Electrooxidative coupling of furans and silyl enol ethers: Application to the synthesis of annulated furans

Article abstract of DOI:10.1021/jo049889i

The preparation of annulated furan systems as key synthetic intermediates through the application of a two-step annulation involving an electrochemical ring closure between a furan and a silyl enol ether has been studied. The reaction was shown to be quite general for the formation of six-membered rings in good yields and was tolerant of a variety of different functional groups. The ring closure was highly stereoselective, leading to the formation of cis-fused systems. Cyclic voltammetry and probe molecules were used to gain mechanistic insight into the reaction. These studies suggested that the key ring closure involved an initial oxidation of the silyl enol ether to a radical cation followed by a furan-terminated cyclization.

Full text of DOI:10.1021/jo049889i

Electr ooxid a tive Cou p lin g of F u r a n s a n d Silyl En ol Eth er s:
Ap p lica tion to th e Syn th esis of An n u la ted F u r a n s
J effrey B. Sperry,‡ Christopher R. Whitehead,§,† Ion Ghiviriga,§,‡ Ryan M. Walczak,§ and
Dennis L. Wright*
Department of Chemistry, Dartmouth College, Hanover, New Hampshire 03755, and
Department of Chemistry, University of Florida, Gainesville, Florida 32611
dennis.l.wright@dartmouth.edu
Received J anuary 19, 2004
The preparation of annulated furan systems as key synthetic intermediates through the application
of a two-step annulation involving an electrochemical ring closure between a furan and a silyl enol
ether has been studied. The reaction was shown to be quite general for the formation of
six-membered rings in good yields and was tolerant of a variety of different functional groups. The
ring closure was highly stereoselective, leading to the formation of cis-fused systems. Cyclic
voltammetry and probe molecules were used to gain mechanistic insight into the reaction. These
studies suggested that the key ring closure involved an initial oxidation of the silyl enol ether to
a radical cation followed by a furan-terminated cyclization.
SCHEME 1. Som e Im p or ta n t Tr a n sfor m a tion s of
F u r a n
In tr od u ction
Simple furan derivatives are readily available and
highly reactive and can serve as direct precursors to a
wide range of different structural moieties.1 Because of
this high degree of synthetic flexibility, furans have
frequently been employed in the total synthesis of natural
products.2The furan ring is highly nucleophilic and
undergoes the typical range of electrophilic aromatic
substitution reactions. However, the furan ring can also
participate in reactions that lead to an overall dearoma-
tization of the nucleus (Scheme 1).
In addition to the products of reduction and oxidation
(2 and 3), oxabicyclo[2.2.1]heptene 4 and oxabicyclo[3.2.1]-
octene derivatives 5 are produced through cycloaddition
reactions.3 We have been especially interested4 in these
bridged systems since they can be readily transformed
into a variety of carbocylic and oxacyclic systems by
cleavage of one of the bridges in the bicyclic structures.
Application of these cycloadditions to annulated furans
would be expected to produce much more complex
systems than those routinely prepared from simple furan
derivatives. Although the synthesis of annulated furans
SCHEME 2. Gen er a l Str a tegy for th e Assem bly of
An n u la ted F u r a n s
has been the focus of several investigations5 and has
found some use in total synthesis,6 general methods for
their construction are lacking. We have been interested
in developing simple methodologies for the rapid as-
sembly of annulated furans for use in the synthesis of
terpenoid natural products. Along these lines, we envi-
sioned a two-step sequence involving the initial conjugate
addition of a furyl appended organometallic 7 to an R,â-
unsaturated carbonyl compound 6 with in situ silylation
to deliver an intermediate silyl enol ether 8 (Scheme 2).
The second step of the ring-forming procedure would
involve the coupling of the R-carbon of the enol ether to
* Corresponding author.
‡ To whom questions regarding NMR structures should be addressed
at Dartmouth College.
§ University of Florida.
† Deceased August 7, 2003.
(1) Lisphutz, B. H. Chem. Rev. 1986, 86, 795.
(2) Wright, D. L. Chem. Innovation 2001, 31, 17.
(3) (a) For a review of furan [4+2] reactions see: Kappe, C. O.;
Murphree, S. S.; Padwa, A. Tetrahedron 1997, 53, 14179. (b) For a
review of furan [4+3] reactions see: Rigby, J . H.; Pigge, F. C. [4+3]
Cycloaddition Reactions. In Organic Reactions; Paquette, L. A., Ed.;
J ohn Wiley & Sons: New York, 1997; Vol. 51, p 351.
(4) (a) Orugunty, R. S.; Wright, D. L.; Battiste, M. A.; Abboud, K.
A. Org. Lett. 2002, 4, 1997. (b) Usher, L. C.; J imenez-Estrella, M.;
Ghiviriga, I.; Wright, D. L. Angew. Chem., Int. Ed. 2002, 4742. (c)
Wright, D. L.; Usher, L. C.; J imenez-Estrella, M. Org. Lett. 2001, 26,
4275.
(5) Padwa, A.; Murphree, S. S. Org. Prep. Proced. Int. 1991, 23, 545.
(6) For some recent examples see: (a) Aso, M.; Ojida, A.; Yang, G.;
Cha, O.-J .; Osawa, E.; Kanematsu, K. J . Org. Chem. 1993, 58, 3960.
(b) Lee, K.; Cha, J . K. J . Am. Chem. Soc. 2001, 123, 5590. (c) Lee, J .
C.; Cha, J . K. Tetrahedron 2000, 56, 10175-84. (d) J acobi, P. A.; Lee,
K.-A. J . Am. Chem. Soc. 1997, 119, 3409.
10.1021/jo049889i CCC: $27.50 © 2004 American Chemical Society
Published on Web 04/28/2004
3726
J . Org. Chem. 2004, 69, 3726-3734

Electrooxidative Coupling of Furans and Silyl Enol Ethers
TABLE 2. Effect of Cu r r en t Den sity a n d Electr od e
Ma ter ia l on Cycliza tion Efficien cy
F IGURE 1. Targets for furan annulation strategy.
current
entrya
anode
cathode
densityb
yield (%)
TABLE 1. Oxid a tive Cycliza tion Stu d ies
1
2
3
4
5
6
7
8
9
carbon
carbon
carbon
carbon
carbon
platinum
platinum
RVC
steel
steel
steel
steel
carbon
carbon
carbon
carbon
carbon
0.5
1.0
5.0
10
0.5
0.5
0.1
N/A
N/A
68
66
35
10
67
27
54
30
35
entry
oxidant/conditionsa
observations yield of 13
1
2
3
4
5
6
2.5 equiv of CAN/MeCN
10 equiv of CAN/MeCN
2 equiv of Mn(OAc)3/Et2O
2 equiv of VO(OCH2CF3)Cl2 decomposition
2 equiv of Ar3NSbCl6
carbon anode (1.5 V)
hydrolysisb
decomposition
hydrolysisb
N/A
N/A
N/A
N/A
68%
76%
RVC
a
Cyclizations were run on a 100-mg scale under the optimized
conditions. Expressed as mA/cm2 and obtained by dividing the
b
current passed by the surface area of the electrode.
cyclization
cyclization
a
been shown to oxidize electron-rich olefins, were ineffec-
tive in promoting cyclization to 13. It was possible to
effect cyclization with tris(4-bromophenyl)aminium hexa-
chloroantimonate (BAHA),15 a stable radical cation popu-
larized by Bauld.16 Exposure of the silyl enol ether to 2
equiv of the aminium salt promoted smooth oxidation to
the annulated furan 13. Although this reagent would
effectively promote the oxidative cyclization, it was very
difficult to remove the amine byproducts from the reac-
tion and the reagent is fairly expensive. The best result
for the conversion of enol ether 12 to the annulated furan
was through the use of direct anodic oxidation on a
carbon anode.17 Moeller18 and others19 have shown that
both furans and enol ethers can be oxidized to the
corresponding radical cations under anodic conditions.
