asymmetric Rhodium-Catalyzed Cyclodimerization of
Oxabenzonorbornadiene 1a. With the optimal conditions for
the racemic cyclodimerization of oxabenzonorbornadienes
determined, the focus was shifted to the development of the
asymmetric version. Using the optimized reaction conditions
for the cyclodimerization of oxabenzonorbornadiene 1a, we
simply substituted (R)-BINAP for the (()-BINAP to test the
initial asymmetric version. We were pleased to find that, along
with an excellent yield, the chiral BINAP ligand provided high
enantioselectivity (96% ee, Table 5, entry 1).
entry
solvent
time (h)
yieldb (%)
1
2
3
4
5
6
7
8
1,2-dichloroethane
toluene
0.25
0.25
0.25
0.25
0.25
0.25
0.25
18
95
91
92
93
83
79
91
29a
THF
dichloromethane
methanol
DMF
acetone
hexanes
a Obtained 20% of 1-naphthol 10. b Isolated yield after column chro-
matography.
TABLE 4. Optimization of Nonchiral Ligands
TABLE 5. Optimization of Chiral Ligands
entry
phosphine12
time (h)
yieldd (%)
1
2
3
4
5
6
(()-BINAP
dppf
dppe
0.25
19.5
18
18
18
95
entry
phosphine13
(R)-BINAP
(R)-Tol-BINAP
(R)-3,5-xylyl-BINAP
(R,R)-Me-Duphos
(S,S)-Chiraphos
(R)-Prophos
(R,S)-Josiphos
(S,S)-DIOP
(S,S)-Norphos
(S)-Phanephos
(R,R)-Me-BPE
Trost ligand
time (h)
yieldf (%)
eeg (%)
46a
0b,c
36a
0
1
2
3
4
5
6
7
8
9
0.5
0.5
2
21
2.5
2
24
22.5
20
21
21.5
23
96
86
96
96
PPh3
PCy3
0a,d
93b
45
P(biphenyl)tBu2
18
0a,b
68
60c
40
37
4
a Recovered 14-42% of 1a. b Obtained 17-48% of 1-naphthol 10.
c Obtained 22% alcohol 9. d Isolated yield after column chromatography.
41b
23b,d
28b
0d
trifluoromethanesulfonate than with silver tetrafluoroborate when
using the chloro(1,5-cyclooctadiene)rhodium(I) dimer as the
catalyst; interestingly, upon using the similar cationic bis(1,5-
cyclooctadiene)rhodium(I) complex, both counterions provide
the same yield (Table 1, entries 4-5).
In general, solvent had only a modest effect on the reaction,
and good to excellent yields of dimer 8a were obtained within
15 min (Table 3). Both methanol and DMF (entries 5-6)
provided slightly lowered yields of 83 and 79% respectively,
although the reaction was still complete within 15 min. All other
entries provided yields of 91% or higher, with the exception of
hexanes (entry 8). Both the starting oxabenzonorbornadiene 1a
and the catalyst dissolved very poorly in hexanes resulting in a
drastically lowered yield of 8a and a significantly increased
reaction time. In addition, 20% of 1a isomerized to give
1-naphthol 10.
10
11
12
0b,d
0b,d
0e
a Obtained 25% of alcohol 9. b Recovered 17-61% of 1a. c Opposite
configuration. d Obtained 3-57% of 1-naphthol 10. e No reaction observed.
f Isolated yield after column chromatography. g The enantiomeric excess
(ee) of 8a was determined by HPLC using a Chiralcel OD-H column.
In order to determine if there was a more suitable ligand for
the asymmetric reaction, a variety of chiral ligands were
evaluated (Table 5).13 As the (R)-BINAP ligand provided
excellent initial results, we were inclined to test other BINAP
ligands. While (R)-p-Tol-BINAP gave identical enantioselec-
tivities as the parent BINAP ligand, the yield was diminished
slightly to 86% (entry 2). When changing the ligand to the (R)-
3,5-xylyl-BINAP, we found that no cyclodimerization product
8a was obtained; instead, up to 25% of the isomerized dimer,
alcohol 9, and 57% of 1-naphthol 10 were produced (entry 3).
