Non-defluorinative electrochemical silylation of ethyl trifluoroacetate: A practical synthesis of trifluoroacetyltrimethylsilane via its ethyltrimethylsilyl ketal

Article abstract of DOI:10.1016/S0040-4039(03)00745-7

An efficient method for the preparation of original trifluoroacetyltrimethylsilane, CF₃COSiMe₃ (3), in two steps from readily available ethyl trifluoroacetate is described. Electrochemical reduction of this ester using a sacrificial anode and performed on a semimolar scale afforded the unprecedented corresponding ketal, CF₃C(SiMe₃)(OSiMe₃)OEt (2) in 30-56% isolated yield. Treated with concentrated sulphuric acid at room temperature, the latter directly led to pure acylsilane 3 in 86% yield.

Full text of DOI:10.1016/S0040-4039(03)00745-7

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 44 (2003) 3741–3744
Non-defluorinative electrochemical silylation of ethyl
trifluoroacetate: a practical synthesis of
trifluoroacetyltrimethylsilane via its ethyltrimethylsilyl ketal
Michel Bordeau,a,* Philipe Clavel,a Alic Barba,b Muriel Berlande,a Claude Birana and
Nicolas Roquesc
aLaboratoire de Chimie Organique et Organome´tallique (UMR 5802 CNRS), Universite´ Bordeaux 1, 351, cours de la Libe´ration,
F-33405 Talence cedex, France
bInstitute of Chemistry, Academy of Sciences, Academiei Str. 3, MD2028 Chisinau, Republic of Moldova
cRhodia Organique Fine, 85, avenue des fre`res Perret, BP 62, 69192 Saint-Fons, France
Received 25 February 2003; revised 21 March 2003; accepted 24 March 2003
Abstract—An efficient method for the preparation of original trifluoroacetyltrimethylsilane, CF3COSiMe3 (3), in two steps from
readily available ethyl trifluoroacetate is described. Electrochemical reduction of this ester using a sacrificial anode and performed
on a semimolar scale afforded the unprecedented corresponding ketal, CF3C(SiMe3)(OSiMe3)OEt (2) in 30–56% isolated yield.
Treated with concentrated sulphuric acid at room temperature, the latter directly led to pure acylsilane 3 in 86% yield. © 2003
Elsevier Science Ltd. All rights reserved.
Trifluoromethyl and difluoromethylene substituted
molecules are of importance in organic synthesis
because of their unique biological activity.1 Conse-
quently, the development of new strategies for the
introduction of these groups into organic molecules has
become a good challenge.2 Among the methodologies
used to make trifluoromethylated compounds,3 tri-
fluoroacetaldehyde remains one of the most useful
building blocks. However, this compound being too
volatile, reactive and commercially unavailable it
almost never is used directly but is usually generated in
situ from its hydrate, hemiacetal, ketal, hemiaminal or
aminal-type precursors using protic or Lewis acid catal-
ysis.4 Here, we report the electrosynthesis, from the
readily available and cheap ethyl trifluoroacetate (1), of
trifluoroacetyltrimethylsilane ethyltrimethylsilyl ketal,
CF3C(SiMe3)(OSiMe3)OEt (2), and its transformation
into the original acylsilane CF3COSiMe3 (3) (Scheme
1). Both are new trifluoroacetaldehyde synthetic
equivalents.
This new two-step-strategy generalises and makes more
practical and less expensive the trifluoroacylsilane route
to difluoroenoxysilanes5 (Scheme 2a) which can be
advantageously (or complementary) compared to the
more classical method from acylsilanes and (tri-
fluoromethyl)trimethylsilane6 (Scheme 2b). Effectively,
the trifluoroacylsilane route which involves the addition
of an organometallic reagent to trifluoroacetyl-
triphenylsilane is up to now limited to the triphenyl-
silyl and phenyldimethylsilyl groups because it re-
quires a silyllithium reagent for its preparation.5,7
Scheme 1.
Keywords: fluorine and compounds; electrochemistry; acylsilanes; ketals; silylation.
