Sodium Chlorodifluoroacetate

[1895-39-2]  · C2ClF2NaO2  · Sodium Chlorodifluoroacetate  · (MW 152.46)

(generation of difluorocarbene and the synthesis of difluorocyclopropanes and difluoroalkenes)

Physical Data: mp 196-198 °C.

Solubility: sol DME, diglyme (0.60 g mL-1), triglyme, DMF.

Preparative Methods: the salt is prepared by careful neutralization of 130.5 g (1.0 mol) of chlorodifluoroacetic acid in 300 mL of Et2O with 53 g (0.5 mol) of anhydrous Na2CO3. The Et2O and water are removed under reduced pressure. The salt is then dried in a vacuum desiccator over P2O5.1

Handling, Storage, and Precautions: is an irritant and is hygroscopic.


The thermal decomposition of sodium chlorodifluoroacetate produces difluorocarbene, which will react with alkenes to form the difluorocyclopropane. Cyclohexene reacts (eq 1) with the carbene to form the difluorocyclopropane in 11% yield.2

There are times when a change from the sodium salt to the lithium salt results in a higher yield of the cyclopropane (eq 2). In addition to the higher yields, the reaction with the lithium salt was also reported to be cleaner. No stereochemistry for the products of this reaction was reported.3

A change from sodium to lithium can also alter the course of the reaction. The reaction of an o-quinone (eq 3) with the sodium salt produces the carbonate, while the lithium salt provides the hydroxy ketone. The exact reason for the different paths followed by the two salts is not known for certain. One explanation was that the sodium salt produced the difluorocarbene directly by a one-step a-elimination and then the carbene underwent a 1,4-addition to the o-quinone. The carbonate was then produced by hydrolysis, which would require only 0.05% water to be present in the diglyme.4

A 1,4-addition had been previously observed in the reaction of a cisoid b-methoxy enone (eq 4). This reaction affords the 1,4-adduct as a mixture of 3-methoxy isomers.5

Transoid enones produce the cyclopropane derivatives. The a,b-unsaturated ketone (eq 5) only underwent addition at a temperature of 150 °C, even though the salt was observed to decompose at 125 °C.6 In the case of a trienone (eq 6), the reaction selectively took place at the carbon-carbon double bond furthest removed from the carbonyl.7

Similar regiospecificity has been seen in the reaction of a dienone (eq 7) from the prostaglandin area. Again, reaction at the more remote double bond was observed. The reaction was not stereoselective in this case, providing a mixture of stereoisomeric cyclopropanes.8

A mixture of cyclopropanes was also observed for the reaction of a diene lactone with the sodium salt in triglyme (eq 8). Cyclopropanation of the more remote double bond is again preferred.9

The selective difluorocyclopropanation of an a-methylene lactone over an unsaturated ketone was also achieved by reaction with sodium chlorodifluoroacetate. Helenalin acetate (eq 9) afforded a mixture of three products, all resulting from cyclopropanation of the a-methylene group.10

Enol acetates will react with difluorocarbene. The enol acetate of diisobutyl ketone (eq 10) leads to the formation of an unspecified 2:1 mixture of acetoxy difluorocyclopropanes.11 The enol acetate (eq 11) was selectively cyclopropanated over the 5,6-double bond by controlling the amount of the sodium salt used (~10 mol equiv). The yield in this reaction was 34%, with 53% of the starting enol acetate being recovered.12

Silyl enol ethers react readily with difluorocarbene. In the case of an a,a-difluoro ketone, the enol acetate was unreactive toward difluorocarbene generated by thermal decomposition of the salt. The trimethylsilyl enol ether reacted rapidly with the carbene to produce the trimethylsilyloxycyclopropane (eq 12). Some of the enol ether underwent hydrolysis to return the ketone under the reaction conditions. The t-butyldimethylsilyl enol ether was stable to hydrolysis but reacted at a much slower rate that the trimethylsilyl enol ether.13

A comparison of methods for difluorocarbene formation has shown the thermal decomposition of the salt to be one of the two most effective for the cyclopropanation of butyl (Z)-propenyl ether (eq 13). The reaction of the enol ether and 2 equiv of the salt in triglyme at 165 °C for 0.5 h results in a 42% yield of the cyclopropane. The cis relationship of the groups on the double bond appears to be maintained in the final product. By using 4 equiv of the salt, the yield could be raised to 53%. The only method found to be superior in terms of yield was the fluoride ion-induced decomposition of (bromodifluoromethyl)triphenylphosphonium bromide.14

Wittig Alkenation.

