Cyclizations of 5-hexenyl, 6-heptenyl, 7-octenyl, and 8-nonenyl radicals. The kinetic and regiochemical impact of fluorine and oxygen substituents

Article abstract of DOI:10.1021/jo990544n

Using competition kinetic methodology, rate constants for cyclizations of a series of hydrofluorocarbon (HFC) and ether 5-hexenyl, 6-heptenyl, and 7-octenyl radicals have been determined. Remarkably large rate constants (> 10⁷ s^-1) have been observed for 6-exo-cyclizations of 1,1,2,2-tetrafluoro- and 1,1,2,2,3,3,4,4,-octafluoro-6-heptenyl radicals (> 10³ those of analogous hydrocarbon radicals), whereas HFC hexenyl and heptenyl ethers exhibit lower cyclization reactivity, as do HFC 7-octenyl radical systems, which cyclize in an endo manner. HFC 8-nonenyl radicals were not observed to cyclize. The results can be rationalized in terms of transition state polar influences, though other factors may also play significant roles.

Full text of DOI:10.1021/jo990544n

J . Org. Chem. 1999, 64, 5993-5999
5993
Cycliza tion s of 5-Hexen yl, 6-Hep ten yl, 7-Octen yl, a n d 8-Non en yl
Ra d ica ls. Th e Kin etic a n d Regioch em ica l Im p a ct of F lu or in e a n d
|
Oxygen Su bstitu en ts
†
†
‡
‡
§
Anrong Li, Alexander B. Shtarev, Bruce E. Smart, Zhen-Yu Yang, J anusz Lusztyk,
§
§
,†
Keith U. Ingold, Anna Bravo, and William R. Dolbier, J r.*
Department of Chemistry, University of Florida, Gainesville, Florida 32611-7200, DuPont Central
Research & Development, Experimental Station, Wilmington, Delaware 19880-0328, and Steacie Institute
for Molecular Sciences, National Research Council of Canada, Ottawa, Ontario, Canada K1A 0R6
Received March 29, 1999
Using competition kinetic methodology, rate constants for cyclizations of a series of hydrofluoro-
carbon (HFC) and ether 5-hexenyl, 6-heptenyl, and 7-octenyl radicals have been determined.
7
-1
Remarkably large rate constants (>10 s ) have been observed for 6-exo-cyclizations of 1,1,2,2-
3
tetrafluoro- and 1,1,2,2,3,3,4,4,-octafluoro-6-heptenyl radicals (>10 those of analogous hydrocarbon
radicals), whereas HFC hexenyl and heptenyl ethers exhibit lower cyclization reactivity, as do HFC
7
-octenyl radical systems, which cyclize in an endo manner. HFC 8-nonenyl radicals were not
observed to cyclize. The results can be rationalized in terms of transition state polar influences,
though other factors may also play significant roles.
Whereas 5-hexenyl radical cyclizations comprise the
ultimate tool for making five-membered rings, use of
cyclizations of 6-heptenyl radicals to make six-membered
rings is a much less important synthetic technique, with
practical cyclizations to form larger rings being almost
absent from the literature.1 The reason for this dispar-
ity in importance lies, of course, in the large difference
in the respective rate constants for such cyclizations
Ta ble 1. Cycliza tion Rea ctivities of Hyd r oca r bon
Ra d ica ls a t 25 °C4
,5
-3
kexo/s-
1
kendo/s-1
5
3
n ) 1
n ) 2
n ) 3
2.7 × 10
5 × 10
(Table 1), the 6-exo process being considerably less
3
2
5.4 × 10
7.5 × 10
1.2 × 10
2
efficient (50 times slower) than the 5-exo process, with
cyclization of the 7-octenyl radical being another 45 times
slower, its cyclization proceeding exclusively endo.5
In an earlier comprehensive study, we demonstrated
that fluorine substitution at the radical center has a
remarkable influence on the cyclization reactivity of
radical systems, including ethers, which provide consid-
erable insight into the factors that govern cyclization
7
reactivity and regiochemistry in such systems.
5
-hexenyl radicals, leading to significant enhancement
of their rate constants for 5-exo cyclization, as well as to
an unexpected impact on the regiochemistry of such
processes, a result not readily rationalized.6
In this paper, we present data for cyclizations of a
series of fluorinated 5-hexenyl, 6-heptenyl, and 7-octenyl
†
The University of Florida.
DuPont Central Research and Development.
Steacie Institute for Molecular Sciences.
Contribution No. 7938.
‡
§
|
(
1) Lee, I.; Yoon, C. H.; Lee, T. H.; Kim, S. Y.; Ha, T. J .; Sung, Y.-S.;
Park, S.-H.; Lee, S. J . Am. Chem. Soc. 1998, 120, 7469.
2) (a) Sato, T.; Ishida, S.; Ishibashi, H.; Ikeda, M. J . Chem. Soc.,
Perkin Trans. 1 1991, 353. (b) Crich, D.; Chen, C.; Hwang, J . T.; Yuan,
H.; Papadatos, A.; Walter, R. I. J . Am. Chem. Soc. 1994, 116, 8937. (c)
Snider, B. B.; Merritt, J . E. Tetrahedron 1991, 47, 8663. (d) Molander,
G. A.; McKie, J . A. J . Org. Chem. 1994, 59, 3186. (e) Pirrung, F. O. H.;
Hiemstra, H.; Speckamp, W. N.; Kaptein, B.; Schoemaker, H. E.
Tetrahedron 1994, 50, 12415. (f) Russell, G. A.; Li, C. Tetrahedron Lett.
(
Resu lts
Absolute rate constants for the cyclizations of the series
of fluorinated radicals 1-7 (see list in Table 2) were
determined by uni- vs bimolecular competition experi-
ments, as depicted for radical 2 in Scheme 1. Reactions
were carried out under pseudo-first-order conditions
designed so that kinetically controlled cyclizations of the
intermediate radicals took place at a rate competitive
with their abstraction of a hydrogen atom from a reduc-
1
996, 37, 2557.
3) Fluorinated examples: (a) Morikawa, T.; Kodama, : Y.; Uchida,
J .; Takano, M.; Washio, Y.; Taguchi, T. Tetrahedron 1992, 41, 8915.
b) Arnone, A.; Bravo, P.; Frigerio, M.; Viani, F.; Cavicchio, G.;
(
(
Crucianelli, M. J . Org. Chem. 1994, 59, 459. (c) Arnone, A.; Bravo, P.;
Viani, F.; Zanda, M.; Cavicchio, G.; Crucianelli, M. J . Fluorine Chem.
1
996, 76, 169.
(
(
(
4) Beckwith, A. L. J .; Schiesser, C. H. Tetrahedron 1985, 41, 3925.
5) Newcomb, M. Tetrahedron 1993, 49, 1151.
6) Dolbier, W. R., J r.; Rong, X. X.; Bartberger, M. D.; Koroniak,
(7) Some of this work has been reported previously in communica-
tion form: Dolbier, W. R., J r.; Li, A. R.; Smart, B. E.; Yang, Z.-Y. J .
Org. Chem. 1998, 63, 5687.
H.; Smart, B. E.; Yang, Z.-Y. J . Chem. Soc., Perkin Trans. 2 1998, 219.
