Table 1. Effects of Various Bases on the Rate of Reaction and Product Distribution
a Reaction conditions: OsO4 (0.02 equiv), NaIO4 (4.0 equiv), dioxane-water (3:1).
that the rate of formation of osmium(VI) ester complexes
could be dramatically increased by the addition of an excess
of tertiary amine such as pyridine.5 When pyridine was
added, the reaction was indeed faster, and the selectivity was
excellent (Table 1, entry 4). Only a trace amount of
compound 3 was isolated. Unfortunately, we also observed
the formation of the epimer of the methyl group at the
R-position of the aldehyde 2, which was formed via the
enolization of the aldehyde by pyridine. To avoid the base-
promoted enolization, 2,6-di-tert-butylpyridine was employed
(Table 1, entry 5). However, no effect was observed. These
findings logically led us to investigate the use of 2,6-lutidine
as the base. To our delight, we found that 2 equiv of 2,6-
lutidine effectively suppressed the formation of compound
3 and dramatically improved the yield of the desired aldehyde
2 to 90% along with only 6% of the side product 3.
Furthermore, the reaction was faster and was complete in 2
h without epimerization of the methyl group at the R-position
of the aldehyde 2.
The amazing inhibiting effect of 2,6-lutidine led us to study
its broad scope in a variety of substrates (Table 2).6 In the
absence of 2,6-lutidine, the reaction of compound 4 with
OsO4-NaIO4 gave only 44% yield of the desired product 5
(Table 2, entry 1). Although structurally compound 4 is very
similar to compound 1 except for the different protecting
group at the homoallylic alcohol, we noticed that the
oxidative cleavage reaction of compound 4 under the classic
conditions (without 2,6-lutidine) was very messy, and many
side products were formed (we did not try to characterize
each side product). Amazingly, the yield of the desired
aldehyde 5 was improved to 83% with the addition of 2,6-
lutidine. It appeared that 2,6-lutidine inhibited several
uncharacterized side reactions at the same time. The reaction
of compound 6 under the classic conditions was quite slow
and gave only 34% yield of the desired product 7 along with
a 48% recovery of the starting material after the reaction
was stirred at room temperature for 20 h (Table 2, entry 2).
It was noted that longer reaction times resulted in even lower
yield due to cleavage of the acid-labile TBS protecting group.
In the presence of 2,6-lutidine, the same reaction was
complete in 20 h and afforded 99% yield of compound 7.
Both inhibiting and rate acceleration effects by 2,6-lutidine
were quite obvious in this instance. Similar effects were also
observed in compound 8 (Table 2, entry 3). Compound 10
gave only 42% yield of the desired aldehyde 11 (Table 2,
entry 4). But the yield was improved to 71% when 2 equiv
of 2,6-lutidine was added. It should be noted that 2,6-lutidine
also served as a weak base to neutralize the acid generated
in the reaction to prevent the cleavage of the TES group.
Compound 12, an internal olefin, gave only 28% yield of
the aldehyde 13 under the classic conditions. However, the
yield was improved to 77% after the addition of 2,6-lutidine
(Table 2, entry 5).
A typical procedure for the improved OsO4-NaIO4
oxidative cleavage reaction follows: To a solution of
compound 1 (296 mg, 0.812 mmol) in dioxane-water (3:1,
8 mL) were added 2,6-lutidine (0.189 mL, 1.62 mmol), OsO4
(2.5% in 2-methyl-2-propanol, 165 mg, 0.016 mmol), and
NaIO4 (695 mg, 3.25 mmol). The reaction was stirred at 25
°C and monitored by TLC. After the reaction was complete,
water (10 mL) and CH2Cl2 (20 mL) were added. The organic
layer was separated, and the water layer was extracted by
CH2Cl2 (10 mL) three times. The combined organic layer
was washed with brine and dried over Na2SO4. The solvent
was removed, and the product was purified with silica gel
column chromatography to afford aldehyde 2 (268 mg, 90%)
as a colorless oil.
In conclusion, we have successfully developed an im-
proved procedure for the classic OsO4-NaIO4-mediated
oxidative cleavage reaction. We have demonstrated that 2,6-
lutidine (or pyridine where epimerization is not likely to be
an issue) can effectively suppress the formation of R-hydroxy
ketone side products, accelerate the rate of the desired
(5) Criegee, R.; Marchand, B.; Wannowlus, H. Justus Liebigs Ann. Chem.
1942, 550, 99. (b) Schroder, M. Chem. ReV. 1980, 80, 187 and references
therein.
(6) All compounds were fully characterized.
3218
Org. Lett., Vol. 6, No. 19, 2004