Synthesis of vinyl 1,2-diketones

Article abstract of DOI:10.1016/j.tetlet.2004.03.157

A new route is outlined for preparation of vinyl 1,2-diketones via a three-step sequence. First, allylic alcohols are photooxidized by ¹O₂ to hydroperoxides, which are reduced to vinyl 1,2-diols. These vinyl 1,2-diols are oxidized to vinyl 1,2-diketones with oxoammonium salts, which are prepared in situ from organic nitroxyl radicals. The new route is short, avoids the use of protecting groups, and is generally applicable to obtain aliphatic or aromatic vinyl 1,2-diketones.

Full text of DOI:10.1016/j.tetlet.2004.03.157

Tetrahedron
Letters
Tetrahedron Letters 45 (2004) 4057–4059
Synthesis of vinyl 1,2-diketonesq
Lothar W. Habel,a Sigrid De Keersmaecker,b Joos Wahlen,a Pierre A. Jacobsa and
Dirk E. De Vosa,*
aCentre for Surface Chemistry and Catalysis, K.U.Leuven, Kasteelpark Arenberg 23, 3001 Leuven, Belgium
bCentre for Microbial and Plant Genetics, K.U.Leuven, Kasteelpark Arenberg 20, 3001 Leuven, Belgium
Received 3 February 2004; revised 24 March 2004; accepted 26 March 2004
Abstract—A new route is outlined for preparation of vinyl 1,2-diketones via a three-step sequence. First, allylic alcohols are
1
photooxidized by O2 to hydroperoxides, which are reduced to vinyl 1,2-diols. These vinyl 1,2-diols are oxidized to vinyl 1,2-di-
ketones with oxoammonium salts, which are prepared in situ from organic nitroxyl radicals. The new route is short, avoids the use
of protecting groups, and is generally applicable to obtain aliphatic or aromatic vinyl 1,2-diketones.
ꢀ 2004 Published by Elsevier Ltd.
As part of a program focusing on the synthesis of
polyfunctional C4 and C5 ketoalcohols, we wanted to
obtain vinyl 1,2-diketones as intermediate compounds.
The vinyl group in these molecules can easily be further
oxyfunctionalized, for example, by epoxidation or by
dihydroxylation. Moreover, 1,2-diketones are widely
employed in the preparation of heterocycles.1 However,
there are so far surprisingly few reports on the synthesis
of vinyl 1,2-diketones (R0 ¼ H) or related compounds.
triazole derivatives, some alkenyl 1,2-diketones have
been obtained in a few steps,4 but this method has not
been demonstrated for vinyl 1,2-diketones. Finally, a
substituted styryl 1,2-diketone was prepared by Wads-
worth–Emmons reaction of an aromatic aldehyde with a
(2-keto-3,3-dimethoxybutyl)phosphonate.5 Summariz-
ing, there is no general route to vinyl 1,2-diketones,
which starts from simple precursors, and is generally
useful for preparative purposes.
We here report an alternative, widely applicable route to
vinyl 1,2-diketones with various substituents. A first key
step is the photooxidation of allylic alcohols, which
brings the double bond at the desired terminal position.
In the last step, vinyl 1,2-diols are oxidized to vinyl 1,2-
diketones using oxoammonium salts.
O
R'
R
O
The only reported preparation of a vinyl 1,2-diketone is
that of methyl vinyl 1,2-diketone (R0 ¼ H; R ¼ CH3),
which was obtained by a pyrolytic retro-Diels–Alder
reaction of 1-(bicyclo[2.2.1]hept-5-ene-2-yl)-1,2-propane-
dione at 600 ꢁC in high vacuum.2 Other routes have
been reported for preparation of alkenyl 1,2-diketones
(R0 ¼ alkyl).3–5 A fairly general method is the conden-
sation of 2,3-butanedione with aldehydes.3 However,
this method only provides satisfactory results with
activated aldehydes such as enals. Starting from benzo-
The newly proposed route starts from simple precursors,
and comprises three steps as outlined in Scheme 1. First,
crotonaldehyde reacts with bromomagnesium or lithium
nucleophiles to obtain the variously substituted unsat-
urated alcohols 1. Under visible light irradiation and in
the presence of tetraphenylporphyrin and O2, these
1
unsaturated alcohols are oxidized by O2 to the unsat-
urated hydroxy hydroperoxides.6;7 Generally, the more
apolar allylic alcohols, for example, 1c, require longer
reaction times to reach acceptable conversion levels.