1
0.1021/jo990544n CCC: $18.00 © 1999 American Chemical Society
Published on Web 07/16/1999

5
994 J . Org. Chem., Vol. 64, No. 16, 1999
Li et al.
Ta ble 2. Com p etition Ra te Da ta for Cycliza tion ver su s Red u ction of F lu or in a ted 5-Hexen yl, 6-Hep ten yl, a n d 7-Octen yl
Ra d ica ls, 1-7, a t 25 ( 2 °C
kH/kc(endo)a
red. agent
kH/10 M s-1 a
6
-1
compound
kH/kc(exo)
1
2
3
4
5
6
7
, CH2dCHCH2OCF2CF2•
, CH2dCH(CH2)3CF2CF2•
, CH2dCHCH2(CF2)3CF2•
, CH2dCHCH2O(CF2)2CF2•
, CH2dCHO(CH2)2CF2CF2•
, CH2dCH(CH2)4CF2CF2•
, CH2dCH(CH2)2O2CCF2CF2•
1.38 ( 0.04
1.25 ( 0.02
2.56 ( 0.08
0.41 ( 0.02
9.77 ( 0.23
(TMS)3SiH
(TMS)3SiH
(TMS)3SiH
Et3SiH
n-Bu3SnH
tBuMe2SiD
Et3SiD
52 ( 6b
c
21.7 ( 0.5
18 ( 1
51 ( 5d
e
0.50 ( 0.04
27.1 ( 0.8
2.84 ( 0.14
3.28 ( 0.15
92 ( 8d
d
0.055 ( 0.0.013
b
0.12 ( 0.01
a
For radicals 5-7, the values are for kD rather than kH. b This work. c Reference 6. d Reference 11. e Reference 12. f Reference 10.
Sch em e 1
Ta ble 3. Cycliza tion Ra te Con sta n ts of F lu or in a ted
5
-Hexen yl, 6-Hep ten yl, a n d 7-Octen yl Ra d ica ls, 1-7, a t 25
(
2 °C
kc/s-
1
%
exo
kc(exo)/
10 s
kc(endo)/
10 s
6
-1
6
-1
a
krelb
radical product
(model)
6
3
1
2
>98
38 ( 4
8.5 × 10
4.5
94.5 14.4 ( 0.9
5.4 × 10 2667
2
0
.83 ( 0.05 7.5 × 10
1107
3
3
4
5
>98
>98
73.4
19.9 ( 2
5.4 × 10 3685
3
1.21 ( 0.12
9.4 ( 0.9
(5.4 × 10 ) 224
3
(5.4 × 10 ) 1740
2
3
.4 ( 0.3
(7.5 × 10 ) 4533
2
6
7
<2
<2
0.020 ( 0.005 1.2 × 10
167
300
2
0.036 ( 0.003 1.2 × 10
a
Appropriate model radical system; in the case of 1, a hydro-
carbon ether model (ref 4); for all others, hydrocarbon models were
used (ref 5). b kc of radical versus kc of model system.
manner, values of kadd [2.08 ((0.2) × 10 M s-1 and 3.10
8
-1
7
-1 -1
(
(0.3) × 10 M
3
s ] were obtained for addition of CH -
ing agent. All of the competition studies except for those
involving radicals, 6 and 7, which were relatively unre-
active toward cyclization, were carried out using C D
6 6
•
OCF
2
CF
2
radical to R-methylstyrene and to pentafluo-
rostyrene, respectively.
as solvent. To avoid complications from reactions of these
radicals with the benzene solvent, 1,2-dichloroethane was
used as solvent for these two radical systems.8,9 Another
tactic used to obtain more effective competitions in those
It was then fortunately possible to devise a good
competition experiment using pentafluorostyrene in com-
3 H
petition with (TMS) SiH, wherein the rate constant k
7
-1 -1
) 5.2 ((0.6) × 10 M
s
was obtained in a manner
described previously.12 This value is virtually identical
two systems was to use the D-transfer agents t-BuMe
and Et SiD, instead of the usual H-transfer analogues,
to take advantage of the slower rates of D-transfer.
2
SiO
to that for the analogous H-transfer to a perfluoro-n-alkyl
3
1
0
7
-1 -1 11
radical [k
H
) 5.1 ((0.5) × 10 M
s ].
Using the study of radical 2 in Scheme 1 as an explicit
example, the ratios of products 11:9 and 13:9 were
determined directly by 19F NMR analysis of the product
mixtures. The rate constants, kexo and kendo, were obtained
by means of plots of experimental concentration data
using eqs 1 and 2 in conjunction with the known values
There remained one rate constant, k , in Table 2 for
D
which there was not a measured value, that of reduction
•
of the radical, RO CCF CF , by Et SiD. Not having lfp
2
2
2
3
data for alkene additions of this radical, we have arbi-
5
-1 -1
trarily chosen to use the k value [1.2 × 10 M
D
s ] for
•
14
Et SiD’s reaction with n-C F . Since radicals R CF -
3
4
9
F
2
•
•
for k
There was one required k
of hydrogen transfer from (TMS)
H
.
2 2
CF2 and ROCF CF2 have identical reactivities and RCF -
•
is only 2-3 times less reactive,15,16 one can assume
H
value not yet known, that
SiH by radicals of the
. To obtain this H-transfer rate constant,
CF
2
•
3
that a radical of the type RO
reactivity more similar to that of n-R
2
CCF
2
CF
F
2
will have a
•
•
type ROCF
2
CF
2
than to that of
•
•
•
an absolute rate constant for the addition of ROCF
2
CF
2
2 2 H F
RCF CF . Thus we have used the k value of n-R to
to some alkene was needed, so that we could carry out
an appropriate addition versus reduction competition
study. The required values of kadd were obtained by 308
approximate the cyclization rate constant for radical 7.
With all required values for kH (or k ) in hand along
D
with the determined ratios of k /k (or k /k ), it was
H
C
D
C
nm laser flash photolysis (lfp) of CH
3
OCF
2
CF
2
I in Freon
therefore possible to calculate the values for kexo and kendo
for each of the radical systems, 1-7. These values are
given in Table 3.
Two other radicals, 14 and 15, were examined under
conditions similar to those used for the less reactive
systems, 6 and 7, those being:
1
3
1
13 at 298 K in a manner described previously. In this
(8) 1,2-Dichloroethane has one of the smallest rate constants that
has been measured for bimolecular H-transfer to a perfluoro-n-alkyl
radical.9
(
9) Shtarev, A. B.; Tian, F.; Smart, B. E.; Dolbier, W. R., J r. J . Am.
Chem. Soc. 1999, in press.
10) Shtarev, A. B.; Dolbier, W. R., J r.; Smart, B. E. J . Am. Chem.
Soc. 1999, 121, 2110.
11) Dolbier, W. R., J r.; Rong, X. X. J . Fluorine Chem. 1995, 72,
35.
(
3
(14) This value was determined by a direct competition between Et -
•
4 9
F , as described in the
(
SiH and Et
Experimental Section. k
(15) 2.2 and 2.8 times less reactive with n-Bu
3
SiD in their reaction with n-C
/k was determined to be 4.04 ( 0.13.