Our optimized reaction conditions involve the use of a
carbon anode in an acetonitrile solution containing
lithium perchlorate as the supporting electrolyte, 2,6-
lutidine as an acid scavanger, and 2-propanol as the
cation trap. Methanol is traditionally used as the cation
trap but these substrates underwent rapid methanolysis
to yield the uncyclized ketones. However, the silyl enol
ethers are stable for several days in the presence of the
more sterically hindered 2-propanol. Another critical
factor in these cyclizations was the current density of the
anode. This is defined as the amount of charge passed
per square centimeter of anode surface (Table 2).
100 mg of the enol ether was treated with the oxidizing agent
at room temperature and the reaction monitored by GC-MS and
b
TLC. Corresponding ketone was isolated.
the 2-position of the pendant heterocycle. This overall
process is an oxidation and involves the formal loss of
an equivalent of hydrogen. The ring closure involves the
coupling of two nucleophilic carbons and requires an
umpolong of one of these carbons to an electrophilic
species through oxidation.
We have been interested in the application of this
annulation strategy to the synthesis of terpenoid natural
products such as erinacine C7 and frullanolide8 (Figure
1). In both cases, an A-ring enone would be annulated
by the above strategy, using a furyl ethyl chain that
would close the six-membered B-rings and install a furan
as a latent form of the C-rings. A [4+3] cycloaddition can
be used to construct the seven-membered C-ring of the
erinacine C9 while oxidation of the furan to a butenolide
and methylenation would give the C-ring lactone of
frullanolide.10
Resu lts a n d Discu ssion
We initially examined the cyclization of a model enol
ether 12, derived from the copper-catalyzed addition of
furylethylmagnesium bromide11 to 3-methylcyclohex-
enone, under a variety of oxidative conditions (Table 1).
Traditional chemical oxidants such as CAN,12 manga-
nese triacetate,13 or vanadium oxide,14 all of which have
A study on the ring closure of the enol ether 14, derived
from cyclohexenone, showed that as the current density
was increased the yield of the cyclized product 15
decreased markedly. Attempted cyclizations at higher
current densities produced large amounts of high molec-
(7) Kawagishi, H.; Shimada, A.; Shiraj, R.; Okamoto, K.; Ojima, F.;
Sakamoto, H.; Ishiguro, Y.; Furukawa, S. Tetrahedron Lett. 1994, 35,
1569.
(8) Perold, G. W.; Ourisson, G.; Muller, J . C. Tetrahedron 1972, 28,
5797.
(9) Wright, D. L.; Whitehead, C. R.; Sessions, E. H.; Ghiviriga, I.;
Frey, D. A. Org. Lett. 1999, 1, 1535.
(13) Snider, B. B. Chem. Rev. 1996, 96, 339.
(14) Ryter, K.; Livinghouse, T. J . Am. Chem. Soc. 1998, 120, 2668.
(15) Walter, R. I. J . Am. Chem. Soc. 1966, 88, 1923.
(16) (a) Gao, D. J .; Bauld, N. L. J . Org. Chem. 2000, 65, 6276. (b)
Bauld, N. L.; Gao, D. J . Chem. Soc., Perkin Trans. 2 2000, 931.
(17) (a) Wright, D. L.; Whitehead, C. R.; Sessions, E. H.; Ghiviriga,
I.; Frey, D. A. Org. Lett. 1999, 1, 1535. (b) Whitehead, C. R.; Sessions,
E. H.; Ghiviriga, I.; Wright, D. L. Org. Lett. 2002, 3763.
(18) (a) Moeller, K. D.; New, D. G. Tetrahedron Lett. 1994, 35, 2857.
(b) New, D. G.; Tesfai, Z.; Moeller, K. D. J . Org. Chem. 1996, 61, 1578.
(19) For an excellent review see: Moeller, K. D. Tetrahedron 2000,
56, 9527.
(10) For
a related strategy see: Tanis, S. P.; Robinson, E. D.;
McMills, M. C.; Watt, W. J . Am. Chem. Soc. 1992, 114, 8349.
(11) Prepared from 3-bromofuran by lithiation and reaction with
ethylene oxide followed by conversion to the bromide and then the
Grignard reagent. See the Experimental Section for details.
(12) (a) Baciocchi, E.; Casu, A.; Ruzziconi, R. Tetrahedron Lett. 1989,
30, 3707. (b) Snider, B. B.; Shi, B.; Quickley, C. A. Tetrahedron 2000,
56, 10127.
J . Org. Chem, Vol. 69, No. 11, 2004 3727

Sperry et al.
TABLE 3. Electr och em ica l Cycliza tion s
a
b
Yield is reported for two steps based on the enone as the limiting agent. Isolated as an inseparable 4:1 mixture of trans:cis isomers.
ular weight materials. This was attributed to the forma-
tion of polymers at the surface of the electrode. Since the
current density is believed to determine the amount of
electroactive species on the surface, higher current densi-
ties would favor bimolecular reactions.20 Interestingly,
reticulated vitrious carbon electrodes,21 which have a very
high surface area because of the porous nature of the
material, were very inefficient at promoting the cycliza-
tion. It was found that the use of a carbon anode and a
stainless steel cathode at a current density of 0.5 mA/
cm2 was the optimal arrangement and all reactions
discussed below were performed under these conditions.
We began our studies of this annulation process by
examining the reaction with a variety of simple enone
starting materials that were devoid of any functionality
that may be competitively oxidized under the electro-
chemical conditions. These studies revealed that the ring
(20) For example, the yield of Kolbe dimerization is enhanced at
high current densities: Coleman, J . P.; Lines, R.; Utley, J . H. P. J .
Chem. Soc., Perkin Trans. 2 1973, 1064.
(21) Frey, D. A.; Wu, N.; Moeller, K. D. Tetrahedron Lett. 1996, 37,
8317.