1
This could well be due to quenching of the O2 by the
C–H bonds in the long alkyl chains.
Keywords: Vinyl diketones; Nitroxyl radicals; TEMPO oxidation.
q Supplementary data associated with this article can be found, in the
online version, at doi:10.1016/j.tetlet.2004.03.157
The hydroperoxides, which were formed in the pho-
tooxygenation were reduced in situ to the vinyl 1,2-diols
2. A stoichiometric amount of triphenylphosphine
* Corresponding author. Tel.: +32-16-32-1465; fax: +32-16-32-1998;
e-mail: dirk.devos@agr.kuleuven.ac.be
0040-4039/$ - see front matter ꢀ 2004 Published by Elsevier Ltd.
doi:10.1016/j.tetlet.2004.03.157

4058
L. W. Habel et al. / Tetrahedron Letters 45 (2004) 4057–4059
monohydrate was added to the cooled solution (0 ꢁC) of
the diol in dichloromethane. Then the 4-acetamido-
TEMPO in dichloromethane was added slowly over a
period of 15–30 min.14
LiR, Et2O, -80°C or
RMgBr, Et2O, 0°C
O
OH
H
R
1
O2, hν, CCl4, 0°C,
5,10,15,20-tetraphenyl-21H,23H-porphine
OH
The reaction takes place in two steps. The first oxidation
can be performed without or with only a slight excess of
the oxidation mixture (Scheme 2); it oxidizes the allylic
alcohol rather than the homoallylic alcohol, and thus
leads to the acyloins 4 in almost quantitative yields.15 To
form the diketones 3, the reaction should be performed
with 2.5 equiv of 4-acetamido-TEMPO and p-toluene-
sulfonic acid monohydrate for a prolonged time, usually
5–7 days. The two successive oxidations can be done in
one pot; or alternatively, the intermediate 4 can be iso-
lated and then subjected to a second reaction. In
another, more lengthy approach, the intermediate 4 may
be formed directly from the hydroperoxide, via acetyl-
ation with Ac2O, base-catalyzed Kornblum-de la Mare
rearrangement to the unsaturated ketone and deacetyl-
ation to form the acyloin 4.16
R
OOH
Ph3P,
0°C
4-acetamido-TEMPO,
pTsOH, CH2Cl2
O
OH
R
R
OH
O
Me
R =
a
b
c
d
e
f
2
3
n-Bu
n-Hexyl
i-Pr
Ph
2-Ph-ethyl
Scheme 1. Synthesis route for vinyl 1,2-diketones.
proved appropriate for this reduction, which was run
overnight. After chromatography, the vinyl 1,2-diols are
obtained in high purity.7 NMR analysis of the reduced
products shows indeed that the attack by the perepoxide
intermediate of the 1O2 reaction proceeds with high
selectivity on the terminal methyl group, resulting in the
migration of the double bond to the desired terminal
position.6
On TLC, the eventual products 3 can be distinguished
from the starting materials 1 and from the intermediates
4 by their more lipophilic character and by their bright
yellow color. In the clear solution the color of the
diketone is hidden by the orange color of the excess 4-
acetamido-TEMPO, but when the excess of this reagent
is destroyed, for example, by the addition of methanol,
the residual yellow color originating from the diketone
can be used as a measure for the progress of the reac-
tion. The diketones can be quite easily purified by
chromatography on a silica gel column with diethyl
ether as the eluent. The desired compounds elute much
before all other compounds of the mixture and are
exceptionally easy to follow by their color.
