New approach to synthesize β,β-difluorohomoallylic alcohols

Article abstract of DOI:10.1016/j.jfluchem.2012.12.011

Treatment of ethyl 4,4-difluoro-3-(triethylsilyl)but-3-enoate (3) with TBAF in the presence of MS4? followed by addition of an aromatic aldehyde afforded ethyl (E)-4,4-difluoro-5-hydroxy-5-arylpent-2-enoate (1) that was a versatile intermediate for constructing a stable isostere of peptide or fluorinated sugar molecules. This strategy offers straightforward approach to synthesize such a versatile intermediate (1).

Full text of DOI:10.1016/j.jfluchem.2012.12.011

Journal of Fluorine Chemistry 147 (2013) 1–4
Contents lists available at SciVerse ScienceDirect
Journal of Fluorine Chemistry
journal homepage: www.elsevier.com/locate/fluor
Short communication
New approach to synthesize
b,b-difluorohomoallylic alcohols
*
Masaaki Omote, Tatsuya Miake, Atsushi Tarui, Kazuyuki Sato, Itsumaro Kumadaki, Akira Ando
Faculty of Pharmaceutical Sciences, Setsunan University, 45-1, Nagaotoge-cho, Hirakata 573-0101, Japan
A R T I C L E I N F O
A B S T R A C T
˚
Article history:
Treatment of ethyl 4,4-difluoro-3-(triethylsilyl)but-3-enoate (3) with TBAF in the presence of MS4A
followed by addition of an aromatic aldehyde afforded ethyl (E)-4,4-difluoro-5-hydroxy-5-arylpent-2-
enoate (1) that was a versatile intermediate for constructing a stable isostere of peptide or fluorinated
Received 19 November 2012
Received in revised form 25 December 2012
Accepted 28 December 2012
Available online 21 January 2013
sugar molecules. This strategy offers straightforward approach to synthesize such
a versatile
intermediate (1).
ß 2013 Elsevier B.V. All rights reserved.
Keywords:
b,b-Difluorohomoallylic alcohol
Triethylsilyl
TBAF
1. Introduction
(TES) group to generate the carbanion that could be isomerized to
dienolate anion, and then an additional aldehyde could be trapped
by the terminal carbanion to give 1. This strategy would realize the
fastest access to 1. Herein, we wish to report the result on
convenient synthesis of 1 from 3 (Scheme 2).
Methods for gem-difluoroalkylation of organic compounds have
been currently the focus of much synthetic interest due to
potential utility of a difluoromethylene unit as a biological and
medicinal material. 3-Bromo-3,3-difluoropropene and bromodi-
fluoroacetic derivatives are markedly accepted as convenient
building blocks for constructing difluoromethylene backbone [1].
Qing and co-workers reported syntheses of gem-difluorinated
sugar nucleosides by way of gem-difluoroalkylation with 3-bromo-
3,3-difluoropropene [2]. During roughly the same period, Percy
and co-workers attained the syntheses of 4-deoxy-4,4-difluor-
2. Results and discussion
Preparation of 3 was attained through three steps from 2,2,2-
trifluoroethanol in moderate yield [6–8]. To explore feasibility of
bond dissociation of the Si-Csp2 bond, the reaction of 3 with
benzaldehyde (4a) was conducted at 0 8C in a THF solvent in the
presence of several classes of fluoride anion (Table 1). The reaction
was sluggish in the condition with potassium fluoride and 18-
crown-6, to give 1a in 12% yield (entry 1). The condition with
cesium fluoride afforded the product in a similar yield (entry 2).
These results suggested that the use of alkali metals as counter
cation were not suitable for the reaction. In contrast, a class of
tetraalkylammonium fluoride such as tris(dimethylamino)sulfo-
nium difluorotrimethylsilicate (TASF), tetrabutylammonium
difluorotriphenylsilicate (TBAT) and tetrabutylammonium fluoride
(TBAF) seemed to promote the bond dissociation of the Si-Csp2
osugars with (3-bromo-3,3-difluoroprop-1-ynyl)benzene as
a
reagent for gem-difluoroalkylation [3]. On the other hand,
monofluoroalkene material such as 2 also constitutes an important
issue in the field of peptide chemistry because 2 is perceived as a
chemically stable isostere of dipeptide. Successful synthesis of 2
was demonstrated by Taguchi [4] and Otaka [5] through same
intermediate 1 in the both cases. While there is an obvious fact that
1 is an important platform to a variety of monofluoroalkene,
construction of 1 is not simple. Indeed, 1 has been synthesized
from ethyl bromodifluoroacetate through many reaction steps in
low chemical yield (Scheme 1).