3728 J . Org. Chem., Vol. 69, No. 11, 2004

Electrooxidative Coupling of Furans and Silyl Enol Ethers
TABLE 4. Electr och em ica l Cycliza tion of F u n ction a lized Der iva tives
a
b
Yield is reported for two steps based on the enone as the limiting agent. Prepared according to ref 23. c Prepared according to ref
24. All other enones are commercially available
closure was fairly general and proceeded in a highly
stereoselective manner (Table 3).
completion of the reaction, the crude mixture was par-
titioned with dilute acid to break down any intermediate
acetals that might form. Isolation of the compounds gave
the annulated furans as single diastereomers, except in
the case of 17g.
Under the optimized conditions, it was found that the
silyl enol ethers 14 and 16a -h could be oxidatively
cyclized to yield the corresponding annulated furans 15
and 17a -h in very good yields (58-78% over two steps)
after a mild acidic workup.22 There were also some
important differences between substrates in terms of the
initial conjugate addition reaction. As expected, addition
to those enones without a â-alkyl substituent was con-
siderably faster than that to those that were â,â-disub-
stituted at the enone. However, silylation was often
incomplete in these cases and resulted in significant
amounts of the nonsilylated enones being formed. Alter-
natively, the â,â-disubstituted enones, although slower
to react, were more easily silylated and no intermediate,
nonsilylated compounds were observed. The addition of
TMEDA as an additive to the conjugate addition reaction
accelerated the silylation resulting in silyl enol ethers
being produced cleanly and in high yield. The crude silyl
enol ethers were utilized directly in the electrochemical
cyclizations without the benefit of further purification.
The oxidations were conducted at a constant current,
using a carbon anode and a stainless steel cathode. Upon
Stu d ies on F u n ction a l Gr ou p Toler a n ce. Our
initial studies revealed that the cyclizations were very
efficient and highly stereoselective with regard to the ring
closure. We next turned to examining the tolerance of
various functional groups to the annulation conditions.
It is possible to oxidize additional functional groups under
the present electrochemical conditions. However, one
advantage of electrochemical reactions is the ability to
tightly control the potential and thus limit the reaction
to only the most easily oxidized or reduced functionality
in a given molecule. To examine the potential for control-
ling the oxidation reactions, we prepared five additional
enol ether substrates with additional functionality and
exposed them to the annulation conditions (Table 4).
These studies demonstrate that the electrooxidative
annulation can be carried out on substrates bearing
potentially reactive functionality. Isolated olefins, acetals,
esters, aryl ethers, and carbamates were all tolerated by
the electrochemical conditions.
Str u ctu r a l Assign m en t of An n u la tion P r od u cts.
When the annulation was onto a preexisting five- or six-
membered ring, the only diastereomer formed was that
resulting from a cis ring closure. The stereochemistry of
the ring closures was assigned exclusively through NMR.
(22) We have found that the furyl ketone products are very
susceptible to air oxidation at the activated bridgehead position and
cannot be stored for long periods of time. Purging of chromatography
and other solvents improves the yield by 10-15%. Subsequent reduc-
tion or ketalization of the ketone eliminates this problem.
J . Org. Chem, Vol. 69, No. 11, 2004 3729

Sperry et al.
1
F IGURE 2. Numbering scheme and H and 13C (in bold) NMR
assignments for adduct 20c.
SCHEME 3. P ossible Mech a n ism s of Oxid a tive
Rin g Closu r e
F IGURE 3. Cyclic voltammograms of model substrates.
SCHEME 4. A Byp r od u ct of th e Electr och em ica l
Cycliza tion
Initial loss of an electron from the silyl enol ether 14
would lead to radical cation 21 that would be attacked
by the pendant furan to generate 22. Trapping of the
oxonium ion with alcohol and the loss of a second electron
from the radical would lead to 24, which generates the
initial product 25 through loss of the silyl group. Alter-
natively, the initial electron loss could occur from the
furan to generate radical cation 26, which would be
subsequently attacked by the nucleophilic silyl enol ether
to produce an intermediate furyl radical 27. Loss of a
second electron would then give the oxonium ion 28,
which is trapped by solvent to give the intermediate 25.
The initial site of oxidation should be governed by the
relative oxidation potentials of the two functional groups
with the most easily oxidized group being that with the
lowest potential. To examine this, silyl enol ether 14 was
studied along with two model substrates by cyclic voltam-
metry (Figure 3).
Model substrate 14 was found to have an oxidation
potential resembling that of simple silyl enol ether 29
rather than a 3-substituted furan 30. The silyl enol ether
of cyclohexanone has an oxidation potential significantly
lower than that of the monosubstituted furan (0.87 V vs
1.31 V as measured against ferrocene). The analysis of
the CV data suggested that the silyl enol ether was more
readily oxidized than the furan. Further evidence that
the silyl enol ether was the primary site of oxidation was
found through the identification of a minor byproduct
(Scheme 4).
The case of compound 20c, a typical example, is discussed
below. The assignment of chemical shifts (shown in
Figure 2) was made on the basis of proton-proton and
proton-carbon couplings seen in the GHMQC, GHMBC,
and DQCOSY spectra. Large vicinal couplings identified
the chemical shifts of the axial protons at positions C4,
C5, C6, C7, and C8 as 2.82, 1.52, 2.91, 2.16, and 2.26,
correspondingly. An nOe between the axial protons at
positions C4 and C6 indicated that the junction of rings
A and B is cis; no such nOe would be expected for the
trans stereochemistry. In the alternative trans configu-
ration, the angular methyl group at C5a would be
positioned between these two axial protons.
Similar strategies were used to determine the relative
stereochemistry of the other adducts shown in Tables 3
and 4.
Mech a n istic Con sid er a tion s. The overall cyclization
reaction leading from the silyl enol ether to the annulated
furan requires the loss of two electrons and two protons.
By eliminating the acidic workup, we have been able to
isolate a dihydrofuran intermediate 25 that results from
the addition of 2-propanol to a furyl cation (see Scheme
9). There are two main mechanistic pathways for the
reaction to follow that are distinguished by the site of
initial electron loss from the cyclization substrate. Al-
though there are several minor variations on these two
mechanistic formulations, the two pathways involve the
same general steps, initiation to form a cationic center
followed by nucleophilic ring closure (Scheme 3).
Enone 31 was identified in the crude reaction mixture
as a minor byproduct from the oxidation of 16h . The
origin of this product was attributed to the loss of a
proton from an intermediate radical cation analogous to
21, which may also rationalize the slightly higher yields
obtained when there is a quaternary center at the
γ-position (see Table 3).
3730 J . Org. Chem., Vol. 69, No. 11, 2004

Electrooxidative Coupling of Furans and Silyl Enol Ethers
SCHEME 5. Rea ctive In ter m ed ia te Der ived fr om
Ver ben on e
SCHEME 6. A Mech a n istic P r obe Molecu le
F IGURE 4. Cyclic voltammograms of probe molecules.
carbinyl cations28 or radicals,29 the ring opening of 36
would be expected to be rapid. However, ring closure to
37 proceeded in excellent yield with no evidence of ring
fragmentation observed. There was some concern that
the absence of fragmentation was due to poor overlap
between the reactive intermediate and the σ bond of the
strained ring. To eliminate this possibility, an acyclic
probe molecule was prepared and subjected to electro-
chemical oxidation (Scheme 7).
