Oxidative Cyclization for THF Ring Construction
A R T I C L E S
Scheme 6
adsorption of the electrolyte onto the electrode surface, a process
that excludes solvent from the area surrounding the electrode.10
The net effect is to lower the concentration of the solvent in
the region where the reactive intermediate is generated and
thereby provide more time for the cyclization reaction. In fact,
cyclizations that employ a very good radical cation trapping
group like an enol ether1b or a furan1e can tolerate the use of
pure methanol as the solvent for the reaction without evidence
for methanol trapping. For less effective terminating groups such
as an allylsilane, cosolvents such as THF and CH2Cl2 are
required for optimizing the cyclizations. In these cases, the use
of pure methanol solvent for the reaction leads to a competition
between the intramolecular cyclization and solvent trapping of
the radical cation. On the basis of these observations, the need
for a cosolvent can be used to gain insight into the relative
reactivity of trapping groups for the radical cation. In the case
of the alcohol trapping groups, optimizing the yield of the
cyclization did require the use of a cosolvent, a result that
initially placed the reactivity of the intramolecular alcohol
trapping group below that of an enol ether or furan and on par
with that of an allylsilane. This conclusion was supported by
the results obtained using substrate 14e. While cyclization
reactions utilizing allylsilane terminating groups also fail to
afford seven-membered ring products, reactions terminated with
the use of either a second enol ether or a furan ring are
compatible with seven-membered ring formation.1b,e
Scheme 7
reticulated vitreous carbon (RVC) anode and a platinum cathode
(Scheme 7).7 A 0.03 M tetraethylammonium tosylate in 30%
MeOH/THF electrolyte solution was used along with 2,6-lutidine
as a proton scavenger.8 The oxidation was continued until 2
F/mol of charge had been passed. These reaction conditions were
selected because of their similarity to the conditions used for
many of the reactions summarized in Scheme 1. The cyclization
reactions leading to five- and six-membered ring products led
to good to excellent yields. As in the earlier carbon-carbon
bond forming reactions, the reactions leading to the formation
of six-membered rings did prove to be less efficient than
reactions leading to five-membered rings. Attempts to generate
a seven-membered ring product (15e) were not successful. In
this example, the product obtained resulted from methanol
trapping of the radical cation.
The initial cyclizations were also compatible with the use of
a 6 V lantern battery as the power supply.11 Using the battery,
the cyclization of 14a led to a 74% isolated yield of the cyclized
product. The cyclization of 14b led to a 55% yield of product
along with 25% of the recovered starting material. While the
yields using the battery were not as high as previous trials, it
was clear that the effectiveness of the cyclization could be
evaluated without the need for specialized equipment.
For the cyclization reactions originating from 14a, 14b, and
14d the major product had trans-stereochemistry. For the five-
membered ring products, the stereochemistry of the substituents
was established with the use of a NOESY experiment. This was
accomplished by noting the larger interaction between methine
protons Hb and Hc in the cis-isomer and the larger interaction
between methines Ha and Hc in the trans-isomer. For the six-
membered ring product, the stereochemistry of the major product
was established using the 9.3 Hz coupling constant observed
for the interaction between Hb and Hc.
The success of the alcohol nucleophiles as trapping groups
for the radical cations was intriguing because earlier anodic
carbon-carbon bond forming reactions had been accomplished
in methanol based solvent mixtures without the methanol solvent
trapping the initial radical cation. This observation is currently
thought to be a result of the electrochemical “double layer” that
forms at the anode surface.9 The double layer forms due to
Building a Quaternary Carbon
The reactivity of a terminating group for an enol ether radical
cation can also be probed by examining the compatibility of
the cyclization reaction with the formation of a quaternary
carbon. This was of special interest in the current case because
of the need to generate a quaternary carbon during the synthesis
of (+)-nemorensic acid. For this reason, substrates 16a and 16b
were synthesized in a fashion directly analogous to the syntheses
of substrates 14a-d (vide supra). In this case, methyllithium
was added to the lactone in order to generate a substrate for the
Wittig reaction.
Scheme 8
(7) The electrolyses were conducted utilizing a Model 630 coulometer, a Model
410 potentiostatic controller, and a Model 420A power supply purchased
from the Electrosynthesis Co., Inc.
(8) A proton scavenger is typically added to enol ether oxidation reactions in
order to assist the removal of acid from the anode surface. This reduces
the chance for methanolysis of the enol ether as it approaches the anode.
It is important to note that the overall reaction remains neutral because the
half-reaction that takes place at the cathode generates an equivalent of
methoxide for every equivalent of acid generated at the anode.
(9) For discussions of the “double layer” see: (a) Anodic Oxidation; Ross, S.
D., Finkelstein, M., Rudd, E. J., Eds.; Academic Press: New York, 1975;
p 13. (b) Synthetic Organic Electrochemistry, 2nd ed.; Fry, A. J., Ed.; John
Wiley and Sons: New York, 1989; p 37.
Once synthesized the substrates were oxidized using the same
conditions described above (Scheme 8). The cyclization orig-
(10) For a discussion of how the adsorption of an electrolyte onto an electrode
surface can exclude solvent and protect a reactive intermediate see: Organic
Electrochemistry, 4th ed., Revised and Expanded; Lund, H., Hemmerich,
O., Eds.; Marcel Dekker: New York-Basel, 2001; p 802.
(11) Frey, D. A.; Wu, N.; Moeller, K. D. Tetrahedron Lett. 1996, 37, 8317.
9
J. AM. CHEM. SOC. VOL. 124, NO. 34, 2002 10103