3
SnH and (TMS) SiH,
2
H
D
(
12) Delest, B.; Shtarev, A. B.; Dolbier, W. R., J r. Tetrahedron 1998,
3
1
6
5
4, 9273.
respectively.
(
13) Avila, D. V.; Ingold, K. U.; Lusztyk, J .; Dolbier, W. R., J r.; Pan,
(16) Bartberger, M. D.; Dolbier, W. R., J r.; Lusztyk, J .; Ingold, K.
U. Tetrahedron 1997, 53, 9857.
H.-Q.; Muir, M. J . Am. Chem. Soc. 1994, 116, 99.

Cyclization of Alkenyl Radicals
J . Org. Chem., Vol. 64, No. 16, 1999 5995
The latter was studied to determine whether a cyclo-
propane ring might act as an addend in an intramolecu-
lar reaction with a fluorinated radical. No cyclization
products were observed for either of these radical systems
5-Hexen yl a n d 6-Hep ten yl Eth er Ra d ica l Sys-
tem s. Beckwith earlier demonstrated that replacement
of the C -methylene in the skeleton of the 5-hexenyl
3
radical with an oxygen atom led to a significantly more
under our experimental conditions, using t-BuMe
2
SiH as
the H-transfer agent.
reactive radical system, 18, exhibiting a 5-exo cyclization
6
-1
rate constant (8.5 × 10 s ) 37 times greater than that
of the parent, non-ether system.4
,18
Discu ssion
This kinetic study of fluorinated 6-heptenyl and 7-octe-
nyl radical cyclizations demonstrates that the earlier-
observed significant enhancing influence of fluorine
substituents on 5-hexenyl cyclization rate constants is
manifested to an even greater extent within 6-heptenyl
systems, with similar regiochemical variations being
observed. Significant enhancements of endo-cyclizations
of 7-octenyl systems are also observed. The single 8-
nonenyl radical system that was examined failed to
cyclize under our experimental conditions, and cyclization
was found not to occur by addition to a cyclopropane ring
in place of a double bond.
Contrary to our expectations based on this result, when
the C -CF group of 6-heptenyl radical 3 was similarly
4 2
replaced by an oxygen atom to give radical (4), cyclization
was 10 times slower than for 3. To determine whether
this detrimental impact of 4’s ether linkage was due to
something intrinsic to the 6-heptenyl system, the reactiv-
ity of an analogous fluorinated 5-hexenyl ether, 1, was
examined. Indeed, ether radical 1 was found to also be
less reactive toward cyclization than its non-ether ana-
logue 17. Therefore, both the 5-hexenyl and the 6-hep-
6
-Hep ten yl Ra d ica l System s. The 6-exo-cyclizations
of tetra- and octafluoroheptenyl radicals 2 and 3 occur
with rate constants 3 orders of magnitude larger than
that of their hydrocarbon analogue. To our surprise these
rate constants are amazingly similar to those of the
fastest 5-hexenyl radical cyclizations that we have re-
6
ported. For example, the total cyclization reactivity of
tetrafluoro-6-heptenyl system 2 is slightly greater than
6
that of tetrafluoro-5-hexenyl system, 16, and octafluoro-
6
-heptenyl system 3 is only 2.5 times less reactive than
6
its most analogous 5-hexenyl system, 17.
tenyl fluorinated radicals 1 and 4, which contained allylic
ether linkages, failed to exhibit the enhanced reactivity
that is characteristic of hydrocarbon ether radical sys-
tems, such as 18.
Why should the presence of an allylic ether linkage be
beneficial to cyclization of hydrocarbon radical 18 but
detrimental to the cyclizations of the hydrofluorocarbon
radicals 1 and 4? The answer probably comes from a
combination of factors deriving from the impact of the
allylic oxygen atom. First, there is the potential impact
of the oxygen atom on the relative energies of ground
state conformations, particularly the way in which it
affects the mole fraction of what Bruice has termed “near
The reason that 6-heptenyl systems 2 and 3 exhibit
greater enhancements in cyclization rates (relative to the
hydrocarbon systems) than 5-hexenyl systems 16 and 17
reflects the fact that, consistent with the reactivity-
selectivity principle, slow reactions (i.e, 6-heptenyl cy-
clizations) should be more responsive to kinetically
beneficial structural change than fast reactions (i.e.,
19
attack conformations.” This must be considered an
unknown factor at this time, since we have not yet
calculated the pertinent conformational distributions.
Second, it was suggested by Beckwith that the enhanced
reactivity of ether 18 (krel ) 37), as compared to its non-
ether analogue, was the result of the impact of the oxygen
atom on bond lengths and bond angles within 18, so as
to make its 5-exo cyclization transition state more easily
5
-hexenyl cyclizations). As it is, the cyclization rate
constant of 3 is by far the largest yet measured for a 6-exo
cyclization. Even the cyclization of 7,7-diphenyl-6-hep-
tenyl radical, which forms a highly stabilized radical
5
-1
upon cyclization, has a rate constant of only 5 × 10 s
,
1
7
4
0 times slower.
(
17) Newcomb, M.; Horner, J . H.; Filipkowski, M. A.; Ha, C.; Park,
(18) Beckwith, A. L. J .; Glover, S. A. Aust. J . Chem. 1987, 40, 157.
(19) Bruice, T. C.; Lightstone, F. C. Acc. Chem. Res. 1999, 32, 127.
S.-U. J . Am. Chem. Soc. 1995, 117, 3674.

5
996 J . Org. Chem., Vol. 64, No. 16, 1999
Li et al.
attained.18 The present results certainly do not exclude
such factors from playing a role for 18, or for that matter
for radicals 1 and 4, but on the basis of our broad studies
of fluorinated radical cyclizations, we have come to
believe that polar influences may be the most important
of the factors involved in all of these ether systems.
If the transition state for cyclization maintains the
C-O bond in a geometry that will allow interactive
overlap of the C-O σ* orbital with the alkenyl π bond,
such perturbation would be expected to lower the ener-
gies of both the HOMO and the LUMO of the alkenyl
segment. In the case of the nucleophilic hydrocarbon
ether radical 18, such an effect would be beneficial to the
cyclization transition state, because the allylic oxygen
effectively makes the alkene segment more electrophilic.
In contrast, for electrophilic fluorinated ether radical
systems 1 and 4, this same influence of the allylic oxygen
should prove detrimental to cyclization. Thus, the dif-
ferent effect of a skeletal allylic oxygen atom on the rate
constants for cyclization of a hydrocarbon versus a
fluorohydrocarbon radical system can be explained sim-
ply in terms of the difference in the dipolar nature of their
respective transition states.
(alkoxycarbony)methyl radicals, such as 18, using sy-
ringe pump addition of n-Bu SnH, conditions that maxi-
1
3
mize the potential efficiency of radical cyclization pro-
cesses.
Unfortunately, although a significant enhancement (a
factor of 167) was observed for the 8-endo-cyclization of
-octenyl radical 6, there was not the same order of
magnitude of acceleration derived from its four fluorine
substituents as there had been for analogous 6-heptenyl
radical 2. A slightly greater enhancement for 8-endo-
cyclization of ester radical 7 (factor of 300) was observed.
