derivative 1b′, as shown by NMR and by chromatographic
comparison with the former series.5 Apparently, a competi-
tion between the intramolecular cyclization and the second
conjugate addition takes place.
appeared to be essential. With 1 mmol of DMAP or with
mixtures of Et3N (1.0-1.3 mmol) and DMAP (0.2-0.5
mmol), 18 h were required to complete the reaction.
Under these conditions, 1 can be chemoselectiVely pro-
tected in the presence of alcohols and diols. Only 1a was
formed by mixing equimolar amounts of 1 and benzyl alcohol
or meso-1,2-diphenyl-1,2-ethanediol and treating the mixture
with tert-butyl propynoate and excess of DMAP as above.
The higher acidity of phenols explains this relevant fact. As
pointed out in Scheme 1, the proton transfer from a phenolic
hydroxy group to the alkyl propiolate-DMAP adduct may
play a significant role in the reaction mechanism.
Catechols can also be selectiVely protected in the presence
of phenols. For example, a mixture of 1, 4-methylphenol,
and 2-naphthol (1 mmol of each), when treated with 1.5
mmol of DMAP and 1.0 mmol of tert-butyl propiolate,
afforded 1a in practically quantitative yields; Bocvinyl
derivatives of 4-methylphenol or 2-naphthol were not
obtained. Apparently, the cyclization to 1a is the driving
force that shifts the DMAP-catalyzed equilibrium among the
different Bocvinyl derivatives.
Deprotection of 1a-3a with bases, via elimination and
addition-elimination reactions, was very easy (Scheme 2,
bottom). With 10 equiv of pyrrolidine, in CH3CN at room
temperature, deprotection was practically quantitative in 3
h (at 0.1 M concentration). In neat pyrrolidine, it was
quantitative in 1 h. Separation of catechols from tert-butyl
3-(1-pyrrolidinyl)propenoate (“N-Bocvinyl-pyrrolidine”) by
extraction or by chomatography was simple. It is worth
noting that Bocdene acetals are not hydrolyzed by 70:30
AcOH-H2O after 24 h at room temperature. Cleavage of
the corresponding Mocdene acetals (1b-3b) was also very
efficient.9
Compounds 1a and 1a′ were not convertible directly to
each other under the reaction conditions; a sample of 1a,6
isolated by column chromatography, did not give 1a′ when
treated with tert-butyl propiolate and DMAP for 3 days,
while pure 1a′ 6 did not afford 1a when treated with DMAP.
Thus, the Bocdene group of 1a and the Bocvinyl groups of
1a′ are stable against DMAP at room temperature. However,
1 plus 1a′ reacted slowly in the presence of DMAP, as they
were mostly converted into 1a after 18 h. This indicates a
slow equilibrium reaction (1 + 1a′ ) 2 × 1a) that is shifted
toward 1a. In other words, DMAP catalyzes a Bocvinyl
transfer from 1a′ to 1 (to give two molecules of monosub-
stituted derivative, which cyclizes to 1a). Therefore, the key
for a quantitative conversion of catechols to acetals had to
be the ratio between alkyl propiolate and DMAP: more
propiolate than DMAP favors disubstitution, while more
DMAP than propiolate favors cyclization, independently of
the use of an excess or deficit of reagents.7
In practice, when 1, 3,4-dihydroxybenzaldehyde (2), and
2,3-dihydroxybenzaldehyde (3) were independently treated
with tert-butyl propiolate (1.1 mmol/mmol) and DMAP (1.5
mmol/mmol), 96-100% yields of Bocdene derivatives 1a-
3a were obtained within 15-30 min (Scheme 2).8 With
Scheme 2
Application to the Bocdene protecting group to a real
example, in which an organometallic reagent was added to
2a, follows. Treatment of 2a with EtMgBr (1.1 equiv, from
a 1.0 M solution in THF) at -78 °C for 5 min afforded 4a
in 90% yield;10 deprotection with pyrrolidine in CH3CN gave
1-(3,4-dihydroxyphenyl)-1-propanol (4). As summarized in
Scheme 2, compound 4 was converted to 4a and 4b under
our standard conditions without touching the alcohol func-
tion.
The case of pyrogallol (5) confirmed the previous results
regarding (i) the competition between cyclization and dis-
ubstitution, (ii) the equilibrium between the starting material
and Bocvinyl derivatives, (iii) the practical significance of
the ratio between HCtC-COOR and DMAP, and (iv) the
methyl propiolate, the corresponding Mocdene derivatives
(1b-3b) were similarly obtained.5 The DMAP excess
(3) Beaulieu, P. L.; Cameron, D. R.; Ferland, J.-M.; Gauthier, J.; Ghiro,
E.; Gillard, J.; Gorys, V.; Poirier, M.; Rancourt, J.; Wernic, D.; Llinas-
Brunet, M.; Betageri, R.; Cardozo, M.; Hickey, E. R.; Ingraham, R.; Jakes,
S.; Kabcenell, A.; Kirrane, T.; Lukas, S.; Patel, U.; Proudfoot, J.; Sharma,
R.; Tong, L.; Moss, N. J. Med. Chem. 1999, 42, 1757.
(4) The yield of 1a increased slowly, with disappearance of 1, when
stirring was continued; after 18 h the yield reached a maximum value of
70%.
(8) Typical Experimental Procedure. To 2 (138 mg, 1.0 mmol) and
DMAP (183 mg, 1.5 mmol) in CH3CN (10 mL), under nitrogen, was added
tert-butyl propynoate (150 µL, 1.1 mmol). Stirring for 30 min, evaporation
of the solvent under vacuum, and separation of the residue by flash
chromatography on silica gel (75:25 hexanes-EtOAc) afforded 2a (259
mg, 98% yield).
(5) Ethyl propynoate also gave identical results.
(6) Data for 1a: 1H NMR (200 MHz, CDCl3) δ 6.80 (br s, 4H), 6.46 (t,
J ) 5.2 Hz, 1H), 2.89 (d, J ) 5.2 Hz, 2H), 1.46 (s, 9H); 13C NMR (50.3
MHz, CDCl3) δ 167.5, 147.1, 121.5, 108.6, 107.9, 81.7, 41.4, 28.0. Data
for 1a′: 1H NMR (200 MHz, CDCl3) δ 7.65 (d, J ) 12.2 Hz, 2H), 7.1-
7.3 (m, 4H), 5.26 (d, J ) 12.2 Hz, 2H), 1.43 (s, 9H); 13C NMR (50.3 MHz,
CDCl3) δ 165.7, 158.2, 145.6, 125.9, 120.4, 103.7, 80.8, 28.0.
(7) Even by using 2.0 mmol of tert-butyl propiolate (per mmol of
substrate), provided that 3.0 mmol of DMAP was present in the reaction
flask, the disubstituted product (1a′) was formed in minute amounts; the
1a/1a′ ratio was 95:5. With 2.0 mmol of tert-butyl propiolate and 1.0 mmol
of DMAP, the 1a/1a′ ratio was 3:97.
(9) Typical Experimental Procedure. To a solution of 2b (91 mg, 0.41
mmol) in CH3CN (4 mL) was added pyrrolidine (0.35 mL, 4.1 mmol).
Stirring at room temperature for 2.5 h, removal of the solvent, and separation
by chromatography on silica gel (from CH2Cl2 to 95:5 CH2Cl2-MeOH)
afforded 2 (55 mg, 97%).
(10) Only one product is seen by TLC and 1H and 13C NMR spectroscopy
(although very small splittings appear when the 13C spectrum is expanded),
as the stereocenters are far away to each other.
1400
Org. Lett., Vol. 3, No. 9, 2001