A practical and user-friendly method for the selenium-free one-step preparation of 1,2-diketones and their monoxime analogs

Article abstract of DOI:10.1055/s-2004-832815

Treatment of α-methylene ketones with excess sodium nitrite and aqueous HCl in THF at reduced temperatures provides an effective tool for the preparation of a variety of 1,2-diketones. The diastereoselective synthesis of the corresponding (Z)-1,2-dione monoximes could be accomplished under similar conditions, but by using only one equivalent of nitrosating reagent.

Full text of DOI:10.1055/s-2004-832815

LETTER
2315
A Practical and User-Friendly Method for the Selenium-Free One-Step
Preparation of 1,2-Diketones and their Monoxime Analogs
P
reparation of 1,2-
e
D
iketones
a
o
nd their Mo
r
noxime
Ag
nalogs Rüedi,*a Matthias A. Oberli,a Matthias Nagel,b Christophe Weymuth,a Hans-Jürgen Hansena
a
Organisch-chemisches Institut, Universität Zürich, Winterthurerstrasse190, 8057 Zürich, Switzerland
Fax +41(1)6356853; E-mail: georg@access.unizh.ch
b
Swiss Federal Laboratories for Materials Testing and Research (EMPA), Überlandstrasse 138, 8600 Dübendorf, Switzerland
Received 18 June 2004
allows for the formation of water-insoluble 1,2-diketones
from a-methylene ketones by modifying the procedure
described by Fileti and Ponzio. In addition, we demon-
strate that this methodology is also applicable to the syn-
Abstract: Treatment of a-methylene ketones with excess sodium
nitrite and aqueous HCl in THF at reduced temperatures provides an
effective tool for the preparation of a variety of 1,2-diketones. The
diastereoselective synthesis of the corresponding (Z)-1,2-dione
monoximes could be accomplished under similar conditions, but by thesis of the initially formed 1,2-dione monoximes by
using only one equivalent of nitrosating reagent.
only marginally changing the reaction conditions. The lat-
ter represent versatile starting compounds for the synthe-
sis of a variety of pharmaceutically active heterocycles
and a-amino acid derivatives (Scheme 1).11
Key words: 1,2-diketones, 1,2-dione monoximes, nitrosation, a-
oximation, oxidations
1,2-Diketones serve as useful intermediates in organic
synthesis and they are frequently present as important
substructures of medicinally active compounds.1,2 Usual-
ly, the employment of selenium dioxide is the almost au-
tomatic choice for 1,2-diketone formation. However, the
concomitantly generated selenium is notoriously difficult
to remove from desired end products. In addition, its tox-
icity makes techniques using selenium compounds inap-
propriate for syntheses of drugs and medicines. Thus, the
development of an efficient synthetic route to 1,2-di-
ketones from simple starting materials continues to be an
attractive task for organic chemists. Representative
methodologies that have recently been exploited include
the rearrangement of a,b-epoxyketones,3 the oxidation of
a-functionalized ketones,4 1,2-diols,5 acetylenes,6 and
epoxides7 as well as intramolecular reductive coupling
reactions of carboxylate precursors.8,9 However, while the
latter is restricted to the preparation of dimers other
methods are limited somewhat by the complexity and
expensiveness of appropriate starting materials and
reagents, unsatisfying yields provided, or their incom-
patibility with large scale applications.
O
O
NOH
R2
O
O
NaNO2, H+
R1
R2
R1
R1
R2
Scheme 1 Direct route to 1,2-diketones via monoxime inter-
mediates.
