A photoprecursor for difluorocarbene

Article abstract of DOI:10.1016/S0022-1139(02)00274-9

4,4-Difluoropyrazolidine-3,5-dione was synthesized as a precursor for the corresponding pyrazolinedione, envisioned as a photochemical source of difluorocarbene. However, this azo compound proved to be far too unstable. In contrast, 10,10-difluorobicyclo[4.3.1]deca-1,3,5-triene, readily synthesized from indane, was found to be a practical source of difluorocarbene for photochemical as well as thermal cyclopropanation reactions.

Full text of DOI:10.1016/S0022-1139(02)00274-9

Journal of Fluorine Chemistry 119 (2003) 75±80
A photoprecursor for di¯uorocarbene
Yigang He1, David M. Lemal*
Department of Chemistry, Dartmouth College, Hanover, NH 03755, USA
Received 22 July 2002; received in revised form 11 October 2002; accepted 12 October 2002
Dedicated to Professor Tatlow on the occasion of his 80th birthday
Abstract
4,4-Di¯uoropyrazolidine-3,5-dione was synthesized as a precursor for the corresponding pyrazolinedione, envisioned as a photochemical
source of di¯uorocarbene. However, this azo compound proved to be far too unstable. In contrast, 10,10-di¯uorobicyclo[4.3.1]deca-1,3,5-
triene, readily synthesized from indane, was found to be a practical source of di¯uorocarbene for photochemical as well as thermal
cyclopropanation reactions.
# 2002 Elsevier Science B.V. All rights reserved.
Keywords: Di¯uorocarbene; Cyclopropanation; Photoprecursor; Photoextrusion; Electrocyclic equilibrium; Pyrazolidine-3,5-diones; [4.3.1]Propellane
1. Introduction
2. Results and discussion
Thermal methods abound for the generation of di¯uor-
ocarbene (1) [1,2], but the literature contains just one
photochemical source that is at all practical from a syn-
thetic perspective;2 namely, di¯uorodiazirine (2) (Eq. (1))
[1]. Since this compound is a gas, and one prone to
detonation, it seemed desirable to develop an alternative
photoprecursor.
Maier et al. found that irradiation of 4,4-dimethylpyrazo-
line-3,5-dione (3) yields dimethylketene (4), which he was
able to generate in matrix isolation (Eq. (2)) [3].
(2)
(1)
Potential advantages of such a reagent over thermal sources
include the ability to generate the carbene under gentle
conditions over a wide range of temperatures, especially
low temperatures for reaction with very sensitive or labile
substrates. The reagent might also be valuable as a CF2
source in ¯ash photolysis and matrix isolation experiments.
We sought a compound that would absorb long wavelength
UV or visible light and, in extruding di¯uorocarbene, pro-
duce stable, relatively unreactive byproduct(s).
Our initial plan was to synthesize 4,4-di¯uoropyrazoline-
3,5-dione (5), which we expected to fragment readily in
analogous fashion upon irradiation with visible light into
nitrogen, carbon monoxide and di¯uoroketene (6) and/or
di¯uorocarbene (Eq. (3)). In light of the exothermicity of the
fragmentation of 53 together with the great lability of
di¯uoroketene [5], it seemed likely that, if formed at all,
the ketene would be born with suf®cient vibrational energy
to fragment into carbon monoxide and carbene 1. If ketene 6
survived, photolysis of 5 would make possible detailed study
of this elusive molecule that has been trapped but never
* Corresponding author. Tel.: ‡1-603-646-2989; fax: ‡1-603-646-3946.
E-mail address: david.m.lemal@dartmouth.edu (D.M. Lemal).
1 Co-corresponding author.
3 At the HF/6-31GÃ//3-21G level of theory, DE (HF) ˆ À33.5 kcal/mol
for fragmentation of 5 into nitrogen, carbon monoxide and difluor-
ocarbene [4].
2 Other photochemical methods employ multiphoton infrared or very
short wavelength ultraviolet radiation [1].
0022-1139/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved.