˚
bond. Among them, TBAF in the presence of MS4A afforded the
In this view, we expected that the advent of convenient
synthetic approach to 1 will offer fruitful advance to the area of
sugar and peptide science. Considering these, we have developed a
new building block (3) for difluoromethylene unit. Several
experiments revealed that fluoride anion attacked to triethylsilyl
product in 51% yield (entry 5). In the conditions using TASF, TABT
and TBAF, fluoride anion attacked 3 completely so that 3 was not
recovered. These results indicated that TBAF was a suitable fluoride
source for this reaction.
Subsequently, we addressed optimization of the reaction
conditions with a different solvent and temperature. The results
are summarized in Table 2. Dimethylformamide worked well in
the reaction to give the product in moderate yield (entries 1 and 2,
Table 2). Other aprotic amide solvent such as DMA or NMP
* Corresponding author.
E-mail address: aando@pharm.setsunan.ac.jp (A. Ando).
0022-1139/$ – see front matter ß 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.jfluchem.2012.12.011

2
M. Omote et al. / Journal of Fluorine Chemistry 147 (2013) 1–4
Table 2
˚
Solvent effect of the reaction of 3 with PhCHO in the presence of TBAF and MS4A.
Entrya
Solvent
Temp.
Time (h)
Yield (%) 1ab
1
2
3
4
5
6
7
8
9
DMF
DMF
DMA
NMP
Et2O
THF
0
À20
0
1.5
1.5
1.5
1.5
5
57
54
41
36
32
51
64
58
n.r.
0
0
Scheme 1. Structure of key intermediate 1 to synthesize peptide isoster 2.
0
2
THF
À20
À40
À78
2
THF
2
THF
12
a
Reaction was carried out with 3 (1 mmol), 4a (2 mmol) and F anion (1.5 mmol).
b
Isolated yield.
Table 3
Scheme 2. Synthetic strategy to synthesize 1 from 3.
Scope and limitations of the reaction with several aromatic and aliphatic aldehydes.
Entrya
R-CHO (4)
Yield (%) 1b
1
2
3
4
5
6
7
4-CH3O–C6H4–
4-CF3–C6H4–
C6H5–CH5CH–
2-Furyl–
4b
4c
4d
4e
4f
1b
1c
1d
1e
1f
27
33
32
61
65
44
n.d.
Table 1
Search for suitable fluoride anion.
2-F-C6H4–
.
2-Cl-C6H4–
4g
4h
1g
1h
n-C7H15
–
a
Entrya
F anion
Time (h)
Yield (%) 1ab
Reaction was carried out with 3 (1 mmol), 4 (2 mmol), F anion (1.5 mmol) and
THF (4 mL) at À20 8C for 2 h.
b
1
2
3
4
5
KF +18-croun-6
10
10
2
12
18
42
29
51
Isolated yield.
CsF
TASF
TBAT
2
2-furylaldehyde, was suitable for the reaction to afford 1e in 61%
yield. The reaction with 2-fluorobenzaldehyde also gave 1f in
moderate yield. However, 2-chlorobenzaldehyde was converted
into 1g in a lower yield. Aliphatic aldehyde was not a competent
substrate in the reaction because basicity of fluoride anion
promoted aldol reaction of such an aliphatic aldehyde leading to
decomposition of it. The disorderly trend of the reaction was not
clear yet.
Plausible reaction mechanism of the reaction is demonstrated
in Scheme 3. The reaction would be triggered off by the
dissociation of Si-Csp2 bond with fluoride attack on the silicon
atom. The carbanion 5 generated could be guided into divergent
ways. One possibility was that another fluoride anion abstracted
˚
TBAF + MS4A
2
a
Reaction was carried out with 3 (1 mmol), 4a (2 mmol) and F anion (1.5 mmol).
b
Isolated yield.
decreased a yield of the product rather than that with DMF. On the
other hand, diethyl ether was not effective due to considerable
retardation of the reaction. THF was found to be the solvent of
choice for the reaction and a 64% yield of 1a was obtained under
the condition at À20 8C (entry 7). Dropping a temperature off to
À40 8C, the reaction was not improved. The lower temperature at
À78 8C suppressed the reaction completely and 3 was recovered
(entry 9).
