Selective Reduction of Halopolyfluorocarbons
J . Org. Chem., Vol. 63, No. 21, 1998 7297
peroxides, and compounds 1, 5, 13, 29, and 32 were commercial
and used without further purification. Compounds 21, 27,
34,16 and 2417 were prepared using known procedures. Prod-
ucts 3, 4, 6, 7, 9-12, 19a ,18 20,3 and 2819 were identified by
comparison of 1H and 19F data with reported values. Materials
15, 16, 22, 23, 30, and 33 were identified by comparison with
authentic samples.
Ca u tion ! We did not experience any problems in this work;
however, use a peroxide initiator and general consideration
of radical chain mechanism of reduction polyfluoroalkanes by
silanes suggest that the reaction under certain conditions can
be hard to control. Reductions should be carried behind a
shield or in a barricade.
Red u ction of P olyh a loflu or oa lk a n es by Or a n osila n es
Hyd r id es. (Gen er a l P r oced u r e). A mixture of polyhalo-
fluoroalkane (0.05-0.2 mol) and silane (0.05-0.4 mol) was
placed in Hastelloy reactor, and initiator/catalyst was added.
The reactor was closed, purged with N2, and kept 1 h at 25 °C
and 6-13 h at 80-120 °C. The reactor was cooled to -78 °C
and unloaded. The crude reaction mixture was analyzed by
GC and 1H and 19F NMR spectroscopy. Reaction conditions,
ratio of reagents, and products are given in Table 1.
is probably the result of a much slower hydrogen transfer
from trichlorosilane to the polyfluoroalkyl radical.
In reductions of chloropolyfluoroalkanes, triethylsilane
is less reactive than tributyltin hydride since the latter
effectively replaces several chlorine atoms in chloropoly-
fluorocarbons under mild conditions.3 This difference in
hydrodechlorination reactivity is not surprising, since the
latter was shown to be more than 200 times more active
as a hydrogen transfer agent in reactions involving
polyfluorinated radicals.2 An advantage of the less
reactive silanes lies in the greater selectivities available
with them in the reduction of chloropolyfluoroalkanes.
The radical chain mechanism proposed for the reduc-
tion of halopolyfluoroalkanes by organosilanes is similar
to that offered for the reduction of polychloroalkanes by
silanes7,8 and for polychloro- and chloropolyfluoroalkanes
or polyfluorinated ethers by 2-propanol10-12 under UV-
irradiation. All of these processes were shown to have a
radical chain mechanism involving the formation of
electrophilic polyhalogenated radicals6,7,12 as intermedi-
ates.
Since compounds 18, 19, and 19a were previously reported
in patent literature only, they were characterized by 19F NMR.
On the other hand, the mechanism for the reduction
of chloropolyfluoroalkanes catalyzed by H2PtCl6 and Pd/C
may be much more complex. There are several possibili-
ties including formation of active metal clusters as a
result of reduction by H2PtCl6, followed by oxidative
addition of the C-Cl bond of chloropolyfluoroalkane to
an electron-rich metal center and further formation of a
polyfluoroalkyl σ-complex of Pt (or Pd). A polyfluoroalkyl
derivative of the metal may be converted to final product
either through the sequence of formation of metal hydride
complex and reductive elimination of polyfluoroalkane or
as a result of the homolysis of a carbon-metal bond with
generation of fluoroalkyl radical. Unfortunately, experi-
mental data are not sufficient to choose between these
mechanisms.
The experimental results obtained in this work are in
excellent agreement with a recently published opinion
based on an analysis of the reactivity of polyfluorinated
radicals that triethylsilane may be “a very useful agent
for relatively slow chain processes involving fluorinated
radicals”.2 Much lower reactivity of silicon hydrides vs
tin hydrides in radical reductions can be compensated
for by much higher reactivity of polyfluorinated radicals
vs the corresponding hydrocarbon radicals, since the
reactivities of fluorinated radicals approach that of the
highly electronegative tert-butoxy radical.2
18: -63.91 (2F, dm, 176 Hz), -64.82 (2F, dm, 176 Hz),
-193.66 (1F, dm, 56 Hz).
19: -61.94 (1F, dm, 187 Hz), -65.38 (1F, dm, 187 Hz),
-129.10 (1F, dm, 326 Hz), -130.75 (1F, dm, 326 Hz), -206.71
(1F, dm, 57 Hz).
19a : -132.55 (4F, dm, 54 Hz), -147.13 (1F, m).
Compounds 25 and 26 were not isolated but were character-
ized in mixture by 19F NMR spectroscopy.
25: -49.77 (2F, m; 11 Hz), -86.82 (3F, s), -118.15 (2F, m).
26: -54.18 (1F, dm; 168 Hz), -57.00 (1F, dm; 168 Hz),
-78.68 (3F, m), -116.08 (1F, dm; 272 Hz), -121.50 (1F, dm;
272 Hz).
Red u ction of 34 by Tr ieth ylsila n e. 10 mL of HSi(C2H5)3
was added dropwise to 10 g of 34 with stirring at 80 °C over
10 min. The reaction mixture was kept at this temperature
for 5 h, collecting the crude product which passed a short reflux
water condenser in a cold trap (-78 °C). Collected product
(based on GC and NMR data 35 was 95% purity) was
redistilled to give 4.8 g (86%) of 35.
Ack n ow led gm en t. The author thanks Dr. V. V.
Grushin and Dr. C. G. Krespan for helpful discussion,
reviewer A for a number useful comments, and Dr. D.
D. Khasnis for technical assistance.
J O9807363
(16) Petrov, V. A.; Krespan, C. G.; Smart, B. E. J . Fluorine Chem.
1998, 89, 125.
(17) Sievert, A. C.; Nappa, M. J . PCT WO 95/16656, 1995 (to
Dupont); Chem. Abstr. 1995, 123, 339128.
(18) Weigert F. J . J . Fluorine Chem. 1990, 46, 375.
(19) Burdon, J .; Ezmirly, S. T.; Huckerby, T. N. J . Fluorine Chem.
1988, 40, 283.
Exp er im en ta l Section
19F and 1H NMR spectra were recorded using CFCl3 as
internal standard and either chloroform-d or acetone-d as a
lock solvent. Silanes, poly(methylhydrosiloxane) (Gelest),