* Corresponding author. Fax: +33-055-684-6286; e-mail: m.bordeau@lcoo.u-bordeaux1.fr
0040-4039/03/$ - see front matter © 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0040-4039(03)00745-7

3742
M. Bordeau et al. / Tetrahedron Letters 44 (2003) 3741–3744
Scheme 2.
A constant current of 100 mA (0.07 A dm−2) was
passed at room temperature under nitrogen gas until 1
was consumed (2.2 F mol−1). Silylketal 2, the new
silylated first reduction product of 1, was the major
product (70%) along with a-silylester 4 (15%) resulting
from monodefluorinative silylation of 1 and silyldiketal
5, as a 50:50 diastereoisomeric mixture (Table 1, entry
2).
Difluoroenoxysilanes, when opposed to an electrophile,
are now well known as excellent building blocks for the
synthesis of gem-difluorinated compounds.6 In addi-
tion, 3 could be a valuable starting material for the
preparation of a-(trifluoromethyl)vinylsilane.8
While searching for anionic difluoromethylene group
transfer agents, we have previously reported the molar
scale electroreductive synthesis of (trimethylsilyldi-
fluoromethyl)benzene from trifluoromethylbenzene9
and of ethyl a-(trimethylsilyl)difluoroacetate, a Refor-
matsky reagent equivalent, from ethyl chlorodifluoroac-
etate (less expensive than bromodifluoroacetate).10 This
second silylated synthon is a more stable alternative to
difluoroketene silylacetals for the transfer of the ethyl
difluoroacetate group to electrophiles.10,11 The elec-
troreduction of ethyl trifluoroacetate is known to take
place at relatively cathodic potentials: −2.36,12a
−2.65,12b −2.5812c V versus SCE.
In order to improve the selectivity, conditions were
varied (Table 1). In every case, conversion of 1 was
quantitative and 2 was the major product. With HMPA
as the co-solvent, varying the excess of chlorosilane at
25°C did not very much affect the GC yield of 2
(68–70%, entries 2 and 3) but a smaller amount of
TMSCl favoured silyl ester 4 to the detriment of 5
(entry 3). Temperature was one of the critical factors
for the yield of 2. If increasing it to 50°C did not result
in a noticeable change (entry 1), lowering it to −25°C is
favourable to 2 (78%, entry 4). The best GC yield (89%)
was obtained combining the lowest temperature with
progressive addition of 1 by syringe pump at the elec-
trolysis rate on a 0.25 mL initial amount (0.03 mol L−1).
Under these conditions, the formation of 5 was sup-
pressed and the isolated yield of 2 was 56% (entry 5).
As shown in Table 1, other co-solvents could be used in
combination with THF: tris(2,5-dioxaheptyl)amine
(THF/TDA-1: 6/1 vol.) gave a similar result at 25°C
under batch conditions (65% of 2, entry 6); N,N-
The electrochemical transformation of 1 to 2 was car-
ried out using the versatile intensiostatic sacrificial
anode technique with an undivided cell fitted with an
aluminium anode and a cylindrical stainless steel
cathode.13 Electroreduction of 1 (21 mmol, 3 g) was
first performed in THF (36 mL) and HMPA (6 mL) as
the co-solvent containing n-Bu4NBr (0.8 mmol) and
excess chlorotrimethylsilane TMSCl (40 mL, 15 equiv.).
Table 1. Influence of co-solvent, amount of TMSCl, temperature and progressive addition of 1
Entrya
THF (mL)/co-solvent (mL)
TMSCl (equiv.)