Subsequently, we assessed the scope and limitation of the
reaction with several aromatic and aliphatic aldehydes under
the optimized condition (Table 3). For an aromatic aldehyde, p-
anisaldehyde with electron-donating methoxy group was con-
verted into adduct (1b) in 27% yield. On the other hand, aromatic
aldehyde with electron-withdrawing trifluoromethyl group at
para-position was also resulted in a low yield of 1c (33%, entry 2).
Conjugated cinnamaldehyde was reacted in the same manner
to yield 1d in 32% yield (entry 3). Heteroaromatic aldehyde,
the
formation of diene intermediate [9]. Another was that the
carbanion of 5 abstracted the -proton of 5 intermolecularly or
a-proton of 5 to generate dianion 6 with slightly stabilized by
a
intramolecularly, in which 5 had to undergo tentative 1,2-proton
shift [10]. Both intermediates 6 and 7 could eventually give the
same tautomer 8 that would be stabilized by tautomerization
through carbonyl group. Subsequently, 8 could add to aldehyde to
afford the product 1.
Scheme 3. Plausible reaction mechanism of the reaction.

M. Omote et al. / Journal of Fluorine Chemistry 147 (2013) 1–4
3
3. Conclusion
4.2.3. Ethyl (E)-4,4-difluoro-5-hydroxy-5-
(4-(trifluoromethyl)phenyl)pent-2-enoate (1c)
In summary, we have synthesized new gem-difluorocarbanion
source (3) from easily available 2,2,2-trifluoroethanol. Treatment
A colorless oil. 1H NMR (400 MHz, CDCl3)
3H), 3.15 (bs, 1H), 4.21 (q, J = 7.6 Hz, 2H), 5.03 (t, J = 9.6 Hz, 1H),
6.20 (dt, J = 15.6, 2.0 Hz, 1H), 6.80 (ddd, J = 15.6, 12.4, 11.2 Hz, 1H),
7.54 (d, J = 8.4 Hz, 2H), 7.64 (d, J = 8.4 Hz, 2H); 13C NMR (100 MHz,
d
: 1.28 (t, J = 7.6 Hz,
˚
of 3 with TBAF in the presence of MS4A and subsequent addition of
aromatic aldehyde afforded the adduct 1. Structural interest of 1
was exemplified by extensive pioneers work. Our method offers
straightforward approach to the significant scaffold of 1. Although
the reaction proceeded in moderate yield, but rapid accessibility
to the scaffold of 1 is fascinating. To get the detailed reaction
character, mechanistic studies are now under investigation.
CDCl3)
d: 14.0, 61.4, 75.0 (t, J = 31.9 Hz), 118.8 (t, J = 244.9 Hz),
123.9 (q, J = 271 Hz), 125.2 (q, J = 3.7 Hz), 126.3, 127.2 (t,
J = 8.1 Hz), 127.9, 131.0 (q, J = 32.5 Hz), 135.7 (t, J = 25.6 Hz),
164.8; 19F NMR (90 MHz, CDCl3)
d
: 0.00 (s, 3F), À43.4 (dt, J = 255.1,
10.2 Hz, 1F), À47.3 (dt, J = 255.1, 10.2 Hz, 1F); IR (neat) max 3472,
2992, 1726 cmÀ1; MS m/z = 324 (M+); HRMS m/z M+ calcd. for
14H13F5O3: 324.078; found: 324.078.