SCHEME 7. An Acyclic P r obe Molecu le
Oxidation of the silyl enol ether 39, prepared from
known30 enone 38, led to a good yield (60% over two steps)
of the ring-closed product without evidence of fragmenta-
tion. A possible explanation for these results lies with
the siloxy group at the cyclopropylcarbinyl center stabi-
lizing the reactive intermediate so effectively that ring
opening was not favored. A further extension of this could
involve a rapid loss of the trimethylsilyl group upon
oxidation to directly produce a ketone. Whatever the
nature of the stabilization, it should not be present at
the adjacent carbon, C2. To examine this position,
known31 enone 41 was converted to silyl enol ether 42
and reacted under standard conditions (Scheme 8).
The presumed radical-cation intermediate 43, derived
from oxidation, contains a reactive cyclopropylcarbinyl
radical cation without the benefit of oxygen stabilization.
When 42 was oxidized under electrochemical conditions,
several compounds were produced. Analysis of the crude
reaction mixture indicated that there was no longer a
cyclopropane moiety among the crude materials. Purifi-
cation gave a low yield of ketone 44. This compound is
believed to result from ring opening of 43 followed by
trapping of a cationic intermediate with 2-propanol. The
same types of ring-opened products could be obtained
from BAHA oxidation. Addition of 2 equiv of the aminium
salt to 43 gave low yields of 44 as well as alcohol 45 that
arises from trapping of the cation by adventitious water.
The difference in behavior between C1 and C2 cyclopropyl
substitution can also be seen in the corresponding cyclic
voltammograms (Figure 4).
SCHEME 8. An Alter n a tive P r obe Molecu le
In light of this mechanistic proposal, the uneventful
cyclization of enol ether 16f seemed surprising because
the analogous intermediate would place a radical or
cationic center adjacent to a relatively strained ring
(Scheme 5).
Loss of an electron from the silyl enol ether moiety of
16f would be expected to produce radical cation 32. It
has been documented both in the case of radicals25 and
cations26 that intermediates of this type promote rapid
scission of the bridging isopropylidine system to give
intermediates such as 33. However, no evidence for this
type of fragmentation was observed in the cyclization of
16f. To examine this process in greater detail, a more
reactive probe molecule 35 was prepared from the
known27 enone 34 (Scheme 6).
Oxidation of 35 would be expected to produce 36, which
places a radical cation at the cyclopropylcarbinyl carbon
C1. Although less well studied than either cyclopropyl-
Addition of a cyclopropane at the C2 position as in 42
causes a significant lowering of the oxidation potential
(23) Kerr, W. J .; McLaughlin, M.; Morrison, A. J .; Pauson, P. L. Org.
Lett. 2001, 3, 2945.
(24) Comins, D. L.; Foley, M. A. Tetrahedron Lett. 1988, 29, 6711.
(25) Nguyen, T. M.; Seifert, R. J .; Mowrey, D. R.; Lee, D. Org. Lett.
2002, 4, 3959.
(26) Kusakari, T.; Ichiyanagi, T.; Kosugi, H.; Kato, M. Tetrahedron:
Asymmetry 1999, 10, 339.
(28) Friedric, L. E.; Schuster, G. B. Tetrahedron Lett. 1971, 34, 3171.
(29) Kantorowki, E. J .; Borhan, B.; Nazarian, S.; Kurth, M. J .
Tetrahedron Lett. 1998, 39, 2483.
(30) Gras, J . L. Tetrahedron Lett. 1978, 2955.
(31) Pattenden, G.; Whybrow, D. J . Chem. Soc., Perkin Trans. 1
1981, 3147.
(27) Corey, E. J .; Chaykorsky, M. J . Am. Chem. Soc. 1965, 87, 1353.
J . Org. Chem, Vol. 69, No. 11, 2004 3731

Sperry et al.
addition, the reaction is tolerant of a wide range of
functionality including protected amines. Mechanistic
studies suggest an initial oxidation of the silyl enol ether
followed by a furan-terminated cyclization. Current ef-
forts are focused on the extension of this annulation to
other ring sizes and other heterocycles. The application
of this method to the synthesis of terpenoid natural
products is also ongoing.
F IGURE 5. Numbering scheme and 1H and 13C chemical
shifts in compounds 46a ,b. The arrows represent a long-range
coupling between a proton and a carbon (at the tip).
Exp er im en ta l Section
SCHEME 9. Isola tion of a Dih yd r ofu r a n
In ter m ed ia te
3-(2-Br om oeth yl)fu r a n . A solution of 3-bromofuran33 (5.4
mL, 60.6 mmol) in THF (100 mL) was cooled to -78 °C, then
a solution of n-BuLi (2.6 M, 23.3 mL, 60.6 mmol) was added
slowly via an addition funnel over 30 min. The resulting
solution was stirred for 1 h. A solution of ethylene oxide (3.2
g, 72 mmol) in THF (20 mL) was added over 5 min, and then
BF3‚OEt2 (9.2 mL, 72.7 mmol) was added via addition funnel
over 30 min. The resulting solution was stirred for 5 h.
Saturated NaHCO3 was added slowly over 15 min, and the
semisolid mixture was allowed to warm slowly to room
temperature over 3 h. The aqueous phase was extracted with
EtOAc (2 × 30 mL), and the combined organic phases were
washed with saturated NaHCO3 and brine. The solution was
dried with Na2SO4, and after evaporation of solvent the crude
product was distilled (64-68 °C, 2.5 mmHg) to provide the
alcohol as a clear liquid (4.3 g, 62%) having physical and
spectroscopic properties consistent with those reported. The
alcohol was then converted to the bromide utilizing the method
of Tanis.34
Gen er a l P r oced u r e for Cu p r a te Ad d ition a n d Elec-
tr olysis. A solution of 2-(3-furyl)ethylbromide (1.0 g, 5.8 mmol)
in THF (6.0 mL) was purged with argon while the solution
was stirred for 10 min. This solution was then added to a flask
containing Mg turnings (0.17 g, 7.0 mmol). After 2.5 h the
majority of the Mg had dissolved and a clear dark solution
had formed. The solution was cooled to 0 °C and CuI (0.165 g,
0.87 mmol) was added. After being stirred for 5 min the turbid
black solution was cooled to -78 °C. A 1:1 mixture of TMSCl/
Et3N (6.0 mL) was added, followed by N,N,N,N-tetramethyl-
ethylenediamine (0.95 mL, 5.8 mmol). Dropwise addition of
the enone (4.5 mmol) produced a bright yellow solution. The
mixture was allowed to warm to room temperature over 5 h
and then placed in the refrigerator overnight. The black
mixture was then poured into an ice-cold mixture of hexanes
(100 mL) and saturated aqueous NH4Cl (50 mL). The hexanes
layer was separated and washed with NaHCO3, then brine,
and dried over Na2SO4. Removal of solvent provided the crude
enol ether (1.2 g). A 250-mL beaker was charged with the crude
enol ether and 225 mL of an electrolyte solution (4:1 MeCN:
2-propanol, 0.1 M LiClO4, and 0.08 M 2,6-lutidine) and the
mixture was stirred until homogeneous. An electrode system
(see the Supporting Information for details) consisting of
alternating plates of carbon (anode) and steel (cathode) was
inserted and charged with a constant current of 45.0 mA.