PII: S 0 0 22- 1 1 3 9 ( 0 2) 0 0 27 4 - 9

76
Y. He, D.M. Lemal / Journal of Fluorine Chemistry 119 (2003) 75±80
Scheme 1.
detected spectroscopically.
instances. As judged by color intensity, the most promising
set of conditions employed a base followed by t-butyl
hypochlorite at À60 to À78 8C. The mixture turned black
and effervesced, but even with a large excess of cyclopen-
tadiene present no Diels±Alder adduct of 5 with the diene
could be obtained. It appears that pyrazolinedione 5 is an
extremely labile compound, clearly unsuitable for our ori-
ginal purpose. In contrast, pyrazolinedione 13, made by
oxidation of diazo compound 10 with t-butyl hypochlorite at
±70 8C, reacts with cyclopentadiene to give adduct 14 in
good yield and is fairly stable in solution at temperatures up
to at least À10 8C (Eq. (4)).
(3)
A precursor for 5, 4,4-di¯uoropyrazolidine-3,5-dione (7),
was synthesized as shown in Scheme 1. Ethoxycarbonylhy-
drazide (8) [6] was prepared with a large excess of malonic
ester to minimize dihydrazide formation and cyclized with
methoxide to pyrazolidine-3,5-dione (9) [7]. Diazo transfer
from tosyl azide gave 10, which required protection as in 11
for the subsequent reaction with ¯uorine to succeed. Ther-
mal deprotection of the ¯uorination product 12 yielded 7 as a
high melting, very moisture-sensitive solid.
Oxidation of 7 to obtain the target compound 5 was
attempted in many different ways. The oxidants tried
included dinitrogen tetroxide, t-butyl hypochlorite, dichlor-
ine, N-bromosuccinimide, lead tetraacetate and lead tetra-
kis(tri¯uoroacetate). The compound was considerably more
dif®cult to oxidize to the pyrazolinedione than un¯uorinated
analogs 9 and 10. In order to facilitate the process by
generating salts of 7, bases both weak (e.g. 2,6-lutidine)
and strong (e.g. 1,8-bis(dimethylamino)naphthalene, potas-
sium t-butoxide) were introduced in many runs prior to
addition of the oxidant. Like other pyrazoline-3,5-diones
and the related triazoline-3,5-diones, 5 was certain to be
highly colored,4 but deep color was observed in only a few
(4)
During our oxidation attempts, we learned that Jacobs had
also synthesized 7 via a different route and tried to transform
it into 5 [8]. She observed effervescence, though no color
changes, in some of her oxidation attempts and concluded as
we did that 5 must be very labile.
Attention was now directed toward a very different type of
potential photoprecursor for di¯uorocarbene. In the 1960s
Scholl synthesized bicyclic triene 15 and showed that it
could function as a thermal source of the carbene [9].
4 Both diazo compound 13 and 4,4-dimethylpyrazoline-3,5-dione (3)
are deep blue.

Y. He, D.M. Lemal / Journal of Fluorine Chemistry 119 (2003) 75±80
77
Scheme 2.
Pyrolysis at 160 8C in cyclohexene as solvent gave 7,7-
di¯uoronorcarane (17) in 72% yield (Eq. (5)).
®nding neutral homoaromatic systems, and attention has
focused on semibullvalenes. These compounds undergo
degenerate Cope rearrangement with barriers of ꢁ5 kcal/
mol or less, and there is strong evidence that appropriately
substituted semibullvalenes and even barbaralanes have
been observed in a bishomoaromatic state (equivalent to a
Cope transition state) [15]. If the calculated results for 15 to
16 are reasonably accurate, it is striking that the ground
states of these molecules come so close to being a transition
state for electrocyclization.
(5)
Before obtaining the Scholl thesis [9], we synthesized 15
by essentially the same route he followed, as shown in
Scheme 2. The triene is readily accessible, as the overall
yield from indane (18) was 55% and no chromatography
was required. We explored the ability of 15 to function as a
thermal precursor for di¯uorocarbene by heating solutions
of it in a variety of alkenes at 143 8C, with the results shown
in Table 1. Most cyclopropane yields were excellent,
though that with cyclohexene was far less than Scholl's.