n
4. Experimental
C
4.1. General
4.2.4. Ethyl (2E,6E)-4,4-difluoro-5-hydroxy-7-phenylhepta-2,
6-dienoate (1d)
1H NMR and 13C NMR spectra were recorded on JNM-GX400
and JEOL-ECA-600SN spectrometers. 19F NMR spectrum was
recorded on Hitachi FT-NMR R-90H spectrometer. Chemical shifts
of 1H NMR and 13C NMR are reported in ppm from tetramethylsi-
lane (TMS) as an internal standard. Chemical shifts of 19F NMR are
A colorless oil; 1H NMR (400 MHz, CDCl3)
d: 1.31 (t, J = 7.2 Hz,
3H), 2.34 (bs, 1H), 4.24 (q, J = 7.2 Hz, 2H), 4.56 (m, 1H), 6.18 (m, 1H),
6.37 (dt, J = 16.0, 2.8 Hz, 1H), 6.80 (m, 1H), 6.92 (ddd, J = 16.0, 12.4,
11.6 Hz, 1H), 7.27–7.43 (m, 5H); 13C NMR (100 MHz, CDCl3)
d: 14.1,
61.2, 74.3 (t, J = 29.3 Hz), 118.9 (t, J = 244.5 Hz), 122.2 (t, J = 2.1 Hz),
126.7, 127.0 (t, J = 8.3 Hz), 128.4, 128.6, 135.2, 135.6, 136.3 (t,
reported in ppm from
a,a,a-trifluorotoluene as an internal
standard. Mass spectra were obtained on JEOL JMS-700T spec-
trometer. Melting points were measured on Yanagimoto micro
melting point apparatus MP-S3.
J = 25.8 Hz), 164.7; 19F NMR (90 MHz, CDCl3)
d
: À43.6 (dt, J = 250.3,
10.7 Hz, 1F), À49.4 (dt, J = 250.3, 10.7 Hz, 1F); IR (neat) max 3472,
2988, 1724 cmÀ1; MS m/z = 282 (M+); HRMS m/z M+ calcd. for
15H16F2O3: 282.107; found: 282.107.
n
C
4.2. Typical procedure for the synthesis of 1
4.2.5. Ethyl (E)-4,4-difluoro-5-(furan-2-yl)-5-hydroxypent-
2-enoate (1e)
Under an argon atmosphere, TBAF (1.5 mL, 1.0 M in THF,
1.5 mmol) was added into a suspension of frame-dried MS4A
A colorless oil. 1H NMR (400 MHz, CDCl3)
d: 1.30 (t, J = 7.2 Hz,
˚
(1.0 g) in THF (4.0 mL) and the mixture was stirred at room
3H), 2.97 (bs, 1H), 4.23 (q, J = 7.2 Hz, 2H), 4.95 (t, J = 10.0 Hz, 1H),
6.32 (dt, J = 15.2, 2.4 Hz, 1H), 6.43 (m, 2H), 6.90 (dt, J = 15.2,
12.4 Hz, 1H), 7.44 (m, 1H); 13C NMR (100 MHz, CDCl3)
d: 14.1, 61.3,
temperature for 2.0 h. Then the mixture was chilled at À20 8C and
solution of 3 (265 mg, 1.0 mmol) and benzaldehyde (204
m
L,
2.0 mmol) was added to the mixture over 1.0 h at À20 8C. After
stirring the mixture for 1 h at the same temperature, the whole was
poured into aqueous 10% HCl, extracted with Et2O and organic
layer was dried with anhydrous MgSO4. After evaporation of the
solvent, the residue was purified by column chromatography (SiO2,
EtOAc:hexane) to give 1.
70.0 (t, J = 31.0 Hz), 109.9, 110.6, 118.1 (t, J = 254.2 Hz), 127.0 (t,
J = 8.3 Hz), 136.2 (t, J = 25.6 Hz), 143.2, 148.6, 164.8; 19F NMR
(90 MHz, CDCl3)
d: À43.1 (dt, J = 252.0, 11.3 Hz, 1F), À46.7 (dt,
J = 252.0, 11.3 Hz, 1F); IR (neat) nmax 3476, 2988, 1728 cmÀ1; MS
m/z = 246 (M+); HRMS m/z M+ calcd. for C11H12F2O4: 246.070;
found: 246.070.
