Direct conversion of aromatic ketones to arenecarboxylic esters via carbon-carbon bond-cleavage reactions

Article abstract of DOI:10.1246/bcsj.81.369

Aromatic methyl ketones, ss-keto esters, and trifluoromethyl-l,3- diketones can be directly converted to arene-carboxylic esters via carbon-carbon bond cleavage of pyridinium iodide intermediates in the presence of copper(II) oxide, iodine, pyridine, and potassium carbonate in alcoholic media. The advantages of the present method in terms of good yields, mild reaction conditions, and inexpensive reagents should make this protocol a valuable alternative to the existing methods.

Full text of DOI:10.1246/bcsj.81.369

Ó 2008 The Chemical Society of Japan
Bull. Chem. Soc. Jpn. Vol. 81, No. 3, 369–372 (2008)
369
Direct Conversion of Aromatic Ketones to Arenecarboxylic
Esters via Carbon–Carbon Bond-Cleavage Reactions
1
1
1
2
ꢀ1
Guodong Yin, Meng Gao, Zihua Wang, Yandong Wu, and Anxin Wu
1
Key Laboratory of Pesticide and Chemical Biology, Ministry of Education, Central China Normal University,
Wuhan 430079, P. R. China
2
College of Chemistry, Xiangtan University, Xiangtan 411105, P. R. China
Received October 12, 2007; E-mail: chwuax@mail.ccnu.edu.cn
Aromatic methyl ketones, ꢀ-keto esters, and trifluoromethyl-1,3-diketones can be directly converted to arene-
carboxylic esters via carbon–carbon bond cleavage of pyridinium iodide intermediates in the presence of copper(II)
oxide, iodine, pyridine, and potassium carbonate in alcoholic media. The advantages of the present method in terms
of good yields, mild reaction conditions, and inexpensive reagents should make this protocol a valuable alternative to
the existing methods.
8
One of the most important tasks in organic chemistry is to
perform functional group inter-conversions. During various
stages in organic synthesis and natural products synthesis, it
is often required to introduce a carboxylic ester group into
molecules, which is usually achieved by oxidation of alcohols
this conversion. Recently, we have found that ꢁ-iodination
of aromatic ketones can be efficiently accomplished by using
a copper(II) oxide/iodine reagent combination in alcoholic
9
media. Herein, we report an efficient method for the direct
conversion of aromatic ketones to their corresponding carbox-
ylic ethyl esters, which involves carbon–carbon bond-cleavage
reactions.
1
and aldehydes. Molecular iodine, as one of the most simple,
least expensive, and least toxic oxidants has been widely used
to achieve the direct oxidation of alcohols or aldehydes to
carboxylic esters.2–4 Generally, the introduction of an acetyl
Results and Discussion
group onto an aromatic ring is much easier than introduction
of a hydroxy or aldehyde group, so the development of an
efficient method for the direct synthesis of carboxylic esters
from carbonyl compounds is significant. To date, few exam-
ples have been reported for performing this conversion. The
Initially, we expected to obtain intermediate 5 in one pot
from p-methoxyacetophenone (1) in the presence of copper(II)
oxide (2 equivalents), iodine (2 equivalents), and appropriate
base (2 equivalents) in methanol (Scheme 1). The results are
summarized in Table 1. Heating the reaction mixture at 100
5
ꢂ
Baeyer–Villiger reaction is the oxidative cleavage of a car-
C in a sealed tube for 24 h without any other base, ꢁ-iodo-
bon–carbon bond adjacent to a carbonyl group, which is the
classic method for the transformation of carbonyl compounds
to the corresponding carboxylic esters in the presence of per-
acids, hydrogen peroxides, or other peroxy compounds. How-
ever, this oxidation reaction often provides a pair of isomers.
ketone 3 was obtained in 90% isolated yield together with
the solvolytic product 4 in 8% yield (Entry 1). When triethyl-
amine was used as base, no reaction occurred (Entry 2). The
desired product 5 was only obtained in 6% yield and the con-
version was about 40% when using potassium carbonate as
base (Entry 3). It was interesting to find that the unexpected
product methyl 4-methoxybenzoate (2) was obtained in 40%
yield when the base was changed to pyridine. As the temper-
ature decreased from 100 to 80 C, the yield of ester 2 was
slightly lower (Entries 4 and 5).