There was ∼90 cm2 of carbon charged as the anode. The
reaction mixture was partially covered with PTFE tape to slow
evaporation of solvent, and after 7.5 h the current was turned
off. The volatiles were removed under vacuum, and the crude
material was dissolved in ether (150 mL) and washed with 1
M HCl. The aqueous phase was extracted with Et2O (2 × 25
mL), and the combined organic phases were washed with
water, NaHCO3, and brine and dried over Na2SO4.
by 0.4 V while a cylcopropane at C1 (35 or 39) alters the
potential little relative to model enol ether 14. These
mechanistic studies suggest that the mechanism occurs
through oxidation of the silyl enol ether function (path
a, Scheme 2). In addition to these studies, it also proved
possible to isolate a key intermediate in the reaction,
dihydrofuran 43 (Scheme 9).
If the standard acidic workup was omitted, a relatively
stable dihydrofuran acetal 46 could be isolated in high
yield without any furan formation. The relative stereo-
chemistry of this compound was determined by NMR
spectroscopy (Figure 5).
Compound 46 was isolated as an equimolar mixture
of diastereomers (46a and 46b). The presence of an
isopropoxy group at C2 and the closure of the six-
membered ring between C9a and C9b were demonstrated
by the proton-carbon long-range couplings seen in the
GHMBC spectrum and presented in Figure 5. The
chemical shifts for positions C4-C8 could not be assigned
due to severe overlap; however, the cross-peak of H9a
and H9b in the DQCOSY spectrum showed that 9a has
a 10 Hz coupling with 9b and a coupling of less than 3
Hz with 5a. These coupling constants indicate that
protons 5a, 9a, and 9b are correspondingly equatorial,
axial, and axial to ring B, therefore the junction of rings
A and B is cis. The coupling constant H2-H9b was 4.0
Hz in 46a and 1.1 Hz in 46b, indicating that these
protons are trans in the former and cis in the latter.32
The exclusive formation of the syn/anti relationship at
the ring junctions suggests that the cyclization proceeds
through a cis-exo transition state such as TS-1 whereby
the furan closes on the radical cation through an exo
orientation.
A general method for the formation of annulated furans
has been developed through a two-step process that
involves an initial conjugate addition of a furyl-substi-
tuted cuprate followed by an electrochemical cyclization.
The reaction has been shown to be effective for the
formation of six-membered rings during the cyclization
and proceeds to give primarily cis-fused products. In
Gen er a l P r oced u r e for Tr is(4-br om op h en yl)a m in iu m
Hexa ch lor oa n tim on a te Cycliza tion . To a solution of tris-
(4-bromophenyl)aminium hexachloroantimonate (295 mg, 0.36
(33) 3-Bromofuran is commercially available or can be prepared from
furan as in: (a) Politis, J . K.; Nemes, J . C.; Curtis, M. D. J . Am. Chem.
Soc. 2001, 123, 2537. (b) Srogi, J .; J anda, M.; Stibor, I. Collect. Czech.
Chem. Commun. 1970, 35, 3478.
(32) (a) Alves, M. J .; Azoia, N. G.; Bickley, J . F.; Fortes, A. G.;
Gilchrist, T. L.; Mendonca, R. J . Chem. Soc., Perkin Trans. 1 2001,
2969. (b) Guizzardi, B.; Mella, M.; Fagnoni, M.; Albini, A. Tetrhedron
2000, 56, 9383.
(34) Tanis, S. P.; Herrinton, P. M. J . Org. Chem. 1985, 50, 3988.
3732 J . Org. Chem., Vol. 69, No. 11, 2004

Electrooxidative Coupling of Furans and Silyl Enol Ethers
mmol) in 10 mL of a 4:1 MeCN:iPrOH solution with 0.2 M 2,6-
lutidine at 0 °C was added the silyl enol ether (0.18 mmol)
dropwise. After 10 min, the solvent was evaporated and the
crude mixture was taken up in 20 mL of 1 M HCl and 15 mL
of diethyl ether. The aqueous layer was extracted thrice and
any solids rinsed with 20 mL of Et2O. The combined organic
layers were washed with 20 mL of saturated sodium bicarbon-
ate and dried over sodium sulfate. The compounds were
purified by silica gel chromatography.
13C NMR δ 20.0, 22.4, 24.7, 25.8, 26.8, 39.2, 45.2, 51.3, 110.5,
116.3, 142.1, 151.3, 211.9; EI-HRMS m/z calcd for C13H16O2
(M+) 204.1150, found 204.1166.
2,2,8a -Tr im eth ylocta h yd r o-1,3-m en th a n on a p h th o[6,7-
b]fu r a n -4-on e (17f). With use of the general procedure as
described above, verbenone (50% ee, 0.031 mL, 0.20 mmol) was
reacted to provide the ketone 17f as a yellow solid (72 mg,
69%) after chromatography: Rf(11:1 hexane:EtOAc) 0.29; mp
121 °C dec; IR (KBr) νmax 2906, 1711, 1633, 1179, 1029 cm-1
;
cis-4,5,5a ,6,7,8a -H e xa h yd r o-1-oxa -a s-in d a ce n -8-on e
(17a ). With use of the general procedure as described
above, 2-cyclopenten-1-one (0.065 mL, 0.78 mmol) was reacted
to give 17a as a pale yellow oil (88 mg, 70%) after chroma-
tography: Rf(10:1 hexanes:EtOAc) 0.30; IR (neat) νmax 2855,
1H NMR δ 7.42 (dd, J ) 1.0, 1.9 Hz, 1H), 6.23 (d, J ) 1.9 Hz,
1H), 3.45 (s, 1H), 2.70 (3 line m, 1H), 2.58 (5 line m, 1H), 2.50
(m, 1H), 2.48 (dd, J ) 1.7, 3.8 Hz, 1H), 2.04 (3 line m, 1H),
1.87 (m, 1H), 1.69 (d, J ) 11.3 Hz, 1H), 1.43 (s, 3H), 1.35 (dt,
J ) 4.3, 3.3 Hz, 1H), 1.24 (d, J ) 1.0 Hz, 3H), 1.23 (s, 3H); 13
C
1
1743, 1627, 1454, 1122, 1094 cm-1; H NMR δ 7.36 (dd, J )
NMR δ 209.0, 146.4, 142.7, 118.5, 110.1, 58.8, 54.2, 50.6, 40.3,
0.6, 2.0 Hz, 1H), 6.21 (d, J ) 2.0 Hz, 1H), 3.38 (d, J ) 7.3 Hz,
1H), 2.75 (m, 1H), 2.51 (m, 2H), 2.36 (m, 2H), 2.12 (m, 1H),
1.90 (m, 2H), 1.68 (m, 1H); 13C NMR δ 215.2, 144.7, 142.6,
118.1, 110.3, 48.3, 36.7, 36.3, 25.7, 25.5, 20.2; EI-HRMS m/z
calcd for C11H12O2 (M+) 176.0837, found 176.0839.