The key question for us was whether the triene would serve
as a photochemical source of carbene 1. Irradiation with a
medium pressure mercury arc at ꢁ15 8C of solutions of
triene 15 in the same substrates did indeed yield the
cyclopropanes, albeit in lower yields than those obtained
thermally (Table 1). In light of di¯uorocarbene's modest
reactivity [1], lower photochemical yields can be attributed
at least in part to the great difference in reaction tempera-
ture. The yields in Table 1 were determined by 19F NMR,
but in the case of 2,3-dimethyl-2-butene isolated yields
Doubtless electrocyclization to the tricyclic diene 16 pre-
ceded extrusion of the carbene with formation of indane
(18). The electrocyclic equilibrium lies on the tricyclic
diene side in the dichloro and dibromo analogs of 15 and 16
as well as the unsubstituted system. This contrast re¯ects
the well-known destabilization of cyclopropane rings by
¯uoro substituents, in particular the selective weakening of
the bond opposite the site of substitution [10].5 The Ger-
man workers deduced the triene structure 15 principally on
the basis of careful analysis of the proton and ¯uorine NMR
spectra, but the distinction between the two isomers is
clearest in the carbon spectrum, which was unavailable to
them in 1969. The bridgehead carbons appear at d À123.1,
more than 70 ppm down®eld from their expected chemical
shift in 16.6
Hybrid density functional calculations at the B3LYP/6-
31GÃà level of theory indicate that DGꢀ ˆ 2:12kcal/mol at
25 8C for 15 to 16 [12]. The calculated transition state free
energy at 25 8C lies just 2.53 kcal/mol above the triene and
0.41 kcal/mol above the diene, signifying an extremely low
barrier for a cycloheptatriene±norcaradiene interconver-
sion.7 There has been much interest for many years in
Table 1
Cyclopropanation with difluorocarbene from 10,10-difluorobicyclo[4.3.1]-
deca-1,3,5-triene (15)
Alkene
D (143 8C)
hn (>280 nm)
Molar
ratio
Yield
(%)
Molar
ratio
Yield
5 Earlier attributed to a p interaction, the lengthening and weakening of
the cyclopropane bond opposite fluorine substituents has been shown to
result from inductive electron withdrawal from a s orbital [11].
6 The bridgehead carbons in 10,10-dichlorotricyclo[4.3.1.01,6]deca-2,4-
diene appear at 50.6 ppm [13], and relative to chlorine, fluorine shifts the
resonances of b-carbons upfield [14].
(%)
2,3-Dimethyl-2-butene [16]
2-Methyl-2-butene [16,17]
a-Methylstyrene [17,18]
3,4-Dihydro-2H-pyran [19]
2-Methoxypropene [17,19]
Norbornene [20]
10
10
10
10
10
99
97
95
94
90
5
5
5
5
5
87
75
75
68
62
7 Triene 15 displays novel dynamic NMR behavior, especially above
room temperature, that requires further study. Clearly, interconversion
of 15 and 16 is orders of magnitude too fast to account for it.
2
20
0
80
2
0
2
Cyclohexene [16,21]
18
20
Trace

78
Y. He, D.M. Lemal / Journal of Fluorine Chemistry 119 (2003) 75±80
were obtained as well: 90% by the thermal method and
77% by photolysis.
mp 218 8C (dec). IR (card): 2158 (C=N2), 1682(C=O);
1H NMR (DMSO-d6) d 10.1 (broad, NH); 13C NMR
(DMSO-d6) d 162.0 (C=O), 59.1 (C=N2). Anal. calcd. for
C3H2N4O2: C 28.58; H 1.60; N 44.44. Found: C 28.76; H
1.62; N 44.22.
In the course of our study, we learned that Platz and
Tippman [22] had also synthesized 15 in the hope of
studying carbene 1 in ¯ash photolysis experiments. They
discovered that the triene does not yield appreciable
amounts of 1 by ¯ash photolysis under conditions where
a variety of bicyclic dienes analogous to 16 do give car-
benes. As Platz has noted, this ®nding strongly suggests that
isomerization of 15 to 16 is exclusively a thermal process,
i.e. that photons are only capable of producing carbene
after 16 has formed spontaneously. This surmise accords
with the dictate of Woodward and Hoffmann that a hexa-
triene should undergo photoelectrocyclization in conrota-
tory fashion [23], a process which is geometrically
impossible for triene 15.