As early as the 1940s, King and Pearson reported that reac-
tion of aryl methyl ketones with iodine and pyridine produced
pyridinium iodide, which could be hydrolyzed to the corre-
sponding carboxylic acid by aqueous alkali.10 On the base of
6
Nikishin and co-workers reported an electrocatalytic method
for transformation of methyl ketones into methyl esters. Zhang
7
and Vozzolo found that (dimethoxymethyl)dimethylamine
(DMF DMA) could promote a carbon–carbon bond-cleavage
ꢂ
ꢁ
reaction of aryl alkyl ketones to arenecarboxylic esters in
refluxing methanol. However, the substrates were limited to
electron-deficient ketones. In addition, the use of strong oxi-
dant such as KOCl (activated aromatic ketones give lower
yield of the products due to ring chlorination) can also perform
COCH3
COOCH3
COCH2I
COCH OMe
2
COCH(OMe)2
CuO, I , base
2
MeOH, Temp.
sealed tube, 24 h
OMe
OMe
OMe
OMe
OMe
1
2
3
4
5
Scheme 1.
Published on the web March 13, 2008; doi:10.1246/bcsj.81.369

3
70 Bull. Chem. Soc. Jpn. Vol. 81, No. 3 (2008)
Carbon–Carbon Bond Cleavage
Table 1. Effect of the Base on the Yield of Ester 2a)
Table 2. Effect of Base on the Cleavage of Carbon–Carbon
Bond of Pyridinium Iodide 6 to Ethyl 4-Methoxybenzoate
b)
d)
Entry
Base
Temp
ꢂ
Conv.
/%
Yield/%
a)
(
7)
c)
/
C
2
3
4
5
I-
e)
N+
1
2
3
4
5
none
NEt3
K2CO3
Py
100
100
100
100
80
100
—
40
95
—
—
—
40
38
90
—
20
—
—
8
—
10
—
—
—
—
6
—
—
EtOH
reflux
OEt
O
MeO
MeO
O
f)
6
7
f)
Py
90
Baseb)
Yield/%c)
Entry
Time/h
a) Reaction conditions: 2 mmol of p-methoxyacetophenone,
CuO (4 mmol), I2 (4 mmol), 24 h, in sealed tube for all cases.
b) NEt3 (8 mmol), K2CO3 (4 mmol), Py (8 mmol). c) Oil bath
temperature. d) Isolated yield. e) Products were not observed.
f) Pyridinium iodide was the major product.
1
2
3
4
none
Py
K2CO3
CuO
24
16
8
NR
trace
99
8
45
a) Reaction conditions: 1 mmol of pyridinium iodide 6, in
refluxing ethanol for all cases. b) 2 mmol base was employed.
c) Isolated yield.
O
O
R2
R1
R1
OR
CuO, I2
N
I
VII
R2
Table 3. Conversion of Methyl Ketone 1 to Esters in Other
a)
VI
I
Alcoholic Media
CuI
COCH3
base, ROH
COOR
CuO (2 equiv), I (2 equiv)
2
Py (4 equiv), ROH, Temp
ROH
O
N
2
4 h then K CO (2 equiv), 8 h
2
3
O
R1
O
R1
R2
OMe
Entry
I
OMe
N
R1
N
ꢂ
Yield/%b)
R2
IV
I
V
R2
ROH
Temp/ C
Product
II
1
2
3
4
5
6
MeOH
EtOH
n-PrOH
i-PrOH
n-BuOH
t-BuOH
65
78
97
100
100
5
7
8
72
80
51
46
50
0
Py, I2
c)
O
9
Py
N
I
10
11
base
R1
c)
100
R2
III
a) Reaction conditions: 2 mmol of p-methoxyacetophenone,
CuO (4 mmol), I2 (4 mmol), Py (8 mmol), ROH (20 mL) for all
cases. b) Isolated yield. c) In sealed tube, oil bath temperature.