In the oxidation of the 1,2-diols 2 to the 1,2-diketones 3,
the first oxidation may be easy, but the second alcohol
group will be more difficult to oxidize because of the
adjacent electron-withdrawing ketone group. Several
procedures were considered for this reaction. A CrO3
oxidation has been reported for a related reaction, but
the reaction was not fully selective, and chromic acid is
an environmentally unfriendly reagent.8 Swern reagents
are also less suitable for such demanding oxidations,
since they require the maintenance of anhydrous con-
ditions and very low temperatures for extended periods
of time.9 In our hands, the oxidation of 4-pentene-2,3-
diol using DMSO and oxalyl chloride following Swern’s
procedure yielded a complex mixture of several com-
pounds. In the search for a more general and straight-
forward method, the relatively mild oxidation of
alcohols with organic oxoammonium salts was investi-
gated next.10;11 In a preferred procedure for this reac-
tion, the free radical 4-acetamido-2,2,6,6-tetramethyl-1-
piperidinyloxy (4-acetamido-TEMPO) is dispropor-
tionated in the presence of a strong acid to form the
hydroxylamine and the oxidizing oxoammonium salt in
situ.
1
The oxidation of 1,2-diols 2a–f was monitored by H
and 13C NMR analysis. This indicated that the mono
and diketones were the only products formed; there was
no evidence, for example, oxidative splitting of 1,2-
diketones. Weighed yields of the analytically pure vinyl
1,2- diketones, obtained after column chromatography,
are given in Table 1. The rather low yield for 3a, methyl
vinyl 1,2-diketone, is due to the volatility of this com-
pound. Not only aliphatic vinyl 1,2-diketones, such as
3a, 3b, 3c, and 3d, are readily obtained; the method is
equally useful for an aromatic diketone such as 3f.
Finally, the diol oxidation even leads to a high yield of
1 eq. 4-acetamido-TEMPO,
1 eq. p-TsOH
OH
OH
R
R
OH
O
2
4
Oxidation of cyclic saturated diols or even unsaturated
diols by oxoammonium salts has been reported by
Banwell et al.12 but other workers have reported that the
presence of an oxygen atom in the b-position of the
alcohol group strongly retards alcohol oxidation.13
Nevertheless, in our hands, the oxoammonium method
proved successful for the formation of vinyl 1,2-dike-
tones. In a typical reaction, solid p-toluenesulfonic acid
O
2.5 eq. 4-acetamido-TEMPO,
2.5 eq. p-TsOH
R
O
3
Scheme 2. Oxidation of vinyl 1,2-diols with oxoammonium salts.

L. W. Habel et al. / Tetrahedron Letters 45 (2004) 4057–4059
4059
Table 1. Oxidation of vinyl 1,2-diols to vinyl 1,2-diketones
chromatographic column. Isolated yields after chroma-
tography are between 11% and 60%.
8. Grove, J. F. J. Chem. Soc., Perkin Trans. 1 1985, 865.
9. Mancuso, A. J.; Huang, S.-L.; Swern, D. J. Org. Chem.
1978, 43, 2480.
10. Ma, Z.; Bobbitt, J. M. J. Org. Chem. 1991, 56, 6110.
11. de Nooy, A. E. J.; Besemer, A. C.; van Bekkum, H.
Synthesis 1996, 1153.
12. Banwell, M. G.; Bridges, V. S.; Dupuche, J. R.; Richards,
S. L.; Walter, J. M. J. Org. Chem. 1994, 59, 6338.
13. Yamaguchi, M.; Takata, T.; Endo, T. Tetrahedron Lett.
1988, 29, 5671.
Compound
R¼
–CH3
Isolated yield (%)
3a
3b
3c
3d
3e
3f
16
51
50
28
75
52
–(CH2)3CH3
–(CH2)5CH3
–CH(CH3)2
–Ph
–(CH2)2Ph
vinyl phenyl 1,2-diketone 3e, in which the phenyl group
is directly conjugated with the vinyl 1,2-diketone moiety.