In contrast to the previously discussed reaction components,
the phosphine ligand showed considerable influence on the
cyclodimerization of oxabenzonorbornadiene 1a. Both mono-
and bidentate phosphine ligands were evaluated (Table 4). It
was found that the BINAP ligand provided significantly better
results than any other ligands studied (entry 1). The only other
ligands to provide the cyclodimerization product 8a were
diphenylphosphinoferrocene (entry 2) and triphenylphosphine
(entry 4); however, these reactions were plagued by low yields
and a lack of complete starting material consumption even after
18 h. Diphenylphosphinoethane (entry 3) and 2-(di-tert-butyl-
phosphino)biphenyl (entry 6) both resulted in the isomerization
of 1a to yield 1-naphthol 10, while the diphenylphosphinoethane
also provided the isomerized dimer, alcohol 9. The only
(13) Abbreviations for various phosphine ligands used: dppf, 1,1′-bis-
(diphenylphosphino)ferrocene; dppe, 1,2-bis(diphenylphosphino)ethane; (R)-
BINAP, (R)-(+)-2,2′-bis(diphenylphosphino)-1,1′-binaphthyl; (R)-Tol-
BINAP, (R)-(+)-2,2′-bis(di-p-tolylphosphino)-1,1′-binaphthyl; (R)-3,5-xylyl-
BINAP, (R)-(+)-2,2′-bis(di(3,5-xylyl)phosphino)-1,1′-binaphthyl; (R,R)-Me-
Duphos, (-)-1,2-bis((2R,5R)-2,5-dimethylphospholano)benzene; (S,S)-
Chiraphos: (2S,3S)-(-)-bis(diphenylphosphino)butane; (R)-Prophos, (R)-
(+)-1,2-bis(diphenylphosphino)propane; (R,S)-Josiphos, (R)-(-)-1-[(S)-2-
(diphenylphosphino)ferrocenyl]ethyldicyclohexylphosphine ethanol adduct;
(S,S)-DIOP, (4S,5S)-(+)-4,5-bis(diphenylphosphinomethyl)-2,2′-dimethyl-
1,3-dioxolane; (S,S)-Norphos, (2S,3S)-(+)-2,3-bis(diphenylphosphino)-
bicyclo[2.2.1]hept-5-ene; (S)-phanephos, (S)-(+)-4,12-bis(diphenylphosphino)-
[2.2]paracyclophane; (R,R)-Me-BPE, (+)-1,2-bis((2R,5R)-2,5-dimethyl-
phospholano)ethane; (R)-Trost ligand: (1R,2R)-(+)-1,2-diaminocyclohex-
ane-N,N′-bis(2′-diphenylphosphinobenzoyl).
7852 J. Org. Chem., Vol. 72, No. 21, 2007

Rhodium-Catalyzed Asymmetric Cyclodimerization
Chiral ligands with other architectures were also screened.
(R,R)-Me-Duphos provided the cyclodimerization product 8a
in an excellent yield of 93%; however, compared to (R)-BINAP,
the reaction was significantly sluggish and considerably less
enantioselective (Table 5, entry 4). Several other chiral ligands
also provided 8a, however, with much lower yields and
selectivities, ranging from 28 to 45% and 4-60% ee, respec-
tively (entries 5-8). In addition, in all of these cases the
oxabenzonorbornadiene 1a was not completely consumed. For
(S,S)-Chiraphos and (R)-Prophos, reaction times of 24 h were
required for complete consumption of the starting material.
Extended reaction times also caused the resultant cyclodimer-
ization product 8a to isomerize to alcohol 9. Taking heed of
this observation, the reaction times were shortened to maximize
the yield of 8a while leaving some oxabenzonorbornadiene 1a
unreacted (entries 5 and 6). For (R,S)-Josiphos (entry 7) and
(S,S)-DIOP (entry 8), starting material still remained despite
allowing the reaction to proceed for up to 24 h. In the case of
(R,S)-Josiphos, a small amount of 1-naphthol 10 was also
obtained.
Other chiral ligands were evaluated, but the reaction did not
provide the cyclodimerization product 8a (Table 5, entries
9-12). Total starting material consumption within 20 h for (S,S)-
Norphos (entry 9) provided only 1-naphthol 10 in 35% yield.
Both (S)-Phanephos (entry 10) and (R,R)-Me-BPE (entry 11)
yielded 1-naphthol 10 and unconsumed starting material after
nearly 24 h of reaction time. In the case of (S)-Phanephos, a
large amount of decomposition was also obtained. The final
chiral ligand investigated, the Trost ligand (entry 12), resulted
in no reaction as only starting oxabenzonorbornadiene 1a was
observed with no sign of decomposition.
As temperature can have a profound effect on the enantiose-
lectivity of a chiral system, various temperatures were inves-
tigated for the most promising chiral ligands (Table 6). The
chiral system utilizing the (R)-BINAP ligand was evaluated at
60, 40, 25, and 0 °C, while the (R)-p-Tol-BINAP ligand was
evaluated at 60, 40, and 25 °C. The results of these experiments
showed that temperature had very little effect on the enantio-
selectivity of the asymmetric cyclodimerization of oxabenzonor-
bornadiene 1a. Cyclodimerizations completed at lower temper-
atures required progressively longer reaction times.