Scheme 2. Proposed reaction mechanism.
related literature,11 our proposed pathway for this reaction is
shown in Scheme 2. Initially, ꢁ-iodo ketone II, obtained in
situ by iodination of ketone I in the presence of copper(II)
tion of pyridinium iodide (Entry 3).14 We also found that
the transformation of compound 6 to 7 could be achieved in
moderate yield even using the weak base copper(II) oxide
(Entry 4). It is concluded that copper(II) oxide can promote
not only ꢁ-iodination of ketones in the first step but also to
a certain extent cleavage of pyridinium iodide to give target
product ester. To ensure the complete cleavage reaction of
pyridinium iodide, it is necessary to add potassium carbonate
to the solution in the second step.
Under optimized conditions, the effect of alcoholic media
on the yield of ester was investigated. The general procedure
was as follows. Treatment of substrate 1 with iodine (2 equiv-
alents), copper(II) oxide (2 equivalents), and pyridine (4 equiv-
alents) in alcoholic media at reflux for 24 h, followed by addi-
tion of potassium carbonate (2 equivalents) for an additional
8 h of reflux provided the corresponding ester was obtained
in 46–80% yield (Table 3, Entries 1–5). Heating the reaction
mixture in refluxing ethanol, desired compound 7 was obtained
in 80% yield. The reaction gave slightly lower yield in meth-
anol probably due to the effect of temperature. However,
in high boiling point alcoholic media, such as 1-propanol
(n-PrOH), 2-propanol (i-PrOH), and 1-butanol (n-BuOH), the
corresponding esters 8–10 were only isolated in moderate
9
b
12
oxide, reacts with pyridine to yield pyridinium iodide III.
Intermediate III deprotonates under mild basic conditions to
give IV, which then reacts with the nucleophilic alcohol to
deliver the carboxylic ester VI and pyridinium salt VII through
a carbon–carbon bond-cleavage reaction. It should be noted
that the possible intermediate V, obtained from III in the pres-
ence of excess iodine and pyridine, can also be transformed
into carboxylic ester VI.
In order to further confirm the reaction mechanism, we
synthesized 1-(4-methoxyphenacyl)pyridinium iodide (6) ac-
cording to a reported method13 and investigated the effect of
base on the yield of ethyl 4-methoxybenzoate (7) in refluxing
ethanol. As shown in Table 2, heating the ethanol solution of
6
without any base at reflux for 24 h, no expected product
was detected by thin-layer chromatography (TLC) analysis
Entry 1). It was found that pyridine did not promote this con-
(
version (Entry 2). Gladly, carboxylic ester 7 was obtained in
near-quantitative yield (99%) using potassium carbonate as
base because potassium carbonate or high temperature could
effectively accelerate the carbon–carbon bond-cleavage reac-

G. Yin et al.
Bull. Chem. Soc. Jpn. Vol. 81, No. 3 (2008)
371
Table 4. Conversion of Aromatic Ketones to Arenecarbox-
ylic Esters
1-Phenyl-1-butanone provided the expected ethyl benzoate
(27) in only 30% yield along with an unidentified mixture of
by-products (Entry 16). We were glad to find that ꢀ-keto esters
and trifluoromethyl-1,3-diketones also gave the corresponding
ethyl esters (7, 17, 21, and 28–30) in excellent yields (Entries
17–24). It was reasoned by the possible reaction mechanism
CuO (2 equiv), I (2 equiv),
2
O
O
Py (4 equiv), EtOH, 24-72h
R2
R1
R1
OEt
then K CO (2 equiv), 8-16h
2
3
2
that the electron-withdrawing group of R (such as –COOEt
and –COCF3) could promote the pyridinium iodide intermedi-
Entry
R1
R2
Time/h Product Yield/%b)
1
2
3
4
5
6
7
8
9
4-CH3C6H4
4-EtOC6H4
4-C6H5C6H4
4-ClC6H4
4-BrC6H4
4-NO2C6H4
3,4-Cl2C6H3
1-Naph
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Et
24/8
24/8
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
7
74
75
60
76
60
83
68
41
56
75
73
74
59
62
71
30
78
75
90
92
82
85
85
64
ate to deprotonate under basic conditions. However, the donor-
2
withdrawing group of R (such as an ethyl group) did not
facilitate this process (see Scheme 2).
