An unexpected equilibrium process associated with a standard approach to ynone synthesis

Article abstract of DOI:10.1021/jo025682i

A direct comparison between Weinreb amides and morpholine amides was made with regard to their reactions with alkynyllithium reagents to form ynones. While treatment with stoichiometric alkynyllithium generally effects complete reaction in the case of Weinreb amides, incomplete reactions are obtained from the corresponding morpholine amides. This difference is attributed to an unexpected equilibrium process in the latter case, and it is shown that the use of excess alkynyllithium reagent with morpholine amides provides a synthetically useful synthesis of ynones at 0 °C.

Full text of DOI:10.1021/jo025682i

TABLE 1. Use of Wein r eb Am id es a t 0 °C
An Un exp ected Equ ilibr iu m P r ocess
Associa ted w ith a Sta n d a r d Ap p r oa ch to
Yn on e Syn th esis
Mary M. J ackson, Carolyn Leverett,
J ennifer F. Toczko, and J ohn C. Roberts*
productsa
% AUC % AUC
GlaxoSmithKline, Chemical Development, Five Moore Drive,
P.O. Box 13398, Research Triangle Park, North Carolina,
27709
amide
R
alkyllithium
R′-Li
entry
1
2
3
ketone byproduct
1
2
3
4
5
6
1a PhCH2CH2 2a BuC≡CLi 3a
1a PhCH2CH2 2b PhC≡CLi 3b
>99
>99
77
>99
>99
69
0
0
jcr99483@gsk.com
1a PhCH2CH2 2c BuLi
3c
7b
0
1b Ph
1b Ph
1b Ph
2a BuCtCLi 3d
2b PhCtCLi 3e
Received March 5, 2002
0
2c BuLi
3f
17b
Abstr a ct: A direct comparison between Weinreb amides
and morpholine amides was made with regard to their
reactions with alkynyllithium reagents to form ynones.
While treatment with stoichiometric alkynyllithium gener-
ally effects complete reaction in the case of Weinreb amides,
incomplete reactions are obtained from the corresponding
morpholine amides. This difference is attributed to an
unexpected equilibrium process in the latter case, and it is
shown that the use of excess alkynyllithium reagent with
morpholine amides provides a synthetically useful synthesis
of ynones at 0 °C.
a
As analyzed by HPLC; area under the curve (AUC) is
measured at 220 nm. Structures confirmed by NMR and LC/MS.
b
Consistent with loss of methoxy group by LC/MS.
amides in ynone synthesis. Ultimately, we developed a
process that permitted the replacement of our specific
Weinreb amide with its morpholine amide counterpart
to produce the desired ynone at 0 °C without the use of
an alkynyl borane derivative. This success prompted us
to develop a general method that might not have scale
limitations. During the course of our studies, we uncov-
ered unexpected differences between the behavior of
Weinreb amides and morpholine amides.
Although somewhat unstable as a general class, ynones
are useful intermediates in the synthesis of a variety of
heterocycles including isoxazoles and pyrazoles.1 One
commonly used method for ynone preparation is the
addition of alkynyllithium reagent, usually in excess, to
a Weinreb amide in THF, usually at low temperatures.2
Our interest in the synthesis of ynones arose from the
need to replace a Weinreb amide intermediate with a
more stable, easily prepared substitute.3 A search of the
literature indicated that morpholine amides may be used
to replace Weinreb amides in ketone synthesis,4 but in
only one reference is this applied to an ynone.4f The
reported reaction is performed at -78 °C with 3 equiv of
an alkynyl borane.4g Although not ideal for our purposes,
this report encouraged us to pursue the use of morpholine
The addition of alkynyllithium reagents to Weinreb
amides at 0 °C was examined. It was determined that
approximately 1 equiv of alkynyllithium is sufficient for
complete conversion to the ynone. For example, when
amide 1a is added to 1.1 equiv of alkynyllithium 2a at 0
°C and stirred for 1 h, ynone 3a is observed after
transferring the resulting mixture into AcOH/MeOH/
water (Table 1, entry 1). It is important that the reaction
mixture is added to an acidic solution (and not vice versa)
due to the base lability of the ynone product. These
reaction conditions hold true for all pairings of Weinreb
amides with alkynyllithium reagents as indicated in
entries 1, 2, 4, and 5. While this observation constitutes
a significant step forward for practical ynone synthesis,
there still remain compelling safety reasons to avoid the
use of Weinreb amides, at least as a general class, on
scale.3 Incidentally, 0 °C is not a suitable temperature
for the reaction of Weinreb amides with more basic
organolithium reagents such as butyllithium (entries 3
and 6).5
(1) (a) Wang, X.; Tan, J .; Zhang, L. Org. Lett. 2000, 2, 3107. (b)
Utimoto, K.; Miwa, H.; Nozaki, H. Tetrahedron Lett. 1981, 22, 4277.