These observed exclusive 8-endo-cyclizations were unre-
markable. Beckwith previously observed that the hydro-
carbon 7-octenyl radical also cyclized in an exclusively
-endo manner. Apparently, Baldwin’s rules do not apply
in the formation of such large rings, and the thermo-
dynamically more-stable cyclization product (from endo-
addition) prevails.
7
4
8
It is interesting to note that when the ether linkage is
vinylic rather than allylic, as in radical 5, the 6-exo
cyclization rate constant is only slightly less (krel ) 0.65)
than that of the non-ether analogue 2, whereas its 7-endo
cyclization rate constant is four times larger! The sig-
nificant regiochemical influence of the vinyl ether group
is undoubtedly a reflection of the greater ability of the
oxygen atom of 5 to provide electrons to the 7-endo-
cyclization transition state than to the 6-exo transition
state. As a consequence, there is 27% endo-cyclization
in the case of radical 5 but virtually none in the
cyclization of closest non-ether analogue 3 (and only 5%
in the cyclization of tetrafluoro radical 2.)
In an attempt to measure the rate constant for cycliza-
tion of a fluorinated 8-nonenyl radical, a kinetic study of
radical 14 using t-BuMe
transfer agent led to no observable cyclization products
i.e., <2%). Thus it appears that, using our current
competition kinetic methodology, the rates of 8-exo- or
-endo-cyclizations of fluorinated radicals are too slow
2
SiH as competing hydrogen
(
9
to be measured. We were also not able to detect the
intramolecular 6-exo-cyclization onto the cyclopropane
ring of radical 15.
Con clu sion s
Regioch em istr y. Although continuing to display a
greater propensity to undergo endo-cyclization than
purely hydrocarbon systems, the 6-heptenyl systems 2
In conclusion, remarkably fast rates have been ob-
served for 6-exo-cyclizations of partially fluorinated 6-hep-
tenyl radicals 2 and 3. The results are consistent with
transition state polar influences being involved in the
intramolecular addition of electrophilic fluorinated radi-
cals to nucleophilic hydrocarbon alkene systems. The
large observed unimolecular rate constants of such
cyclizations make it likely that these processes will
function quite efficiently as propagation steps in cyclo-
polymerization processes.20
Although similarly fluorinated 7-octenyl radicals do
cyclize (in an endo manner) with measurable rate con-
stants, the rates are considerably slower. Finally, fluori-
nated 8-nonenyl radicals were not observed to cyclize
under the competition conditions used in this study.
(
5.5%) and 3 (<4%) exhibit a lesser tendency to undergo
such addition than their 5-hexenyl analogues 16 (18%)
and 17 (11%).6 Although the data for the 6-heptenyl
system are more limited, similar trends in endo vs exo
reactivity seem to be observed for the 5-hexenyl and
6
-heptenyl radical systems. For example, less endo-
addition is observed when fluorine substitution extends
to the homoallylic position (as in 3 and 17) than for the
situation (as in 2 and 16) where the fluorine atoms
inducing radical electrophilicity are more distant from
the alkene site. Apparently, the impact of fluorines on
regiochemistry results from two effects, the first relating
to the structure and electrophilicity of the fluorinated
radicals and the second relating to the influence of the
neighboring fluorine substituents on the regiochemical
reactivity of the double bond. A detailed theoretical
examination of both the 5-hexenyl and 6-heptenyl flu-
orinated radical systems, currently nearing completion,
will hopefully provide more definitive insight into the
nature of these factors.
Cycliza tion s To F or m Even La r ger Rin gs. With
the huge accelerations observed in cyclizations of the
fluorinated 6-heptenyl radical systems 2 and 3, it was
hoped that similar, if not greater, enhancements would
be observed for the 7-octenyl radical systems 6 and 7.
The latter system was examined in response to the recent
report of a reasonably efficient 8-endo cyclization of
Exp er im en ta l Section
Gen er a l. H and 19F NMR (300 and 282 MHz respectively)
1
3
were measured in CDCl using TMS as internal standard for
1
19
3
H and CFCl or trifluorobenzene for F. All reagents, unless
otherwise specified, were purchased from Aldrich, PCR, or
Synquest and used as received. Benzene was distilled from
lithium aluminum hydride and stored over 4 Å molecular
sieves; triglycol methyl ether (TG) was distilled from NaH.
1
,1,2,2-Tetrafluoroethyl methyl ether has been synthesized
2
1
and characterized previously. Preparative gas chromatogra-
(20) Smart, B. E.; Feiring, A. E.; Krespan, C. G.; Yang, Z.-Y.; Hung,
M.-H.; Resnick, P. R.; Dolbier, W. R., J r.; Rong, X. X. Macromol. Symp.
1995, 98, 753.

Cyclization of Alkenyl Radicals
J . Org. Chem., Vol. 64, No. 16, 1999 5997
phy was carried out on a 20 ft × 0.25 in. copper column packed
with 20% SE-30 on Chromosorb P.
Syn th esis of P a r tia lly-F lu or in a ted Ra d ica l P r ecu r -
to a 100 mL Pyrex tube in which allyl bromide (1.21 g, 10
mmol), 1,4-diiodo-perfluorobutane (4.54 g, 10 mmol), and
benzene (50 mL) were charged. The mixture was degassed
three times and sealed with a rubber septum under nitrogen
sor s. 1,1,2,2-Tetr a flu or o-2-iod oeth yl Meth yl Eth er . ICF
2
-
CF I (3.54 g, 10 mmol) was added to sodium methoxide (0.54
2
and then photolyzed using a Rayonet reactor for 7 h. The
1
g, 10 mmol) in 50 mL of TG at -78 °C, and then the mixture
was warmed to room temperature in 20 min, with stirring
continued for 18 h. The product was purified by preparative
product was purified by preparative GC (1.98 g, 54%):
H
NMR δ 2.84 (d of t, 2H, J ) 7 Hz, J ) 18 Hz), 5.29-5.36 (m,
2H), 5.80 (m, 1H); 19F NMR δ -58.90 (m, 2F), -113.33 (m,
1
19
GC (1.93 g, 75%): H NMR δ 3.68 (s, 3H); F NMR δ -62.97
s, 2F), -92.63 (s, 2F); HRMS calcd for C IO, 257.9165,
found 257.9146.
,1,2,2-Tetr a flu or o-2-iod oeth yl Allyl Eth er (P r ecu r sor
to Ra d ica l 1). ICF CF I (3.54 g, 10 mmol) was added to
2F), -113.83 (m, 2F), -122.69 (m, 2F); HRMS calcd for
+
(
3
H
3
F
4
C
7
H
5
F
7
I (M - F), 348.9324, found 348.9369.