40/8
48/8
48/8
Conclusion
48/10
48/10
72/12
48/12
24/8
In conclusion, we have developed an efficient method for the
direct conversion of aromatic methyl ketones, ꢀ-keto esters,
and trifluoromethyl-1,3-diketones to the corresponding arene-
carboxylic esters via carbon–carbon bond-cleavage reaction
of pyridinium iodide intermediates. The mild reaction condi-
tions, inexpensive reagents, and good yields are advantages
of the present procedure.
c)
2-Naph
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
6-Methoxy-2-naphthyl
2-Furyl
2-Thienyl
24/16
24/16
24/8
5-Cl-2-thienyl
5-Br-2-thienyl
3-Thienyl
24/8
24/16
24/10
C6H5
Experimental
4-MeOC6H4
3,4,5-(OMe)3C6H2
4-NO2C6H4
3-NO2C6H4
4-MeOC6H4
COOEt 18/10
COOEt 16/6
COOEt 16/6
COOEt 16/6
COCF3 24/10
General. Finely powdered copper(II) oxide was purchased
from commercial sources (>98%). Ethanol was freshly distilled
from sodium metal. Melting points were determined on an
28
17
29
7
1
XT4A Meltemp apparatus and are uncorrected. H spectra were
recorded on a Varian Mercury 400 spectrometer operating at
400 MHz. 13C spectra were recorded on a Varian Mercury 600
spectrometer operating at 150 MHz. Chemical shifts are reported
in ppm, relative to the internal standard of tetramethylsilane
6-Methoxy-2-naphthyl COCF3 24/10
4-NO2C6H4
21
17
30
COCF3 24/8
COCF3 18/10
PhCH2OC6H4
a) Reaction conditions: ketone (2.5 mmol), CuO (5 mmol), I2
(
TMS) or 3-(trimethylsilyl)propanesulfonate (DSS). IR spectra
(
5 mmol), Py (10 mmol), K2CO3 (5 mmol), EtOH (15 mL), at reflux
ꢂ
of samples as KBr pellets were recorded on a PE-983 spectro-
photometer. MS was carried out on a Finnigan Trace MS spec-
trometer. Column chromatography was performed on silica gel
(200–300 mesh).
for all cases. b) Isolated yield. c) In sealed tube at 100 C.
yields. When t-butyl alcohol (t-BuOH) was used as the nucleo-
philic alcohol, the expected product tert-butyl 4-methoxyben-
zoate (11) was not detected (Entry 6). It could be concluded
that the yield of ester was influenced not only by the reaction
temperature but also steric hindrance of the nucleophilic alco-
hol. In refluxing ethanol, the reaction gave the best result.
Next, we explored the scope of substrates and the limitation
of this reaction by treating a variety of aromatic ketones with
copper(II) oxide, iodine, pyridine, and potassium carbonate in
refluxing ethanol. The results are summarized in Table 4. The
reaction yield was insensitive to the electron density of the car-
bonyl group of substrates. As for compounds bearing electron-
donating or electron-withdrawing groups in the phenyl rings,
the corresponding ethyl esters 12–18 were obtained in good
yields (60–83%) although a longer reaction time was required
for the latter case (Table 4, Entries 1–7). Under refluxing etha-
nol, we found that substrate methyl 1-naphthyl ketone gave
ethyl 1-naphthoate in only 10% yield. Even when the reaction
1-(4-Methoxyphenacyl)pyridinium Iodide (6).13 A mixture
of p-methoxyacetophenone 1 (7.5 g, 0.05 mol), iodine (12.7 g,
ꢂ
0.05 mol), and pyridine (50 mL) was heated at 100 C for 3 h
and allowed to stand overnight. The reaction mixture was filtered
then thoroughly washed with 40 mL ether to remove unreacted
starting substrate. The solid was washed with 50 mL of cold water
to remove pyridine hydroiodide and compound 6 was obtained in
ꢂ
13
8
2
6
8
0% (14.2 g, 0.04 mol) isolated yield. mp 213–215 C (lit. 218–
ꢂ
1
ꢂ
19 C); H NMR (400 MHz, DMSO-d6, 20 C): ꢂ 8.98 (d, J ¼
:8 Hz, 2H), 8.73 (t, J ¼ 7:6 Hz, 1H), 8.27 (t, J ¼ 6:8 Hz, 2H),
.05 (d, J ¼ 8:8 Hz, 2H), 7.19 (d, J ¼ 8:8 Hz, 2H), 6.42 (s, 2H,
1
CH2), 3.90 (s, 3H, OCH3); H NMR (400 MHz, D2O/DSS, 20
ꢂ
C): ꢂ 8.78 (d, J ¼ 7:0 Hz, 2H), 8.68 (t, J ¼ 7:6 Hz, 1H), 8.17
(
t, J ¼ 6:8 Hz, 2H), 8.09 (d, J ¼ 8:8 Hz, 2H), 7.17 (d, J ¼ 8:8
Hz, 2H), 3.95 (s, 3H, OCH3), the signal of CH2 was not observed
due to the fast hydrogen–deuterium exchange in neutral D2O
13
ꢂ
at room temperature; C NMR (150 MHz, DMSO-d6, 20 C):
88.8, 164.2, 146.1, 130.7, 127.8, 126.2, 114.4, 66.0, 55.9 (10
resonances expected, 9 observed); IR (KBr): 1677, 1605, 1247,
176, 1027 cm
Typical Procedure for the Preparation of Ester. A mixture
ꢂ
temperature was elevated to 100 C and the reaction time was
1
prolonged to 72 h, ethyl 1-naphthoate (19) was obtained in
only 41% yield probably owing to steric hindrance (Entry 8).