(c) Arcadi, A.; Marinelli, F.; Rossi, E. Tetrahedron 1999, 55, 13233. (d)
Hojo, M.; Tomita, K.; Hosomi, A. Tetrahedron Lett. 1993, 34, 485.
(2) (a) Dixon, D. J .; Ley, S. V.; Tate, E. W. Synlett 1998, 1093. (b)
Molander, G. A.; McWilliams, J . C.; Noll, B. C. J . Am. Chem. Soc. 1997,
119, 1265. (c) Trost, B. M.; Schmidt, T. J . Am. Chem. Soc. 1988, 110,
2301. (d) In the original paper on N-methoxy-N-methylamides the
synthesis of an ynone is reported with 1.1 equiv of alkynyllithium at
65 °C. Nahm, S.; Weinreb, S. M. Tetrahedron Lett. 1981, 22, 3815.
(3) N,O-Dimethylhydroxylamine hydrochloride decomposes with the
liberation of 1.5 kJ /g. The free base begins thermal decomposition at
∼50 °C. For a report of a fatal explosion due to detonation of the
related, high-energy compound, hydroxylamine, see Chem. Eng. News
1999, Aug 16, 23.
(4) (a) Sengupta, S.; Mondal, S.; Das, D. Tetrahedron Lett. 1999,
40, 4107. (b) Brown, J . D. Tetrahedron: Asymmetry 1992, 3, 1551. (c)
Tasaka, A.; Tamura, N.; Matsushita, Y.; Kitazaki, T.; Hayashi, R.;
Okonogi, K.; Itoh, K. Chem. Pharm. Bull. 1995, 43, 432. (d) Monteith,
M. J .; Bailey, K. D.; Crosby, J . PCT Int. Appl. WO 9954272, 1999. (e)
Gomtsyan, A. Org. Lett. 1999, 2, 11. (f) Martin, R.; Romea, P.; Tey, C.;
Urp´ı, F.; Vilarrasa, J . Synlett, 1997, 1414 (g) The method was originally
reported for N,N-dimethylamides and N-methyl-N-phenylamides.
Yamaguchi, M.; Waseda, T.; Hirao, I. Chem. Lett. 1983, 35.
The performance of morpholine amides at 0 °C was
similarly assessed. In this case incomplete reaction was
consistently observed (Table 2) when conditions analo-
gous to those used with the Weinreb amides were
employed.6 For example, when amide 4a is added to 1.1
equiv of alkynyllithium reagent 2a at 0 °C, and after 1 h
the resulting mixture is transferred to the acidic medium
(5) At this temperature, a major byproduct forms as well as multiple
lower level impurities. We speculate that this observation is the origin
of the excessive care that has historically been taken in the conversion
of Weinreb amides to ynones.
10.1021/jo025682i CCC: $22.00 © 2002 American Chemical Society
Published on Web 06/11/2002
5032
J . Org. Chem. 2002, 67, 5032-5035

TABLE 2. Use of Mor p h olin e Am id es a t 0 °C
3d is observed in an 18:1 ratio versus amide 4b (Table
3, entry 3). If, however, this -20 °C mixture is treated
with Weinreb amide 1a prior to quenching, warmed to 0
°C, stirred for 1 h, and then quenched, ynone 3a becomes
the only ynone detected, and morpholine amide 4b is
cleanly regenerated (Scheme 1). This experiment was
carried out for all four -20 °C entries in Table 3, and
the same phenomenon was observed, clearly demonstrat-
ing an equilibrium process.8
productsa
amide
R
alkyllithium
R′-Li
% AUC % AUC % AUC
ketone byproduct
entry
4
2
3
4
With the equilibrium behavior qualitatively character-
ized, it follows that a synthetically viable process could
be developed at 0 °C through the use of excess alkynyl-
lithium and, indeed, this is the case (Table 4). For
example, in entry 1 the ynone/amide ratio increased from
3:1 (1 equiv of 2a ) to 61:1 (3 equiv) to 135:1 (5 equiv).