1,1,2,2,3,3-Hexa flu or o-3-iod op r op yl Allyl Eth er (P r e-
cu r sor to Ra d ica l 4). ICF CF COF (2.74 g, 10 mmol), allyl
1
2
2
2
2
bromide (6.05 g, 50 mmol), and KF (2.90 g, 50 mmol) were
mixed in TG (50 mL). The mixture was stirred at 80 °C for 1
day. After workup, the crude products were purified by
preparative GC to obtain 1,1,2,2,3,3-hexafluoro-3-iodopropyl
sodium allyloxide (0.8 g, 10 mmol) in 50 mL of TG at -20 °C,
and the mixture was warmed to room temperature in 10 min,
with stirring continued for 18 h. After workup, the product
was purified by preparative GC to give 1,1,2,2-tetrafluoro-2-
1
allyl ether (1.33 g, 40%): H NMR δ 3.49 (d, 2H, J ) 6 Hz),
1
iodoethyl allyl ether (2.27 g, 80%): H NMR δ 3.51 (d of t, 2H,
4.40 (d of d, 1H, J ) 11 Hz, J ) 2 Hz), 4.51 (d of d, 1H, J ) 17
J ) 8 Hz, J ) 2 Hz), 4.44 (m of d, 1H, J ) 11 Hz), 4.56 (m of
Hz, J ) 2 Hz), 4.96 (m, 1H); 19F NMR δ -58.60 (m, 2F), -83.04
1
9
d, 1H, J ) 17 Hz), 5.01 (m, 1H); F NMR δ -62.20 (s, 2F),
6 5 6
(s, 2F), -117.14 (m, 2F); HRMS calcd for C H F IO, 333.9289,
-
88.50 (s, 2F); HRMS calcd for C
83.9321.
,5,6,6-Tetr a flu or o-6-br om oh exa n ol. A total of 2.35 g (10
5
H
5
F
4
IO, 283.9321, found
found 333.9278.
2
4-Br om o-3,3,4,4-tetr a flu or obu ta n oic a cid was prepared
5
according to the literature.22
mmol) of 5,5,6,6-tetrafluoro-6-bromohexene and 0.11 g (2.9
mmol) of sodium borohydride were dissolved in 6 mL of dry
3,3,4,4-Tetr a flu or o-4-br om obu ta n ol. BH
3
‚Me
2
2
S (7.3 g,
COOH (3.58
0.096 mol) was added dropwise to BrCF CF CH
2
2
triglyme under nitrogen atmosphere, BF
3
.OEt
2
(0.57 g, 4.0
g, 0.015 mol) in 15 mL of THF at room temperature, and the
mixture was then refluxed for 4 h. The reaction mixture was
then poured into 40 mL of a saturated K CO solution and
2 3
mmol) in 1 mL of triglyme was added dropwise to the mixture,
and the mixture was stirred at room temperature for 1 h. To
the resultant solution was added 1 mL of water to decompose
excess sodium borohydride. Then 1.33 mL of NaOH (3 M) was
extracted with ethyl ether (3 × 60 mL). The concentrated
residue was distilled under reduced pressure (30 mmHg, 70
°C) to give 2.83 g (84% yield) of 3,3,4,4-tetrafluoro-4-bromo-
butanol: 1H NMR δ 1.66 (s, 1H), 2.38 (t of t, 2H, J ) 6 Hz, J
2 2
added followed by the addition of 1.33 mL of 30% H O . The
mixture was stirred for another 2 h and then poured into 40
mL of ice water and extracted with ethyl ether (3 × 100 mL).
The combined ethereal extracts were washed with two 50 mL
1
9
) 18 Hz), 3.97 (t, 2H, J ) 6 Hz); F NMR δ -66.77 (s, 2F),
-111.66 (t, 2F, J ) 18 Hz).
portions of water and dried over MgSO
4
. The concentrated
3,3,4,4-Tetr a flu or o-4-br om obu tyl Vin yl Eth er (P r ecu r -
sor to Ra d ica l 5). To a dry 45 mL of vinyl ethyl ether were
added 1.9 g (8.44 mmol) of 3,3,4,4-tetrafluoro-4-bromobutanol
residue was distilled under reduced pressure to give 5,5,6,6-
tetrafluoro-6-bromohexanol (2.0 g, 80% yield): bp 110 °C/30
1
mmHg; H NMR δ 1.65 (m, 4H), 2.10 (m, 2H), 3.66 (t, 2H, J )
and 1.49 g (4.67 mmol) of Hg(OAc)
for 18 h. Then the mixture was diluted with hexane (100 mL)
and washed sequentially with 5% NaOH, H O, and brine. The
organic layer was dried over MgSO and concentrated. Final
2
. The mixture was refluxed
1
9
6
Hz); F NMR δ -66.06 (s, 2F), -112.62 (t, 2F, J ) 18 Hz);
+
HRMS calcd for C
6
H
8
BrF
4
O (M - 1), 250.9695, found
2
2
50.9656.
,5,6,6-Tetr a flu or o-6-br om oh exa n a l. 5,5,6,6-Tetrafluoro-
-bromohexanol (10.12 g, 40 mmol) in 8 mL of CH Cl was
4
5
purification by column chromatography gave 3,3,4,4-tetrafluoro-
4-bromobutyl vinyl ether (0.84 g, 40% yield): 1H NMR δ 1.97
(t of t, 2H, J ) 6.6 Hz, J ) 18 Hz), 3.44 (t, 2H, J ) 6.6 Hz),
3.87 (d of d, 1H, J ) 7 Hz, J ) 2 Hz), 3.97 (d of d, 1H, J ) 14
6
2
2
added to pyridium chlorochromate (12.90 g, 60 mmol) sus-
pended in 80 mL of CH Cl . The mixture was stirred at room
2
2
1
9
temperature for an additional 6 h and filtered through silica
gel. Concentration of the solvent followed by fractional distil-
lation at reduced pressure afforded 4.2 g of 5,5,6,6-tetrafluoro-
Hz, J ) 2 Hz), 6.18 (d of d, 1H, J ) 7 Hz, J ) 14 Hz);
NMR δ -66.46 (s, 2F), -111.08 (t, 2F, J ) 18 Hz); HRMS calcd
for C BrF O, 249.9616, found 249.9608.
7,7,8,8-Tetr a flu or o-8-br om oocten e (P r ecu r sor to Ra d i-
ca l 6). Fe (1.68 g, 30 mmol) and CrCl ‚6H O (1.06 g, 4 mmol)
were mixed in 40 mL of EtOH, and then BrCF CF Br (5.2 g,
F
6
H
7
4
1
6
1
-bromohexanal (42% yield): bp 88 °C/ 30 mmHg; H NMR δ
.85 (m, 2H), 2.02 (m, 2H), 2.50 (t, 2H, J ) 7 Hz), 9.70 (s, 1H);
3
2
1
9
F NMR δ -66.31 (s, 2F), -112.69 (t, 2F, J ) 17 Hz); HRMS
2
2
+
calcd for C
6
H
8
BrF
4
O (M + 1), 250.9695, found 250.9689.
