Alkoxide-catalyzed addition of terminal alkynes to ketones

Full text of DOI:10.1021/jo9515283

416
J . Org. Chem. 1996, 61, 416-417
Sch em e 1
Alk oxid e-Ca ta lyzed Ad d ition of Ter m in a l
(CH3)3COK
Alk yn es to Keton es
+
R′CR′′
C(OH)R′′
R′
RC
CH
DMSO
O
C
CR
J ames H. Babler,* Vincent P. Liptak, and Nguyen Phan
1
2
3
Department of Chemistry, Loyola University of Chicago,
Chicago, Illinois 60626
R
2
cyclohexanone
3-pentanone
cyclohexanone
cyclohexanone
cyclohexanone
2-butanone
a
b
c
d
e
f
(CH2)5CH3
C6H5
HOC(CH3)2
CH2OTHP
CH2OH
CH2CH(OH)CH3
CH2CH(ONa)CH3
Received August 18, 1995
The coupling of a terminal alkyne (1) to a ketone to
generate a tertiary alkynol (3) is a well-established
alkylation methodology in organic synthesis. This pro-
cess is usually initiated by conversion of the alkyne to
the corresponding acetylide anion (RCtC-) using 1 equiv
of an air-sensitive, strong base such as methylmagnesium
bromide,1 butyllithium,2 or sodium amide3 prior to ad-
dition of the ketone. The notion that 1 equiv of a strong
base is required for this alkynylation process is not
surprising in view of the known acidities4 of a terminal
alkyne and other weak acids whose conjugate bases one
might utilize to generate an acetylide anion: R3COH
(pKa ) 17); RCOCH2R (an enolizable ketone, pKa ) 19-
20); RCtCH (pKa ) 25); NH3 (pKa ) 38); and CH3CH2H
(pKa ) 50). These data suggest that, in contrast to strong
bases such as ethylmagnesium bromide or sodium amide,
an alkoxide derived from a tertiary alcohol should be too
weakly basic to convert (appreciably) a terminal alkyne
to an acetylide anion. There are several patents5 that
relate to the use of potassium hydroxide, which is
comparable in basicity to an alkoxide, as a catalyst in
the reaction of gaseous acetylene with carbonyl com-
pounds in liquid ammoniasa process referred to as the
Favorskii reaction. However, subsequent research6 dem-
onstrated that the Favorskii conditions do not involve an
acid-base reaction to generate an alkali metal acetylide,
but instead the reaction proceeds via formation of an
acetylene-alkali hydroxide complex.
It occurred to us that a more economical and conceptu-
ally simple methodology should be feasible to effect this
alkynylation process, after examining the following known7
equilibrium acidities measured in dimethyl sulfoxide
(DMSO): cyclohexanone (pKa ) 26.4); phenylacetylene
(pKa ) 28.7); and tert-butyl alcohol (pKa ) 32.2). Such
data suggest that tert-butoxide should be capable of
converting a terminal alkyne to the corresponding acetyl-
ide anion in DMSO. Moreover, since the initial product
when the alkynylation occurs is a tertiary alkoxide, the
process should be catalytic with respect to the base.
Although we were concerned that a competitive process
(i.e., an aldol condensation) was possible for enolizable
ketones, such a reaction is known to be reversible and
therefore perhaps could be minimized.8
2-butanone
g
In order to test our proposed methodology, a catalytic
amount (10-20 mol %) of potassium tert-butoxide was
added to a DMSO solution containing equimolar amounts
of cyclohexanone and 1-octyne (Scheme 1). After several
hours at room temperature, we were pleased to obtain
the corresponding tertiary alkynol 3a in 80-90% distilled
yield. Remarkably, the process also occurred, albeit at
a slower rate, in nonpolar solvents such as benzene.9 To
assess further the scope of this methodology, several
representative alkynes and ketones were subjected to the
standard reaction conditions (i.e., catalytic potassium
tert-butoxide in DMSO at room temperature). As can be
seen from the data in the Experimental Section, distilled
yields of tertiary alkynols were generally 70-90% and
no significant side reactions were observed. Although the
process was successful even in the presence of an
unblocked tertiary alcohol functionality (1c f 3c), we
found it essential to convert a reactant alkynol possessing
a secondary alcohol functionality (1f) in situ to the
corresponding alkoxide salt (1g)10 prior to initiating the
alkynylation.