We started our investigations with the readily available
water-insoluble cyclododecanone (1, Table 1). When a
suspension of 1 and a one molar equivalent of sodium ni-
trite in THF was treated with excess concentrated HCl at
room temperature for six hours, we obtained solely cy-
clododecane-1,2-dione monoxime 2 possessing a Z-dou-
ble bond in 75% yield (entry 1).12 Apparently, the dilute
mineral acid medium is unable to mediate subsequent hy-
drolysis to the corresponding diketone 3. On the other
hand, using a threefold excess of the nitrosating reagent
enabled the formation of 3. Besides 5% of 2, the desired
product 3 was isolated in 70% yield (entry 2). A reason-
able mechanistic explanation might involve the disruption
of the intramolecular hydrogen bond in 2 mediated by an
additional amount of the reactive nitrosyl cation. It is fur-
ther noteworthy to state that oximation occurred only at
one a-position, although both would be accessible. Add-
ing the acid at low temperatures significantly increased
the yield of both 2 (98%) and 3 (85%, entry 3).13 Interest-
ingly, as the 1,2-dione monoxime formation occurs faster
than the subsequent hydrolysis step an increase in the re-
action rate of monoxime 2 could be accomplished by us-
ing excess sodium nitrite and quenching the reaction after
30 minutes (entry 4). However, applying these conditions
in conjunction with prolonged reaction times adversely
affected the yield of 3.14 Exchanging THF for dioxane had
only marginal effects on the reaction (entry 5). Similar re-
sults but lower conversion rates were obtained when HCl
was replaced with H2SO4 (entry 6). On the other hand,
using formic acid or acidic acid resulted in essentially no
On the other hand, Fileti and Ponzio reported more than a
century ago that 1,2-diketones can be prepared by treating
a-methylene ketones with sodium nitrite in aqueous
HCl.10 However, this procedure has been limited to a few
applications of water-soluble substrates. Although a range
of alkyl nitrites has been applied in order to run the reac-
tion in organic solvents, the desired 1,2-diketones could
only be obtained via additional hydrolysis of the isolated
monoxime precursors. Herein, we describe a surprisingly
simple, but highly effective and practical method that
SYNLETT 2004, No. 13, pp 2315–2318
0
3
.1
1
.2
0
0
4
Advanced online publication: 24.09.2004
DOI: 10.1055/s-2004-832815; Art ID: G21104ST
© Georg Thieme Verlag Stuttgart · New York

2316
G. Rüedi et al.
LETTER
reaction (entries 7 and 8). To our very surprise, replacing give the oxidized products in good-to-excellent yields at
HCl with HBr dramatically affected the course of the re- comparable conversion rates (entries 6 and 7).
action. Neither of the expected products 2 and 3 could be
observed, instead we obtained 2-bromo cyclododecanone
We have further extended our investigations to acyclic
substrates (Scheme 2). Oximation of 2-decanone 7 under
in almost quantitative yields (entry 9). In striking contrast
the same conditions proceeded smoothly via the thermo-
to the in situ generated nitrosyl chloride acting as a N-
dynamically more stable enol form to give only the meth-
yl ketone derivatives 8a and 8b, respectively. On the other
hand, the isomeric 3-decanone 9 failed to undergo regio-
electrophile, its bromide analog offers an excellent Br+-
source.
With optimized conditions in hand, the scope of this reac- selective oxidation. The decane-2,3-dione 10a,b and de-
tion was investigated with a range of cyclic ketones by cane-3,4-dione counterparts 11a,b were formed in
varying ring size and a-substituents (Table 2).15,16 All comparable yields.
reactions were carried out at room temperature in THF
O
O
using equimolar amounts of sodium nitrite for the prepa-
ration of 1,2-dione monoximes 5 while three equivalents
were employed to furnish 1,2-diketones 6. The eight-
membered substrate 4a turned out to be the by far most
reactive. Whereas diketone 6a was obtained quantitative-
ly after 12 minutes the corresponding monoxime 5a
could only be isolated when the reaction was conducted at
–20 °C (entry 1). This behavior was even more pro-
nounced by passing to smaller carbocyclic systems. Treat-
ing cyclopentanone and cyclohexanone under similar
reaction conditions resulted in the exclusive formation of
oligo- and/or polymers. However, these substrates were
successively converted to the corresponding 1,2-dike-
tones by following the procedure described by Fileti and
Ponzio.10 The medium and large ring cyclic ketones 4b–e
broadly followed the above mentioned trends, although
longer reaction times were required and the 1,2-diketone
formation did not go to completion (ca. 80% conversion).
Both the 1,2-diketones 6b–e and their respective mon-
oxime precursors 5b–e were obtained in good yields (en-
tries 2–5). The reaction of unsymmetrical substrates 4f,g
having a substituent in a-position also performed well to
i) or ii)
C6H13
C6H13
X
7
8
via i) X = NOH: 8a (86%)
via ii) X = O:
8b (81%)
X
X
i) or ii)
+
C6H13
C6H13
C6H13
11
O
O
O
9
10
via i) X = NOH: 10a (41%) + 11a (44%)
via ii) X = O: 10b (35%) + 11b (39%)
Scheme 2 Reagents and conditions: i) NaNO2 (1 equiv), HCl, THF,
0 °C; ii) NaNO2 (3 equiv), HCl, THF, 0 °C.