37.8, 34.8, 28.1, 26.8, 25.8, 25.7, 18.7; EI-HRMS m/z calcd for
C
16H20O2 (M+) 244.1463, found 244.1463.
4,5,5a ,6,7,8,9,10a -Octa h yd r o-1-oxa -cycloh ep ta [e]in d en -
10-on e (17 g). With use of the general procedure as described
above, 2-cyclohepten-1-one (0.405 mL, 3.64 mmol) was reacted
to give 17g as a mixture of two isomers in a 4:1 ratio isolated
as a colorless oil (430 mg, 58%) after chromatography. The
two isomers could not be separated but the major isomer could
be characterized by NMR. Major (trans-17g): Rf(9:1 hexanes:
5a -Meth yl-4,5,5a ,6,7,8a -h exa h yd r o-1-oxa -a s-in d a cen -8-
on e (17b). With use of the general procedure as described
above, 3-methylcyclopentenone (0.065 mL, 0.78 mmol) was
reacted to give 17b as a white solid (116 mg, 78%): mp 42-
43 °C; Rf(9:1 hexanes: EtOAc) 0.27; IR (KBr pellet) νmax 2959,
EtOAc) 0.25; IR (neat) νmax 2925, 1711, 1637, 1504, 1158 cm-1
;
1
2862, 1742, 1625, 1550, 1050, cm-1; H NMR δ 7.35 (dd, J )
1H NMR δ 7.29 (dd, J ) 0.9, 1.7 Hz, 1H), 6.21 (d, J ) 1.7 Hz,
1H), 3.92 (d, J ) 9.7 Hz), 2.76 (m, 1H), 2.50 (m, 2H), 2.08-
1.94 (m, 2H), 1.94-1.76 (m, 2H), 1.76-1.55 (m, 2H), 1.38-
1.55 (m, 2H), 1.38-1.20 (m, 2H); 13C NMR δ 212.2, 148.1,
141.9, 120.3, 110.2, 52.1, 44.7, 40.8, 38.7, 32.3, 27.7, 23.2, 21.9;
EI-HRMS m/z calcd for C13H16O2 (M+) 204.2649, found 204.2661.
1.9, Hz, 1H), 6.21 (J ) 1.9 Hz, 1H), 3.00 (s, 1H), 2.51 (m, 2H),
2.41 (m, 2H), 2.00 (m, 1H), 1.81 (m, 1H), 1.62 (m, 2H), 1.19 (s,
3H); 13C NMR δ 215.1, 144.7, 142.6, 116.7, 110.0, 54.9, 40.2,
35.5, 33.1, 31.7, 25.5, 19.0; EI-HRMS m/z calcd for C12H14O2
(M+) 190.0994, found 190.0993.
cis-8a-Meth yl-4,5,5a,6,7,8a-h exah ydr o-1-oxa-a s-in dacen -
8-on e (17c). With use of the general procedure as described
above, 2-methylcyclopentenone (0.023 mL, 0.24 mmol) was
reacted to give the ketone 17c as a colorless oil (29 mg, 64%)
after chromatography: Rf(9:1 hexanes:EtOAc) 0.27; IR (neat)
νmax 2927, 2854, 1745, 1629, 1227, 1054 cm-1; 1H NMR δ 7.31
(d, J ) 1.8 Hz, 1H), 6.17 (d, J ) 1.8 Hz, 1H), 2.55-2.45 (m,
2H), 2.40-2.22 (m, 3H), 2.00-1.75 (m, 2H), 1.38 (s, 3H); 13C
NMR δ 18.9, 21.3, 22.4, 23.2, 36.2, 44.3, 49.9, 110.2, 117.5,
142.5, 147.8, 217.8; EI-HRMS m/z calcd for C12H14O2 (M+)
190.0994, found 190.0989.
1-(4,5,6,7-Tetr a h yd r oben zofu r a n -7-yl)eth a n on e (17h ).
With use of the general procedure as described above, methyl
vinyl ketone (87 mg, 1.24 mmol) was reacted to give ketone
17h (124 mg, 0.76 mmol, 61%) as a colorless oil after chroma-
tography: Rf(3:1 pentane:diethyl ether) 0.50; IR (neat) νmax
2934, 2855, 1715, 1356 cm-1 1H NMR (CDCl3, 500 mHz) δ
;
7.32 (dd, J ) 2.0, 1.2 Hz, 1H), 6.25 (d, J ) 2.0 Hz, 1H), 3.66 (t,
J ) 6.1 Hz, 1H), 2.42-2.53 (m, 2H), 2.20 (s, 3H), 2.06-2.13
(m, 1H), 1.90-1.97 (m, 1H), 1.69-1.84 (m, 2H); 13C δ 208.3,
147.2, 141.7, 119.5, 110.7, 48.1, 28.9, 26.3, 22.1, 22.4; EI-HRMS
m/z calcd for C10H12O2 (M+) 164.0837, found 164.0837.
cis-4,5a ,6,7,8,9a -Hexa h yd r o-5H-n a p h th o[1,2-b]fu r a n -9-
on e (15). With use of the general procedure as described
above, cyclohexenone (0.039 mL, 0.4 mmol) was reacted to give
the ketone 15 as a white solid (52 mg, 68%): Rf(9:1 hexanes:
EtOAc) 0.26; IR (neat) νmax 2930, 1715, 1631, 1503, 1253, 1107
cm-1; 1H NMR δ 7.32 (d, J ) 1.9 Hz, 1H), 6.23 (d, J ) 1.9 Hz,
1H), 3.63 (d, J ) 5.7 Hz, 1H), 2.61-2.43 (m, 3H), 2.42-2.28
(m, 2H), 2.00-1.84 (m, 3H), 1.84-1.72 (m, 1H), 1.72-1.60 (m,
2 H); 13C NMR δ 209.8, 147.0, 142.2, 117.7, 110.4, 50.6, 40.7,
38.8, 28.6, 26.7, 24.1, 20.5; EI-HRMS m/z calcd for C12H14O2
(M+) 190.0994, found 190.0999.