Our results demonstrate that a variety of tertiary
alkynols can indeed be prepared via a catalytic process
that avoids the use of air-sensitive bases such as butyl-
lithium. This process may prove to be especially useful
in large-scale alkynylation reactions.
Exp er im en ta l Section
Gen er a l Meth od s. Methyl sulfoxide (DMSO, HPLC grade)
was purchased from Aldrich Chemical Co. and used without
further purification. Unless specified otherwise, all organic
reagents were purchased from Aldrich Chemical Co. All reac-
tions were protected from atmospheric moisture and CO2 by
connecting the reaction flask to an apparatus similar to that
(8) Only in reactions involving a methyl ketone was a competitive
aldol condensation an observable side reaction. Even then, the latter
could be prevented by slow addition of the methyl ketone to the DMSO
solution containing the terminal alkyne and alkoxide catalyst. At-
tempts to utilize an enolizable aldehyde (e.g., propionaldehyde and
isobutyraldehyde) in this process, however, led to a complex mixture
of products.
(9) Use of other aprotic solvents in lieu of DMSO is feasible, although
the rate at which alkynylation occurs decreases in less polar solvents.
For example, use of tetrahydrofuran as the solvent in lieu of DMSO
afforded alkynol 3a in approximately 50% yield after a reaction time
of 20 h at room temperature. The rest of the product mixture consisted
of unreacted starting materials. A similar experiment using benzene
as the solvent afforded alkynol 3a in 35% yield after a reaction time
of 20 h.
(10) We observed that the alkoxide 1g derived from 4-pentyn-2-ol
isomerized rapidly to 3,4-pentadien-2-ol (characterized by an IR
absorption at 1960 cm-1) when treated with KOC(CH3)3 in DMSO at
room temperature. Since the latter allenyl alcohol was the major
component in the reaction mixture after 2 h, its formation appears to
be reversible. In none of the systems examined was the tertiary
alkynol 3 contaminated with any allenic impurities.
(1) Viehe, H. G.; Reinstein, M. Chem. Ber. 1962, 95, 2557.
(2) Eaton, P. E.; Srikrishna, A.; Uggeri, F. J . Org. Chem. 1984, 49,
1728.
(3) Saunders, J . H. Organic Syntheses; Wiley: New York, 1955;
Collect. Vol. III, p 416.
(4) March, J . Advanced Organic Chemistry: Reactions, Mechanisms,
and Structure; J ohn Wiley & Sons: New York, 1992; Table 8.1, pp
250-2.
(5) These patents are cited by Tedeschi, R. J .; Casey, A. W.; Clark,
G. S., J r.; Huckel, R. W.; Kindley, L. M.; Russell, J . P. J . Org. Chem.
1963, 28, 1740.
(6) Tedeschi, R. J . J . Org. Chem. 1965, 30, 3045. For a more
extensive discussion of the ethynylation of carbonyl compounds, see:
Brandsma, L. Preparative Acetylenic Chemistry, 2nd ed.; Elsevier:
Amsterdam, 1988; pp 79-96.
(7) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456.
0022-3263/96/1961-0416$12.00/0 © 1996 American Chemical Society

Notes
J . Org. Chem., Vol. 61, No. 1, 1996 417
described by J ohnson and Schneider.11 Products were recovered
from the organic extracts after drying over anhydrous sodium
sulfate and removal of the solvent with a rotary evaporator
under reduced pressure. Unless indicated otherwise, IR analysis
of the crude alkynol product confirmed the absence of any
unreacted starting ketone. Evaporative distillation refers to
bulb-to-bulb (Kugelrohr) short-path distillation. The physical
and spectral properties of all alkynols were minimally consistent
with the data in the literature.