Even the ∼1:1 mixture exhibited when amide 4a and
alkynyllithium 2b are stoichiometrically combined is
boosted to a synthetically useful ratio of 23:1 by use of 5
equiv of alkynyllithium (entry 2). Due to the known
instability of ynones, the generated product would ideally
be used in a subsequent step without isolation, but it is
possible to obtain good isolated yields, when necessary,
through the use of chromatography on alumina.
1
2
3
4
5
6
4a PhCH2CH2 2a BuCtCLi 3a
4a PhCH2CH2 2b PhCtCLi 3b
75
42
34
94
70
96
0
0
22
0
0
4
25
58
37
6
30
0
4a PhCH2CH2 2c
BuLi 3c
4b
4b
4b
Ph 2a BuCtCLi 3d
Ph 2b PhCtCLi 3e
Ph 2c
BuLi 3f
a
As analyzed by HPLC, area under the curve (AUC) is
measured at 220 nm. Structures confirmed by NMR and LC/MS.
specified, 25% AUC of the starting amide is observed by
HPLC (entry 1). Longer reaction times do not increase
conversion. Although reactions of morpholine amides
with alkynyllithium reagents do not go to completion, the
reactions appear to proceed cleanly with no detectable
impurities by HPLC. As in the case of Weinreb amides,
the reaction with butyllithium produces a major byprod-
uct (entries 3 and 6). It is worth noting that in the case
of aryl amides the morpholine amide outperforms the
Weinreb amide on treatment with butyllithium at 0 °C
(ca. Table 1 entry 6 versus Table 2 entry 6).
Due to the incomplete conversion to product, there
arose the possibility that morpholine amides and alky-
nyllithium reagents are in equilbrium with their corre-
sponding adducts. By measuring the ynone/amide ratios
using HPLC with an internal standard, the existence of
an equilibrium process was demonstrated. As would be
expected for equilibrium, the ynone/amide ratio obtained
after quenching proved to be temperature dependent. For
example, when a mixture of amide 4a and alkynyllithium
2a is brought to a steady state at 0 °C (i.e., the quenched
ynone/amide ratio remains constant over time, ∼1 h), the
ynone/amide ratio is 3:1 (Table 3, entry 1). When the
mixture is warmed to 20 °C and stirred for an additional
1 h the ratio decreases to 2:1. Conversely, when the same
mixture is cooled to -20 °C and allowed to achieve a
steady state (∼2 h), this ratio increases to 4:1. This
behavior is observed for most pairings of morpholine
amides and alkynyllithium reagents (note that in entry
2 the ratio at 0 °C approaches 1:1 and therefore it is
unclear what would be expected from warming or cool-
ing).
There is an important limitation to the preparative use
of morpholine amides in ynone synthesis. Specifically,
R-alkyl-substituted morpholine amides may not be used
in practical ynone synthesis since the equilibrium lies
too far to the side of the amide. For example, when 4-(2-
ethylbutanoyl)morpholine was subjected to 2a (3 or 5
equiv), ynone/amide ratios of <0.2:1 were obtained.
Presumably this is a simple steric effect in which the
electronically stabilized adduct becomes too crowded (i.e.,
destabilized) for it to compete energetically with the
amide and alkynyllithium mixture. This sensitivity to
R-alkyl substitution appears to be unique to the morpho-
line amides as Weinreb amides of this type do convert to
ynones when treated with alkynyllithium reagents.9
We conclude that alkynyllithium reagents and mor-
pholine amides, when stirred together, are in equilibrium
with their corresponding adduct10 and that appropriate
choice of stoichiometry and temperature allows for
preparative use of morpholine amides in ynone synthesis.
Furthermore, we suggest that comparative ease of syn-
thesis and the relative thermal stability might be ad-
vantages of morpholine amides over Weinreb amides
when large-scale preparation is necessary. In contrast,
the principle advantage of Weinreb amides is that they
may be used in ynone synthesis with a single equivalent
of alkynyllithium reagent.
It was recognized that identification of a competitive
reactant that could cleanly consume amide 4 or alkynl-
lithium 2 and thereby drive the reaction back toward
starting materials would secure characterization of the
proposed equilibrium. Weinreb amides proved to be ideal
competitive reactants as previously demonstrated by
their complete reactions with alkynyllithium reagents.7
When a mixture of morpholine amide 4b and alkynyl-
lithium 2a is stirred for 1 h at 0 °C, cooled to -20 °C,
stirred for an additional 2 h, and then quenched, ynone
(7) When the equilibrium mixtures were treated with butyllithium
(to consume the amide), the results were consistent with equilibrium,
but multiple impurities complicated analysis and rendered the data
less conclusive than desired. Use of pivalaldehyde (to consume the
alkynyllithium) was also consistent, but with similar problems.