4.2.1. Ethyl (E)-4,4-difluoro-5-hydroxy-5-phenylpent-2-enoate
(1a) [11]
4.2.6. Ethyl (E)-4,4-difluoro-5-(2-fluorophenyl)-5-hydroxypent-
2-enoate (1f)
A colorless oil. 1H NMR (400 MHz, CDCl3)
d
: 1.27 (t, J = 7.6 Hz,
A colorless oil. 1H NMR (400 MHz, CDCl3)
d: 1.29 (t, J = 7.2 Hz,
3H), 2.60 (d, J = 4.0 Hz, 1H), 4.21 (q, J = 7.6 Hz, 2H), 4.96 (td, J = 11.2,
4.0 Hz, 1H), 6.20 (dt, J = 15.6, 2.4 Hz, 1H), 6.81 (ddd, J = 15.6, 12.8,
3H), 2.92 (bs, 1H), 4.22 (q, J = 7.2 Hz, 2H), 5.32 (td, J = 10.0, 4.8 Hz,
1H), 6.23 (dt, J = 15.6, 2.4 Hz, 1H), 6.88 (dt, J = 15.6, 12.0 Hz, 1H),
7.03–7.55 (m, 4H); 13C NMR (100 MHz, CDCl3)
d: 14.1, 61.2, 69.1 (t,
11.2 Hz, 1H), 7.32–7.44 (m, 5H); 13C NMR (100 MHz, CDCl3)
d
: 14.0,
61.1, 75.5 (t, J = 28.9 Hz), 119.0 (dd, J = 244.2, 243.4 Hz), 126.6 (t,
J = 8.3 Hz), 127.5, 128.2, 128.8, 135.6 (d, J = 3.8 Hz), 136.5 (t,
J = 31.1 Hz), 115.3 (d, J = 20.1 Hz), 118.8 (t, J = 243.5 Hz), 122.8 (d,
J = 11.8 Hz), 124.2, 126.9 (t, J = 8.2 Hz), 128.9 (d, J = 3.3 Hz), 130.6
J = 25.5 Hz), 165.0; 19F NMR (90 MHz, CDCl3)
d
: À44.1 (dt, J = 252.0,
10.0 Hz, 1F), À47.0 (dt, J = 252.0, 10.0 Hz, 1F); IR (neat) max 3450,
3020, 1743 cmÀ1; MS m/z = 256 (M+); HRMS m/z M+ calcd. for
13H14F2O3: 256.091; found: 256.092.
(d, J = 8.9 Hz), 136.2 (t, J = 24.8 Hz), 160.2 (d, J = 247 Hz), 164.8; 19
F
n
NMR (90 MHz, CDCl3)
d
: À45.8 (ddt, J = 251.5, 9.3, 9.3 Hz, 1F),
À48.4 (ddt, J = 251.5, 9.3, 9.3 Hz, 1F), À54.5 (m, 1F); IR (neat) nmax
3468, 2992, 1718 cmÀ1; MS m/z = 274 (M+); HRMS m/z M+ calcd.
for C13H13F3O3: 274.082; found: 274.082.
C
4.2.2. Ethyl (E)-4,4-difluoro-5-hydroxy-5-(4-methoxyphenyl)pent-
2-enoate (1b)
4.2.7. Ethyl (E)-5-(2-chlorophenyl)-4,4-difluoro-5-hydroxypent-
2-enoate (1g)
A colorless oil. 1H NMR (400 MHz, CDCl3)
d: 1.29 (t, J = 7.2 Hz,
3H), 2.60 (bs, 1H), 3.82 (s, 3H), 4.21 (q, J = 7.2 Hz, 2H), 4.90 (t,
J = 10.0 Hz, 1H), 6.20 (dt, J = 16.0, 2.0 Hz, 1H), 6.81 (ddd, J = 16.0,
12.8, 11.6 Hz, 1H), 6.89 (d, J = 8.4 Hz, 2H), 7.32 (d, J = 8.4 Hz, 2H);
A colorless oil. 1H NMR (400 MHz, CDCl3)
d: 1.30 (t, J = 7.2 Hz,
3H), 2.76 (bs, 1H), 4.23 (q, J = 7.2 Hz, 2H), 5.54 (t, J = 10.0 Hz, 1H),
6.22 (dt, J = 15.6, 2.4 Hz, 1H), 6.88 (dt, J = 15.6, 11.2 Hz, 1H), 7.27–
7.68 (m, 4H); 13C NMR (100 MHz, CDCl3)
d: 14.1, 61.2, 71.4 (t,
13C NMR (100 MHz, CDCl3)
d: 14.1, 55.2, 61.1, 75.3 (t, J = 19.6 Hz),
113.8, 119.0 (t, J = 214.0 Hz), 126.8 (t, J = 8.5 Hz), 127.4 (d,
J = 3.6 Hz), 128.7, 136.3 (t, J = 26.4 Hz), 160.0, 164.8; 19F NMR
J = 29.1 Hz), 119.0 (t, J = 244.9 Hz), 126.9 (t, J = 8.1 Hz), 127.0,
129.2, 129.5, 130.1, 133.2, 133.6, 136.3 (t, J = 25.9 Hz), 164.7; 19F
(90 MHz, CDCl3)
J = 251.4, 9.6 Hz, 1F); IR (neat)
z = 286 (M+); HRMS m/z M+ calcd. for C14H16F2O4: 286.102; found:
d
: À44.3 (dt, J = 251.4, 9.6 Hz, 1F), À47.3 (dt,
NMR (90 MHz, CDCl3)
d
: À45.4 (dt, J = 249.3, 10.8 Hz, 1F), À48.1
n
max 3480, 2988, 1716 cmÀ1; MS m/
(dt, J = 249.3, 10.8 Hz, 1F); IR (neat) nmax 3496, 2988, 1716 cmÀ1
;
MS m/z = 290 (M+); HRMS m/z M+ calcd. for C13H13ClF2O3: 290.052;
286.102.