Temp. (°C)
2 (%)
4 (%)
5 (%)
1
2
3
(36)/HMPA (6)
(36)/HMPA (6)
(36)/HMPA (6)
(36)/HMPA (6)
(48)/HMPA (8)
(36)/TDA-1 (6)
(30)/DMPU (15)e
(30)/DMPU (15)e
(0)/DMPU (45)g
(42)/DMPU (7)e
15
15
3
15
15
15
15
15
15
15
50
25
25
−25
−25
25
69b
15b
15b
27b
3b
16b
15b
3b
70b
68b
4
78b
19b
0b
5c
6
89b (56)d
65b
11b
16b
13b
6f
19b
0b
7c
8c
9c
10c
10
74b
−20
−10
−25
81f (20)d
76f (20)d
75f (33)d
0f
12f
8f
4f
1f
a Entries 1–4, 6 were performed on 21 mmol of 1 and 5, 7–10 on 42 mmol (6 g) of 1.
b GC yields using nonane as the internal standard.
c Syringe pump addition of 1 at the reaction rate corresponding to 0.1 A constant current.
d Isolated.
e 3 (5%, entries 7 and 8; 8%, entry 10) and HCF2COOEt (8%) were formed as by-products.
f 19F NMR yields.
g 3 (8%) was a by-product.

M. Bordeau et al. / Tetrahedron Letters 44 (2003) 3741–3744
3743
Acknowledgements
dimethylpropylene urea (THF/DMPU: 2/1 or 6/1 vol.)
also gave a good selectivity for 2 (74–81% at +10, −10
and −25°C: entries 7, 8, 10) and lowering or suppres-
sion of 5 under progressive addition of 1. With DMPU
as the sole solvent, 76% of 2 were obtained at −10°C
(entry 9). However, with DMPU as the co-solvent or
solvent, 2 was more difficult to isolate likely due to its
partial transformation into 3 (20–33% isolated yields).14
Ministe`re de l’Education Nationale, Association
Nationale de la Recherche Technique (ANRT), Elec-
tricite´ de France, Re´gion Aquitaine are acknowledged
for their financial support and Rhodia Organique for
its scientific and financial support.
The reaction here observed in THF/co-solvent medium
with an aluminium anode is totally different from these
obtained in acetonitrile solution by Stepanov12c and
Uneyama,15 respectively. As a matter of fact, Stepanov,
also using an aluminium anode, but a much higher
concentration of 1 (75 mmol in 60 mL of ACN) and
only 1.1 equiv. of TMSCl, essentially obtained the
Claisen-type product 6 (Scheme 3). Uneyama, with a
two-compartment cell (carbon anode, lead cathode),
obtained at 0°C a small amount of 4 (<5%) together
with 21% of the actual Claisen product 7, but at higher
temperature (50°C), 4 became the sole formed product
(47% yield). These authors did not report any side
reaction with acetonitrile. For our part we observed
side reactions in this medium at 50°C due to its relative
acidity.
References
1. (a) Hudlicky, M. Chemistry of Organic Fluorine Com-
pounds; 2nd ed.; Ellis Horwood: New York, 1992; (b)
Banks, R. E. Organofluorine Chemicals and their Indus-
trial Applications; Ellis Horwood: Chichester, 1979; (c)
Banks, R. E. Preparation, Properties and Industrial Appli-
cations of Organofluorine Compounds; Ellis Horwood:
Chichester, 1982; (d) Olah, G. A.; Chambers, R. D.;
Prakash, G. K. Synthetic Fluorine Chemistry; Wiley-
Interscience: New York, 1992; (e) Organofluorine Com-
pounds in Medicinal Chemistry and Biomedical
Applications; Filler, R.; Kobayashi, Y.; Yagupolskii, L.
M., Eds.; Elsevier: Amsterdam, 1993.
2. For reviews, see: (a) McClinton, M. A.; McClinton, D.
A. Tetrahedron 1992, 48, 6555–6666; (b) Singh, R. P.;
Shreeve, J. M. Tetrahedron 2000, 56, 7613–7632.
3. Percy, J. M. Top. Curr. Chem. 1997, 193, 131–195.
4. (a) Kubota, T.; Iijima, M.; Tanaka, T. Tetrahedron Lett.
1992, 33, 1351–1354; (b) Xu, Y.; Dolbier, W. R., Jr.;
Rong, X. X. J. Org. Chem. 1997, 62, 1576–1577; (c) Xu,
Y.; Dolbier, W. R., Jr. Tetrahedron Lett. 1998, 39, 9151–
9154; (d) Xu, Y.; Dolbier, W. R., Jr. J. Org. Chem. 2000,
65, 2134–2137; (e) Mispelaere, C.; Roques, N. Tetra-
hedron Lett. 1999, 40, 6411–6414; (f) Billard, T.; Bruns,
S.; Langlois, B. R. Org. Lett. 2000, 2, 2101–2103; (g)
Billard, T.; Langlois, B. R.; Blond, G. Tetrahedron Lett.