cis-5a -Meth yl-4,5a ,6,7,8,9a -h exa h yd r o-5H-n a p h th o[1,2-
b]fu r a n -9-on e (17d ). With use of the general procedure as
described above, 3-methyl-2-cyclohexen-1-one (0.77 mL, 7.2
mmol) was reacted to give the ketone 17d as a pale yellow oil
(1.12 g, 76%) after chromatography: Rf(9:1 hexanes:EtOAc)
0.27; IR (neat) νmax 2924, 1716, 1632, 1455, 1230 cm-1; 1H NMR
δ 7.31 (d, J ) 1.8 Hz, 1H), 6.23 (d, J ) 1.8 Hz, 1H), 3.23 (s,
1H), 2.52 (m, 2H), 2.30 (m, 2H), 1.88 (m, 3H), 1.71 (m, 1H),
1.48 (m, 2H), 1.09 (s, 3H); 13C NMR δ 210.3, 147.3, 142.4, 116.5,
110.4, 56.9, 39.8, 39.7, 34.3, 33.4, 26.7, 22.3,19.0; EI-HRMS
m/z calcd for C13H16O2 (M+) 204.1150, found 204.1172.
7-Isop r op en yl-9a -m et h yl-4,5a ,6,7,8,9a -h exa h yd r o-5H-
n a p h th o[1,2-b]fu r a n -9-on e (20a ). With use of the general
procedure as described above, carvone (50 mg, 0.53 mmol) was
reacted to give ketone 20a (77 mg, 0.32 mmol, 60%) as a
viscous, colorless oil after chromatography: Rf(20:1 pentane:
diethyl ether) 0.31; IR (neat) νmax 3084, 2975, 2935, 1714, 1499,
1
1266 cm-1; H NMR (500 mHz, CDCl3) δ 7.25 (d, J ) 1.9 Hz,
1H), 6.21 (d, J ) 1.9 Hz, 1H), 4.72 (m, 1H), 4.64 (m, 1H), 2.60-
2.64 (m, 1H), 2.47-2.54 (m, 2H), 2.37-2.41 (m, 1H), 1.71-
1.85 (m, 2H), 1.69 (s, 3H), 1.40 (s, 3H); 13C δ 210.3, 150.5, 147.5,
141.8, 115.3, 110.5, 109.5, 49.8, 44.5, 44.2, 43.5, 32.2, 24.5, 22.4,
20.6, 18.4; EI-HRMS m/z calcd for C16H20O2 (M+) 244.1463,
found 244.1458.
cis-4,5a ,6,7,8,9a -H exa h yd r o-5H -n a p h t h o[1,2-b]fu r a n -
4,9-d ion e-4-m on oeth ylen e Keta l (20b). With use of the
general procedure as described above, ketal 18b (380 mg, 2.5
mmol) was reacted to give ketone 20b (420 mg, 1.7 mmol, 69%)
as a white solid after chromatography and recrystallization
from pentane:ether: Rf(1:1 hexane:ethyl acetate) 0.48; mp
116-118 °C; IR (KBr) νmax 3054, 2954, 1715, 1501, 1324 cm-1
;
1H NMR (500 mHz, CDCl3) δ 7.36 (d, J ) 1.9 Hz, 1H), 6.23 (d,
J ) 2.0 Hz, 1H), 4.04-4.09 (m, 4H), 3.93 (d, J ) 5.6 Hz, 1H),
2.81 (dt, 1H), 2.53-2.58 (m, 1H), 2.36-2.45 (m, 2H), 2.30-
2.34 (m, 1H), 2.09-2.12 (m, 2H), 1.95-2.00 (m, 1H), 1.31-
1.40 (m, 1H); 13C δ 208.3, 146.9, 142.5, 118.5, 110.2, 108.7,
65.3, 64.6, 47.7, 46.7, 38.0, 30.9, 23.4, 21.7; EI-HRMS m/z calcd
for C14H16O4 (M+) 248.1049, found 248.1050.
cis-9a -Meth yl-4,5a ,6,7,8,9a -h exa h yd r o-5H-n a p h th o[1,2-
b]fu r a n -9-on e (17e). With use of the general procedure as
described above, 2-methyl-2-cyclohexen-1-one (20 mg, 0.18
mmol) was reacted to give 17e as a colorless oil (24 mg, 65%)
after chromatography: Rf(9:1 hexanes:EtOAc) 0.28; IR (neat)
1
νmax 2927, 2851, 1707, 1628, 1163 cm-1; H NMR δ 7.31 (d, J
5a -Met h yl-9-oxo-4,5,5a ,6,7,8,9,9a -oct a h yd r on a p h t h o-
[1,2-b]fu r a n -6-ca r boxylic Acid Eth yl Ester (20c). With use
of the general procedure as described above, Hageman’s ester
) 1.9 Hz, 1H), 6.24 (d, J ) 1.9 Hz, 1H), 2.62-2.40 (m, 4H),
2.22 (m, 1H), 2.09 (m, 1H), 2.20-1.66 (m, 5H), 1.47 (s, 3H);
J . Org. Chem, Vol. 69, No. 11, 2004 3733

Sperry et al.
(98 mg, 0.54 mmol) was reacted to give ketone 20c (114 mg,
0.41 mmol, 76%) as a white solid after chromatography and
recrystallization from pentane:ether: Rf(10:1 pentane:ethyl
acetate) 0.45; mp 72-74 °C; IR (KBr) νmax 3056, 2960, 2916,
from pentane:ether: Rf(5:1 pentane:ethyl acetate) 0.33; mp
69-70 °C; IR (KBr) νmax 3013, 2930, 2860, 1682, 1501 cm-1
;
1H NMR (500 mHz, CDCl3) δ 7.28 (dd, J ) 2.0, 1.2 Hz, 1H),
6.19 (d, J ) 2.0 Hz, 1H), 3.56 (br s, 1H), 2.62-2.67 (m, 1H),
2.55-2.60 (m, 1H), 2.41-2.46 (m, 1H), 2.00-2.06 (m, 1H),
1.65-1.81 (m, 5H), 1.10-1.15 (m, 1H), 0.30 (dd, J ) 11.0, 4.8
Hz, 1H); 13C δ 207.6, 146.0, 141.9, 116.0, 110.6, 48.5, 36.9, 27.4,
24.5, 22.9, 18.8, 18.5, 18.1; EI-HRMS m/z calcd for C13H14O2
(M+) 203.1073, found 203.1071.
1754, 1719, 1476, 1301 cm-1 1H NMR (500 mHz, CDCl3) δ
;
7.40 (d, J ) 1.8 Hz, 1H), 6.28 (d, J ) 1.8 Hz, 1H), 4.18-4.25
(m, 2H), 3.39 (s, 3H), 2.91 (dd, J ) 11.4, 3.9 Hz, 1H), 2.80-
2.87 (m, 1H), 2.50-2.58 (m, 1H), 2.45 (dt, J ) 13.9, 4.1 Hz,
1H), 2.26 (td, J ) 13.4, 6.2 Hz, 1H), 2.13-2.22 (m, 1H), 2.00-
2.06 (m, 1H), 1.88-1.94 (m, 1H), 1.50-1.59 (m, 1H), 1.30 (t, J
) 7.2 Hz, 3H), 1.12 (s, 3H); 13C δ 208.5, 173.5, 146.3, 142.7,
117.0, 110.6, 60.8, 57.8, 44.0, 41.4, 38.0, 34.7, 25.0, 22.5, 19.0,
14.5; EI-HRMS m/z calcd for C16H20O2 (M+) 276.1362, found
276.1367.