(8) Under forcing conditions, Weinreb amides also show evidence
of reversibility, but the equilibrium constant is presumably so large
that reaction reversal becomes synthetically irrelevant.
(9) For examples of alkynyllithium additions to R-alkyl substituted
Weinreb amides, see Masquelin, T.; Obrecht, D. Tetrahedron Lett.
1994, 35, 9387. Trost, B. M.; Schmidt, T. J . Am. Chem. Soc. 1988, 110,
2301.
(10) For
a related phenomenon in which 2,2-dimethylpropargyl
(6) This behavior has been noted for N,N-dimethylamides. Verk-
ruijsse, H. D.; Heus-Kloos, Y. A.; Brandsma, L. J . Organomet. Chem.
1988, 338, 289-294.
alcohols are thermally deprotected to produce acetone and a terminal
alkyne, see Boyall, D.; Lopez, F.; Sasaki, H.; Frantz, D.; Carreira, E.
M. Org. Lett. 2000, 2 (26), 4233-4236.
J . Org. Chem, Vol. 67, No. 14, 2002 5033

TABLE 3. Tem p er a tu r e Dep en d en ce of Yn on e/Am id e Ra tio
amide
R
alkynyllithium
R′CtCLi
steady-state ynone/amide ratioa
entry
4
2
3
-20 °C
0 °Cb
f
20 °C
1
2
3
4
4a
4a
4b
4b
PhCH2CH2
PhCH2CH2
Ph
2a
2b
2a
2b
BuCtCLi
PhCtCLi
BuCtCLi
PhCtCLi
3a
3b
3d
3e
4:1
r
r
r
r
3:1
f
f
f
f
2:1
0.5:1
5:1
0.8:1
18:1
2.2:1
0.9:1
12:1
1.5:1
Ph
0.04:1
a
b
Product ratios were obtained by HPLC and calculated versus an internal standard. Arrows indicate that the ratios reported in the
adjacent column were first brought to steady state at 0 °C.
SCHEME 1
TABLE 4. In cr ea se in Yn on e/Am id e Ra tio w ith Excess Alk yn yllith iu m
amide
alkynyllithium
R′CtCLi
ynone/amide ratio vs equiv alkynyl-Li
entry
4
R
2
3
1 equiv 2
3 equiv 2
5 equiv 2
1
2
3
4
5
6
4a
4a
4a
4b
4b
4b
PhCH2CH2
PhCH2CH2
PhCH2CH2
Ph
2a
2b
2d
2a
2b
2d
BuCtCLi
PhCtCLi
TMSCtCLi
BuCtCLi
PhCtCLi
TMSCtCLi
3a
3b
3g
3d
3e
3h
3:1
1:1
-
12:1
1.5:1
-
61:1 (80%)a
8:1
135:1
23:1
6:1
17:1
>200:1 (86%)a
110:1
>200:1
>200:1
-
Ph
Ph
76:1 (71%)a
a
Isolated yield after chromatography on neutral alumina.
with methanol prior to injection. The AUC ratio of 4a to internal
standard was measured for each sample by HPLC, and thus the
consumption of 4a could be calculated. This consumption of 4a
was directly attributed to the production of ynone (byproducts
were ,1% for all cases).
Exp er im en ta l Section
Gen er a l. Unless otherwise noted, starting materials were
obtained from commercial suppliers and used without further
purification. HPLC analyses were carried out using a Phenom-
enex Luna 3 µm C18(2) column (50 × 2.0 mm).
1-P h en yl-4-n on yn -3-on e (3a ). A flask containing 1-hexyne
(0.79 mL, 6.8 mmol) was charged with THF (3.0 mL) under N2.
After cooling to 0 °C, n-BuLi (2.8 mL, 6.9 mmol) was added
dropwise over 19 min, maintaining an internal temperature <11
°C. The reaction mixture was stirred at 0 °C for 15 min, and
then amide 4a (0.51 g, 2.3 mmol) in THF (1.0 mL) was added.