In the presence of Triphenylphosphine, sodium chlorodifluoroacetate produces the intermediate difluoro ylide. Treatment of benzaldehyde (eq 14) with this reagent combination delivers the difluoroalkene in 74% yield.15 The reaction reportedly does not work well for ketones when triphenylphosphine is used. Substitution of Tri-n-butylphosphine and changing the solvent to N-methylpyrrolidone allows cyclohexanone (eq 15) to be transformed in 46% yield to the difluoroalkene.16

With fluorines on the a-carbon of the carbonyl, the reaction with the salt and triphenylphosphine in DME provides the perfluoroalkenes in good yield. Treatment of 2,2,2-trifluoroacetophenone under these conditions affords the alkene (eq 16) in 68% yield.1 Changing solvents to DMF causes a dramatic increase in the rate of decomposition of the salt. It also results in an improvement in yield to 84%.17

When the perfluoroalkyl chain of the ketone is two or more carbons, fluoride ion-catalyzed rearrangement of the initially produced alkene can occur (eq 17). In these cases, the use of the lithium salt is recommended in order to suppress rearrangement.18


The acyl hypochlorite of chlorodifluoroacetic acid can be prepared by reaction of the sodium salt and Cl-F at low temperature.19 A substitution product has been isolated from the reaction of a tungsten dihydride with the salt.20 The crystal structures of the products from the reaction of Vaska's complex, IrCl(CO)(PPh3)2, with the salt have also been reported.21 An iridobetaine intermediate has been proposed to account for the observed products.21c

1. Herkes, F. E.; Burton, D. J. OSC 1973, 5, 949.
2. Birchall, J. M.; Cross, G. W.; Haszeldine, R. N. Proc. Chem. Soc. 1960, 81.
3. Slagel, R. C. CI(L) 1966, 848.
4. Derenberg, M.; Hodge, P. JCS(P1) 1972, 1056.
5. Hodge, P.; Edwards, J. A.; Fried, J. H. TL 1966, 5175.
6. Beard, C.; Dyson, N. H.; Fried, J. H. TL 1966, 3281.
7. Beard, C.; Berkoz, B.; Dyson, N. H.; Harrison, I. T.; Hodge, P.; Kirkham, L. K.; Lewis, G. S.; Giannini, D.; Lewis, B.; Edwards, J. A.; Fried, J. H. T 1969, 25, 1219.
8. Crabbé, P.; Cervantes, A. TL 1973, 1319.
9. Shibata, K.; Takegawa, S.; Koizumi, N.; Yamakoshi, N.; Shimazawa, E. CPB 1992, 40, 935.
10. Díaz, E.; Salazar, I. T 1979, 35, 815.
11. Crabbé, P.; Cervantes, A.; Cruz, A.; Galeazzi, E.; Verlarde, E. JACS 1973, 95, 6655.
12. Kobayashi, Y.; Taguchi, T.; Terada, T.; Oshida, J.-i.; Morisaki, M.; Ikekawa, N. JCS(P1) 1982, 85.
13. Erni, B.; Khorana, H. G. JACS 1980, 102, 3888.
14. Bessard, Y.; Müller, U.; Schlosser, M. T 1990, 46, 5213.
15. (a) Fuqua, S. A.; Duncan, W. G.; Silverstein, R. M. JOC 1965, 30, 1027. (b) Fuqua, S. A.; Duncan, W. G.; Silverstein, R. M. OSC 1973, 5, 390.
16. Fuqua, S. A.; Duncan, W. G.; Silverstein, R. M. JOC 1965, 30, 2543.
17. Herkes, F. E.; Burton, D. J. JOC 1967, 32, 1311.
18. Herkes, F. E.; Burton, D. J. JOC 1968, 33, 1854.
19. Tari, I.; DesMarteau, D. D. JOC 1980, 45, 1214.
20. Chen, K. W.; Kleinberg, J.; Landgrebe, J. A. IC 1973, 12, 2826.
21. (a) Schultz, A. J.; Khare, G. P.; McArdle, J. V.; Eisenberg, R. JACS 1973, 95, 3434. (b) Schultz, A. J.; Khare, G. P.; McArdle, J. V.; Eisenberg, R. JOM 1974, 72, 415. (c) Schultz, A. J.; Khare, G. P.; Meyer, C. D.; Eisenberg, R. IC 1974, 13, 1019.

Michael J. Taschner

University of Akron, OH, USA

Copyright 1995-2000 by John Wiley & Sons, Ltd. All rights reserved.