4.2. 1,2-Bis (t-butoxycarbonyl)-4-diazopyrazolidine-
3,5-dione (11)
A suspension of 4-diazopyrazolidine-3,5-dione (0.252 g,
2.00 mmol), (BOC)2O (di-t-butyl dicarbonate) in THF
(6 ml, 1 M, 6 mmol) and DMAP (24.4 mg, 0.200 mmol)
in 15 ml of acetonitrile was stirred at room temperature for
24 h. The solvent was removed and the residue was taken up
in ether (20 ml); the resulting solution was washed with 1 M
KHSO4 (20 ml) and saturated NaCl (20 ml), then dried with
anhydrous sodium sulfate. Removal of solvent yielded a
yellow solid which was washed with petroleum ether, and
pale yellow crystals were obtained. Yield: 0.44 g (68%), mp
136±138 8C. IR (card) 2177 (C=N2), 1798 (C=O), 1755
3. Conclusions
1
Synthesized from malonic ester, 4,4-di¯uoropyrazoli-
dine-3,5-dione (7) was subjected to a wide variety of oxida-
tion conditions to obtain pyrazolidinedione 5. However, the
labile 5 was not isolated, nor even trapped. As an alternative
di¯uorocarbene source, 10,10-di¯uorobicyclo[4.3.1]deca-
1,3,5-triene (15) was prepared in 55% overall yield from
indane. This triene proved to be not only an effective thermal
source of di¯uorocarbene, as shown earlier by Scholl, but a
good photochemical precursor as well.
(C=O); H NMR (CDCl3) d 1.53 (s); 13C NMR (CDCl3)
d 156.1 (C=O), 146.5 (C=O), 86.5 (C±O), 62.9 (C=N2), 27.9
(CH3). Anal. calcd. for C13H18N4O6: C 47.85; H 5.56; N
17.17. Found: C 47.94; H 5.50; N 17.30.
4.3. 1,2-Bis(t-butoxycarbonyl)-4,4-difluoropyrazolidine-
3,5-dione (12)
1,2-Bis(t-butoxycarbonyl)-4-diazopyrazolidine-3,5-dione
(11) (2.0 g, 6.1 mmol) in 15 ml dry acetonitrile and 20 ml
CFCl3 was cooled to À50 8C [25]. The yellow solution was
purged with N2 for 10 min, then 30% F2 in He was passed in
until the yellow color disappeared (4±5 bubbles/s). Fluor-
ination could also be monitored with TLC until the starting
material disappeared. Residual F2 was purged with N2 for
15 min. The solvent was removed by rotary evaporator and
the residue was recrystallized from CH2Cl2 and hexane to
obtain pure 12 as very water-sensitive white crystals (1.6 g,
78%), mp 85 8C (dec). 19F NMR (CDCl3) d À116.9 (s); 1H
NMR (CDCl3) d 1.58 (s); 13C NMR (CDCl3) d 154.7 (t,
J ˆ 29 Hz, CF2C=O), 145.7 (s, O±C=O), 100.4 (t,
J ˆ 260 Hz, CF2), 88.7 (s, O±C±(CH3)3), 27.9 (s, CH3).
Anal. calcd. for C13H18N2F2O6: C 46.43; H 5.40; N 8.33; F
11.30. Found: C 45.69; H 5.43; N 8.51; F 11.36.
4. Experimental
NMR spectra were recorded on a Varian Unity Plus 300 or
500 FT NMR spectrometer. CFCl3 was used as reference for
19F NMR. Infrared spectra were recorded on a Perkin-Elmer
1600 Series FTIR instrument, and UV-Vis spectra on a
Perkin-Elmer Lambda 9 UV/VIS/NIR Spectrophotometer.
Melting points were measured with a Thomas-Hoover
Unimelt capillary melting point apparatus without correc-
tion. Elemental analyses were performed by Atlantic Micro-
lab Inc.