Much to our satisfaction, methyl 2-naphthyl ketone, methyl
ꢃ1
1
.
6-methoxy-2-naphthyl ketone, methyl furyl ketone, and methyl
thienyl ketones all delivered the corresponding products 20–26
in good yields (56–75%) under the above optimized conditions
(Entries 9–15). With these successful results in hand, we set
out to extend the scope of substrates to non-methyl ketones.
of ketone (2.5 mmol), powdered copper(II) oxide (0.40 g, 5.0
mmol), iodine (1.27 g, 5.0 mmol), and pyridine (10 mmol) in 15
mL of anhydrous ethanol was heated at reflux for 24–72 h, then
potassium carbonate (0.69 g, 5.0 mmol) was carefully added to
the mixture and stirring was continued for 8–16 h. The mixture

3
72 Bull. Chem. Soc. Jpn. Vol. 81, No. 3 (2008)
Carbon–Carbon Bond Cleavage
was cooled to room temperature and filtered, and the solvent was
removed under reduced pressure. The residue was poured into
S. Oscarson, J. Org. Chem. 1995, 60, 2200. c) M. J. Taschner,
A. Shahripour, J. Am. Chem. Soc. 1985, 107, 5570.
3
0 mL 5% Na2S2O3 solution and extracted with EtOAc (3 ꢄ 30
2
3
H. Togo, S. Iida, Synlett 2006, 2159.
a) N. Mori, H. Togo, Tetrahedron 2005, 61, 5915. b) N.
mL), and then the organic layer was dried over anhydrous
Na2SO4. Removal of the solvent and purification of the residue
by column chromatography over silica gel with petroleum ether
as eluent to give the corresponding products.
Mori, H. Togo, Synlett 2004, 880.
a) S. Yamada, D. Morizono, K. Yamamoto, Tetrahedron
4
Lett. 1992, 33, 4329. b) K. Yamamoto, M. Shimizu, S. Yamada,
S. Iwata, O. Hoshino, J. Org. Chem. 1992, 57, 33.
1
5
ꢂ
13
4
-(Ethoxycarbonyl)biphenyl (14):
Mp 47–48 C (lit. 48–
ꢂ
1
ꢂ
4
2
2
7
1
9 C); H NMR (400 MHz, CDCl3, 20 C): ꢂ 8.12 (d, J ¼ 8:4 Hz,
H), 7.67 (d, J ¼ 8:4 Hz, 2H), 7.65–7.62 (m, 2H), 7.49–7.46 (m,
H), 7.42–7.40 (m, 1H), 4.41 (q, J ¼ 7:2 Hz, 2H), 1.42 (t, J ¼
5
a) G. A. Olah, Q. Wang, N. J. Trivedi, G. K. S. Prakash,
Synthesis 1991, 739. b) T. Kitazume, J. Kataoka, J. Fluorine
Chem. 1996, 80, 157. c) S. I. Murahashi, S. Ono, Y. Imada,
Angew. Chem., Int. Ed. 2002, 41, 2366. d) Y. Imada, H. Iida,
S. I. Murahashi, T. Naota, Angew. Chem., Int. Ed. 2005, 44,
1704.
