Mechanisms for the alkaline hydrolysis of dibromodifluoromethane-alkene adducts to α,β-unsaturated carboxylates

Article abstract of DOI:10.1016/S0022-1139(99)00232-8

Alternative mechanisms for the title reactions are probed. The previously-reported pathway involving double dehydrobromination to a difluorodiene is operative in at least one case, but this route is specifically excluded for systems that cannot dehydrobrominate to dienes yet still yield carboxylates upon alkaline hydrolysis. Although S_N2, S_N1, S_RN1, and monoelimination-addition processes appear formally possible, an S_N2′ mechanism is implicated by studies on model compounds.

Full text of DOI:10.1016/S0022-1139(99)00232-8

Journal of Fluorine Chemistry 102 (2000) 3±9
Mechanisms for the alkaline hydrolysis of dibromodi¯uoromethane-alkene
adducts to a,b-unsaturated carboxylates
*
Seth Elsheimer , JoAnne L. Swanson, Javier Gonzalez
Department of Chemistry, University of Central Florida, PO Box 162366, Orlando, FL 32816-2366, USA
Received 24 March 1999; received in revised form 5 May 1999; accepted 18 June 1999
Dedicated to Professor Paul Tarrant on the occasion of his 85th birthday
Abstract
Alternative mechanisms for the title reactions are probed. The previously-reported pathway involving double dehydrobromination to a
di¯uorodiene is operative in at least one case, but this route is speci®cally excluded for systems that cannot dehydrobrominate to dienes yet
still yield carboxylates upon alkaline hydrolysis. Although S 2, S 1, SRN1, and monoelimination±addition processes appear formally
N
N
0
possible, an S
N
2 mechanism is implicated by studies on model compounds. # 2000 Elsevier Science S.A. All rights reserved.
Keywords: Dehydrobromination; Hydrolysis; Halomethane
1
. Introduction
Bromo¯uoro and chloro¯uoro organics have historically
dehydro¯uorinates to acyl ¯uoride 9 and ultimately 10. This
mechanism was supported by the isolation of intermediates
6, 7, and 9 during the reaction. Each of these were subse-
quently converted quantitatively to 10 under the reaction
conditions [6].
Alternatives to the Scheme 2 pathway are implied by the
observed conversion of norbornene adducts 11 into 16 [7].
Following a path analogous to Scheme 2, conversion of 11 to
carboxylic acid 16 would require the anti-Bredt intermediate
14. A previous report that 1-norbornene (bicyclo[2,2,1]hept-
1-ene) was observed indirectly in a trapping experiment [12]
suggests that 14 may be a short-lived intermediate
eventhough it cannot be isolated (Scheme 3).
been employed in a variety of industrial applications [1±3],
and some of these ubiquitous compounds are signi®cant or
persistent environmental hazards [4,5]. A detoxi®cation or
remediation strategy based on simple alkaline hydrolysis
may exist provided the scope and limitations of the
process are understood and predictable. Our interest in
the chemistry of dibromodi¯uoromethane-alkane adducts
1 [6±10] prompted a mechanistic study of their alkaline
hydrolysis.
Compounds 1 are generally accessible via radical addition
of dibromodi¯uoromethane to alkenes [11]. Adducts 1 were
found to be cleanly converted to a,b-unsaturated carboxylic
acids 2 when treated with aqueous hydroxide followed by
acidic work-up [6,7]. The generality of this process was
shown for a variety of terminal, internal, and cyclic alkenes
that were similarly converted to the corresponding halo-
methane adducts then carboxylic acids [7] (Scheme 1).
Experiments with the cyclohexene adducts 3 and 4 sug-
gest the mechanism shown in Scheme 2 which involves
double dehydrobromination to the intermediate di¯uoro-
diene 7. Base-catalyzed conjugate addition of water to 7
produces an unstable di¯uoroalcohol 8 that spontaneously
Conversion of 12 to 15 by an S 2 or S 1 mechanism on a
N
N
¯uoroalkyl center appears unlikely. Such reactions are gen-
erally not observed [13] although formal substitution on
1
¯uoroalkyl carbons can arise from other mechanism.
0
Also conceivable is the S 2 process on 13 to yield 15.