Through the screening of various chiral ligands, the optimal
conditions for the asymmetric cyclodimerization of oxaben-
zonorbornadiene 1a was found to be the initial chiral catalytic
system: [RhCl(COD)]2 (2 mol %), AgBF4 (4 mol %), and (R)-
BINAP (4 mol %), in 1,2-dichloroethane (0.5 M) at 60 °C.
Scope of the Reaction. With the optimized conditions for
the enantioselective cyclodimerization of oxabenzonorborna-
diene 1a in hand, the scope of the reaction was then investigated
using various oxabicyclic substrates. The benzobicyclic alkenes
evaluated were prepared through Diels-Alder cycloadditions
between furan or pyrroles and the appropriate benzyne species.
In each case, the benzynes were generated in situ from either
anthranilic acid14a,b,15a or halogenated benzenes.14c-g,15b We
found that the reaction was general for most oxabenzonorbor-
nadienes that contain substituents on the benzene ring. A wide
range of yields from 37 to 95% was obtained; however, the
enantioselectivities were high, ranging from 88 to 98% ee (Table
7, entries 2-8). In comparison to results reported by Hayashi,
divergent results were seen for the dimerizations of 1e and 1f.
In our study, the yields of these reactions were noticeably lower,
and for 1e, the enantioselectivity was slightly diminished.
Interesting reactivity was observed for some substrates not
reported by Hayashi. Surprisingly, oxabenzonorbornadiene 1d
did not provide the corresponding cyclodimerization product,
and only 1-naphthol 11a was obtained in 69% yield (Table 7,
entry 4; Scheme 5). This result was not expected as oxaben-
zonorbornadiene 1c (entry 3), which only differs in the
placement of the methoxy substituents, easily underwent the
cyclodimerization to provide the corresponding product 8c with
excellent yield and enantioselectivity. Thus, the position of the
substituents on the benzene ring can crucially modify the
outcome of this reaction. The naphthol obtained was determined
to be the 1-naphthol through conversion to the 1-acetylnaphthol
11b and subsequent NMR analysis (GOESY).8 It is also
noteworthy that oxabenzonorbornadiene 1g provided the cor-
TABLE 7. Rh-Catalyzed Asymmetric Cyclodimerization of
Benzo-Substituted Oxabenzonorbornadienes
entry
substrate
product
time (h)
yieldc (%)
eed (%)
1
2
3
4
5
6
7
8
1a
1b
1c
1d
1e
1f
8a
8b
8c
8d
8e
8f
0.5
0.5
0.5
0.5
0.5
0.25
0.5
1
96
77
94
0a
46
79
37b
95
96
91
98
88
97
98
94
1g
1h
8g
8h
TABLE 6. Effect of Temperature on Enantiomeric Ratio
a Obtained 69% of 11a when using (()-BINAP. b Obtained 32% of 12.
c Isolated yield after column chromatography. d The enantiomeric excess
(ee) of 8 was determined by HPLC using a Chiralcel OD-H column.
entry
phosphine
T (°C)
time (h)
yielda (%)
eeb (%)
(14) For preparations of oxabenzonorbornadienes, see: (a) Nakayama,
J.; Sakai, A.; Hoshino, M. J. Org. Chem. 1984, 49, 5084. (b) Friedman, L.;
Logullo, F. M. J. Org. Chem. 1969, 34, 3089. (c) Jung, K. Y.; Masato, K.
J. Org. Chem. 1989, 54, 5667. (d) Lautens, M.; Fagnou, K.; Yang, D. J.
Am. Chem. Soc. 2003, 125, 14884. (e) Lai, Y.-H.; Yong, Y.-L.; Wong,
S.-Y. J. Org. Chem. 1997, 62, 4500. (f) Caster, K. C.; Keck, C. G.; Walls,
R. D. J. Org. Chem. 2001, 66, 2932. (g) Gribble, G. W.; LeHoullier, C. S.;
Sibi, M. P.; Allen, R. W. J. Org. Chem. 1985, 50, 1611. (h) Hart, H.; Bashir-
Hashemi, B.; Luo, J.; Meador, M. A. Tetrahedron 1986, 42, 1641.
(15) For preparations of azabenzonorbornadienes, see: (a) Lautens, M.;
Fagnou, K.; Zunic, V. Org. Lett. 2002, 4, 3645. (b) Cho, Y.-H.; Zunic, V.;
Senboku, H.; Olsen, M.; Lautens, M. J. Am. Chem. Soc. 2006, 128, 6837.