(t, J ) 7.0 Hz, CH2CtC), 1.86 (m, 2H), 1.67 (m, 2H), 1.52 (m,
7H), 1.39 (m, 2H), 1.29 (m, 5H), 0.89 (t, J ) 7.2 Hz, CH3); 13C
NMR (75 MHz, CDCl3) δ 84.586, 83.888, 68.633, 40.186, 31.196,
28.655, 28.374, 25.188, 23.360, 22.435, 18.536, 13.885.
1-(3-Hyd r oxy-3-m eth yl-1-bu tyn yl)cycloh exa n ol (3c). Re-
action of equivalent amounts of 2-methyl-3-butyn-2-ol (1c) and
cyclohexanone (2c) in the presence of a catalytic amount of
potassium tert-butoxide as described in the procedure19 for the
preparation of alkynol 3d afforded alkynediol 3c in 53% yield,20
the mp and IR and proton NMR spectral properties of which
were consistent with those previously reported:21 13C NMR (75
MHz, CDCl3) δ 89.002, 85.383, 68.254, 64.712, 39.633, 31.319,
25.038, 23.399.
1-[3-[(Tetr a h yd r op yr a n -2-yl)oxy]-1-p r op yn yl]cycloh ex-
a n ol (3d ). To a solution of 704 mg (5.02 mmol) of 3-[(tetrahy-
dropyran-2-yl)oxy]propyne (1d ),12 prepared from 2-propyn-1-ol
using a procedure described by Grieco and co-workers,13 and 0.50
mL (4.82 mmol) of cyclohexanone (2d ) in 3.00 mL of DMSO was
added 112 mg (1.0 mmol) of potassium tert-butoxide.14 This
mixture was subsequently stirred, while being protected11 from
atmospheric moisture and CO2 at room temperature for 15 h.
The mixture was then diluted with 25 mL of 10% (w/v) aqueous
sodium chloride, and the product was isolated by extraction with
30 mL of 1:1 (v/v) hexane:ether. The organic layer was washed
in succession with 10% (w/v) aqueous sodium chloride (3 × 25
mL) and saturated brine (1 × 25 mL). The product was then
isolated from the organic extract in the usual manner and
purified by evaporative distillation: bp 120-138 °C (bath
temperature, 0.25 Torr) [lit.15 bp 124-127 °C at 0.30 Torr],
affording 1.047 g (91%)16 of alkynol 3d as a viscous oil, shown
6-Meth yl-4-octyn e-2,6-d iol (3f). A solution of 0.50 mL (5.3
mmol) of 4-pentyn-2-ol (1f) in 3.00 mL of DMSO was added
dropwise slowly over 10 min to 5.3 mmol of sodium hydride (60%
dispersion in mineral oil, which was removed prior to the
reaction by washing with hexane), protected from atmospheric
moisture11 and maintained at a temperature of 15 °C by use of
an external cold water bath. Once hydrogen evolution had
ceased, 0.05 mL (0.55 mmol) of 2-butanone and 55 mg (0.49
mmol) of potassium tert-butoxide were added to initiate the
alkynylation process.22 This mixture was stirred at room
temperature for 45 min, after which three additional portions23
(3 × 0.05 mL) of 2-butanone were added at 45 min intervals.