N
In that case the observed near-quantitative conversion of
1
Bromophilic reactions on dibromodifluoromethane or 1,2-dibromo-
1,1,2,2-tetrafluoroethane produce carbanions that subsequently eliminate
an alpha or beta bromide to produce difluorocarbene or tetrafluoroethene,
respectively. Nucleophilic additions to these intermediates yield products
that formally appear to result from nucleophilic substitution on a
fluoroalkyl carbon. These pathways, however, require geminal or vicinal
dibromides, respectively. The 1,3 relationship of the bromines on adducts 1
precludes an analogous route between 11 and 15 [14].
*
Corresponding author.
0022-1139/00/$ ± see front matter # 2000 Elsevier Science S.A. All rights reserved.
PII: S 0 0 2 2 - 1 1 3 9 ( 9 9 ) 0 0 2 3 2 - 8

4
S. Elsheimer et al. / Journal of Fluorine Chemistry 102 (2000) 3±9
Scheme 1.
Scheme 2.
Scheme 3.
11 to 16 would require either high regioselectivity for
the initial dehydrobromination of 11 to 13 or possibly
®nally dehydrobrominated to produce 16 upon acidic work-
2
up.
A single electron transfer chain process known as SRN
the isomerization of 12 to 13 under the reaction condi-
0
1
tions. Products formally resulting from S 2 attack on 1-
N
has been observed for some per¯uoroalkyl iodides [19,20].
Such a mechanism could conceivably be operative here as
shown in Scheme 5.
bromo-1,1-di¯uoro-2-alkenes have recently been reported
[
15].
A simple monoelimination±addition mechanism is also
possible and has been observed in other R CHCF X systems
2
2
[
16] (Scheme 4). Initial dehydrobromination of 11 to 13
2. Results and discussion
could be followed by hydroxide-catalyzed addition of water
to the di¯uoroalkene to yield a di¯uoroalcohol that decom-
poses via an acyl ¯uoride to a bromocarboxylate that is
The adduct 17 does not undergo the same alkaline
hydrolysis observed for nearly all other halomethane-alkene
adducts studied [6,7]. Rather than the completely dehalo-
2
Initial dehydrofluorination (vs. dehydrobromination) of 11 would give
-bromo-3-(bromofluoromethylene)bicyclo[2,2,1]heptane which
2
could, in turn, add water yielding an alpha-bromofluoroalcohol, etc.
This variation of the monoelimination±addition pathway shown in
Scheme 4 has not been rigorously ruled out; however, we note that the
greater nucleofugicity of bromide over fluoride favors dehydrobromi-
nation. In bimolecular reactions, HBr elimination is generally several
orders of magnitude faster than HF elimination (see [17]). For
2 2
example, at 308C eliminations of HX from PhCH CH X in ethoxide/
ethanol have relative rates for X ˆ F and Br of 1 and 4300,
Scheme 4.
respectively (see [18]).

S. Elsheimer et al. / Journal of Fluorine Chemistry 102 (2000) 3±9
5
Scheme 8.
[
21]. The observation of 22 provides indirect evidence for
2
conversion of 11 to 16 thus leaving the S 2 , monoelimina-
1 and further argues against a diene intermediate in the
0
N
tion±addition, and SRN1 pathways for further consideration.
Monodehydrobromination of 11a with DBU in ether and
re¯uxing for 24 h yielded exo-13, endo-13, and 12, in a
6
5:15:20 ratio, respectively. A low temperature reaction
yielded the kinetic product endo-13 which was found to
equilibrate to the same 65:15:20 mixture at 1008C with a
half-life of approximately 5 min. A literature precedent
exists for isomerization of the corresponding bromodichloro
analogs of 13 between endo and exo [22] but not to
Scheme 5.