2000, 41, 8777–8780; (h) Billard, T.; Langlois, B. R.;
Blond, G. Eur. J. Org. Chem. 2001, 1467–1471.
As we can see, the mechanism of electroreduction of 1
is much more complex than the one of ethylchlorodi-
fluoroacetate which only gives compound 4.10 This
difference of behaviour could be searched in the radi-
cal-anion which is formed first. Effectively, as fluorine
is a worse leaving group than chlorine, the radical-
anion arising from trifluoroacetate is likely more stable
than the one formed from chlorodifluoroacetate. A
more complete study of this mechanism including the
role of all experimental parameters will be reported in a
forthcoming paper.
The best conditions without HMPA were applied to the
synthesis of 2 at a larger scale in a THF/DMPU/
TMSCl (400 mL/200 mL/400 mL) medium using a
tubular flow cell16 and progressive introduction of 75
mL of 1 by syringe pump under a current of 1 A. The
reaction, carried out at 10°C and monitored by GC and
19F NMR, led to 55 g of pure silyl ketal 2 (bp20=70°C,
74% selectivity, 30% isolated yield).14
5. Jin, F.; Jiang, B.; Xu, Y. Tetrahedron Lett. 1992, 33,
1221–1224.
6. (a) Portella, C.; Brigaud, T.; Lefebvre, O.; Plantier-
Royon, R. J. Fluorine Chem. 2000, 101, 193–198; (b)
Lefebvre, O.; Brigaud, T.; Portella, C. J. Org. Chem.
2001, 66, 1941–1946; (c) Lefebvre, O.; Brigaud, T.;
Portella, C. J. Org. Chem. 2001, 66, 4348–4351; (d)
Berber, H.; Brigaud, T.; Lefebvre, O.; Plantier-Royon,
R.; Portella, C. Chem. Eur. J. 2001, 7, 903–909; (e)
Chorki, F.; Crousse, B.; Bonnet-Delpon, D.; Be´gue´, J.-P.;
Brigaud, T.; Portella, C. Tetrahedron Lett. 2001, 42,
1487–1489.
Pure trifluoroacetyltrimethylsilane 3 was then readily
obtained in 86% yield by mild protolysis of ketal 2
(Scheme 1).17
7. Bonini, B. F.; Comes-Franchini, M.; Fochi, M.; Maz-
zanti, G.; Nanni, C.; Ricci, A. Tetrahedron Lett. 1998, 39,
6737–6740.
8. Under investigation in our group. This vinylsilane has
already been obtained as a minor product by photochem-
ical addition of trimethylsilane to 3,3,3-trifluoropropyne.
See: Haszeldine, R. N.; Pool, C. R.; Tipping, A. E. J.
Chem. Soc., Perkin Trans. 1 1974, 20, 2293–2296.
9. Clavel, P.; Le´ger-Lambert, M.-P; Biran, C.; Serein-Spi-
rau, F.; Bordeau, M.; Roques, N.; Marzouk, H. Synthesis
1999, 829–834.
Scheme 3.

3744
M. Bordeau et al. / Tetrahedron Letters 44 (2003) 3741–3744
2
10. Clavel, P.; Biran, C.; Bordeau, M.; Roques, N.; Trevin, S.
Tetrahedron Lett. 2000, 41, 8763–8767.
15.5 (s, CH3), 60.4 (s, CH2), 98.6 (q, JCF=34.1 Hz, Cq);
124.8 (q, JCF=289.3 Hz, CF3); 29Si NMR (CDCl3, 39.77
MHz) l 5.0 (q, JSiF=3.1 Hz, CSiMe3), 14.8 (s, OSiMe3);
1
3
11. (a) Amii, H.; Kobayashi, T.; Uneyama, K. Synthesis
2000, 2001–2003; (b) Uneyama, K.; Mizutani, G.; Maeda,
K.; Kato, T. J. Org. Chem. 1999, 64, 6717–6723; (c)
Uneyama, K.; Amii, H. J. Fluorine Chem. 2002, 114,
127–131.
