A direct and useful route to difluoroacylsilanes and difluoroacylstannanes and their potential for the generation of structurally diverse difluoroketones

Article abstract of DOI:10.1016/S0040-4039(01)01220-5

Allyl trifluoroethyl ethers 2a-2d were prepared; each underwent efficient dehydrofluorination/metallation and trapping to afford a range of difluoroenol silanes and stannanes which rearranged easily and efficiently to the acyl metals. Fluoride-mediated be

Full text of DOI:10.1016/S0040-4039(01)01220-5

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 42 (2001) 6377–6380
A direct and useful route to difluoroacylsilanes and
difluoroacylstannanes and their potential for the generation of
structurally diverse difluoroketones
Maxime R. Garayt and Jonathan M. Percy*
School of Chemistry, University of Birmingham, Edgbaston, Birmingham B15 2TT, UK
Received 6 June 2001; accepted 5 July 2001
Abstract—Allyl trifluoroethyl ethers 2a–2d were prepared; each underwent efficient dehydrofluorination/metallation and trapping
to afford a range of difluoroenol silanes and stannanes which rearranged easily and efficiently to the acyl metals. Fluoride-medi-
ated benzylation and allylation of an acylsilane, and Stille coupling of the vinyl- and acyl-stannane were achieved in moderate to
good yield. © 2001 Elsevier Science Ltd. All rights reserved.
In 1985, Metcalf, Jarvi and Burkhart published a semi-
nal paper which showed how the [3,3]-Claisen rear-
rangement could be used to generate a range of
a,a-difluoroketone structures from allyl trifluoroethyl
ethers (Scheme 1).1
had progressed a simple allyl trifluoroethyl ether to a
metallated difluoroenol ether which they were able to
quench with water or a silicon electrophile to afford an
aldehyde, or more efficiently, an acyl silane after rear-
rangement. We were unable to find many examples of
this chemistry or of reactions of the acylsilane, apart
from a single example reported by McCarthy,3a while
Normant, Sauveˆtre and Dubuffet3b had isolated a simi-
lar acylsilane species from the solvolysis of a difluori-
nated epoxysilane. We wondered if useful coupling or
transfer reactions could be available from the
difluoroacyl metals or from the difluorovinyl metal
precursors (as none had been reported); we were also
aware that acylstannanes possessed useful properties
for CꢀC bond forming reactions,4 and that the a,a-
difluoro species would be novel. Furthermore, the vinyl
stannane intermediates would offer the possibility of
coupling reactions5 allowing access to a wide range of
aryl difluoroketones. We therefore synthesised a range
of allyl trifluoroethyl ethers 2a–2d designed to test the
scope of the dehydrofluorination/metallation chemistry
(Scheme 2).
The idea was revisited recently by Taguchi and co-
workers2 as they sought to develop an enantioselective
Claisen rearrangement. Both groups used a dehy-
drofluorination reaction to create the vinyl component
of the Claisen system. We noted that the former group
SiMe3
F
F
Cl
F3C
O
O
b,c
a
Scheme 1. (a) F3CCH2ONa, THF, reflux, 63%; (b) 3.3 equiv.
LDA/TMSCl, THF, −100°C, 88%; (c) CCl4, reflux, unknown
yield.
OH
OR
Cl
BnO
Br
Br
OH
a or b
F3C
F3C
2a-2d
1a
1b
1c
1d
Scheme 2. (a) 50% NaOH, Bu4NHSO4, Bu4NI, 1a,1b or 1d; (b) 1,1%-(azodicarbonyl)dipiperidine, Bu3P, PhH, 6 h, rt, 1c.
* Corresponding author.
0040-4039/01/$ - see front matter © 2001 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(01)01220-5