3
trihalomethyl compounds such as 12. Allylic shifts of
bromine have been observed in similar bromodi¯uoro sys-
tems, and an analogous equilibration of 3-bromo-1,1-
di¯uoropropene and 3-bromo-3,3-di¯uoropropene has been
Scheme 6.
reported [24] (Scheme 8). This isomerization may be
0
genated carboxylic acid 19, the monodehydrobromination
product 18 was observed exclusively (Scheme 6). As
expected, this is evidence against a nucleophilic substitution
explained as resulting from S 2 attack of bromide on either
N
isomer. This can be extrapolated to isomers 12 and 13 and
0
may pertain to the possibility of S 2 attack by hydroxide on
N
at the trihalomethyl carbon but is consistent with the diene,
0
13 to yield 15 and ultimately 16.
S 2 , or monoelimination±addition pathways described
N
An alkaline hydrolysis experiment was attempted with 1-
bromo-1,1-di¯uorononane 23. As previously shown, the
S 1 and S 2 mechanisms are not likely to occur nor could
above each of which require the carbon bearing the triha-
lomethyl group to be an alkene carbon at some point. This is
not possible for molecules such as 17 in which the triha-
lomethyl group is attached to a quaternary carbon.
N
N
0
23 react via an S 2 mechanism. If 23 underwent alkaline
N
hydrolysis to yield carboxylate, that would be evidence for
monoelimination±addition, or SRN1 pathway (Scheme 9).
After 29.5 h of heating in 16% aqueous potassium hydro-
xide, starting material 23 was recovered unchanged. This
may be evidence that in order for a bromodi¯uoromethyl
compound to hydrolyze under these conditions, a second
bromine must be present in the gamma position. The gamma
bromine may increase the acidity of the hydrogen on the
beta carbon to allow for beta-gamma dehydrobromination to
yield the previously-observed [6,10] monoelimination pro-
ducts such as 6 which subsequently rearrange to 3-bromo-
1,1-di¯uoroalkenes such as 5.
We had rejected the diene pathway for the conversion of
1 to 16 based on the predicted instability of bridgehead
1
ole®n intermediate 14. Additional experimental evidence
for the existence of a non-diene path was found through
compound 20, which provides a rough structural model for
1
1. Model compound 20 undergoes the usual alkaline
hydrolysis to 21 (Scheme 7) even though the methyl sub-
stituent on bridgehead carbon-1 precludes a diene inter-
mediate analogous to 14. Although stable in boiling alkali,
compounds such as 21 containing the 1,4-dihydro-1,4-
dimethyl-1,4-epoxynaphthalene moiety undergo facile
acid-catalyzed rearrangement to 1,4-dimethyl-2-naphthols
Di¯uoromethylene cyclohexane 24 was prepared and
used as a model for the proposed intermediate 13 except
that, lacking a bromine on carbon-3, it cannot undergo an
0
S 2 process. Compound 24 also serves to test the ``addi-
N
tion'' phase of the elimination±addition pathway. After
being treated under our standard alkaline hydrolysis con-
ditions, 24 was recovered unchanged and cyclohexane
carboxylic acid 25 was not observed (Scheme 10).
To test the possibility of a single electron transfer, SRN
pathway operating in the conversion of 11a to 16, the
1
3
The significant amount of trihalomethyl isomer 12 in addition to exo-
and endo-13 in this equilibrium mixture differs from what was
observed for the bromodichloro analogs [22] and may reflect the
stabilization reported for gem-difluoro substituents on saturated vs.
unsaturated carbons [23].
Scheme 7.

6
S. Elsheimer et al. / Journal of Fluorine Chemistry 102 (2000) 3±9
Scheme 9.
Scheme 10.
4
. Experimental section
Except where otherwise speci®ed, all reagents were
commercial samples obtained from Aldrich Chemical Com-
pany and used without puri®cation. Analytical and prepara-
tive GC was carried out on a Gow-Mac 69-350 gas
chromatograph using 8 ft  0.25 inch copper columns
packed with 10% SE 30 or 10% OV-210 on Chromosorb
WHP (He ¯ow ˆ 60 ml/min). All NMR spectra were
Scheme 11.
alkaline hydrolysis was carried out both with and without
the inhibitor p-dinitrobenzene. After 25 h at 1368C, no
unreacted starting material or simple monoelimination pro-
ducts were observed in the inhibited reaction. In addition to
the expected acid 16, a signi®cant amount of alcohol side
product 26 was detected (Scheme 11). A similar result was
obtained in the absence of p-dinitrobenzene although the
yields of 16 and 26 were slightly higher in that case [7]. This
is attributable to the dif®culty associated with the work-up
of the small scale reaction that contained p-dinitrobenzene.
obtained in CDCl solution on Varian Gemini 200 spectro-
3
1
13
meter operating at 200 MHz for H and 50 MHz for C.