These data show that, even if reaction times are long, the
newly proposed route provides access to virtually all
aliphatic or aromatic vinyl 1,2-diketones.
14. For the synthesis of vinyl 1,2-diketones, 0.5 mmol of the
diols 2a–f were dissolved in 3 mL dichloromethane and
cooled to 0 ꢁC. Then 475 mg (2.5 mmol) of p-toluenesulf-
onic acid were added and the flask was closed with a
rubber septum. Subsequently, the solution of 533 mg
(2.5 mmol) 4-acetamido-TEMPO in 5 mL dichloro-
methane was added via a needle over 15–30 min. The
reaction mixture was allowed to warm-up to room
temperature overnight and stirred for 4–7 days until TLC
control showed completion. The solvent was evaporated
and the residue chromatographed on a silica gel column
(1 · 20 cm) using diethyl ether as eluent. Methyl vinyl 1,2-
diketone, 4-penten-2,3-dione (3a). 1H NMR (CDCl3,
300 MHz): d 7.01 (dd, 1H, J ¼ 10:6, 17.6 Hz), 6.53 (dd,
1H, J ¼ 1:5, 17.6 Hz), 6.04 (dd, 1H, J ¼ 1:5, 10.6 Hz), 2.40
(s, 3H); 13C NMR (CDCl3) 198.7, 187.6, 134.1, 129.1, 24.7;
MS (EI) m=z 98 (Mþ, 8), 55 (34), 43 (100). Butyl vinyl 1,2-
diketone, 1-octene-3,4-dione (3b). 1H NMR (CDCl3): d
6.99 (dd, 1H, J ¼ 10:2, 17.6 Hz), 6.52 (dd, 1H, J ¼ 1:5,
17.6 Hz), 6.04 (dd, 1H, J ¼ 1:5, 10.2 Hz), 2.80 (t, 2H,
J ¼ 7:3 Hz), 1.60 (quintet, 2H, J ¼ 7:3 Hz), 1.36 (sextet,
2H, J ¼ 7:3 Hz), 0.93 (t, 3H, J ¼ 7:3 Hz); 13C NMR
(CDCl3): d 201.2, 188.3, 133.8, 129.6, 36.8, 25.4, 22.6, 14.1;
MS (EI) m=z, 140 (Mþ, 2), 111 (3), 85 (35), 57 (60), 55
(100). Hexyl vinyl 1,2-diketone, 1-decene-3,4-dione (3c).
1H NMR (CDCl3): d 6.99 (dd, 1H, J ¼ 10:6, 17.6 Hz), 6.52
(dd, 1H, J ¼ 1:5, 17.6 Hz), 6.03 (dd, 1H, J ¼ 1:5, 10.6 Hz),
2.80 (t, 2H, J ¼ 7:3 Hz), 1.61 (quintet, 2H, J ¼ 7:3 Hz),
1.35–1.26 (m, 6H), 0.88 (t, 3H, J ¼ 7:3 Hz); 13C NMR
(CDCl3): d 201.4, 188.4, 133.9, 129.7, 37.2, 31.9, 29.2, 23.3,
22.8, 14.4; MS (EI) m=z 168 (Mþ, 1), 139 (1), 125 (2), 113
(82), 111 (2), 85 (28), 57 (19), 55 (100). Isopropyl vinyl 1,2-
diketone, 5-methyl-1-hexene-3,4-dione (3d). 1H NMR
(CDCl3): d 6.93 (dd, 1H, J ¼ 11:0, 17.6 Hz), 6.47 (dd,
1H, J ¼ 1:5, 17.6 Hz), 6.05 (dd, 1H, J ¼ 1:5, 11.0 Hz),
2.09–2.19 (m, 1H), 1.12 (d, 6H, J ¼ 7:3 Hz); 13C NMR
(CDCl3): d 201.3, 189.4, 133.5, 130.5, 34.8, 17.4; MS (EI)
m=z 126 (Mþ, 5), 71 (34), 55 (100). Phenyl vinyl 1,2-
diketone, 1-phenyl-3-butene-1,2-dione (3e). 1H NMR
Summarizing, the new route to vinyl 1,2-diketones is
short; it starts from simple and cheap molecules, and
there is no need to use protecting or activating groups.