20 mmol) and 1,5-hexadiene (2.05 g, 25 mmol) were added in
one portion. The mixture was stirred at 65 °C for 14 h and
then poured into 40 mL of 1 M HCl. The solution was extracted
6
,6,7,7-Tetr aflu or o-7-br om oh epten e (P r ecu r sor to Radi-
ca l 2). Methyltriphenylphosphonium bromide (6.84 g, 19
mmol) was dissolved in 15 mL of anhydrous THF at 0 °C. Then
with CH
2
Cl
2
(3 × 100 mL). The combined extracts were washed
7
.06 mL of a 2.5 M solution of butyllithium in hexane was
with two 50 mL portions of water and dried over MgSO
4
. The
added dropwise to the mixture, which was stirred for an
additional 0.5 h at 0 °C. 5,5,6,6-Tetrafluoro-6-bromohexanal
concentrated residue was purified by preparative GC to give
7,7,8,8-tetrafluoro-8-bromooctene (1.5 g, 29% yield): H NMR
1
(
4.02 g, 16 mmol) in THF (10 mL) was added dropwise to the
(C D ) δ 0.95 (p, 2H, J ) 7.5 Hz), 1.26 (m, 2H), 1.55-1.73 (m,
6 6
19
mixture. After addition, the mixture was warmed to room
temperature and stirred for an additional 6 h. The contents
were poured into 50 mL of water and extracted with ethyl
ether (3 × 100 mL). The combined ethereal extracts were
washed with two 50 mL portions of water and dried over
MgSO . The concentrated residue was purified by preparative
4
GC to give 6,6,7,7-tetrafluoro-7-bromoheptene (1.59 g, 40%
yield): H NMR δ 1.69 (p, 2H, J ) 8 Hz), 2.03-2.16 (m, 4H),
5
4H), 4.88-4.94 (m, 2H), 5.58 (m, 1H); F NMR δ -66.01 (s,
8 11 4
2F), -112.55 (t, 2F, J ) 18 Hz); HRMS calcd for C H F Br
261.9980, found 261.9947.
3-Bu ten yl 3-iod o-2,2,3,3-tetr a flu or op r op ion a te (P r e-
cu r sor to Ra d ica l 7). 3-Iodo-2,2,3,3-tetrafluoropropionyl
chloride (1.72 g, 5.91 mmol) was added dropwise to 3-buten-
1-ol (0.72 g, 10 mmol) at 0 °C. After addition, the mixture was
stirred for another 3 h. The usual workup gave the product
1
1
9
1
.00-5.06 (m, 2H), 5.75 (m, 1H); F NMR δ -65.47 (s, 2F),
which was purified by chromatography (1.21 g, 63%): H NMR
-
111.92 (t, 2F, J ) 18 Hz); HRMS calcd for C
47.9824, found 247.9872.
,4,5,5,6,6,7,7-Octa flu or o-7-iod o-1-h ep ten e (P r ecu r sor
to Ra d ica l 3). Bis(tributyltin) (1.4 g, 2.43 mmol) was added
7
H
9
BrF
4
δ 2.43 (q, 2H, J ) 6 Hz), 4.35 (t, 2H, J ) 6 Hz), 5.06-5.12 (m,
2
2H), 5.70 (m, 1H); 19F NMR δ -60.59 (t, 2F, J ) 8 Hz), -111.78
4
7 7 4 2
(t, 2F, J ) 8 Hz); HRMS calcd for C H F IO , 325.9427, found
325.9428.
(
21) Terrell, R. C.; Speers, L.; Szur, A. J .; Treadwell, J .; Vacciardi,
(22) Huang, W. Y.; Lu, L.; Zhang, Y. F. Chinese J . Chem. 1990, 68,
281.
T. R. J . Med. Chem. 1971, 14, 517.

5
998 J . Org. Chem., Vol. 64, No. 16, 1999
,5,6,6-Tetr a flu or o-6-br om oh exyl Vin yl Eth er (P r ecu r -
Li et al.
5
0.57 mmol) in 1,2,4-trimethylbenzene (0.2 mL) was added to
a Pyrex NMR tube and photolyzed for 0.5 h. The product was
obtained by preparative GC: 1H NMR δ 2.29 (d of t, 2H, J )
17 Hz, J ) 6 Hz), 4.80 (d, 1H, J ) 18 Hz), 4.90 (d, 1H, J ) 10
sor to Ra d ica l 14). To dry 54 mL of vinyl ethyl ether were
added 2.53 g (10 mmol) of 5,5,6,6-tetrafluoro-6-bromohexanol
and 1.77 g (5.55 mmol) of Hg(OAc)
for 18 h, then diluted with hexane (100 mL), and washed
sequentially with 5% NaOH, H O, and brine. The organic layer
was dried over MgSO and concentrated. The final purification
by column chromatography gave 5,5,6,6-tetrafluoro-6-bromo-
hexyl vinyl ether (1.00 g, 36% yield): H NMR δ 1.72 (m, 4H),
2
2
. The mixture was refluxed
1
9
Hz), 5.16 (t of t, 1H, J ) 51 Hz, J ) 5 Hz), 5.38 (m, 1H);
F
2
NMR δ -113.67 (m, 2F), -125.53 (s, 2F), -130.32 (m, 2F),
-137.47 (d, 2F, J ) 51 Hz); HRMS calcd for C H F , 242.0342,
7 6 8
4
found 242.0345.
1
1-Meth yl-2,2,3,3,4,4,5,5-octa flu or ocycloh exa n e. AIBN
(azo-bis-isobutyronitrile) (0.02 g, 0.14 mmol) was added to
.10 (m, 2H), 3.69 (t, 2H, J ) 5.7 Hz), 3.98 (d of d, 1H, J ) 2.1
Hz, J ) 6.6 Hz), 4.16 (d of d, 1H, J ) 2.1 Hz, J ) 14 Hz), 6.44
I(CF
The solution was refluxed for 2 days to carry out an iodine
transfer cyclization.23 Then Bu
SnH (0.17 g, 0.57 mmol) was
added to the mixture, and it was photolyzed for 20 additional
2 4 2 2 2 2
) CH CHdCH (0.21 g, 0.57 mmol) in 0.5 mL of CH Cl .
1
9
(
2
2
d of d, 1H, J ) 6.6 Hz, J ) 14.1 Hz); F NMR δ -66.10 (s,
F), -112.64 (t, 2F, J ) 19 Hz); HRMS calcd for C
77.9929, found 277.9882.
8
H
11BrF
4
O,
3
1
3
,3,4,4-Tetr a flu or o-4-br om obu tylcyclop r op a n e (P r e-
hours. The product was then purified by preparative GC: H
cu r sor to Ra d ica l 15). 5,5,6,6-Tetrafluoro-6-bromohexene
NMR δ 0.92 (d, 3H, J ) 7 Hz), 1.60 (m, 1H), 2.18 (m, 1H),
2.65 (m, 1H); 19F NMR δ -115.47 (m, F), -119.73 (d, 1F, J )
271 Hz), -125.60-126.93 (m, 3F), -130.82 (m, 1F), -143.86
(d, 1F, J ) 102 Hz), -144.82 (d, 1F, J ) 102 Hz); HRMS calcd
(
(
1.18 g, 5 mmol) was mixed with diazomethane in diethyl ether
from 5.1 g of diazald), then Pd(OAc) (25 mg) was added in
2
portions to the mixture, and the mixture was stirred for 10
min. Removal of the solvent afforded 1.02 g (82% yield) of
for C
1,1,2,2,3,3-Hexa flu or op r op yl Allyl Eth er . ICF
OCH CHdCH (0.21 g, 0.63 mmol) and Bu SnH (0.18 g, 0.63
7 6 8
H F , 242.0342, found 242.0365.
