HFC-134a as a fluorinated building block: Short syntheses of α- fluoroenones

Article abstract of DOI:10.1016/S0040-4039(99)02184-X

1,1,1,2-Tetrafluoroethane (HFC-134a) has been transformed into a range of α-fluoroenones via a direct route. Dehydrofluorination/metallation allowed trifluorovinylsilanes to be prepared; these could be converted in situ, or isolated and treated further, with alkyllithium reagents to give difluorovinylsilanes. Fluoride-catalysed addition to aldehydes followed by acidic work-up allowed the isolation of a range of α-fluoroenones in two- or three-pots. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(99)02184-X

TETRAHEDRON
LETTERS
Pergamon
Tetrahedron Letters 41 (2000) 971–974
HFC-134a as a fluorinated building block: short syntheses of
α-fluoroenones
J. Marie Bainbridge,a Stuart Corr,b Masatomi Kanai a and Jonathan M. Percy a,∗
aSchool of Chemistry, University of Birmingham, Edgbaston, Birmingham, B15 2TT, UK
bICI Klea R&T, PO Box 8, The Heath, Runcorn, Cheshire, WA7 4QD, UK
Received 20 October 1999; accepted 24 November 1999
Abstract
1,1,1,2-Tetrafluoroethane (HFC-134a) has been transformed into a range of α-fluoroenones via a direct route.
Dehydrofluorination/metallation allowed trifluorovinylsilanes to be prepared; these could be converted in situ, or
isolated and treated further, with alkyllithium reagents to give difluorovinylsilanes. Fluoride-catalysed addition to
aldehydes followed by acidic work-up allowed the isolation of a range of α-fluoroenones in two- or three-pots.
© 2000 Elsevier Science Ltd. All rights reserved.
The development of the chemistry of new CFC replacements is of current interest as the implications
of the Montreal Protocol erode the range of available versatile fluorinated starting materials or building
blocks. We became interested in the possibility of using HFC-134a in a three component assembly
synthesis of α-fluoroenones using readily available aldehydes and organolithium reagents. The title
compounds have considerable potential for manipulation through Michael and related reactions and we
had failed to reproduce the sequence reported by Normant and co-workers (Scheme 1) that appeared to
offer a general route.1
Scheme 1.
The route was subverted by an extremely sluggish displacement (in our hands) of fluoride upon
treatment of the trifluorovinyl alcohol intermediate with n-butyllithium, a reaction that we were unable
to drive to completion even after refluxing the solution. We therefore decided to utilise a trialkylsilyl
group to facilitate the addition/elimination reaction, and then to use a fluoride-catalysed or -mediated
∗
Corresponding author.
0040-4039/00/$ - see front matter © 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(99)02184-X
tetl 16112

972
C–C bond forming reaction with an aldehyde to complete the carbon skeleton. Hydrolysis would then
complete enone formation.
We therefore took HFC-134a 1 through a dehydrofluorination/metallation sequence2 and trapped the
organolithium reagent with chlorotrimethylsilane. The vinylsilane 2a underwent further reaction in situ
with a second equivalent of n-butyllithium to afford the vicinal difluoride 3a in good yield (82%). Similar
compounds have been prepared recently by Burton and co-workers with low(er) temperature (−90°C)
halogen–metal exchange of trifluoroalkenyl halides as an initial step.3 Exposure to catalytic (20 mol%)
TAS-F in the presence of benzaldehyde according to the excellent precedent established by Hiyama,4
followed by mild acid work-up led (via the allylic alcohol 4a) to the isolation of known enone 5a in an
acceptable 55% yield as a crystalline solid5 according to Scheme 2.
Scheme 2.
3
The alkene configuration was assigned on the basis of the large JH–F coupling constant (36.8 Hz)
consistent only with a mutual trans arrangement between the nuclei. Similar results were obtained when
cinnamaldehyde and furfural were used as electrophiles in the reaction with 3a and 5b and 5c were
isolated in comparable yields. The dienone 5b is a particularly interesting product because of the presence
of two electronically very different double bonds. We were also able to trap hexanal, pivalaldehyde and
acrolein with 3a (Fig. 1); however, work up with dilute HCl led to the isolation of the difluoroallylic
alcohols 4d–f, implying that a mechanism with well developed dissociative character is followed during
the hydrolysis,1a the ease of reaction reflecting acutely the facility of carbenium ion formation. In the
case of acrolein, the yield of allylic alcohol product 4f was particularly poor and crotonaldehyde failed
to react at all under these conditions (Table 1).
Table 1
Syntheses of difluoroallylic alcohols 4 and enones 5
Though an effective catalyst here, TAS-F is expensive so we sought an alternative. Soluble fluoride
ion sources are of some interest and a number of non-hygroscopic species have been described recently
including the Gingras reagent (Ph3SnF2·NBu4)6 and De Shong’s silicate (TBAT, Ph3SiF2·NBu4);7 we
explored the effectiveness of these and TBAF, KF, TBADHTF (Bu4N·H2F3) and CsF as promoters of