6378
M. R. Garayt, J. M. Percy / Tetrahedron Letters 42 (2001) 6377–6380
Table 2. Synthesis of vinyl metals and acyl metals
Different substitution patterns in the allyl system were
of interest and we wished to use the chemistry to attach
chains to rings. We were able to use phase transfer-
Ether
2a
Vinyl metal
(%)a
Acyl metal
(%)d
Time (h)
catalysed
allylation
of
trifluoroethanol
very
3a
4a
3b
4b
3c
4c
3d
4d
90b
85
85
77
78
70
90
80
5a
6a
5b
6b
5c
6c
5d
6d
100
100c
100
96
100
88e
95
4
3.5
10
10
10
successfully6 and also the Mitsunobu procedure
described by Falck7 in the case of 2c for which the
halide was less readily available than the alcohol. The
former method was also used to prepare 2a on a 0.1
mol scale in quantitative yield (Table 1).
2b
2c
8
2d
120
108
90
The dehydrofluorination/metallation (Scheme 3) could
be achieved8 at −78°C with n-BuLi (2 equivalents) for
the preparation of the vinylsilanes 3a–3d or -stannanes
4a–4d, in contrast to the Metcalf report in which it was
necessary to use a larger excess of strong base (3.3
equivalents) and trapping conditions (LDA/TMSCl) at
lower reaction temperature (−100°C) to prepare the
silane shown in Scheme 1 in good yield. This observa-
tion indicates that the thermal stability of the simple
metallated difluoroenol ether is sufficient at −78°C for
rapid trapping to be achieved even in the absence of
inductively electron-withdrawing9 or co-ordinating sub-
stituents,10 or chelation within a part-polyether struc-
ture,11 features we believed necessary to reduce the rate
of the antiperiplanar elimination of LiF.
a At 5 mmol scale.
b 66% at 100 mmol scale.
c Yield for rearrangement in toluene.
d Yield for neat rearrangement.
e Yield for purified stannane from unpurified vinylmetal.
with several other (minor) fluorinated products, sug-
gesting that a one-pot preparation of the acylsilanes
may be available. However, we have not yet optimised
these procedures and recommend at least isolation of
the vinylmetal intermediates.
Acylsilane 5a was transformed into ketone 7 (54%)
when treated with TBAF in the presence of benzyl
bromide.13 The reaction with allyl bromide afforded a
mixture of ketones 8a and 8b (46%, 1:1.2), indicating
that double bond migration occurs under the reaction
conditions; further applications of this chemistry are
under investigation (Scheme 4).
19F NMR spectra of the crude material in each case
revealed relatively clean product mixtures; flash column
chromatography on silica gel or alumina delivered pure
silane or stannane, respectively (to 98% purity by GC).8
Rearrangements of 3a–3c and 4a–4c occurred smoothly
(Table 2) neat or in toluene at 110°C in Ace® tubes,12
whereas 3d and 4d rearranged much more slowly, con-
sistent with steric hindrance in the transition states as
the bulky trialkylmetal and cyclic array are forced
together.
The vinyl stannane seems an obvious substrate for Stille
coupling, given our findings that related difluoroenol
carbamates can be used successfully.5b The issue of
interest here is the possibility of competitive and
destructive Pd(II)-catalysed reactions of the allyl ether
moiety. Attempting to minimise the possibility of com-
peting rearrangement, we worked at room temperature
and though the reaction was slow (72 h for consump-
tion of starting material) we were able to isolate
ketones 9 and 11 cleanly and in good yield (60 and
Purification at the vinylmetal stage is not necessary;
isolation may even be avoided. For example, crude 4c
rearranged to 6c successfully (88% based on starting
ether). In one case, the THF reaction solution of 3b was
refluxed for 20 h to afford 5b in ca. 50% yield along
72%,
respectively)
following
exposure
(5%
Pd2dba3·CHCl3, 20% CuI, 10% tris(2-furyl)phosphine,
DMF, rt then KF work-up)14 of 4a to iodobenzene or
10, respectively (Scheme 5).
Table 1. Synthesis of ethers
Allyl fragment
Trifluoroethyl ether
(%)
Method
Product 9 was also obtained in 52% yield from
iodobenzene and 6a under the same conditions. Expo-
sure of 4a to the Stille catalyst/co-catalyst but in the
absence of an electrophile consumed the stannane but
formed a complex product mixture which contained
only a small amount of 6a, suggesting that there is no
1a
1b
1c
1d
2a
2b
2c
2d
100a
85
100
100
a
a
b
a
a At 5 and 100 mmol scales.
OR
OR
O
OBn
a or b
c
F
R'
a, R' =
c, R' =
F3C
X
X
F
F
F
2a-2d
3a-3d, X = SiMe3;
4a-4d, X = SnBu3.
5a-5d, X = SiMe3;
6a-6d, X = SnBu3.
b, R' =
d, R' =
Scheme 3. (a) i, 2.0 n-BuLi, THF, −78°C; ii, 1.0 Me3SiCl; (b) i, 2.0 n-BuLi, THF, −78°C; ii, 1.0 Bu3SnCl; (c) D (110°C, Ace®
tube).