Infrared spectra were obtained as capillary ®lms of neat
liquids unless otherwise noted and were recorded on a
Perkin-Elmer 1600 series FTIR spectrophotometer. Ele-
mental analyses were performed by Atllantic Microlab
(Norcross, GA) except for those of 12 and 13 which were
performed by Robertson Laboratories (Madison, NJ).
3
. Conclusions
These observations support an S 2 mechanism for the
4.1. Preparation of
1,3-dibromo-1,1-difluoro-2,2,3-trimethylbutane (17)
0
N
conversion of 11 to 16. This general pathway is likely
operative in the alkaline hydrolysis of other halomethane-
alkene adducts although it is not the sole path for all cases
since diene 7 was observed for the conversion of 3 and/or 4
to 10 [6]. Although we have not rigorously ruled out
allylically-stabilized S 1, S 2, or SRN1 paths for all cases,
our results, and those of others, argue against those mechan-
isms. The monelimination±addition pathway was dis-
counted by the attempted alkaline hydrolyses of 23 and
Dibromodi¯uoromethane was added to 2,3-dimethyl-2-
butene according to our adaptation [8] of a literature method
[11]. A 20-ml, screw-capped reaction tube equipped with
poly(tetra¯uoroethylene)-taped threads and a magnetic stir
bar was charged with 2,3-dimethyl-2-butene (2.99 g,
35.5 mmol), t-butyl alcohol (2.44 g), ethanolamine
(1.07 g), copper(I) chloride (0.05 g, 0.5 mmol), and dibro-
modi¯uoromethane (15.12 g, 72 mmol). The vessel was
tightly capped and the reaction mixture was stirred and
heated over an oil bath at 758C for 16 h during which the
mixture changed from clear blue to dark brown. A resinous
layer also appeared. The mixture was cooled to ambient
temperature and diluted with 10 ml of cyclohexane where-
upon it became cloudy. The cyclohexane solution was
separated and the residue was extracted twice with 5-ml
portions of cyclohexane. The cyclohexane extracts were
N
N
2
4 which were unreactive and were not converted into
carboxylates under these conditions. Bromodi¯uoromethyl
groups attached to quaternary carbons are unreactive to
alkaline hydrolysis. Nucleophilic substitution by hydroxide
does not occur on such systems by any mechanism. Bro-
modi¯uoromethyl groups are generally unreactive towards
alkaline hydrolysis unless they are activated by adjacent
unsaturation.

S. Elsheimer et al. / Journal of Fluorine Chemistry 102 (2000) 3±9
7
combined and ®ltered through a 20 cc bed of 70±230 mesh
silica gel in a 4 cm diameter B uÈ chner funnel equipped with
Whatmann No. 1 ®lter paper yielding a clear, colorless
solution which was washed three times with deionized
water, dried over anhydrous calcium chloride, ®ltered,
and rotary evaporated which left 0.82 g (7.9%) of 17 that
out at 75±858C for 17 h. Half the starting material (1,4-
dihydro-1,4-dimethyl-1,4-epoxynaphthalene) was recov-
ered by vacuum distillation. Compound 20 which was
puri®ed by vacuum distillation (1108C, 2 torr) or by pre-
2
parative TLC using 20 Â 20 cm , 500 mm silica coated
plates and 5:4 CH Cl :hexane eluent. Yield ˆ 11% based
2
2
1
1
appeared >95% pure by GC and H NMR analysis. The
product was combined with that from another run at the
on consumed starting material. HNMR: d 7.28 (mult, 2H),
7.20 (mult, 2H) 4.17 (dd, J ˆ 4.7 and 1.3 Hz, 1H), 2.68
(ddd, J ˆ 19.4, 6.6, 4.7 Hz, 1H), 1.90 (mult, 3H), 1.85 (s,
1
same scale and vacuum distilled at 778C (14 Torr). H NMR
1
3
13
d 1.97 (t, JFH ˆ 1.9 Hz, 6H), 1.52 (s, 6H); C NMR d 129.2
3H). CNMR: d 147.1, 144.9, 128.0, 127.0, 122.0, 121.8
(
3
1
t, JFC ˆ 318.2 Hz, CF Br), 72.4, 54.0 (t, J ˆ 16.1 Hz),
(dd, JFC ˆ 312 and 307 Hz, CF Br), 117.5, 87.1, 86.0, 66.1
2
FC
2
2.7 (t, J ˆ 3.5 Hz), 24.5. IR: 2997, 2953, 1462, 1389,
FC
(t, JFC ˆ 21 Hz, C2), 52.0, 15.4, 15.0. IR: 3086, 2990, 2932,
�
1
� 1
380, 1222, 1102 (st), 897 (st), 879 (st), 546 cm
.