After 60 min at 0 °C, the reaction mixture was transferred via
cannula into a solution of AcOH (4.8 mL) and water (2.4 mL) at
0 °C. After 10 min, the organic layer was washed with H2O, dried
over Na2SO4, filtered, and concentrated to an oil (0.61 g).
Chromatography on neutral alumina (10% EtOAc/hexanes)
produced 0.40 g (80%) of 3a , 99% pure by HPLC analysis. 1H
NMR (300 MHz): δ 0.95 (t, J ) 7.3, 3H), 1.20-1.64 (m, 4H),
2.39 (t, J ) 7.0, 2H), 2.86-2.91 (m, 2H), 2.98-3.03 (m, 2H),
7.18-7.37 (m, 5H). 13C NMR (75 MHz): δ 13.40, 18.56, 21.87,
29.63, 46.90, 80.77, 94.80, 126.14, 128.14, 128.41, 140.30, 186.98.
The NMR spectra are in good agreement with the literature.11
Gen er a l P r oced u r e for Ca lcu la tion of [Yn on e]/[Am id e]
r a tios. The ratio for the 4a + 2a equilibrium is described. Amide
4a (0.52 g, 2.4 mmol) and biphenyl (0.17 g, 1.1 mmol as an
internal standard) were dissolved in THF (1.0 mL), and HPLC
analysis was used to calculate the initial AUC ratio of 4a and
internal standard. In a separate reaction vessel, 1-hexyne (0.79
mL, 6.8 mmol) and THF (3.0 mL) were introduced into a flask
under N2. After the mixture was cooled to 0 °C, n-BuLi (2.8 mL,
6.9 mmol) was added dropwise over 17 min, thereby generating
alkynyllithium 2a , while maintaining an internal temperature
<10 °C. The reaction mixture was stirred at 0 °C for 15 min,
and the 4a /internal standard solution was added. HPLC analysis
was performed at 30, 60, and 90 min after addition of the amide
solution. HPLC samples were prepared by placing the reaction
mixture (0.10 mL) into a solution of 1:1:10 AcOH:H2O:MeOH
(1.0 mL). The resulting quenched solution was diluted further
5034 J . Org. Chem., Vol. 67, No. 14, 2002

1-P h en yl-2-h ep tyn -1-on e (3d ). In a procedure analogous to
that above, amide 4b and 3 equiv of alkynyllithium reagent 2a
were combined to prepare ynone 3d (85% yield, 99% purity by
alkynyllithium reagent 2d were combined to prepare ynone 3h
(85% yield, 99% purity by HPLC). 1H NMR (300 MHz): δ 0.33
(s, 9H), 7.47-7.52 (m, 2H), 7.60-7.65 (m, 1H), 8.14-8.17 (m,
2H). 13C NMR (75 MHz): δ 0.00, 101.27, 101.50, 129.26, 130.33,
134.87, 137.14, 178.39. The NMR spectra are in good agreement
with the literature.13
1
HPLC). H NMR (400 MHz): δ 0.97 (t, J ) 7.3, 3H), 1.48-1.54
(m, 2H), 1.63-1.70 (m, 2H), 2.51 (t, J ) 7.2, 2H), 7.46-7.50 (m,
2H), 7.57-7.61 (m, 1H), 8.13-8.15 (m, 2H). 13C NMR (100
MHz): δ 13.46, 18.86, 22.02, 29.79, 79.63, 96.79, 128.43, 129.49,
133.80, 136.90, 178.20. The NMR spectra are in good agreement
with the literature.12
Ack n ow led gm en t. The authors would like to thank
Purnima Narang for her assistance with analytical
issues faced during the course of this study and Tom
Roper, Matthew Sharp, Shiping Xie, and Ken Ingold for
their useful comments on the first draft of this manu-
script.
1-P h en yl-3-(tr im eth ylsilyl)-2-p r op yn -1-on e (3h ). In a pro-
cedure analogous to that above, amide 4b and 3 equiv of
(11) Mattson, M. N.; Rapoport, H. J . Org. Chem. 1996, 61, 6071.
Note: In the citation the referenced compound (9g) is named incor-
rectly; 1-phenyl-4-heptyn-3-one rather then 1-phenyl-4-nonyn-3-one.
(12) Van den Hoven, B. G.; Bassam, E. A.; Alper, H. J . Org. Chem.
2000, 65, 4131.
J O025682I
(13) Logue, M. W.; Moore, G. L. J . Org. Chem. 1975, 40, 131.
J . Org. Chem, Vol. 67, No. 14, 2002 5035