:2 Hz, 3H); 13C NMR (150 MHz, CDCl3, 20 C): 166.4, 145.4,
ꢂ
39.9, 129.9, 129.1, 128.8, 128.0, 127.1, 126.9, 60.8, 14.2; IR
ꢃ1
(
KBr): 1710, 1606, 1277, 1115, 748 cm . MS (EI, 70 eV): m=z
%) = 226 (89), 198 (59), 181 (98), 152 (100), 127 (8), 91 (11),
(
6
a) G. I. Nikishin, M. N. Elinson, I. V. Makhova, Angew.
7
6 (29).
Ethyl 4-(Phenoxymethyl)benzoate (30):
Chem., Int. Ed. 1988, 27, 1716. b) G. I. Nikishin, M. N. Elinson,
I. V. Makhova, Tetrahedron 1991, 47, 895.
Pale white oil;
1
ꢂ
H NMR (400 MHz, CDCl3, 20 C): ꢂ 8.01 (d, J ¼ 8:8 Hz, 2H),
7
8
N. Zhang, J. Vozzolo, J. Org. Chem. 2002, 67, 1703.
a) H. R. Bjørsvik, K. Norman, Org. Process Res. Dev.
7
.45–7.35 (m, 5H), 6.99 (d, J ¼ 8:8 Hz, 2H), 5.12 (s, 2H), 4.35
13
(q, J ¼ 7:2 Hz, 2H), 1.38 (t, J ¼ 7:2 Hz, 3H); C NMR (150
MHz, CDCl3, 20 C): 166.3, 162.3, 136.2, 131.5, 128.6, 128.1,
1999, 3, 341. b) R. V. Stevens, K. T. Chapman, C. A. Stubbs,
W. W. Tam, K. F. Albizati, Tetrahedron Lett. 1982, 23, 4647.
c) R. T. Arnold, R. Buckles, J. Stoltenberg, J. Am. Chem. Soc.
1944, 66, 208.
ꢂ
1
1
27.4, 123.1, 114.3, 70.0, 60.6, 14.3; IR (KBr): 1708, 1607, 1510,
ꢃ1
278, 1258, 1168, 1107 cm . MS (EI, 70 eV): m=z (%) = 256
(
62), 211 (26), 91 (100).
9
a) G. D. Yin, B. H. Zhou, X. G. Meng, A. X. Wu, Y. J. Pan,
Org. Lett. 2006, 8, 2245. b) G. D. Yin, M. Gao, N. F. She, S. L.
Hu, A. X. Wu, Y. J. Pan, Synthesis 2007, 3113.
This work was supported by Central China Normal Univer-
sity, the National Natural Science Foundation of China (Grant
No. 20472022 and No. 20672042), and the Scientific Research
Foundation for the Returned Overseas Chinese Scholars, State
Education Ministry (No. [2005]383).
10 a) L. C. King, J. Am. Chem. Soc. 1944, 66, 894. b) L. C.
King, J. Am. Chem. Soc. 1948, 70, 242. c) R. G. Pearson,
J. Am. Chem. Soc. 1947, 69, 3100.
1
007, 9, 371.
1
B. Maki, A. Chan, E. M. Phillips, K. A. Scheidt, Org. Lett.
2
Supporting Information
12 A. D. C. Parenty, Y. F. Song, C. J. Richmond, L. Cronin,
Spectral data and 1H NMR (or 13C NMR) spectra of compounds
, 7, 12–26, and 28–30. This material is available free of charge
on the web at http://www.csj.jp/journals/bcsj/.
Org. Lett. 2007, 9, 2253.
L. C. King, M. McWhirter, R. L. Rowland, J. Am. Chem.
Soc. 1948, 70, 239.
a) K. W. Ratts, R. K. Howe, W. G. Phillips, J. Am. Chem.
1
3
6
1
4
Soc. 1969, 91, 6115. b) T. M. V. D. Pinho e Melo, Eur. J. Org.
Chem. 2006, 2873.
References
1
a) M. Hudlicky, Oxidations in Organic Chemistry, ACS
Monograph, Washington DC, 1990. b) P. J. Garegg, L. Olsson,
15 M. Rottl a¨ nder, P. Knochel, J. Org. Chem. 1998, 63,
203.