4.1. 4-Diazo-3,5-pyrazolidinedione (10)
To a slurry of 3,5-pyrazolidinedione (2.0 g, 20 mmol) [7]
in 150 ml of acetonitrile at 0 8C was added 2.02 g
(20.0 mmol) of triethylamine, and the mixture was stirred
for 15 min at room temperature [24]. To the mixture was
added 3.94 g (20.0 mmol) of tosyl azide at 0 8C. With
stirring the mixture was allowed to warm to room tempera-
ture over 2h and maintained there overnight. The solution
was ®ltered to remove the insoluble solid and the ®ltrate
was concentrated to dryness. Acetonitrile (30 ml) was added
to the residue, and the resulting mixture was ®ltered and
washed with acetonitrile (2 Â 5 ml). Yield: 1.80 g (71.4%),
4.4. 4,4-Difluoropyrazolidine-3,5-dione (7)
1,2-Bis(t-butoxycarbonyl)-4,4-di¯uoropyrazolidine-3,5-
dione (12) (0.80 g, 2.4 mmol) was placed in a dry round-
bottom ¯ask that was then evacuated and ®lled with argon
[26]. The solid was heated to 105 8C until no more gas
evolved; it was cooled to room temperature and stirred with
20 ml dry ether for 2 h. The ether was decanted and the solid
was dried under vacuum to obtain the highly water-sensitive
compound 7 (0.31 g, 95%), mp 185±187 8C. 1H NMR
(CD3CN) d 9.15 (Br); 19F NMR (CD3CN) d À113.2(s);

Y. He, D.M. Lemal / Journal of Fluorine Chemistry 119 (2003) 75±80
79
13C NMR (CD3CN) d 160.4 (t, J ˆ 27 Hz, C=O), 109.3 (t,
J ˆ 265 Hz, CF2).
4.7. trans-3,4-Dibromo-10,10-difluoro[4.3.1]propellane
(5,6-dibromo-8,8-difluorohexahydro-3a,7a-methanoindene)
4.5. Bicyclo[4.3.0]nona-1(6),3-diene
10,10-Di¯uoro[4.3.1]propell-3-ene (34.4 g, 0.202 mol) in
250 ml dry CH2Cl2 was cooled to À78 8C, and Br2
(ꢁ10.4 ml, 0.20 mol) was added dropwise with stirring till
the color of Br2 remained. The solution was stirred for
30 min, then stripped on a rotary evaporator. The resulting
solid was recrystallized (MeOH:EtOEt, 1:1) to give 46.2g
colorless crystals (69.2%), mp 85±86 8C. 1H NMR (CDCl3,
500 MHz) d 4.18 (m, 2H), 2.83 (m, 1H), 2.60 (m, 1H), 2.39
(m, 1H), 2.26 (m, 1H), 2.20 (m, 2H), 1.98 (m, 2H), 1.79 (m,
1H), 1.58 (m, 1H); 13C NMR (CDCl3, 125.7 MHz) d 116.4
(dd, J ˆ 307, 285 Hz), 52.0 (dd, J ˆ 4:4, 3.3 Hz), 50.3 (d,
J ˆ 3:7 Hz), 34.7, 34.2, 33.4 (dd, J ˆ 12, 8.8 Hz), 33.0 (dd,
J ˆ 13, 9.3 Hz), 32.8, 32.3 (d, J ˆ 3:3 Hz), 24.3 (d,
J ˆ 12Hz); 19F NMR (CDCl3, 282.2 MHz) d À131.3 (d,
J ˆ 159 Hz), À143.7 (d, J ˆ 159 Hz); Anal. calcd. for
C10H12F2Br2: C 36.40; H 3.67; F 11.51; Br 48.43. Found:
C 36.18; H 3.45; F 11.32; Br 48.71.