12. (a) Inesi, R.; Rampazzo, L. J. Electroanal. Chem. 1974,
49, 85; (b) Mayranovsky, V. G.; Ponomarev, A. M. Zh.
Obsch. Khim. 1977, 47, 2343–2346; (c) Stepanov, A. A.;
Minyaeva, T. V.; Martinov, B. I. Tetrahedron Lett. 1999,
40, 2203–2204.
IR (neat) 2980.3, 2962.0, 1287.4, 1254.8, 1174.1, 1136.0,
1067.1 cm−1; MS m/z 273 (M−15, 1), 259 (M−29, 15), 215
(M−73, 2), 165 (2), 151 (5), 117 (3), 103 (7), 77 (15), 73
(Me3Si+, 100), 55 (8), 45 (15), 43 (8). Anal. calcd for
C10H23F3O2Si2: C, 41.89; H, 8.14; F, 19.30. Found: C,
41.64; H, 8.04; F, 19.76.
15. Uneyama, K.; Mizutani, G. J. Chem. Soc., Chem. Com-
mun. 1999, 613–614.
16. For this technique, see: Chaussard, J.; Troupel, M.;
Robin, Y.; Jacob, G.; Juhasz, J. P. J. Appl. Electrochem.
1989, 19, 345.
13. Deffieux, D.; Bordeau, M.; Biran, C.; Dunogues, J.
Organometallics 1994, 13, 2415–2422.
17. Typical procedure for 3: Pure 2 (3 g, 10.5 mmol) was
added at room temperature to 1.5 mL (28 mmol, 2.8
equiv.) of concentrated sulphuric acid. After stirring at
room temperature for 3 h, then heating at 40°C under
15–20 mbar reduced pressure, 1.54 g of pure product 3,
as a light-yellow liquid, were condensed into a cold trap
14. Isolation of 2 from THF/DMPU medium (tubular flow
cell): After electrolysis, pentane was added (500 mL) and
the salts were decanted. Pentane, THF and excess TMSCl
were removed in vacuo (100 Torr). Pentane (400 mL) was
added and the mixture was poured into 200 mL of
ice-cold brine. The aqueous DMPU layer was extracted
with pentane (3×100 mL) and the organic layer washed to
neutrality by NaHCO3 solution and water, then dried
(MgSO4). Distillation gave 55 g (30%) of 2.
1
(86% yield). H NMR (CDCl3, 250 MHz) l 0.3 (s, 9H);
19F NMR (CDCl3, 188 MHz, PhCF3 as an internal
standard −63.0 relative to CFCl3) l −80.3 (s, 3F); 13C
NMR (CDCl3, 62.86 MHz) l −3.3 (s, SiMe3), 115.8 (q,
Silyl ketal 2: colourless liquid; bp 70°C/20 mmHg; 1H
NMR (CDCl3, 250 MHz) l 0.18 (s, 9H, CSiMe3), 0.19 (s,
9H, OSiMe3), 1.21 (t, J=7.0 Hz, 3H), 3.68 (m, 2H); 19F
NMR (CDCl3, 188 MHz, PhCF3 as an internal standard
−63.0 relative to CFCl3) l −75.5 (s, 3F); 13C NMR
(CDCl3, 62.86 MHz) l −2.5 (s, CSiMe3), 1.6 (s, OSiMe3),
2
1JCF=296.5 Hz, CF3), 223.7 (q, JCF=35 Hz, CꢀO); 29Si
NMR (CDCl3, 39.77 MHz) l −2.0 (s, SiMe3); IR (neat)
1690 cm−1 (wCO); MS m/z 170 (M+, 0.2), 155 (M−15,
0.04), 101 (M−CF3, 62), 77 (Me2SiF+, 66), 73 (Me3Si+,
100), 69 (CF+3, 35), 58 (27), 45 (71), 43 (59).