1
2
3
4
5
6
7
(R)-BINAP
(R)-BINAP
(R)-BINAP
(R)-BINAP
(R)-Tol-BINAP
(R)-Tol-BINAP
(R)-Tol-BINAP
60
40
25
0
60
40
25
0.5
1.5
5
22
0.5
2
96
97
90
98
86
92
98
96
96
96
96
96
97
97
2.5
a Isolated yield after column chromatography. b The enantiomeric excess
(ee) of 8a was determined by HPLC using a Chiralcel OD-H column.
J. Org. Chem, Vol. 72, No. 21, 2007 7853

Allen et al.
SCHEME 8. Cyclodimerization Attempts of Non-Benzo
SCHEME 5. Cyclodimerization Attempt of
Oxabenzonorbornadiene 1d
Oxabicyclic Alkenes
SCHEME 6. Cyclodimerization of Oxabenzonorbornadiene
1g
TABLE 8. Evaluation of the Cyclodimerization of Unsymmetrical
Oxabenzonorbornadienes
SCHEME 7. Cyclodimerization Attempt of
Oxabenzonorbornadiene 13
entry
substrate
time (h)
yield 20a (%)
recovered 19a (%)
1
2
3
19a
19b
19c
26
1.5
26
30
70
20
70
0
70
responding cyclodimerization product 8g with excellent enan-
tioselectivity (98% ee) in only 37% yield (entry 7). This low
yield can be attributed to the generation of a second dimeric
product, naphthol dimer 12, which was obtained in 32% yield
(Scheme 6).
a Isolated yield after column chromatography.
SCHEME 9. Generation of Naphthols 22 and 23 from
Unsymmetrical Oxabenzonorbornadiene 21
Along with oxabenzonorbornadiene substrates with substi-
tuted benzenes, we investigated the cyclodimerization of other
symmetrical oxanorbornadienes. We attempted the cyclodimer-
ization of C1,C4-substituted oxabenzonorbornadiene 13; how-
ever, no dimer product was observed and only naphthol 14 was
isolated in 86% yield (Scheme 7). As only oxabenzonorborna-
diene substrates had been investigated, we decided to test the
applicability of the reaction to some non-benzo oxabicyclic
alkenes. Oxanorbornadiene 15 and oxanorbornene 16 were both
subjected to the optimized reaction conditions, but it was found
that neither substrate provided the desired cyclodimerization
product (Scheme 8). Oxanorbornadiene 15 resulted in two
different dimer products, phenol 17 and cyclobutane 18, obtained
in 9 and 39% yields, respectively, also seen in the Hayashi study.
Interestingly, phenol 17 has a structure very similar to that of
fluorinated naphthol dimer 12 and most likely results through
a similar mechanism. In the cases of bicyclic alkenes 1g and
15, it seems that the electron-withdrawing abilities of the
fluorinated benzene ring of 1g and the two ester moieties of
15 allowed the reaction to proceed through an alternate path-
way, providing the corresponding naphthol or phenol dimer-
ization product. In contrast, oxanorbornene 16 did not provide
any kind of dimer, and 39% starting material was recovered
after 32 h.
bearing a substituent at the C1 bridgehead position, simply
isomerized, leading to the isolation of the corresponding
1-naphthol and, in some cases, recovered oxabenzonorborna-
diene (Table 8; Scheme 9). In the case of 21, the 2-naphthol 23
was also obtained in a 10% yield (Scheme 9). Two non-benzo
unsymmetrical oxanorbornadiene substrates (24a,b) were also
evaluated (Table 9). In both cases, no dimerization was
observed, quite possibly due to the presence of the C1
substituents. In the case of 24a, which bears a C1 ester moiety
(entry 1), only starting material was isolated after decomposition
was observed. However, oxanorbornadiene 24b, with a C1
methyl substituent, provided the corresponding 1- and 2-naph-
thols (25b and 26b, respectively) in a 4:1 ratio (25b/26b) and
65% combined yield.
Along with the symmetrical oxabicyclic alkenes evaluated,
a number of unsymmetrical oxabicyclic alkenes were also
investigated using the optimized cyclodimerization conditions.
Unsymmetrical oxabenzonorbornadienes 19a-d and 21, each
7854 J. Org. Chem., Vol. 72, No. 21, 2007

Rhodium-Catalyzed Asymmetric Cyclodimerization
TABLE 9. Evaluation of the Cyclodimerization of Unsymmetrical
Oxanorbornadienes
Inspired by the isolation of the alternate isomerized dimer,
oxabenzonorbornenol 9, from the cyclodimerization product
8a, we attempted to generate this species from the isolated
cyclodimerization product. Although disappointing conver-
sion was obtained under basic conditions (no reaction with
n
NaOMe, 20% yield with BuLi), we were pleased to find that
subjecting 8a to p-toluenesulfonic acid provides 9 in quanti-
tative yield with no degradation of the enantiopurity (Scheme
11).
entry
substrate
time (h)
yield 25 + 26a (%)
0b
SCHEME 11. Isomerization of the Cyclodimerization
Product 8a to Oxabenzonorbornenol 9
1
2
24a
24b
1.5
1.5
65c
a Isolated yield after column chromatography. b Recovered 75% of 24a.
c Products isolated in a 4:1 (25/26) ratio.