2865, 1696, 1465, 1389, 1312, 1245, 1091,756, 698 cm
.
Elemental analysis was not performed due to the
observed decomposition of 17 upon standing at room
temperature.
Analysis calculated for C H Br F O: C, 40.87; H 3.16.
13 12 2 2
Found: C, 41.37; H 3.21.
4.4. Alkaline hydrolysis of 20. Preparation of
3-hydroxy-1,4-dimethyl-2-naphthalenecarboxylic acid (22)
4
4
.2. Attempted alkaline hydrolysis of 17. Preparation of
-bromo-4,4-difluoro-2,3,3-trimethyl-1-butene (18)
The procedure described for the hydrolysis of 17 to 18
was followed except where noted and using 1.88 g
(4.9 mmol) 20 as the starting material. The reaction was
run for 9 h at 1358C. After cooling the reaction mixture was
diluted with a small amount of water and then extracted four
times with 5-ml portions of methylene chloride. The organic
portion was set aside and the aqueous phase was acidi®ed
with 6 M HCl. A milky white precipitate and an yellow oil is
formed at the bottom of ¯ask. The mixture was extracted
four times with 5-ml portions of CH Cl . The combined
A 20-ml, screw-capped reaction tube equipped with
poly(tetra¯uoroethylene)-taped threads and a magnetic stir
bar was charged with 1,3-dibromo-1,1-di¯uoro-2,2,3-tri-
methylbutane 17 (0.58 g, 1.97 mmol), potassium hydroxide
(
0.55 g, 9.8 mmol) and 3 ml of water was stirred and heated
over a 1818C oil bath for 6 h. The cooled reaction mixture
was extracted twice with 1-ml portions of dichloromethane
and the combined organic extracts were dried over anhy-
drous sodium sulfate and condensed by rotary evaporation
2
2
4
1
to yield 0.35 g of alkene 18. (83%). H NMR: d 5.1 (two
partially resolved singlets at 5.06 and 5.08 ppm), 1.9 (s, 3H),
extracts were washed three times with water, dried over
anhydrous sodium sulfate, and rotary evaporated leaving a
brown solid which was not soluble in water but did dissolve
1
3
1
.35 (s, 6H); C NMR: d 145, 129.2 (t, JFC ˆ 311 Hz,
2 FC
CF Br), 116, 51.3 (t, J ˆ 19 Hz, C3), 23, 21; IR: 2992,
in 5% aqueous solutions of NaOH or NaHCO (efferves-
3
2
8
925, 2958, 2855, 1639 (w), 1451, 1384, 1261, 1104, 1020,
84, 808 cm
cence). After dissolving the solid in aqueous NaHCO the
3
�
1
.
solution was re-acidi®ed with 6 M HCl and became cloudy.
After chilling for several hours a light yellow solid appeared
which was identi®ed as 22 (0.28 g, 27% yield). During
the m.p. determination the compound decomposed at
4
1
.3. Endo-2-bromo-exo-3-(bromodifluoromethyl)-
,4-dimethyl-1,2,3,4-tetrahydro-1,4-epoxynaphthalene (20)
1
residue. NMR spectra were obtained in acetone-d solution.