In an effort to determine whether this procedure is also ap-
plicable to monoterpene derivatives, we treated a-cam-
pholanic acid derivative 1217 under similar reaction
conditions (Scheme 3). In agreement with the results ob-
tained above, the desired diketone 13 was formed in 78%
yield (based on recovered starting material),18 whereas its
Table 1 Formation of 2 and 3 as a Function of Nature and Stoichiometry of the Nitrosating Reagent
1,2-Dione Monoxime 2
1,2-Diketone 3
NaNO2 (equiv) Time (h)
Entry Solvent
Acid
HCl
Temp (°C) NaNO2 (equiv) Time (h)
Yield (%)
Yield (%)
1
2
3
4
5
6
7
8
9
THF
THF
THF
THF
Dioxane
THF
THF
THF
THF
25
25
0
1
3
1
5
1
1
3
3
1
2
75
5
1
3
3
5
3
3
6
6
3
6
8
0
70
85
73
84
80b
0
HCl
8
HCl
3
98
97
96
90a
3
8
HCl
0
0.5
2
5
HCl
0
8
H2SO4
AcOH
HCOOH
HBr
0
24
24
48
48
10
0
30
30
4
0
4
0
5
0
0
a Based on recovered 1 (33%).
b Based on recovered 1 (15%) and unhydrolyzed 2 (46%).
Synlett 2004, No. 13, 2315–2318 © Thieme Stuttgart · New York

LETTER
Preparation of 1,2-Diketones and their Monoxime Analogs
2317
Table 2 Ring Size and Substituent Variation
1,2-Dione Monoxime 5
NaNO2 (equiv) Time (h)
1,2-Diketone 6
Entry
n
R
Product
Yield (%)a NaNO2 (equiv) Time (h)
Yield (%)
1
2
3
4
5
6
7
8
H
a
b
c
d
e
f
1
1
1
1
1
1
1
0.1
5
55b
84
87
81
85
89
90
3
3
3
3
3
3
3
0.2
10
98a
87c
89c
83c
84c
89c
94c
10
11
14
15
12
12
H
H
4
6
12
12
8
H
5
H
6
Me
OMe
4
g
3
8
a Isolated yield.
b The reaction was conducted at –20 °C.
c Based on recovered material (ca. 80% conversion).
O
(2) (a) Angelastro, M. R.; Mehdi, S.; Burkhart, J. P.; Peet, N. P.;
Bey, P. J. Med. Chem. 1990, 33, 11. (b) Khan, F. A.;
Prabhudas, B.; Dash, J.; Sahu, N. J. Am. Chem. Soc. 2000,
122, 9558. (c) Nicolaou, K. C.; Gray, D. L. F.; Tae, J. J. Am.
Chem. Soc. 2003, 126, 613.
NaNO2, HCl
dioxane, r.t.
Ph
Ph
O
O
78%
12
13
Scheme 3
(3) (a) Rao, T. B.; Rao, J. M. Synth. Commun. 1993, 23, 1527.
(b) Suda, K.; Baba, K.; Nakajima, S.; Takanami, T. Chem.
Commun. 2002, 2570. (c) Chang, C.-L.; Kumar, M. P.; Liu,
R.-S. J. Org. Chem. 2004, 69, 2793.
(4) (a) Williams, D. R.; Robinson, L. A.; Amato, G. S.;
Osterhout, M. H. J. Org. Chem. 1992, 57, 3740. (b) Chang,
S.; Lee, M.; Ko, S.; Lee, P. H. Synth. Commun. 2002, 32,
1279. (c) Rao, T. V.; Dongre, R. S.; Jain, S. L.; Sain, B.
Synth. Commun. 2002, 32, 2637.
(5) (a) Khurana, J. M.; Kandpal, B. M. Tetrahedron Lett. 2003,
44, 4909. (b) Jain, S. L.; Sharma, V. B.; Sain, B.
Tetrahedron Lett. 2004, 45, 1233.
(6) (a) Yusubov, M. S.; Filimonov, V. D.; Vasilyeva, V. P.; Chi,
K. Synthesis 1995, 1234. (b) Rogatchov, V. O.; Filimonov,
V. D.; Yusubov, M. S. Synthesis 2001, 1001. (c) Yusubov,
M. S.; Zholobova, G. A.; Vasilevsky, S. F.; Tretyakov, E. V.;
Knight, D. W. Tetrahedron 2002, 58, 1607.