M. R. Garayt, J. M. Percy / Tetrahedron Letters 42 (2001) 6377–6380
6379
a
O
b
O
O
+
5a
Ph
F
F
F
F
F
F
7
8a
8b
Scheme 4. TBAF, THF, rt, 12 h and (a) PhCH2Br or (b) H2CꢁCHCH2Br.
I
O
O
a
a
O
SnBu3
F
OTf
F
F
F F
SnBu3
X
F
4a
9, X = H;
11, X = OTf.
6a
10
Scheme 5. (a) PhI or 10 (from 4a only), 5% Pd2dba3·CHCl3, 20% CuI, 10% P(2-furyl)3, DMF, rt, 72 h.
X
NMeOMe
O
F
R1
F
R1MgBr
F
F
O
F3C
F
O
R2
R2
(Uneyama)
X = F (Gelb) or OEt (Lang)
b
R1
+
X
a
R1
OSiMe3
R2
b
F
F
F
(Portella)
O
12
a
R1
CF3SiMe3
SiMe3
R2
R1 = alkyl or aryl
O
Scheme 6.
Acknowledgements
simple catalysis of the rearrangement reaction system
under the coupling reaction conditions. The timing of
the two events (coupling versus rearrangement) is
unknown and we are trying to ascertain their sequence.
We thank the EPSRC and the University of Birming-
ham for a studentship (M.R.G.) and Ms C. Powell for
repeating and scaling the syntheses of 5a–6d.
These short sequences provide a very general route to
fluoroketones of general type 12 from trifluoroethanol.
An alternative synthesis could start from an aryl
acylsilane and prepare the difluoroenol silyl ether, sub-
jecting the intermediate to allylation, as described by
Portella15 (Scheme 6).
References
1. Metcalf, B. W.; Jarvi, E. T.; Burkhart, J. P. Tetrahedron
Lett. 1985, 26, 2861–2684. For a review of rearrange-
ment-based methods, see: Percy, J. M.; Prime, M. E. J.
Fluorine Chem. 1999, 100, 147–156.
A similar approach can be made using the reductive
defluorination chemistry of Uneyama.16 Various meth-
ods have been described for the synthesis of 2,2-
difluoropent-4-enoic acid derivatives which could then
be elaborated through Weinreb amides and reaction
with the Grignard reagents; Gelb’s route17 uses tetra-
fluoroethylene which is not particularly convenient,
while the Reformatsky–Claisen reaction described by
Lang18 has proved both very difficult to repeat and
non-general in our hands. We contend that the method
described herein is at least competitive with these more
established approaches and represents a direct and reli-
able route to 2,2-difluoropent-4-enones with variable
substitution at C-1, C-3, C-4 and C-5.
2. Ito, H.; Sato, A.; Kobayashi, T.; Taguchi, T. Chem.
Commun. 1998, 2441–2442.
3. (a) McCarthy, P. A.; Wint, L. T.; Diaz, C. L. Bioorg.
Med. Chem. Lett. 1992, 2, 119–122; (b) Dubuffet, T.;
Sauveˆtre, R.; Normant, J.-F. Bull. Soc. Chim. Fr. 1989,
677–682.
4. (a) Shirakawa, E.; Nakao, Y.; Hiyama, T. Chem. Com-
mun. 2001, 263–264; (b) Shirakawa, E.; Nakao, Y.;
Yoshida, H.; Hiyama, T. J. Am. Chem. Soc. 2000, 122,
9030–9031.
5. (a) For recent couplings of a-metallated enol ethers, see:
Yu, W. S.; Jin, Z. D. J. Am. Chem. Soc. 2000, 122,
9840–9841; (b) for couplings to difluoroenol carbamates,