After addition of the last portion of 2-butanone, the reaction
mixture was stirred for an additional 2 h. The mixture was then
diluted with 25 mL of 10% (w/v) aqueous sodium chloride, and
the product was isolated by extraction with 25 mL of 4:1 (v/v)
ether:dichloromethane. The organic layer was washed in suc-
cession with 10% (w/v) aqueous sodium chloride (3 × 25 mL)
and saturated brine (1 × 25 mL). The product was then isolated
from the organic extract in the usual manner and purified by
evaporative distillation: bp 102-120 °C (bath temperature, 0.25
Torr) [lit.24 bp 112 °C at 2 Torr], affording 218 mg (63%)25 of
alkynediol 3f shown by NMR analysis to be a mixture of
diastereomers: IR νmax (film) 3340 (OH), 1210, 1158, 1120, 1085,
1
by H and 13C NMR to be essentially free of impurities: IR νmax
(film) 3420 (OH), 1200, 1182, 1130, 1117, 1075, 1055, 1025, 965,
945, 900, 870, 815 cm-1; 1H NMR (300 MHz, CDCl3) δ 4.87 [t, J
) 3.0 Hz, OCHO], 4.31 [s, CH2CtC], 3.84 [t, J ) 9.6 Hz, 1H],
3.56 (m, 1H), 3.12 (s, OH), 1.52-1.92 (m, 14H), 1.25 (m, 2H);
13C NMR (75 MHz, CDCl3) δ 96.205, 89.757, 79.463, 68.114,
61.628, 54.065, 39.607, 30.018, 25.163, 24.996, 23.032, 18.753.
3-Eth yl-1-p h en yl-1-p en tyn -3-ol (3b). Reaction of equiva-
lent amounts of phenylacetylene (1b) and 3-pentanone (2b) in
the presence of a catalytic amount of potassium tert-butoxide
as described in the procedure for the preparation of alkynol 3d
afforded adduct 3b in 70% distilled yield: bp 97-112 °C (bath
temperature, 0.25 Torr) [lit.17 bp 138-142 °C at 12 Torr]; IR
νmax (film) 3390 (OH), 1597, 1485, 1140, 955, 753, 688 cm-1; 1H
NMR (300 MHz, CDCl3) δ 7.43 (d, J ) 3.6 Hz, 2 ortho H’s), 7.29
(m, 3H), 2.32 (s, OH), 1.778 (2H, q, J ) 7.2 Hz, one of the
diastereotopic methylene protons), 1.768 (2H, q, J ) 7.2 Hz),
1.11 (t, J ) 7.2 Hz, 2 × CH3); 13C NMR (75 MHz, CDCl3) δ
131.522, 128.068, 128.000, 122.791, 91.692, 84.387, 72.508,
34.475, 8.735.
1
988, 938, 910 cm-1; H NMR (300 MHz, CDCl3) δ 3.95 (sextet,
1H, J ) 6.0 Hz), 3.50 (broad s, OH), 3.32 (broad s, OH), 2.36 (m,
2H), 1.67 (m, 2H), 1.452 and 1.445 (s, 3H), 1.26 (d, 3H, J ) 6.0
Hz), 1.023 and 1.019 (t, 3H, J ) 7.6 Hz); 13C NMR (75 MHz,
CDCl3) δ 86.268, 79.911, 68.540, 66.249, 36.566, 29.245, 28.949,
22.175, 9.067; other diastereomer 86.245, 79.858, 68.540, 66.302,
36.490, 29.298, 28.934, 22.160, 9.067.
1-(1-Octyn yl)cycloh exa n ol (3a ). Reaction of equivalent
amounts of 1-octyne (1a ) and cyclohexanone (2a ) with a catalytic
amount of potassium tert-butoxide18 in DMSO9 as described in
the procedure for the preparation of 3d afforded alkynol 3a in
84% distilled yield: bp 110-122 °C (bath temperature, 0.25 Torr)
[lit.1 bp 97 °C at 0.015 Torr]; IR νmax (film) 3370 (OH), 1060,
J O9515283
(19) The following modification was made: alkynediol 3c was
extracted from the reaction mixture, after dilution with 10% aqueous
NaCl, using 4:1 (v/v) ether:dichloromethane in lieu of 1:1 (v/v) hexane:
ether.