(7) Antoniotti, S.; Dunach, E. Chem. Commun. 2001, 2566.
(8) For recent transition and lanthanide metal-mediated
examples, see: (a) Baruah, B.; Bourah, A.; Prajapati, D.;
Sandhu, J. S. Tetrahedron Lett. 1997, 38, 6717. (b) Baek,
H. S.; Lee, S. J.; Yoo, B. W.; Ko, J. J.; Kim, S. H.; Kim, J.
H. Tetrahedron Lett. 2000, 41, 8097. (c) Saikia, P.; Laskar,
D. D.; Prajapati, D.; Sandhu, J. S. Tetrahedron Lett. 2002,
43, 7525. (d) Wang, X.; Zhang, Y. Tetrahedron 2003, 59,
4201.
monoxime analog could not be observed. Fortunately, 13
suffered no epimerization to give the corresponding cis-
compound.
In conclusion, we have presented a very convenient a-ox-
imation method that allows for the selective preparation of
1,2-dione monoximes and/or their hydrolyzed 1,2-dike-
tone derivatives simply by using the appropriate amount
of the in situ generated nitrosyl chloride. The versatility
and very high efficiency coupled with the synthetic
significance of 1,2-dicarbonyl compounds makes this pro-
cedure a powerful tool in organic synthesis. In addition,
the reaction can be run on a preparative scale with no sig-
nificant change in yield,19 rendering this procedure also
very promising for industrial applications.
Acknowledgment
This work was generously supported by the Swiss National Science
Foundation (SNF).
(9) For recent electroreductive coupling reactions, see:
(a) Hebri, H.; Dunach, E.; Heintz, M.; Troupel, M.;
Perichon, P. Synlett 1991, 901. (b) Kashimura, S.; Murai,
Y.; Ishifune, M.; Masuda, H.; Murase, H.; Shono, T.
Tetrahedron Lett. 1995, 36, 4805. (c) Kise, N.; Ueda, N.
Bull. Chem. Soc. Jpn. 2001, 74, 755. (d) Heintz, M.;
Devaud, M.; Hebri, H.; Dunach, E.; Troupel, M.
Tetrahedron 1993, 49, 2249.
References
(1) (a) Wright, M. W.; Welker, M. J. Org. Chem. 1996, 61, 133.
(b) Trost, B. M.; Schroeder, G. M. J. Am. Chem. Soc. 2000,
122, 3785.
Synlett 2004, No. 13, 2315–2318 © Thieme Stuttgart · New York

2318
G. Rüedi et al.
LETTER
(10) Fileti, M.; Ponzio, G. Gazz. Chim. Ital. 1895, 25, 239.
(11) (a) Campillo, N.; Garcia, C.; Goya, P.; Paez, J. A.; Carrasco,
E.; Grau, M. J. Med. Chem. 1999, 42, 1698. (b) Fröhlich, L.
G.; Kotsonis, P.; Traub, H.; Tagavi-Moghadam, S.; Al-
Masoudi, M.; Hoffmann, H.; Strobel, H.; Matter, H.;
Pfleiderer, W.; Schmidt, H. H. H. W. J. Med. Chem. 1999,
42, 4108.
36.47 (t, C12), 29.86 (t), 26.34 (t), 26.17 (t), 24.95 (t), 23.79
(t), 23.70 (t), 22.09 (t), 14.45 (q, CH3) ppm. MS (EI): m/z (%)