973
Fig. 1.
C–C bond formation in the reaction of 3a and PhCHO. Only TBAF afforded any product at all despite
an exhaustive survey of reaction stoichiometries, a full equivalent being required to achieve a 28% yield
of enone 5a after acid work-up. This observation suggests strongly that the alternative fluoride sources
were not able to form the reactive silicate intermediate effectively.
The less reactive difluoroallylic alcohols could be hydrolysed upon treatment with carefully controlled
amounts of concentrated sulfuric acid and water and 5d and 5e were obtained in good yield by this
method. Again, the acrolein adduct 4f was recalcitrant and only the method of Normant (iodine in diethyl
ether at reflux)8 allowed progression to the dienone 5f. Montmorillonite K-10, an effective catalyst for
the hydrolysis of related systems9 was quite ineffective for this reaction.
Next, we attempted to use methyllithium as the organometallic nucleophile; due to difficulties in
isolating the adduct with chlorotrimethylsilane due to volatility, we prepared the dimethylphenyl silane
2b attempting to preserve the high level of reactivity in silic-ate formation. Treatment of 2b with
methyllithium under a range of conditions afforded only products of attack at silicon and none of the
difluorocompound. However, the triethylsilane 3c could be synthesised readily in an unoptimised 75%
yield and TAS-F-catalysed addition to benzaldehyde and cinnamaldehyde was successful. Treatment
of the crude products with 10 M HCl afforded the enone 5g and dienone 5h in 46% and 53% yield,
respectively. The requirement for stronger acid conditions reflects the lower degree of substitution on
reducing the length of the sidechain. The success of the sequence with the less reactive methyllithium
suggests that generality of the substituent attached to the acyl carbon could be achievable, though
access to difluorovinylsilanes via one-pot reactions of the trifluorosilane were unsuccessful with this
nucleophile. The use of metallated alkoxymethanes will form the subject of future study. Related
compounds have been synthesised by Schlosser,10 McCarthy,11 Yamanaka and co-workers,12 Haufe,13
and Funabiki.9
However, the procedure described herein offers a direct (at best two-, and at worst, three-pot) access to
the target enones from a particularly readily available and inexpensive starting material HFC-134a. This
sequence is strongly competitive with all the alternatives because of its directness and ready availability
of starting material. Optimisation of the individual reaction steps is in progress.
Acknowledgements
We thank Central Glass Co., Ltd (Kawagoe, Japan) for a secondment (to M.K.), ICI Klea and the
EPSRC (Engineering and Physical Sciences Research Council of Great Britain) for a CASE Studentship
(to J.M.B.) and ICI Klea for a generous donation of HFC-134a.

974
References
1. (a) Normant, J.-F.; Foulon, J. P.; Masure, D.; Sauvêtre, R.; Villieras, J. Synthesis 1975, 122–125. (b) Chuit, C.; Sauvêtre, R.;
Masure, D.; Baudry, M.; Normant, J.-F.; Villieras, J. J. Chem. Res. (S) 1977, 104. (c) Sauvêtre, R.; Masure, D.; Chuit, C.;
Normant, J.-F. Synthesis 1978, 128–130.
2. Burdon, J.; Coe, P. L.; Haslock, I. B.; Powell, R. L. Chem. Commun. 1996, 49–50.
3. Lu, L.; Pedersen, S. D.; Liu, Q.; Narske, R. M.; Burton, D. J. J. Org. Chem. 1997, 62, 1064–1071 and references cited
therein.
4. Fujita, M.; Obayashi, M.; Hiyama, T. Tetrahedron 1988, 44, 4135–4145.
5. Preparation of 3a. n-Butyllithium (20.0 ml of a 1.8 M solution in hexane, 36 mmol) was added dropwise to a stirred solution
of HFC-134a (600 ml, 25 mmol) in THF (30 ml) at −78°C. After 1 h, trimethylsilyl chloride (2.3 ml, 18 mmol) was added
dropwise, then 1 h later, n-butyllithium (11.0 ml of a 1.8 M solution in hexane, 20 mmol) was added dropwise. The reaction
was stirred at −78°C for 1 h and then warmed to −30°C, then quenched after 30 min with saturated aqueous ammonium
chloride. Extractive work-up and concentration then filtration though an alumina column (Brockmann, activity 1) eluting
with pentane afforded silane 3a (2.82 g, 81%) as a colourless oil. Preparation of 5a. TASF (0.43 g, 1.4 mmol) was added
gradually to a stirred solution of silane 3a (1.3 g, 7.0 mmol) and benzaldehyde (0.82 ml, 7.7 mmol) in THF (14 ml) at
−30°C under argon over 24 h. Dilute HCl (15 ml of a 1 M solution, 2.1 equiv.) was added and the mixture was stirred for 15
min before extractive work-up. Column chromatography (Rf 0.26, 10ether, 90% petroleum ether) and recrystallisation from
diethyl ether/hexane afforded 5a (0.74 g, 51%) as white needles, mp 52–53°C.
6. Gingras, M. Tetrahedron Lett. 1991, 32, 7381–7384.
7. Pilcher, A. S.; DeShong, P. J. Org. Chem. 1996, 61, 6901–6902.
8. Tellier, F.; Sauvêtre, R. Tetrahedron Lett. 1995, 36, 4223–4226.
9. Funabiki, K.; Murata, E.; Fukushima, Y.; Matsui, M.; Shibata, K. Tetrahedron 1999, 55, 4637–4642.
10. Bessiere, T.; Savary, D. N.-H.; Schlosser, M. Helv. Chim. Acta 1977, 173, 1739–1746.
11. Chen, C.; Wilcoxen, K.; Zhu, Y. F.; Kim, K. I.; McCarthy, J. R. J. Org. Chem. 1999, 64, 3476–3482.
12. Park, H. M.; Uegaki, T.; Konno, T.; Ishihara, T.; Yamanaka, H. Tetrahedron Lett. 1999, 40, 2985–2988.
13. Ernet, T.; Haufe, G. Synthesis 1997, 953–956.