We then turned our attention to the cyclodimerization of
azabenzonorbornadienes (Table 10), providing a wider scope
than reported by Hayashi. All substrates evaluated, with the
exception of one, provided the corresponding cyclodimerization
product 28 in moderate to good yields (54-88%) and with
excellent selectivities (94-99% ee). Azabenzonorbornadiene
27b (entry 2), the only azabicyclic alkene evaluated that did
not bear an ester moiety on the bridging nitrogen, did not
successfully undergo the cyclodimerization reaction. After 23
h at 60 °C, 76% of azabenzonorbornadiene 27b was recovered
with no apparent product formation; heating to 80 °C for 50 h
produced 1-naphthyltosylamine 29 in 30% yield (Scheme 10).
A plausible mechanism has been devised for the formation
of the cyclodimerization product B1, and the side products A1
(or A2) and B2 (Scheme 12). The active catalyst, [Rh(BINAP)]+,
is formed through treatment of the precatalyst, [RhCl(COD)]2,
with AgBF4 to remove the chloride and then addition of the
BINAP ligand, which replaces the COD ligand. The active
catalyst is chelated by 1 to form complex 30. Strain in the
oxabicyclic structure is then released through an oxidative
insertion of the rhodium into the C-O bond to form complex
31. This intermediate is in equilibrium with π-allyl complex
32 and metallocycle 33. As rhodium has the ability to coordinate
five species, this allows a second molecule of 1 to coordinate
to complex 33 (pathway B), generating complex 35. A carbo-
metalation can then occur to form the six-membered metallo-
cycle 36. It is at this point where the mechanistic pathway
diverges to provide either the dimerization product B1 or
naphthol B2. Reductive elimination (pathway B1) of metallocycle
36 would yield the cyclodimerization product B1. However, a
â-hydride elimination (pathway B2) can also occur with met-
allocycle 36, which leads to the rhodium hydride species 37.
Reductive elimination of 37 would lead to ketone 38 and
regeneration of the active catalyst. Compound 38 will then
rearomatize via keto-enol tautomerism to provide naphthol B2.
The corresponding naphthol side products A1 (or A2) could be
formed according to pathway A, which begins with the reductive
elimination of complex 33 generating epoxide 34. Completion
of the production of naphthols A1 (or A2) from 34 has been
previously described.6
When comparing the reactivity of the oxa- and azabenzonor-
bornadienes toward the Rh-catalyzed cyclodimerization, it can
be seen that the oxabenzonorbornadiene substrates are signifi-
cantly more reactive than their aza analogues. The oxaben-
zonorbornadiene substrates require reaction times of just 15 min
to 1 h, while the azabenzonorbornadienes required 4-47 h to
go to completion. It should also be noted that the less sterically
hindering methyl carbamate 27c (Table 10, entry 3) reacted
significantly faster than the corresponding azabenzonorborna-
diene 27a bearing the Boc moiety (entry 1), which indicates
that sterics may also be at play.
TABLE 10. Rh-Catalyzed Asymmetric Cyclodimerization of
Benzo-Substituted Azabenzonorbornadienes
The formation of naphthol B2 was only observed during the
cyclodimerization of oxabenzonorbornadiene 1g, providing 12.
Through control experiments, it was determined that 12 was
formed directly from 1g. Dimerization product 8g was subjected
to the both the catalyst conditions and acidic conditions, and in
either case, no further reaction was observed.
entry
substrate
product
time (h)
yieldb (%)
eec (%)
98
1
2
3
4
5
27a
27b
27c
27d
27e
28a
28b
28c
28d
28e
20
23
4
47
16
88
0a
76
54
61
98
99
94
a Recovered 76% of 27b. Heating at 80 °C for 50 h produced 30%
1-naphthyltosylamine 29. b Isolated yield after column chromatography.c The
enantiomeric excess (ee) of 28 was determined by HPLC using a Chiralcel
OD-H column.
Cyclodimerization product B1 could also be formed through
an alternate mechanism, proposed by Hayashi and co-workers.7b
In this mechanism, two molecules of 1 coordinate to the active
catalyst, forming 39, which undergoes an oxidative cyclization
to form metallocycle 40. Subsequent â-oxygen elimination
provides metallocycle 36, which upon reductive elimination
provides B1. Hayashi has also proposed a mechanism for the
formation of 17. Their proposed pathway is analogous to what
we have proposed for the formation of dimerization product B2
(pathway B2).