C13H22O2 (210.32) = 210 (7) [M+], 192 (1), 167 (3), 153 (6),
139 (8), 125 (22), 112 (29), 98 (37), 83 (24), 69 (51), 55
(100). Anal. Calcd for C13H22O2 (210.32): C, 74.24; H,
10.54. Found: C, 74.44; H, 10.60.
(Z)-3-Methylcyclododecane-1,2-dione-1-oxime (5f):
colorless crystals (0.99 g); mp 97–98 °C (from hexane–
EtOAc). 1H NMR (300 MHz, CDCl3): d = ca. 8 (s, 1 H,
NOH), 3.62 (qdd, 3J = 7.0 Hz, 3J = 6.9 Hz, 3J = 3.6 Hz, 1 H,
H–C12), 2.89–2.80 (m, 1 H), 2.65–2.57 (m, 1 H), 1.84–1.76
(m, 1 H), 1.66–1.42 (m, 3 H), 1.37–1.16 (m, 12 H), 1.09 (d,
3J = 6.9 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3):
d = 203.15 (s, C1), 160.06 (s, C2), 39.64 (d, C12), 31.38 (t,
C3), 26.09 (t, 2 C), 25.04 (t), 23.42 (t), 22.87 (t), 22.65 (t),
21.89 (t), 21.19 (t), 15.12 (q, CH3) ppm. MS (EI): m/z (%)
C13H23NO2 (225.32) = 225 (4) [M+], 208 (26), 197 (5), 180
(42), 168 (11), 138 (18), 124 (26), 110 (28), 96 (34), 82 (37),
69 (45), 55 (100). Anal. Calcd for C13H23NO2 (225.33): C,
69.29; H, 10.29; N, 6.22. Found: C, 69.18; H, 10.19; N, 6.11.
3-Methoxycyclododecane-1,2-dione (6g): yellow oil (1.23
g). 1H NMR (300 MHz, CDCl3): d = 4.85 (t, 3J = 4.4 Hz, 1
H, H–C3), 3.38 (s, 3 H, OCH3), 2.23–2.14 (m, 1 H), 1.90–
1.83 (m, 2 H), 1.73–1.66 (m, 2 H), 1.53–1.49 (m, 1 H), 1.39–
1.02 (m, 12 H) ppm. 13C NMR (75 MHz, CDCl3): d = 202.10
(s), 199.00 (s), 81.14 (d, C3), 57.61 (q, OCH3), 36.71 (t,
C12), 27.30 (t), 26.21 (t), 26.17 (t), 24.33 (t), 23.29 (t), 22.03
(t), 21.83 (t), 19.48 (t) ppm. MS (EI): m/z (%) C13H22O2
(210.32) = 226 (1) [M+], 198 (8), 148 (5), 138 (8), 127 (6),
109 (13), 96 (19), 82 (43), 71 (100), 55 (34). Anal. Calcd for
C13H22O3 (226.32): C, 68.99; H, 9.80. Found: C, 69.12; H,
9.85.
(Z)-3-Methoxycyclododecane-1,2-dione-1-oxime (5g):
colorless crystals (1.86 g); mp 100–102 °C (from hexane–
EtOAc). 1H NMR (300 MHz, CDCl3): d = 7.50 (s, 1 H,
NOH), 4.88–4.86 (m, 1 H, H–C12), 3.37 (s, 3 H, OCH3),
2.94–2.84 (m, 1 H), 2.65–2.58 (m, 1 H), 1.97–1.86 (m, 2 H),
1.55–1.37 (m, 2 H), 1.36–1.02 (m, 12 H) ppm. 13C NMR (75
MHz, CDCl3): d = 198.16 [s, C(1)], 159.63 [s, C(2)], 82.52
[d, C(12)], 57.46 (q, OCH3), 29.03 (t), 26.24 (t), 26.16 (t),
24.02 (t), 22.81 (t), 22.50 (t), 21.91 (t), 21.78 (t), 18.80
(t)ppm. MS (EI): m/z (%) C13H23NO2 (225.33) = 241 (2)
[M+], 224 (58), 196 (95), 182 (17), 166 (14), 120 (23), 110
(21), 96 (27), 71 (100), 55 (52). Anal. Calcd for C13H23NO3
(241.33): C, 64.70; H, 9.61; N, 5.80. Found: C, 64.83; H,
9.70; N, 5.88.
(12) The 1H NMR of 2 showed a significantly low field shifted
singlet at d = 8.53 ppm accounting for an OH group being
part of a hydrogen bridge.
(13) MS analysis provided evidence for the formation of a minor
amount of a dehydrogenated form of 2, that was purified by
silica gel chromatography, characterized and proved to be 2-
nitroso-cyclododec-2-enone (1% yield), colorless oil. 1H
NMR (300 MHz, CDCl3): d = 6.91 (t, 3J = 7.7 Hz, 1 H), 2.74
(t, 3J = 6.7 Hz, 2 H), 2.46 (dt, 3J = 7.7 Hz, 3J = 6.2 Hz, 2 H),
1.80–1.66 (m, 2 H), 1.46 (quint., 3J = 6.2 Hz, 2 H), 1.34–
1.23 (m, 10 H) ppm. 13C NMR (75 MHz, CDCl3): d = 198.23
(s, C1), 140.99 (d, C3), 133.57 (s, C2), 37.53 (t), 29.07 (t),
26.55 (t), 25.66 (t), 24.79 (t), 24.75 (2 C, t), 24.72 (t), 23.95
(t)ppm. MS (EI): m/z (%) for C12H19NO2 (209.29) = 209 (19)
[M – NO]+, 161 (16), 149 (10), 143 (23), 132 (27), 111 (32),
98 (100), 84 (67), 55 (71).