1
3
,3,4,4-tetrafluoro-4-bromobutylcyclopropane: H NMR (C
6
D
6
)
2 2 2
CF CF -
δ 0.07 (m, 2H), 0.47 (m, 2H), 0.71 (m, 1H), 1.49 (m, 2H), 2.17
2
2
3
1
9
(
1
m, 2H); F NMR (C
6
D
6
) δ -66.06 (s, 2F), -112.09 (t, 2F, J )
mmol) were dissolved in 1,2,4-trimethylbenzene (0.2 mL). The
solution was photolyzed for 0.5 h. The product was purified
by preparative GC: 1H NMR δ 3.43 (d of t, 2H, J ) 6 Hz, J )
2 Hz), 4.36-4.47 (m, 2H), 4.81 (t of t, 1H, J ) 51 Hz, J ) 7
Hz), 4.92 (m, 1H); 19F NMR δ -87.86 (m, 2F), -133.78 (m,
2F), -138.02 (d of t, 2F, J ) 51 Hz, J ) 7 Hz); HRMS calcd
8 Hz); HRMS calcd for C BrF , 247.9824, found 247.9819.
7
H
9
4
Syn t h esis of P r od u ct s fr om Kin et ic E xp er im en t s.
,1,2,2-Tetr a flu or o-4-(2,3,4,5,6-p en ta flu or op h en yl)bu tyl
1
Meth yl Eth er . CH
3
OCF
2 2 3
CF I (0.11 g, 0.43 mmol), Et SiH
(0.23 g, 1.94 mmol), and 2,3,4,5,6-pentafluorostyrene (0.45 g,
2
.33 mmol) were added to a Pyrex NMR tube. The mixture
for C
3-Meth yl-4,4,5,5,6,6-h exaflu or o-1-oxacycloh exan e. AIBN
(0.025 g, 0.16 mmol) was added to I(CF OCH CHdCH (0.21
g, 0.63 mmol) in CH Cl (1 mL). The solution was refluxed for
3
2 days to carry out the iodine transfer cyclization. Then Bu -
6 6 6
H F O, 208.0323, found 208.0323.
was degassed and photolyzed for 4 days; 80% of iodide was
1
converted. The product was purified by preparative GC:
H
F
2
)
3
2
2
NMR δ 2.25 (m, 2H), 2.96 (t, 2H, J ) 7 Hz), 3.65 (s, 3H); 19
2
2
2
3
NMR δ -94.75 (s, 2F), -119.28 (t, 2F, J ) 18 Hz), -144.52 (d
of d, 2F, J ) 7 Hz, J ) 22 Hz), -157.07 (t, 1F, J ) 22 Hz),
SnH (0.18 g, 0.63 mmol) was added to the mixture, and it was
photolyzed for 20 h. The product was purified by preparative
GC: 1H NMR δ 0.41 (d of t, 3H, J ) 7 Hz, J ) 2 Hz), 1.56 (m,
-
162.88 (t of d, 2F, J ) 22 Hz, J ) 7 Hz); HRMS calcd for
C
11
H
7
F
9
O, 326.0353, found 326.0362.
1
9
1
,1,2,2-Tetr a flu or oeth yl Allyl Eth er . This product was
1H), 2.67-2.74 (m, 1H), 2.87 (t, 1H, J ) 12 Hz); F NMR δ
-89.37 (d of q, 1F, J ) 17 Hz, J ) 154 Hz), -93.62 (m of d,
1F, J ) 154 Hz), -125.66 (d, 1F, J ) 259 Hz), -128.59 (d, 1F,
J ) 281 Hz), -131.07 (m of d, 1F, J ) 281 Hz), -131.07 (m of
d, 1F, J ) 281 Hz), -145.09 (d, 1F, J ) 260 Hz); HRMS calcd
1
also isolated by GC from the above reaction mixture: H NMR
δ 3.96 (d of t, 2H, J ) 6 Hz, J ) 2 Hz), 4.83-4.88 (m, 1H),
4
3
.96 (m of d, 1H, J ) 17 Hz), 5.05 (t of t, 1H, J ) 54 Hz, J )
19
Hz), 5.47 (m, 1H); F NMR δ -91.56 (m, 2F), -136.85 (d of
O,
t, 2F, J ) 54 Hz, J ) 5 Hz); HRMS calcd for C
58.0355, found 158.0357.
-Met h yl-4,4,5,5-t et r a flu or o-1-oxa cyclop en t a n e. The
5
H
6
F
4
for C
3,3,4,4-Tetr a flu or obu tyl Vin yl Eth er . BrCF
OCHdCH (0.10 g, 0.40 mmol) and Bu SnH (0.13 g, 0.45 mmol)
6 6 6
H F O, 208.0323, found 208.0311.
1
2 2 2 2
CF CH CH -
3
2
3
same procedure was followed as for the preparation of 1-meth-
were dissolved in 1,2,4-trimethylbenzene (1.5 mL). The solu-
tion was photolyzed for 8 h. The product was purified by
1
yl-2,2,3,3,4,4,5,5-octafluorocyclohexane: H NMR δ 0.52 (d, 3H,
J ) 7 Hz), 2.00 (m, 1H), 3.08 (m, 1H), 3.41 (m, 1H); 19F NMR
δ -78.08 (d, 1F, J ) 139 Hz), -89.29 (d of d, 1F, J ) 148 Hz,
preparative GC: H NMR δ 2.44 (m, 2H), 3.97 (t, 2H, J ) 6
1
Hz), 4.02 (d of d, 1H, J ) 2 Hz, J ) 7 Hz), 4.25 (d of d, 1H, J
) 2 Hz, J ) 14.4 Hz), 6.22 (t of t, 1H, J ) 4.8 Hz, J ) 53 Hz),
J ) 12 Hz), -119.58 (d of t, 1F, J ) 234 Hz, J ) 12 Hz),
+
19
-
127.14 (d, 1F, J ) 237 Hz); HRMS calcd for C
5
H
6
F
3
O (M
-
6.50 (d of d, 1H, J ) 7 Hz, J ) 14.4 Hz); F NMR δ -117.26
F), 139.0371, found 139.0389.
,6,7,7-Tetr a flu or oh ep ten e (9). 6,6,7,7-Tetrafluoro-7-bro-
moheptene (0.13 g, 0.51 mmol) and tris(trimethylsilyl)silane
6 8 4
(m, 2F), -137.63 (d, 2F, J ) 53 Hz); HRMS calcd for C H F O,
6
172.0511, found 172.0553.
2-Meth yl-3,3,4,4-tetr a flu or o-1-oxa cycloh exa n e. AIBN
(0.025 g, 0.16 mmol) was added to BrCF CF CH CH O-
CHdCH (0.16 g, 0.63 mmol) in 1 mL of CH Cl . The solution
was refluxed for 2 days to carry out iodine transfer cycliza-
tion.23 Then Bu
SnH (0.18 g, 0.63 mmol) was added to the
(0.17 g, 0.68 mmol) were dissolved in 1,3,5-trimethylbenzene
2
2
2
2
(
6
0.9 mL). The mixture was degassed and photolyzed for 6 h.
,6,7,7-Tetrafluoroheptene was separated by GC: H NMR
2
2
2
1
(
(
(
CD
3
COCD
3
) 1.59 (p, 2H, J ) 8 Hz), 1.87-2.02 (m, 2H), 2.12
3
q, 2H, J ) 7 Hz), 4.91-5.02 (m, 2H), 5.69-5.83 (m, 1H), 6.12
t of t, 1H, J ) 53 Hz, J ) 3.6 Hz); F NMR δ -117.56 (t, 2F,
mixture, and it was photolyzed for 20 h. The product was
1
9
purified by preparative GC: 1H NMR δ 1.25 (d, 3H, J ) 6 Hz),
1
9
J ) 19 Hz), -137.0 (d, 2F, J ) 53 Hz); HRMS calcd for C
70.0719, found 170.0736.