H NMR: d 8.16 (d with unresolved ®ne structure,
83±1858C in an open capillary leaving a dark brown
Compound 20 was prepared by the addition of dibromo-
di¯uoromethane to 1,4-dihydro-1,4-dimethyl-1,4-epoxy-
6
1
5
naphthalene according to the method described above
for the synthesis of 17. The synthesis of 20 was carried
J ˆ 8.8 Hz, 1H), 7.96 (d with unresolved ®ne structure,
J ˆ 8.4 Hz, 1H) 7.58 (ddd, J ˆ 8.4, 6.8, and 1.4 Hz, 1H),
7
.41 (ddd, J ˆ 8.4, 6.8, and 1.4 Hz, 1H), 2.90 (s, 3H) 2.51
13
4
1
The H NMR chemical shifts we observed do not match a literature
(s, 3H). C NMR: d 173.0, 152.1, 136.5, 136.2, 128.7,
128.4, 126.7, 124.3, 121.0, 115.7, 17.5, 11.0. IR: (KBr
report: d 5.78, 1.05 (s, 3H), 1.956 (s, 6H) but do agree with those of the
analogous iodo compound 4,4-difluoro-2,3,3-trimethyl-4-iodo-1-butene
pellet) 3324±2605 (br), 1654 (st, sharp), 1512, 1441,
� 1
[
25,26].
5
1376, 1294, 180, 1098, 885, 826, 738, 635 cm . Analysis
calculated for C H O : C, 72.21; H, 5.59. Found: C,
The precursor to compound 20 was 1,4-dihydro-1,4-dimethyl-1,4-
epoxynaphthalene which was prepared via a Diels±Alder reaction of
benzyne and 2,5-dimethylfuran according to a published procedure
13 12 3
7
1.62; H, 5.59.
[
27] and has properties consistent with those previously reported [28].
13
13
CNMR and IR spectral data were not previously reported:
NMR: d 152.1 (C9, C10), 146.2 (C6, C7), 118.0 (C1, C4), 88.0 (C2,
C3), 15.0 (CH ); IR: 3072, 2977, 1453, 1384, 1307, 1236, 1141, 908 (st,
sharp), 859, 743 (st,b) 693, 648, 603, 509 cm . H NMR: d 7.0 (mult,
C
4.5. Monodehydrobromination of 11a with DBU.
Preparation and equilibration of 12, exo-13, and endo-13
3
�
1 1
A 100-ml, round-bottomed, three-necked ¯ask was ®tted
with a re¯ux condenser, thermometer, and magnetic stirrer.
4
H), 6.7 (s, 2H), 1.84 (s, 6H). The literature value for the 6H singlet is t
.66 (d 1.34) (see [28]).
8

8
S. Elsheimer et al. / Journal of Fluorine Chemistry 102 (2000) 3±9
The system was ¯ushed with dry nitrogen and charged with
-exo-(bromodi¯uoromethyl)-3-endo-bromobicyclo[2,2,1]-
heptane, 11a, (6.08 g, 20.0 mmol), DBU (3.35 g, 22 mmol,
.1 equiv.), and 30 ml of diethyl ether. The mixture was
4.8. Alkaline hydrolysis of 11a in the presence of
p-dinitrobenzene
2
1
A 20-ml, screw-cap reaction vessel was charged with
1.09 g (3.59 mmol) of 11a [7], 1.37 g (>21 mmol) KOH,
7 ml of water, and 0.64 g (3.81 mmol) of p-dinitrobenzene.