SCHEME 10. Generation of 1-Naphthyltosylamine 29
J. Org. Chem, Vol. 72, No. 21, 2007 7855

Allen et al.
SCHEME 12. Proposed Mechanisms
Cyclodimerization Product 8a (Table 5, Entry 1). Performed
according to the general procedure using oxabenzonorbornadiene
1a (40.9 mg, 0.284 mmol) and heating for 15 min. The crude
product was purified by column chromatography (1:4 EtOAc/
hexanes, Rf ) 0.49) to obtain 8a as an off-white solid (39.2 mg,
0.136 mmol, 96%): mp 164-165 °C; HPLC using a Chiralcel
OD-H column, retention times of 24.90 min (minor enantiomer)
and 28.79 min (major enantiomer), 96% ee; [R]23D +208 (c ) 0.57,
CHCl3); 1H NMR (CDCl3, 400 MHz) δ 7.37 (d, J ) 7.2 Hz, 1H),
7.32-7.23 (m, 4H), 7.20-7.12 (m, 3H), 6.62 (dd, J ) 9.9, 2.3 Hz,
1H), 6.03 (dd, J ) 9.9, 3.0 Hz, 1H), 5.43 (s, 1H), 5.21 (s, 1H),
5.01 (d, J ) 7.2 Hz, 1H), 4.36 (d, J ) 6.4 Hz, 1H), 3.31-3.24 (m,
1H), 2.94 (dd, J ) 9.6, 6.5 Hz, 1H); 13C NMR (APT, CDCl3, 100
MHz) δ 146.0, 143.0, 132.4, 131.5, 128.8 (2), 127.4, 127.2, 127.1,
126.8, 126.5, 126.0, 120.5, 119.4, 83.1, 83.0, 81.6, 79.1, 52.4, 39.2;
IR (NaCl, cm-1) 3049 (m), 3030 (m), 3003 (m), 2951 (m), 2894
(m), 2850 (m), 1492 (m), 1457 (m), 1320 (w), 1264 (w), 1216 (w),
1061 (s); HRMS (EI) calcd for C20H16O2 (M+) 288.1150, found
288.1146. Characterized also by 1D GOESY, 2D COSY NMR
experiments and X-ray crystal structure analysis.
Conclusion
In summary, we have demonstrated that the asymmetric Rh-
catalyzed cyclodimerization of oxa- and azabicyclic alkenes can
be applied to a wide variety of benzonorbornadiene substrates
achieving moderate to excellent yields and excellent enantiose-
lectivities. In our studies, we have also discovered classes of
substrates that fail to undergo this reaction, such as those bearing
substituents on the bridgehead carbons and non-benzo oxabi-
cyclic alkenes. We have also found substrates bearing strong
electron-withdrawing substituents may show differing reactivity
and undergo an alternate dimerization pathway.
Experimental
Only representative procedures and characterization of the
products are described here. Full details can be found in the
Supporting Information.
General Procedure for Rh-Catalyzed Dimerization of Bicyclic
Alkenes. Inside an inert atmosphere (Ar) glovebox, [RhCl(COD)]2
(0.02 equiv, 0.00555 mmol) and AgBF4 (0.04 equiv, 0.0111 mmol)
were added to an oven-dried vial, dissolved in 1,2-dichloroethane
(0.15 mL) and allowed to stir for 20 min. (R)-BINAP (0.04 equiv,
0.0111 mmol) was added to the vial as a solid followed by 1,2-
dichloroethane (0.15 mL), and the mixture was stirred for an
additional 20 min. The bicyclic alkene (1.0 equiv, 0.277 mmol)
was then added to the vial as a solid followed by 1,2-dichloroethane
(0.30 mL). The mixture was heated to 60 °C and allowed to stir
for 15 min to 48 h. The crude mixture was purified by column
chromatography.