6380
M. R. Garayt, J. M. Percy / Tetrahedron Letters 42 (2001) 6377–6380
10. (a) Lee, J.; Tsukazaki, M.; Snieckus, V. Tetrahedron Lett.
see: DeBoos, G. A.; Fullbrook, J. J.; Owton, W. M.;
1993, 34, 415; (b) Tsukazaki, M.; Snieckus, V. Tetra-
hedron Lett. 1993, 34, 411–414; (c) Howarth, J. A.;
Owton, W. M.; Percy, J. M.; Rock, M. H. Tetrahedron
1995, 51, 10289–10302.
Percy, J. M.; Thomas, A. C. Synlett 2000, 963–966.
6. Schlosser, M.; Strunk, S. Tetrahedron 1989, 45, 2649–
2664.
7. Falck, J. R.; Yu, J.; Cho, H.-S. Tetrahedron Lett. 1994,
35, 5997–6000.
11. Patel, S. T.; Percy, J. M.; Wilkes, R. D. Tetrahedron
1995, 51, 9201–9216.
8. Dehydrofluorination/metallation/trapping of 2c to 3c: n-
Butyllithium (5 mL of a 2 M solution in hexane, 2 equiv.)
was added dropwise to a solution of trifluoroethyl allyl
ether 2c (1.3 g, 5 mmol) in tetrahydrofuran (15 mL) at
−78°C under nitrogen. The mixture was stirred a further
20 minutes at −78°C then chlorotrimethylsilane (0.63 mL)
was added in one portion. The mixture was allowed to
warm to −40°C over 1 hour, then quenched at −40°C
with saturated aqueous ammonium chloride solution (20
mL) and extracted with diethyl ether (3×40 mL). The
combined organic extracts were washed with brine (20
mL), dried (MgSO4) and concentrated in vacuo to an
orange oil which was purified by flash column chro-
matography on silica gel to afford silane 3c (1.21 g, 78%)
12. Rearrangement of 3c and data for 5c: Neat silane 3c was
heated in an Ace® tube at 110°C to afford 5c which was
purified by Kugelrohr distillation bp 125°C/0.1 mmHg
(96% by GC); 
(KBr/cm−1): 3033, 2960, 2870, 1673,
max
847; lH (CDCl3, 300 MHz) 7.35–7.27 (5H, m), 5.82–5.69
3
3
(1H, m), 5.31 (1H, d, JH–H 7.35 Hz), 5.28 (1H, d, JH–H
16.9 Hz), 4.42 (2H, s), 3.70–3.45 (2H, m), 3.36–3.21 (1H,
m), 0.21 (9H, s); lC (CDCl3, 75 MHz) 137.6, 130.0 (t,
1
3JC–F 4.2 Hz), 128.4, 127.8, 127.7, 121.3, 118.7 (dd, JC–F
2
252.4, 252.9 Hz), 72.9, 68.6 (dd, JC–F 3.96, 3.95 Hz),
65.9, 15.3; the CꢁO carbon was too weak to assign
2
properly; lF (CDCl3, 282 MHz) −106.3 (1F, dd, JF–F
3
2
3
275.0, JF–H 9.9 Hz), −115.6 (1F, dd, JF–F 275.0, JF–H
20.0 Hz); m/z (ES) 335 (100%, M+Na+). HRMS calcd for
C16H22F2NaO2Si: 335.1255. Found: 335.1245.
13. Schinzer, D.; Heathcock, C. H. Tetrahedron Lett. 1981,
22, 1881–1884.
as a colourless oil (98% by GC); 
(KBr/cm−1): 3032,
max
2959, 2860, 1688 (CꢁCF2), 843; lH (CDCl3, 300 MHz)
7.38–7.30 (5H, m), 5.83–5.78 (2H, m), 4.53 (2H, s), 4.24
3
3
(2H, d, JH–H 4.8 Hz), 4.11 (2H, d, JH–H 4.4 Hz), 0.20
14. Farina, V.; Baker, S. R.; Benigni, D. A.; Sapino, C.
1
(9H, s); lC (CDCl3, 75 MHz) 160.9 (dd, JC–F 283.16,
Tetrahedron Lett. 1988, 29, 5739–5742.
15. Lefebvre, O.; Brigaud, T.; Portella, C. Tetrahedron 1999,
55, 7233–7242.
16. Amii, H.; Kobayashi, T.; Hatamoto, Y.; Uneyama, K.
Chem. Commun. 1999, 1323–1324.
17. Yuan, W.; Berman, R. J.; Gelb, M. H. J. Am. Chem. Soc.
283.15 Hz), 138.1, 129.8, 128.7, 128.5, 127.8, 127.7, 117.4
3
4
(dd, JC–F, 7.1, 7.1 Hz), 72.4, 68.6 (dd, JC–F 2.5, 2.5 Hz),
2
65.7, 1.1; lF (CDCl3, 282 MHz) −81.3 (d, JF–F 57.9),
−104.8 (d, JF–F 57.9); m/z (ES) 335 (100%, M+Na+).
2
HRMS calcd for C16H22F2NaO2Si: 335.1255. Found:
335.1250.
1987, 109, 8071–8081.
9. Nakai, T.; Tanaka, K.; Ishihawa, N. Chem. Lett. 1976,
1263–1266.
18. Greuter, H.; Lang, R. W.; Romann, A. J. Tetrahedron
Lett. 1988, 29, 3291–3294.
.