Cyclop r op yl(4,5,6,7-tetr a h yd r oben zofu r a n -7-yl)m eth -
a n on e (40). With use of the general procedure as described
above, cyclopropyl vinyl ketone (62 mg, 0.65 mmol) was reacted
to give ketone 40 (75 mg, 0.39 mmol, 60%) as a colorless oil
Rf(20:1 pentane:ether) 0.22; IR (neat) νmax 2930, 2849, 1713,
1366 cm-1; 1H NMR (500 mHz, CDCl3) δ 7.32 (dd, J ) 2.0, 0.7
Hz, 1H), 6.25 (d, J ) 1.9 Hz, 1H), 3.81 (t, J ) 6.1 Hz, 1H),
2.42-2.53 (m, 2H), 2.12-2.16 (m, 1H), 1.97-2.07 (m, 2H),
4,5,5a ,11a -Tetr a h yd r o-1,6-d ioxa -cyclop en ta [a ]a n th r a -
cen -11-on e (20d ). With use of the general procedure as
described above, chromone (240 mg, 1.6 mmol) was reacted to
give ketone 20d (253 mg, 1.1 mmol, 67%) after chromatogra-
phy: Rf(5:1 hexane:ethyl acetate) 0.25; mp 153 °C dec; IR (KBr)
1.70-1.89 (m, 1H), 1.02-1.10 (m, 2H), 0.84-0.93 (m, 2H); 13
C
δ 210.3, 147.6, 141.8, 119.6, 110.7, 48.4, 26.8, 22.3, 21.5, 19.3,
11.8, 11.6; EI-HRMS m/z calcd for C12H14O2 (M+) 190.0944,
found 190.0943.
1
νmax 3016, 2954, 2928, 1688, 1607 cm-1; H NMR (500 MHz,
CDCl3) δ 7.96 (dd, J ) 7.9, 1.8 Hz, 1H), 7.47 (ddd, J ) 8.6,
7.2, 1.9 Hz, 1H), 7.34 (dd, J ) 1.8, 1.1 Hz, 1H), 7.00 (ddd, J )
7.9, 7.2, 1.1 Hz, 1H), 6.94 (dd, J ) 8.4, 1.0 Hz, 1H), 6.25 (d, J
) 1.9 Hz), 5.04 (ddd, J ) 6.2, 4.3, 2.0, 1H), 3.82-3.90 (m, 1H),
2.84-2.92 (m, 1H), 2.56-2.65 (m, 1H), 2.39-2.48 (m, 1H),
1.91-2.00 (m, 1H); 13C NMR δ 189.5, 161.1, 143.1, 142.3, 136.5,
127.8, 121.6, 119.5, 118.3, 118.2, 110.6, 75.6, 46.8, 26.9, 18.2;
EI-HRMS m/z calcd for C15H12O3 (M+) 240.0786, found 240.0788.
9-Oxo-5,5a ,7,8,9,9a -h exa h yd r o-4H-fu r o[2,3-f]qu in olin e-
6-ca r boxylic Acid Meth yl Ester (20e). With use of the
general procedure as described above, pyridone 18e (186 mg,
1.2 mmol) was reacted to give ketone 20e (198 mg, 0.80 mmol,
65%) as a viscous, colorless oil. Rf(3:1 diethyl ether:pentane)
0.26; IR (neat) νmax 3114, 2950, 2860, 1698, 1509 cm-1; 1H NMR
(500 mHz, CDCl3) δ 7.36 (d, J ) 2.0 Hz, 1H), 6.23 (d, J ) 2.0
Hz), 4.86 (br, 1H), 4.45 (br, 1H), 3.75 (s, 3H), 3.73 (d, J ) 3.2
Hz, 1H), 3.31 (t, J ) 12 Hz, 1H), 2.62 (m, 1H), 2.55 (dd, J )
5.5, 5.1 Hz), 2.39 (d, J ) 14.1 Hz), 1.77-1.84 (m, 2H); 13C NMR
δ 205.7, 155.9, 145.1, 143.4, 118.3, 110.2, 54.4, 53.4, 49.7, 40.7,
25.8, 20.5; EI-HRMS m/z calcd for C13H15NO4 (M+) 249.1001,
found 249.1010.
3-F u r a n -3-y lm e t h y l-3-(3-is o p r o p o x y p r o p y lid e n e )-
cycloh exa n on e (44). With use of the general procedure as
described above, enone 41 (44 mg, 0.32 mmol) was reacted to
give enone 44 (22 mg, 25%) as a colorless oil. Rf(3:1 pentane:
ether) 0.25; 1H NMR (300 MHz, CDCl3) δ 7.35 (d, J ) 1.7 Hz,
1H), 7.24 (s, 1H), 6.3 (t, J ) 7.4 Hz, 1H), 6.24 (s, 1H), 3.57 (7
line multiplet, J ) 6.1 Hz, 1H), 3.53 (m, 2H), 3.01 (m, 1H),
2.57 (m, 1H), 2.31-2.48 (m, 6H), 1.69-1.82 (m, 5H), 1.54-
1.58 (m, 1H), 1.18 (d, J ) 6.3 Hz, 6H); 13C δ 203.0, 143.3, 143.1,
139.0, 134.7, 124.7, 111.0, 71.9, 66.9, 40.7, 35.7, 34.3, 29.2, 28.0,
22.6, 22.4, 22.3, 19.3; IR (neat) νmax 2968, 2933. 1687, 1920,
1501.
Ack n ow led gm en t. The authors thank the National
Science Foundation, Dartmouth College, and the Uni-
versity of Florida for generous support of this work. We
would also like to thank Prof. Mark McMills (Ohio
University) for helpful discussions.
Su p p or tin g In for m a tion Ava ila ble: Electrode schematic
and copies of 1H and 13C NMR spectra for compounds 15, 17a -
h , 20a -e, 37, and 40. This material is available free of charge
via the Internet at http://pubs.acs.org.
4,5,5a ,6,6a ,7,7a ,8a -Oct a h yd r o-1-oxa -cyclop e n t a [a ]-
cyclop r op a [g]n a p h th a len e-8-on e (37). With use of the
general procedure as described above, enone 34 (120 mg, 1.1
mmol) was reacted to give ketone 37 (156 mg, 0.77 mmol, 69%)
as a white solid after chromatography and recrystallization
J O049889I
3734 J . Org. Chem., Vol. 69, No. 11, 2004