Ammonia (500 ml) was condensed in a three-necked ¯ask
equipped with a mechanical stirrer and a cold ®nger con-
denser [27]. A dry ice±acetone bath was used to maintain the
temperature at À78 8C. Lithium (10.5 g, 1.5 mol) was intro-
duced into the ¯ask in small pieces, resulting in a deep blue
solution. This solution was then stirred vigorously for
additional 2h to ensure that lithium was dissolved com-
pletely in the liquid ammonia. A solution of 59.0 g
(0.50 mol) of indane in 120 ml absolute ethanol was then
added dropwise over 1 h, and the mixture was stirred for an
additional 7 h at À78 8C. The ¯ask was opened to air and
warmed to room temperature with stirring to evaporate most
of the ammonia. Methanol (100 ml) was added to the ¯ask
carefully to react with excess lithium. The resulting brown
solution was quenched with 300 ml water and extracted with
three 200 ml portions of pentane. Pentane layers were
combined and dried over Na2SO4. Evaporation of the pen-
tane followed by vacuum distillation gave bicyclo[4.3.0]-
nona-1(6),3-diene (56.9 g, 94.8%) as a colorless liquid, bp
4.8. 10,10-Difluorobicyclo[4.3.1]deca-1,3,5-triene (15)
trans-3,4-Dibromo-10,10-di¯uoro[4.3.1]propellane
1
33±35 8C/3 Torr. H NMR (500 MHz, CDCl3) d 5.78 (s,
(33.0 g, 0.100 mol) in 100 ml dry THF was added dropwise
to a solution of potassium tert-butoxide (22.4 g, 0.200 mol)
in 300 ml dry THF at À45 8C. The solution became cloudy
immediately and was warmed to room temperature slowly
over a period of 8 h. Most solvent was removed on a rotary
evaporator and pentane (300 ml) was added to the residue;
the resulting organic layer was washed with ®ve 150 ml
portions of water. After drying over Na2SO4, the solvent was
removed on a rotary evaporator to give a yellowish liquid.
Vacuum distillation gave the triene (15.8 g, 94.2%) as a
2H), 2.68 (s, 4H), 2.29 (t, J ˆ 7 Hz, 4H), 1.87 (quintuplet,
J ˆ 7 Hz, 2H)
4.6. 10,10-Difluoro[4.3.1]propell-3-ene (8,8-difluoro-
2,3,4,7-tetrahydro-1H-3a,7a-methanoindene)
Dibromodi¯uoromethane (131.5 g, 0.625 mol) was added
to a solution of triphenylphosphine (164 g, 0.625 mol) in dry
triglyme (400 ml), and a white precipitate formed instanta-
neously [28]. The suspension was stirred at room tempera-
ture for 0.5 h. The reaction vessel was placed in a water bath,
then bicyclo[4.3.0]nona-1(6),3-diene (30.0 g, 0.25 mol), 18-
crown-6 (6.60 g, 0.025 mol) and dried KF (72.5 g, 1.25 mol)
were added. After being stirred at room temperature for 2
days, the mixture had become black. Slow vacuum distilla-
tion at 3 Torr gave about 100 ml colorless liquid, which was
quenched with 300 ml water and extracted with two 200 ml
portions of pentane. Pentane layers were combined and dried
over Na2SO4. The solvent was removed on a rotary eva-
porator to give 37.7 g colorless liquid (88.8%, >94% purity).
The product can be used for the next step without further
puri®cation. Distillation yielded adduct with bp 33±34 8C/
4 Torr. IR (®lm) 1648 cmÀ1; 1H NMR (CDCl3, 500 MHz) d
5.59 (s, 2H), 2.44 (m, 2H), 2.28 (m, 2H), 2.15 (d, J ˆ 17 Hz,
2H), 1.86 (m, 2H), 1.75 (m, 1H), 1.63 (m, 1H); 13C NMR
(CDCl3, 125.7 MHz) d 122.9 (t, J ˆ 2:3 Hz), 118.4 (dd,
J ˆ 306, 287 Hz), 34.3, 30.5 (dd, J ˆ 12, 8.6 Hz), 23.6 (d,
J ˆ 1:1 Hz), 23.5 (dd, J ˆ 4:9, 1.6 Hz); 19F NMR (CDCl3,
282.2 MHz) d À138.8 (d, J ˆ 156 Hz), À143.1 (d,
J ˆ 156 Hz). Anal. calcd. for C10H12F2: C 70.57; H 7.11;
F 22.32. Found: C 70.39; H 7.29; F 22.08.