The singlet oxygen reaction used to obtain the diols is
only one of several possible ways to obtain the appro-
priate starting material 2 for the oxidation. The present
route to vinyl 1,2-diketones will enable further trans-
formation of these synthons, for example, via selective
oxidation, or by incorporation in heterocycles.
Supplementary information available
NMR data for vinyl 1,2-diols and acyloins.
Acknowledgements
We thank IWT (L.W.H., J.V.) for support. P.A.J. and
J.V. are indebted to the Belgian Federal Government for
support in the frame of an IAP project Supramolecular
Catalysis.
References and notes
1. Katritzky, A. R.; Rees, C. W. Comprehensive Heterocyclic
Chemistry; Pergamon: Oxford, 1996.
2. Kramme, R.; Martin, H.-D.; Mayer, B.; Weimann, R.
Angew. Chem. 1986, 98, 1134.
3. Pictet, G.; Karrer, P. Helv. Chim. Acta 1954, 37, 1720.
4. Katritzky, A. R.; Wang, Z.; Lang, H.; Feng, D. J. Org.
Chem. 1997, 62, 4125.
5. Ciufolini, M. A.; Roschangar, F. J. Am. Chem. Soc. 1998,
118, 12082.
6. (a) Adam, W.; Nestler, B. J. Am. Chem. Soc. 1993, 115,
(CDCl3):
d
7.97 (d, 2H, J ¼ 7:3 Hz), 7.66 (t, 1H,
J ¼ 7:3 Hz), 7.51 (t, 2H, J ¼ 7:3 Hz), 6.75 (dd, 1H,
J ¼ 11:0, 17.6 Hz), 6.41 (d, 1H, J ¼ 17:6 Hz), 6.24
(d, 1H, J ¼ 11:0 Hz); 13C NMR (CDCl3): d 193.7, 193.2,
135.2 (2C), 133.2, 132.9, 130.4, 129.3; MS (EI) m=z 160
(Mþ, 13), 105 (95), 77 (100), 55 (55). (2-Phenyl)ethyl vinyl
1,2-diketone, 1-phenyl-5-hexene-3,4-dione (3f). 1H NMR
(CDCl3): d 7.28–7.20 (m, 5 H), 6.98 (dd, 1H, J ¼ 10:6,
17.6 Hz), 6.50 (dd, 1H, J ¼ 1:5, 17.6 Hz), 6.02 (dd, 1H,
J ¼ 1:5, 10.6 Hz), 3.16 (t, 2H, J ¼ 7:5 Hz), 2.95 (t, 2H,
J ¼ 7:5 Hz); 13C NMR (CDCl3): d 200.0, 187.9, 140.7,
134.1, 129.4, 128.9, 128.7, 126.7, 38.7, 29.3; MS (EI) m=z
188 (Mþ, 5), 91 (15), 77 (30), 55 (100).
ꢀ
5041; (b) Adam, W.; Dıaz, M. T.; Saha-Moller, C. R.
Tetrahedron: Asymmetry 1998, 9, 589.
€
7. For the photooxidation, 20 mmol olefin was dissolved in
20 mL CCl4 at 0 ꢁC, together with 10 mg 5,10,15,20-
tetraphenyl-21H,23H-porphine. A pure oxygen atmo-
sphere was maintained, and the flask was irradiated by a
400 W lamp for 3–7 days. After reduction with PPh3, the
allylic alcohols were isolated by work-up on a SiO2
15. In a test reaction we could isolate 3-hydroxy-1-phenyl-
hex-5-en-4-one 4f with a yield of 98%.
€
16. Adam, W.; Diaz, M. T.; Saha-Moller, C. R. Tetrahedron:
Asymmetry 1998, 9, 791.