Cyclodimerization Product 8g (Table 7, Entry 7, R1 ) R2 )
F, Scheme 6). Performed according to the general procedure using
oxabenzonorbornadiene 1g (60.9 mg, 0.282 mmol) and heating for
30 min. The crude product was purified by column chromatography
to obtain 8g (1:9 EtOAc/hexanes, Rf ) 0.32): 22.3 mg, 0.052 mmol,
37%, white solid, mp 194-196 °C dec, HPLC using a Chiralcel
OD-H column, retention times of 11.67 min (major enantiomer)
and 13.76 min (minor enantiomer), 98% ee; [R]24 ) +133 (c )
D
1
0.28, CHCl3); H NMR (CDCl3, 400 MHz) δ 6.81 (d, J ) 10.1
Hz, 1H), 6.13 (dd, J ) 10.1, 2.4 Hz, 1H), 5.70 (s, 1H), 5.47 (s,
7856 J. Org. Chem., Vol. 72, No. 21, 2007

Rhodium-Catalyzed Asymmetric Cyclodimerization
1H), 5.26 (d, J ) 7.3 Hz, 1H), 4.42 (d, J ) 6.4 Hz, 1H), 3.35 (m,
1H), 3.06 (dd, J ) 9.5, 6.5 Hz, 1H); 13C NMR (APT, CDCl3, 100
MHz, observed signals) δ 147.3 (apt d), 144.7 (apt t), 142.2 (m),
141.3 (m), 140.7 (m) 139.7 (m), 138.7 (m), 138.2 (m), 127.4, 127.2
(minor enantiomer) and 13.14 min (major enantiomer), 98% ee;
[R]25D ) +47.0 (c ) 0.52, CHCl3); 1H NMR (acetone-d6, 400 MHz,
40 °C) δ 7.45-6.85 (m, 8H), 6.57 (br s, 1H), 5.94 (dd, J ) 9.7,
5.5 Hz, 1H), 5.52 (d, J ) 11.2 Hz, 1H), 4.82 (s, 1H), 4.75 (br s,
1H), 4.17 (br s, 1H), 3.23 (m, 1H), 2.55 (apt t, J ) 7.4 Hz, 1H),
1.57 (s, 9H), 0.99 (s, 9H); 13C NMR (APT, HSQC, acetone-d6,
100 MHz, 40 °C) δ 205.8 (2), 155.6, 137.3, 132.3 (2), 129.7, 128.6,
127.6, 127.4, 127.2, 127.0, 124.8, 122.2, 121.3, 120.5, 80.2, 79.1,
67.4, 64.6, 63.7, 60.1, 53.5, 40.0, 28.8, 28.2; IR (NaCl, cm-1) 3055
(w), 2976 (m), 2931 (m), 1693 (s), 1478 (m), 1456 (m), 1289 (w),
1252 (w), 1229 (w), 1169 (s), 1152, 1113 (m), 941 (w), 785 (w),
752 (w), 737 (w), 702 (w). Anal. Calcd for C30H34N2O4: C, 74.05;
H, 7.04. Found: C, 74.41; H, 6.98.
Oxabenzonorbornenol 9 (Scheme 11). To an oven-dried, round-
bottom flask were added 8a (200.1 mg, 0.694 mmol), p-toluene-
sulfonic acid monohydrate (55.7 mg, 0.293 mmol), and 1,2-
dichloroethane (3.5 mL). The mixture was stirred under argon for
5.5 h and then concentrated by rotary evaporation. The crude
product was purified using column chromatography (1:4 EtOAc/
hexanes, Rf ) 0.29) to obtain 9 as a white solid (199.8 mg, 0.693,
99%). Spectroscopic data for 9 was in agreement with values
reported in the literature:7b 1H NMR (CDCl3, 400 MHz) δ 7.93 (s,
1H), 7.89 (m, 3H), 7.50 (m, 3H), 7.41 (m, 1H), 7.28 (m, 3H), 5.60
(s, 1H), 5.35 (s, 1H), 4.31 (apt t, J ) 7.8 Hz, 1H), 3.44 (d, J ) 7.0
Hz, 1H), 1.58 (d, J ) 8.8 Hz, 1H); 13C NMR (APT, CDCl3, 75
MHz) δ 146.6, 142.6, 135.0, 133.5, 132.6, 128.6, 127.8, 127.6,
127.54, 127.48 (2), 127.0, 126.4, 126.0, 120.9, 119.1, 86.0, 84.5,
74.2, 51.8; IR (NaCl, cm-1) 3555 (w), 3436 (w), 3052 (w), 2999
(w), 2941 (w), 1632 (w), 1600 (w), 1507 (w), 1460 (w), 1347 (w),
1269 (w), 1213 (w), 1191 (w), 1153 (w), 1068 (m), 995 (w), 982
(w), 942 (w), 907 (w), 826 (m), 810 (m), 784 (w), 757 (m), 741
(m); HRMS (CI) calcd for C20H16O2 (M+) 288.1150, found
288.1140.