colorless liquid, bp 38±39 8C/3 Torr. IR (®lm) 1567,
1448 cmÀ1; UV l
258 nm (e 9376). 1H NMR
cyclohexane
max
(CDCl3, 500 MHz) d 6.66 (dd, J ˆ 4:0, 1.5 Hz, 2H), 6.14
(dd, J ˆ 4:0, 1.5 Hz, 2H), 2.81 (m, 2H), 2.07 (m, 2H), 1.99
(m, 1H), 1.73 (m, 1H); 13C NMR (CDCl3, 125.7 MHz) d
127.4 (d, J ˆ 4:9 Hz), 123.1 (dd, J ˆ 31, 20 Hz), 120.7 (dd,
J ˆ 258, 238 Hz), 116.8 (dd, J ˆ 4:4, 2.9 Hz), 48.7, 32.7
(dd, J ˆ 4:0, 1.6 Hz); 19F NMR (CDCl3, 282.2 MHz) d
À116.2(d, J ˆ 188 Hz), À125.9 (d, J ˆ 188 Hz). Anal.
calcd. for C10H12F2: C 71.41; H 5.99; F 22.59. Found: C
71.58; H 6.19; F 22.30.
4.9. 1,1-Difluoro-2,2,3,3,-tetramethylcyclopropane
4.9.1. Part A: via thermolysis
A solution of 10,10-di¯uorobicyclo[4.3.1]deca-1,3,5-tri-
ene (0.588 g, 3.50 mmol) in 2,3-dimethyl-2-butene (1.48 g,
17.5 mmol) was sealed in a thick-walled glass tube, which
was immersed in an oil bath at 143 8C for 27 h [16]. The
cooled tube was then opened and the mixture was diluted
with 15 ml of freshly distilled triglyme. The solution was
cooled to À40 8C and Br2 (0.75 ml, 15 mmol) was added.

80
Y. He, D.M. Lemal / Journal of Fluorine Chemistry 119 (2003) 75±80
[4] Spartan, W. Hehre, Wavefunction Inc., 18401 Von Karman, Suite
370, Irvine, CA 92717.
After the resulting solution had been stirred for 20 min,
trans-stilbene (2.0 g, 11 mmol) was added to consume
residual bromine. Vacuum transfer of the mixture (0.06 Torr,
10 8C) gave the cyclopropane as a colorless liquid (0.42g,
[5] D.C. England, C.G. Krespan, J. Org. Chem. 33 (1968) 816.
[6] C.N. O'Callaghan, J. Chem. Soc. (C) (1971) 207.
[7] G. Fritsch, G. Zinner, M. Beimel, D. Mootz, H. Wunderlich, Arch.
Pharm. (Weinheim) 319 (1986) 70.
1
90%). H NMR (CDCl3, 300 MHz) d 1.11 (t, J ˆ 2:1 Hz);
19F NMR (CDCl3, 282.2 MHz) d À148.7.
[8] C.A. Jacobs, Ph.D. Dissertation, University of Pennsylvania, 1994.
È
È
[9] H.J. Scholl, Dissertation, Universitat Koln, 1969.
[10] W.R. Dolbier Jr., Acc. Chem. Res. 14 (1981) 195.
[11] T. Clark, G.W. Spitznagel, R. Klose, P.V.R. Schleyer, J. Am. Chem.
Soc. 106 (1984) 4412.
4.9.2. Part B: via photolysis
A solution of 10,10-di¯uorobicyclo[4.3.1]deca-1,3,5-tri-
ene (0.588 g, 3.50 mmol) and 2,3-dimethyl-2-butene
(1.48 g, 17.5 mmol) in 15 ml freshly distilled triglyme
was sealed in a quartz tube, which was then irradiated with
a 450 W Hanovia medium-pressure mercury arc for 55 h.
The work-up was exactly the same as above. Vacuum
transfer of the ®nal mixture (0.06 Torr, 108C) gave the
cyclopropane as a colorless liquid (0.36 g, 77%).