-Meth yl-2,2,3,3-tetr aflu or ocycloh exan e (11). The prepa-
7
H
10
F
4
2.29-2.38 (m, 2H), 3.67-3.82 (m, 2H), 4.02 (m, 1H); F NMR
δ -134.77 (d of m, 1F, J ) 252 Hz), -137.43 (d, 1F, J ) 257
Hz), -117.56 (d of m, 1F, J ) 252 Hz), -120.46 (d, 1F, J )
1
1
1
ration method was the same as above: H NMR (CD
3
COCD
3
)
6 8 4
257 Hz); HRMS calcd for C H F O, 172.0511, found 172.0501.
δ 1.07 (d, 3H, J ) 6.9 Hz), 1.38 (m, 1H), 1.58 (m, 1H), 1.77-
4,4,5,5-Tetr a flu or o-1-oxa cycloh ep ta n e. This product was
1
1
.80 (m, 2H), 2.04-2.21 (m, 3H); 19F NMR δ -115.26 (d of d,
F, J ) 33.8 Hz, J ) 7.3 Hz), -116.15 (d of d, 1F, J ) 34 Hz,
also isolated by GC from the above reaction mixture: H NMR
1
1
9
δ 3.79 (t, 4H, J ) 6 Hz), 2.38 (m, 4H); F NMR δ -111.81 (m,
J ) 7.3 Hz), -119.38 (s, 1F), -120.27 (s, 1F); HRMS calcd or
170.0719, found 170.0731.
,1,2,2-Tetr a flu or ocycloh ep ta n e (13). The preparation
4F); HRMS calcd for C O, 172.0511, found 172.0536.
6 8 4
H F
7 10 4
C H F
1
Kin etic Exp er im en ts
1
method was the same as above: H NMR (CD
3
COCD
m, 2H), 2.12 (m, 4H), 2.43 (m, 4H); F NMR (CD COCD
110.80 (m, 4F); HRMS calcd for C 170.0719, found
70.0746.
,4,5,5,6,6,7,7-Oct a flu or o-1-h ep t en e. The mixture of
I(CF CH CHdCH (0.21 g, 0.57 mmol) and Bu SnH (0.17 g,
3
) δ 1.20
Deter m in a tion of th e Ra te Con sta n t for Ad d ition of
1
9
(
-
1
3
3
) δ
•
CH
3
OCF
2
CF
2
to r-Meth ylstyr en e a n d P en ta flu or osty-
7 10 4
H F
(23) (a) Curran, D. P. In Free Radicals in Synthesis and Biology;
4
Minisci, F., Ed.; Kluwer Academic Publishers: 1989; pp 37-51. (b)
Curran, D. P.; Kim, D. Tetrahedron Lett. 1986, 27, 5821.
2
)
4
2
2
3

Cyclization of Alkenyl Radicals
J . Org. Chem., Vol. 64, No. 16, 1999 5999
r en e by La ser F la sh P h otolysis of CH
3
OCF
2
CF
2
I in
was changed over the set of 6 tubes. Ratios of n-C
4
F
9
H/n-C
4
F
9
2
D
D
F r eon 113 a t 298 K. Values of kadd were determined by 308
nm laser flash photolysis (lfp) of CH OCF CF I in Freon 113
at 298 K following procedures described in detail in our
were determined by the ratio of integrals of -CF H and -CF
2
resonances in the 19F NMR at 138.7 and 139.4 ppm, respec-
tively.
3
2
2
1
3
previous publications. They were obtained as slopes of the
plots of the observed rate constants for the growth of the
absorption at ca. 320 nm of benzyl-like addition product
radicals vs concentration of the styrene quencher.
With a slope of 4.04 ((0.13), and using the known rate
5
-1 -1
constant for k
H
(5.0 ( 0.4 × 10 M
s , a value of 1.2 ((0.1)
5
-1 -1
×
10 M
s
D
for k was calculated.
Com p etition Kin etics: Un im olecu la r Cycliza tion vs
Deter m in ation of th e Rate Con stan t for H-atom Tr an s-
fer to 1,1,2,2,-Tetr aflu or o-2-m eth oxyeth -1-yl Radical fr om
Hyd r ogen Atom Abstr a ction . Gen er a l P r oced u r e for
Com p etition Stu d ies. Into each of a set of six Pyrex NMR
(
TMS)
added C
fluorotoluene as an internal F NMR standard, and varying
amounts of (TMS) SiH and 2,3,4,5,6-pentafluorostyrene. Each
3
SiH. Into each of a set of six Pyrex NMR tubes were
6 6
tubes were added C D or dichloroethane, iodide, trifluoro-
6
D
6
, 1-iodo-1,1,2,2-tetrafluoroethyl methyl ether, tri-
19
toluene as an internal F NMR standard, and varying
amounts of hydrogen atom donors. Each sample was sealed
with rubber septa, frozen in a dry ice-2-propanol slush, and
degassed (freeze and thaw) three times; then the tubes were
photolyzed using a Rayonet reactor until the starting material
was consumed completely. Product ratios for varied concentra-
tions of hydrogen atom donor allow the determination of the
1
9
3
sample was sealed with rubber septa, frozen in a dry ice-2-
propanol slush, and degassed (freeze and thaw) three times;
then the tubes were photolyzed using a Rayonet reactor until
the starting material was consumed completely. Product ratios
for varied concentrations ratios of (TMS)
allow the determination of the ratio k /kadd. Ratios of reduction
and addition products were determined by the integration of
the -CF H and -CF - (signals at δ -137.06 and -119.28,
3 6 5 2
SiH/C F CHdCH
H
ratio k /kexo(endo) applying eqs 1 and 2. Yields are determined
H
by integration of product resonances versus that of internal
19
standard (δ -63.24) in the F NMR spectrum.
2
2 6 5
C F
1
9
respectively) in the F NMR. The hydrogen transfer rate
7
-1 -1
Ack n ow led gm en t. Support of this work in part by
the National Science Foundation is acknowledged with
thanks.
constant (5.2 ( 0.6) × 10 M
s
was obtained on the basis
•
of the addition rate constant of ROCF
2
CF
2
to 2,3,4,5,6-
as described
7
-1
-1
pentafluorostyrene (3.1 ( 0.3) × 10
M
s
1
6
previously.
Com p etition of H ver su s D Abstr a ction fr om Tr ieth -
ylsila n e a n d Tr ieth ylsila n e-d . Deter m in a tion of th e Ra te
Con sta n t for D-Tr a n sfer fr om Et SiD. The reactions were
3
conducted in 1,3-bis(trifluoromethyl)benzene initiated with
UV-light in Pyrex tubes. The overall concentration of the
silanes was kept constant, and only their relative concentration
Su p p or tin g In for m a tion Ava ila ble: Tables of raw ki-
netic data and 1H and 19F NMR spectra of new compounds.
This material is available free of charge via the Internet at
http://pubs.acs.org.
J O990544N