1
958, 898 cm-1; H NMR (300 MHz, CDCl3) δ 2.34 (s, OH), 2.20
(20) The moderate yield of alkynediol 3c in this experiment may
arise from the difficulty in separating 3c from DMSOsi.e., removal of
the latter with several aqueous NaCl washes undoubtedly results in
partial loss of the polar diol 3c. The unblocked tertiary alcohol moiety
in 2-methyl-3-butyn-2-ol did not appreciably slow down the alkynyla-
tion process since only a minor amount (5-10%) of unreacted cyclo-
hexanone was detected in the crude product mixture.
(11) The reaction flask was connected to an apparatus similar to
that described by J ohnson, W. S.; Schneider, W. P. Org. Synth. 1950,
30, 18.
(12) Picard, P.; Moulines, J . Bull. Soc. Chim. Fr., Part II 1974, 2256.
(13) Miyashita, N.; Yoshikoshi, A.; Grieco, P. A. J . Org. Chem. 1977,
42, 3772.
(14) Purchased from Lancaster Synthesis, Windham, NH.
(15) Claesson, A.; Bogentoft, C. Acta Chem. Scand. 1972, 26, 2540.
(16) In an identical experiment using 2-propyn-1-ol as the alkyne
in lieu of 3-[(tetrahydropyran-2-yl)oxy]propyne (1d ), the expected
adduct 3e was obtained, but the reaction rate was too slow at room
temperature (less than 30% conversion after 2 days). This result
indicates that the alkynylation process requires the blockingsor
conversion to the corresponding alkoxide saltsof primary alcohols,
although not tertiary alcohols (vide infra).
(21) Saimoto, H.; Hiyama, T.; Nozaki, H. Bull. Chem. Soc. J pn. 1983,
56, 3078.
(22) A subsequent experiment demonstrated that addition of KOC-
(CH3)3 is unnecessary for the success of this reactionsan indication
that the alkoxide derived from 4-pentyn-2-ol can serve as the catalyst
for the process. However, if KOC(CH3)3 is present as a catalyst but
4-pentyn-2-ol is not converted to an alkoxide derivative, the alkyny-
lation process is very sluggish.
(23) In a similar experiment that involved the addition of 0.20 mL
(2.2 mmol) of 2-butanone in one portion to the reaction mixture, aldol
condensation was observed as a side reaction.
(24) Favorskaya, T. A.; Medvedeva, A. S.; Vlasov, V. M.; Chich-
kareva, G. G. Izv. Akad. Nauk SSSR, Ser. Khim. 1967, 2107-9.
(25) The moderate yield in this experiment may arise from the
difficulty in separating the polar alkynediol 3f from DMSO. In a
similar experiment, the sodium alkoxide derivative 1g of 4-pentyn-2-
ol was coupled with a higher-molecular-weight methyl ketone (6-
methyl-5-hepten-2-one) and a 93% distilled yield of the less polar C-13
alkynediol (6,10-dimethyl-9-undecen-4-yne-2,6-diol) was obtained.
See: Babler, J . H. U.S. Patent 5,349,071 (Sept. 20, 1994), Example
XV.
(17) Papa, D.; Villani, F. J .; Ginsberg, H. F. J . Am. Chem. Soc. 1954,
76, 4446.
(18) Potassium methoxide, as well as powdered potassium hydroxide
(ACS reagent, 85%), can be used as the catalyst in this alkynylation
process. However, these bases were less effective than potassium tert-
butoxide since conversion to the desired adduct 3a was only ap-
proximately 40% after a reaction time of 10 h. Sodium alkoxide bases
were even less efficient catalysts for this process. For example,
replacement of KOC(CH3)3 with NaOCH3 resulted in a product mixture
containing only 15% of the desired tertiary alkynol 3a after 1 day at
room temperature.