[12] M.J. Frisch, G.W. Trucks, H.B. Schlegel, G.E. Scuseria, M.A. Robb,
J.R. Cheeseman, V.G. Zakrzewski, J.A. Montgomery Jr., R.E.
Stratman, J.C. Burant, S. Dapprich, J.M. Millam, A.D. Daniels,
K.N. Kudin, M.C. Strain, O. Farkas, J. Tomasi, V. Barone, M. Cossi,
R. Cammi, B. Mennucci, C. Pomelli, C. Adamo, S. Clifford, J.
Ochterski, G.A. Petersson, P.Y. Ayala, Q. Cui, K. Morokuma, D.K.
Malick, A.D. Rabuck, K. Raghavachari, J.B. Foresman, J. Cioslows-
ki, J.V. Ortiz, B.B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I.
Komaromi, R. Gomperts, R.L. Martin, D.J. Fox, T. Keith, M.A. Al-
Laham, C.Y. Peng, A. Nanayakkara, C. Gonzalez, M. Challacombe,
P.M.W. Gill, B. Johnson, W. Chen, M.W. Wong, J.L. Andres, M.
Head-Gordon, E.S. Replogle, J.A. Pople, Gaussian 98 Revision A.5,
Gaussian Inc., Pittsburgh, PA, 1998.
Acknowledgements
[13] H. GuÈnther, T. Keller, Chem. Ber. 103 (1970) 3231.
[14] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identi-
fication of Organic Compounds, 5th ed., Wiley, New York, 1991.
[15] A.C. Goren, D. Hrovat, M. Seefelder, H. Quast, W.T. Borden, J. Am.
Chem. Soc. 124 (2002) 3469, and references therein.
The authors thank Prof. E. Vogel and Prof. W.P. Dailey for
copies of the Ph.D. theses of R. Scholl and C.A. Jacobs,
respectively. We also thank Prof. M.S. Platz and E. Tippman
for valuable information, and Prof. Robert Ditch®eld for
very helpful advice regarding the computations. The
National Science Foundation is gratefully acknowledged
for support of this research.
[16] G.A. Wheaton, D.J. Burton, J. Fluorine Chem. 9 (1977) 25.
[17] M. Kamel, W. Kimpenhaus, J. Buddrus, Chem. Ber. 109 (1976) 2351.
[18] O.Z. Glemser, Naturforsch 36 (b) (1981) 1375.
[19] Y. Bessard, U. MuÈller, M. Schlosser, Tetrahedron 46 (1990) 5213.
[20] M.A. Battiste, R.G. Posey, J. Fluorine Chem. 102 (2000) 285.
[21] D. Seyferth, H. Dertouzos, R. Suzuki, J.Y.-P. Mui, J. Org. Chem. 32
(1967) 2980.
References
[22] M.S. Platz, private communication.
[23] R.B. Woodward, R. Hoffmann, Angew. Chem. Int. Ed. Engl. 8
(1969) 781.
[1] D.L.S. Brahms, W.P. Dailey, Chem. Rev. 96 (1996) 1585.
[2] Q.-Y. Chen, J. Fluorine Chem. 72 (1995) 241;
R.N. Buravtsev, Yu.A. Kolbanovskii, Dokl. Akad. Nauk. 358
(1998) 57;
[24] M. Rosenberger, P. Yates, Tetrahedron Lett. (1964) 2285.
[25] T.B. Patrick, J.J. Scheibel, G.L. Cantrell, J. Org. Chem. 46 (1981)
3917.
N.V. Kirii, S.V. Pazenok, Yu.L. Yagupolskii, D. Naumann, W.
Turra, Russ. J. Org. Chem. (Engl. transl. Zh. Org. Khim.) 37 (2001)
207.
[26] V.H. Rawal, R.J. Jones, M.P. Cava, J. Org. Chem. 52 (1987) 19.
[27] N. Chen, M. Jones, W.R. White, M.S. Platz, J. Am. Chem. Soc. 113
(1991) 4981.
[3] G. Maier, M. Heider, C. Sierakowski, Tetrahedron Lett. 32(1991)
1961.
[28] D.J. Burton, D.G. Naae, J. Am. Chem. Soc. 95 (1973) 8467.