(14) In a control experiment we could show that diketone 3 did
not react further to give the corresponding 1,2,3-trione, even
if large excess of sodium nitrite in combination with longer
reaction times was used.
(15) General Procedure for the Prearation of 1,2-Diketones
and 1,2-Dione Monoximes.
A suspension of the starting ketone (50 mmol) and NaNO2
(10.35 g, 150 mmol) in THF (100 mL) was cooled to 0 °C.
To this mixture, concd HCl (65 mL) was added in such a way
that the temperature did not exceed 10 °C. In order to avoid
the evolution of nitrous gases the acid was added via cannula
that was immerged into the reaction mixture. After the
addition the cooling bath was removed and the suspension
turned dark yellow. The progress of the reaction was
monitored by GC. After the starting material had vanished
(0.1–12 h) the reaction mixture containing the crude 1,2-
diketone was poured into a separatory funnel containing
crushed ice (200 g) and Et2O (100 mL). The organic layer
was separated, and the aqueous phase extracted with Et2O (3
× 100 mL). The combined organic layers were washed with
a sat. aq solution of NaHCO3 (100 mL) and with brine (100
mL), dried over MgSO4, filtered and concentrated under
reduced pressure. The residue was purified by filtration over
a pad of silica gel (5:1) using hexane–EtOAc (50:1) as
eluent. The corresponding 1,2-dione monoximes were
prepared accordingly by using only 3.45 g (50 mmol) of
NaNO2 and 25 mL of concd HCl. The crude product was
purified either by filtration over a pad of silica gel (5:1) using
hexane–EtOAc (50:1) as eluent or by crystallization from
hexane–EtOAc.
(17) Rüedi, G.; Nagel, M.; Hansen, H.-J. Org. Lett. 2003, 5, 2691.
(18) (+)-trans-2-(2,2,3-Trimethylcyclopentyl)-1-phenyl-
ethanedione (13): bright yellow oil; [a]D23 +39.2 (c 1.6,
CH2Cl2). 1H NMR (300 MHz, CDCl3): d = 8.03 (dd, 3J = 8.2
Hz, 4J = 1.4 Hz, 2 H), 7.64–7.59 (m, 1 H), 7.52–7.47 (m, 2
H), 3.61 (dd, 3J = 8.3, 6.4 Hz, 1 H), 2.10–1.24 (m, 5 H), 0.97
(s, 3 H), 0.95 (s, 3 H), 0.87 (d, 3J = 6.9 Hz, 3 H) ppm. 13
C
(16) All new compounds were fully characterized. Selected data
for novel compounds:
NMR (75 MHz, CDCl3): d = 205.83 (s), 192.16 (s), 135.10
(s), 134.17 (d), 130.40 (d), 128.70 (d), 55.64 (d), 45.75 (s),
44.27 (d), 31.74 (t), 24.92 (q), 24.81 (t), 24.22 (q), 14.44 (q)
ppm. MS (EI): m/z (%) C16H20O2 (244.33) = 244 (3) [M+],
139 (59) [M – COPh]+, 111 (97), 105 (82) [COPh+], 77 (75),
69 (100), 55 (94).
3-Methylcyclododecane-1,2-dione (6f): yellowoil(1.42 g).
1H NMR (300 MHz, CDCl3): d = 3.52 (qdd, 3J = 7.0 Hz,
3J = 6.9 Hz, 3J = 3.5 Hz, 1 H, H–C3), 3.36–3.27 (m, 1 H),
2.26–2.17 (m, 1 H), 1.82–1.68 (m, 3 H), 1.38–1.17 (m, 13
H), 1.10 (d, 3J = 6.9 Hz, 3 H, CH3) ppm. 13C NMR (75 MHz,
CDCl3): d = 204.07 (s, C2), 201.93 (s, C1), 38.34 (d, C3),
(19) Several reactions were conducted on a 20 g scale.
Synlett 2004, No. 13, 2315–2318 © Thieme Stuttgart · New York