27 and 28 in 98% and 99% yields, respectively (entries 11
and 12). Methyl esterification of Boc-L-Thr-OH (17) gave
rise to Boc-L-Thr-OMe (29) in 95% yield (entry 13).
Benzoylation, acetylation, and formylation reactions of
DL-1,2-isopropylideneglycerol (18) provided the corre-
sponding esters 30aꢀc in greater than 95% yields (entries
14ꢀ16). It should be noted that (2,2-dimethyl-1,3-dioxo-
lan-4-yl)methyl formate (30c) was not stable to silica gel;
thus, its yield was determined based on 1H NMR analysis
of the crude product. On the other hand, formylation of
(3,5-bis(benzyloxy)phenyl)methanol (19) afforded 31 in
95% yield after silica gel chromatography (entry 17).
Selective esterifications of diols were also demonstrated,
and selected examples are summarized in Table 2. The
primary alcohol of butane-1,3-diol (20) was selectively
benzoylated to afford 32 in 80% yield. Esterifications of
glycerol (21) with benzoic acid and n-hexanoic acid furn-
ished the corresponding diesters 33a and 33b in 85% and
90% yield, respectively (entries 19 and 20). Benzoylation of
benzyl 2-(acetylamino)-2-deoxy-R-D-glucopyranoside (22)
was achieved selectively at the C6-position to afford the
monobenzoate 34 in 90% yield (entry 21).
Finally, acylations of the diol of a complex muramic acid
derivative 35 were demonstrated as selective esterifications
of primary alcohols (Scheme 1).11 Acetylation and benzoy-
lation of 35 using 5b (1.5 equiv), acid (2 equiv), EDCI
(1.5 equiv), and NaHCO3 (6 equiv) at 0 °C gave rise to the
primary acetate 36a and benzoate 36b in greater than 95%
yield without the formation of diacylated products. In the
reactions summarizedin Scheme 1, it is a significant benefit
to use glyceroacetonide-Oxyma 5b. Although the same
reaction with Oxyma 5a gave an equal conversion yield as
observed in Scheme 1, separation of 5a from the product
was extremely difficult via silica gel chromatography.
On the other hand, 5b could be removed completely via
standard acidic and basic workups.
Scheme 1. Selective Acylations of 35
superior to those in the other solvent systems tested. Under
the optimized conditions [acid (1 equiv), alcohol (2 equiv),
5a or 5b (1.5 equiv), EDCI (1.5 equiv), and NaHCO3
(6 equiv)], isopropanol and tert-butanol did not form the
corresponding esters with 6 even after a prolonged reaction
time.8
In order to understand the scope and limitations of the
selective esterification reactions of primary alcohols with
EDCI, Oxyma 5a, and NaHCO3 in 5% H2OꢀCH3CN,
these conditions were applied to esterifications of a wide
variety of acids with alcohols. Selected examples are
summarized in Table 2. Esterifications of 6 with methanol,
primary alcohols, and phenols furnishedthe corresponding
esters in greater than 90% yield without detectable race-
mization (entries 1ꢀ7). Significantly, an allyl alcohol could
be esterified to provide 23c in 98% yield. It is worth
pointing out that esterifications of carboxylic acid with
allyl alcohols have never been successfully performed using
carbodiimide-mediated reaction conditions (entry 3).9
Unlike 4-(dialkylamino)pyridine-catalyzed DCC-mediated
esterification conditions, the Fmoc-group was not cleaved
during the benzyl esterifications of the Fmoc-protected
amino acids, 12 and 13 (entries 8 and 9).10 Esterifications
of N-sulfonylated R-amino acids using carbodiimide cou-
pling reagents often result in low conversion with signifi-
cant racemization. However, under the conditions in
Table 2, the benzyl esterification of 14 furnished 26 in
98% yield with >99% ee (entry 10). The chiral carboxylic
acids possessing secondary alcohols, 15, 16, and 17, could
be esterified efficiently with the primary alcohols. Benzyl
esterifications of (S)-mandelic acid (15) and 3-hydroxybu-
tanoic acid (16) furnished the corresponding benzyl esters
Inconclusion, wehaveoptimized selectiveesterifications
of primary alcohols using Oxyma 5a or glyceroacetonide-
Oxyma 5b, EDCI, and NaHCO3 in 5% H2OꢀCH3CN.
The selective esterification conditions described here do
not require the strict anhydrous conditions necessary
for ordinal esterification reactions. The coupling additive
5b can be removed easily after the reactions via acidic and
basic workups. The new esterification conditions reported
here should be a valuable asset in organic synthesis and for
selective modifications of polyol molecules.
Acknowledgment. We thank the National Institutes of
Health (NIAID Grant AI AI084411) and University of
Tennessee for generous financial support.
(8) Esterifications of 6 with (þ)-menthol and cholesterol also did not
provide the corresponding esters.
(9) (a) Monagle, J. J. J. Org. Chem. 1962, 27, 3851–3855. (b) Steglich,
€
W.; Hofle, G. Angew. Chem., Int. Ed. Engl. 1969, 8, 981. (c) Boden, E. P.;
Supporting Information Available. Experimental pro-
cedures and copies of NMRs. This is available free of
Keck, G. E. J. Org. Chem. 1985, 50, 2394–2395.
(10) Spivey, A. C.; Arseniyadis, S. Angew. Chem., Int. Ed. 2004, 43,
5436–5441.
(11) Under the optimized conditions, acetylation of 4a furnished 3a
in greater than 95% yield without the formation of the diacetate
(Figure 1).
The authors declare no competing financial interest.
D
Org. Lett., Vol. XX, No. XX, XXXX