found: 290.059.

4
M. Omote et al. / Journal of Fluorine Chemistry 147 (2013) 1–4
(b) Y. Nakamura, M. Okada, A. Sato, H. Horikawa, M. Koura, A. Saito, T. Taguchi,
References
Tetrahedron 61 (2005) 5741–5753.
[5] (a) A. Otaka, J. Watanabe, A. Yukimasa, Y. Sasaki, W. Watanabe, T. Kinoshita, S.
Oishi, H. Tamamura, N. Fujii, J. Org. Chem. 69 (2004) 1634–1645;
(b) Y. Yamaki, A. Shigenaga, J. Li, Y. Shimohigashi, A. Otaka, J. Org. Chem. 74 (2009)
3278–3285.
[6] S. Higashiya, W.J. Chung, D.S. Lim, S.C. Ngo, W.H. Kelly IV, P.J. Toscano, J.T. Welch, J.
Org. Chem. 69 (2004) 6323–6328.
[7] S.C. Ngo, W.J. Chung, D.S. Lim, S. Higashiya, J.T. Welch, J. Fluorine Chem. 117
(2002) 207–211.
[8] M. Omote, T. Miake, A. Tarui, K. Sato, J.T. Welch, I. Kumadaki, A. Ando, J. Fluorine
Chem. 144 (2012) 114–117.
[1] (a) M. Kirihara, T. Takuwa, S. Takizawa, T. Momose, H. Nemoto, Tetrahedron 56
(2000) 8275–8280;
(b) K. Sato, A. Tarui, T. Kita, Y. Ishida, H. Tamura, M. Omote, A. Ando, I. Kumadaki,
Tetrahedron Lett. 45 (2004) 5735–5737;
(c) R.J. Kloetzing, T. Thaler, P. Knochel, Org. Lett. 8 (2006) 1125–1128;
(d) K.Sato, A.Tarui, M.Omote,A.Ando,I.Kumadaki,Synthesis11(2010)1865–1882.
[2] (a) X. Zhang, H. Xia, X. Dong, J. Jin, W.-D. Meng, F.-L. Qing, J. Org. Chem. 68 (2003)
9026–9033;
(b) X.-H. Xu, X.-L. Qiu, X. Zhang, F.-L. Qing, J. Org. Chem. 71 (2006) 2820–2824;
(c) Y.-Y. Wu, X. Zhang, W.-D. Meng, F.-L. Qing, Org. Lett. 6 (2004) 3941–3944.
[3] C.A. Audouard, I. Barsukov, J. Fawcett, G.A. Griffith, J.M. Percy, S. Pintat, C.A. Smith,
Chem. Commun. (2004) 1526–1527.
[4] (a) M. Okada, Y. Nakamura, A. Saito, A. Sato, H. Horikawa, T. Taguchi, Tetrahedron
Lett. 43 (2002) 5845–5847;
[9] T. Nihei, N. Iwai, T. Matsuda, T. Kitazume, J. Org. Chem. 70 (2005) 5912–5915.
[10] A.E. Mattson, A.M. Zuhl, T.E. Reynolds, K.A. Scheit, J. Am. Chem. Soc. 128 (2006)
4932–4933.
[11] T. Tsukamoto, T. Kitazume, Synlett 12 (1992) 977–979.