2
2
(d, JC-F ) 17.9 Hz), 124.6 (d, JC-F ) 18.8 Hz), 116.8, 116.7,
114.2 (2JC-F ) 13.7 Hz), 81.8, 80.8, 76.9, 73.7, 51.6, 38.3; 19F
NMR (CDCl3, 377 MHz) δ -147.9 (apt t, J ) 20.4 Hz, 1F), -149.3
(apt t, J ) 20.6 Hz, 1F), -152.7 (dd, J ) 13.6, 20.5 Hz, 1F), -155.4
(dd, J ) 13.4, 20.5 Hz, 1F), -160.7 (apt t, J ) 19.8 Hz, 1F), -161.1
(apt t, J ) 19.7 Hz, 1F), -162.8 (apt t, J ) 19.2 Hz, 1F), -164.1
(apt t, J ) 20.3 Hz, 1F); IR (NaCl, cm-1) 2958 (m), 2925 (m),
2854 (m), 1507 (m), 1488 (m), 1466 (w), 1423 (w), 1387 (w), 1378
(w), 1292 (w), 1125 (w), 1073 (w), 1046 (w), 1012 (w), 975 (w),
944 (w), 877 (w), 823 (w); HRMS (CI) calcd for C20H8F8O2 (M+)
432.0397, found 432.0390.
Perfluoronaphthol 12 (Scheme 6). Performed according to the
general procedure using oxabenzonorbornadiene 1g (60.9 mg, 0.282
mmol) and heating for 30 min. The crude product was purified by
column chromatography to obtain 12 (1:9 EtOAc/hexanes, Rf )
0.51): 19.5 mg, 0.045 mmol, 32%, white solid, mp 156-157 °C,
HPLC (Chrial Technologies Chiralcel OB-H column, 99:1 hexanes/
2-propanol, 0.05 mL/min, 220 nm) retention times of 194.86 min
(major enantiomer) and 235.11 min (minor enantiomer), 93% ee;
1
[R]26 ) +17.9 (c ) 0.27, CHCl3); H NMR (CDCl3, 400 MHz)
D
δ 7.86 (d, JH-F ) 9.9 Hz, 1H, OH), 7.56 (dd, J ) 8.7, 1.6 Hz, 1H),
7.49 (d, J ) 8.8 Hz, 1H), 5.95 (m, 1H), 5.70 (d, JH-F ) 1.6 Hz,
1H), 3.46 (dd, J ) 4.7, 8.7 Hz, 1H), 2.29 (m, 1H), 2.20 (m, 1H);
13C NMR (APT, CDCl3, 75 MHz) δ 148.6 (m), 144.9 (m), 143.7
(m), 141.7 (m), 140.4 (m), 139.4 (m), 138.5 (m), 136.0 (m)
2
(preceding signals account for 8 ipso-F C), 128.8, 127.0 (d, JC-F
) 17.8 Hz), 126.9 (d, 2JC-F ) 17.0 Hz), 124.9 (ipso-OH C), 120.7
2
(d, JC-F ) 13.7 Hz), 111.9, 110.6 (m), 81.7, 77.8, 40.3, 35.2; IR
(NaCl, cm-1) 3174 (m), 2958 (w), 2924 (m), 2854 (w), 1665 (w),
1622 (w), 1591 (w), 1524 (w), 1507 (m), 1490 (m), 1463 (w), 1411
(m), 1333 (w), 1319 (w), 1309 (w), 1287 (w), 1273 (w), 1252 (w),
1184 (w), 1136 (w), 1114 (w), 1075 (w), 1043 (w), 1020 (w), 987
(w), 972 (w), 923 (w), 903 (w), 876 (w), 854 (w), 816 (w), 804
(w), 761 (w), 719 (w); HRMS (CI) calcd for C20H8F8O2 (M+)
432.0397, found 432.0395. Characterized also by HSQC and 2D
COSY NMR experiments.
Acknowledgment. This work was supported by NSERC
(Canada) and Boehringer Ingelheim (Canada). W.T. thanks
Boehringer Ingelheim (Canada) Ltd. for a Young Investigator
Award, A.A., P.L., and K.V. thank NSERC for postgraduate
scholarships (CGS M, PGS M, and CGS D), and R.B. thanks
the Ontario Government for a postgraduate scholarship (OGS).
Cyclodimerization Product 28a (Table 10, Entry 1, R1 ) R2
) H, R3 ) Boc). Performed according to the general procedure
using azabenzonorbornadiene 27a (67.6 mg, 0.278 mmol) and
heating for 20 h. The crude product was purified by column
chromatography (1:4 EtOAc/hexanes, Rf ) 0.39) to obtain 28a as
a light brown solid (59.6 mg, 0.122 mmol, 88%): mp 70 °C dec;
HPLC using a Chiralcel OD-H column, retention times of 8.83 min
Supporting Information Available: Detailed experimental
procedures and full characterization of all new compounds. This
material is available free of charge via the Internet at http://
pubs.acs.org.
JO7012884
J. Org. Chem, Vol. 72, No. 21, 2007 7857