The reaction vessel was heated in an oil bath while stirring
for 13 h at 1368C. After cooling, the contents were diluted
with 7 ml of water and rinsed into a separatory funnel with
7 ml of methylene chloride. The entire mixture was acidi®ed
to pH 2 with 6 M HCl. The organic layer was drawn off and
the aqueous phase was extracted with 5 ml of methylene
chloride which was combined with the organic phase. The
combined organic extracts were washed twice with 10 ml
portions of deionized water and then the carboxylic acid 16
was converted to the salt and extracted out of the methylene
chloride solution with 15 ml of 5% aqueous sodium bicar-
bonate. The methylene chloride layer was dried over anhy-
drous calcium chloride and condensed by rotary evaporation
to yield 0.37 g of reddish-brown oil that was identi®ed as
crude exo-3-(di¯uoromethylene)bicyclo[2,2,1]heptan-2-ol,
stirred and re¯uxed for 24 h. The resulting reaction mixture
was diluted with 20 ml of water and the ether layer was
separated, washed with 20 ml of 10% aqueous NaCl, dried
over anhydrous magnesium sulfate and the ether was
removed by rotary evaporation. Vacuum distillation of
the residue yielded a colorless oil in 69% based on unrec-
overed starting material. (b.p. 62±668C, 5 torr). In the course
of isolating the product by distillation, it rearranged to a
mixture of isomers. The isomerization, which was found to
have a half-life of 5 min at 1008C, led to a 65:15:20 mixture
of exo-13, endo-13, and 12, respectively. The kinetic pro-
duct obtained at low temperatures was endo-13. The proton
on carbon-3 for this compound (bearing a secondary carbon
bromine bond) appeared at 4.5 ppm as a broad singlet. By
1
3
C NMR, both di¯uoromethylene carbons appeared as
unresolved doublets of doublets near 155 ppm (JFC approxi-
mately 300 Hz). In addition, the ole®nic carbon appeared as
a small triplet (J ˆ 10 Hz) at 137 ppm. For the mixture,
analysis calculated for C H BrF : C, 43.08; H 4.08. Found:
1
13
26. H and C NMR analysis showed that the oil did not
contain any unreacted 11a or the monoelimination products
12 or 13.
8
9
2
C, 43.11; H, 4.17.
The aqueous bicarbonate extract was slowly acidi®ed
with 6 M HCl (caution! effervescence) and a small amount
of ®nely divided white precipitate was observed. The mix-
ture was extracted twice with 4-ml portions of methylene
chloride. The combined extracts were dried over anhydrous
calcium chloride and condensed by rotary evaporation to
yield 0.32 g of residue identi®ed as the carboxylic acid 16
contaminated by some p-dinitrobenzene that was carried
4
.6. Attempted alkaline hydrolysis of 23
The procedure described for the conversion of 17 to 18
was followed except where noted. The reaction vessel was
charged with 0.50 g (2.1 mmol) of 23 [8], and 0.58 g KOH
dissolved in 3 ml of water. The reaction mixture was stirred
and heated over a 1568C oil bath for 30 h, cooled to ambient
temperature and extracted with three 2-ml portions of
methylene chloride. The combined extracts dried over
anhydrous sodium sulfate and concentrated by rotary eva-
poration to yield 0.48 g (83%) of recovered starting mate-
rial.
1
through the work-up. The yield of 16 was 51% by H NMR
7
integration.
Acknowledgements
We gratefully acknowledge the donors of the Petroleum
Research Fund, administered by the American Chemical
Society, for partial support (21424-B1), Research Corpora-
tion (C-2666) for partial support, and the National Science
Foundation for purchase of our NMR spectrometer (No.
CHE-8608881). Thanks to G.F. Olde Jr., and J. Gillin for
literature search assistance.
4
(
.7. Attempted alkaline hydrolysis of
difluoromethylene)cyclohexane, 24
The procedure described for the conversion of 17 to 18
was followed except where noted. The reaction tube was
6
charged with 0.59 g (4.5 mmol) of 24, 1.20 g (21 mmol)
KOH and 2.1 g of water. After heating at 1338C for 18 h the
mixture was cooled to ambient temperature and worked up
to yield 0.32 g (54%) recovered 24. The aqueous phase was
acidi®ed and extracted as usual but no carboxylic acid was
isolated or observed.
7
When this reaction is carried out in the absence of p-dinitrobenzene
the work-up is much simpler and the products can be more easily isolated
13
and characterized [30]. Compound 26:
C NMR: d 153.3 (t,
JFC ˆ 287 Hz, CF ), 99 (dd, J ˆ 19.2 and 17.5 Hz, C3), 74.0 (d,
2
FC
1
J
FC ˆ 6 Hz, COH) 44.2, 36.8, 36.0, 28.7, 23.6. H NMR: d 4.23 (s, 1H,
on C2), 2.92 (s, 1H, on C1), 2.33 (s, 1H, on C4), 2.04 (s, 1H, OH), 1.82
(d of mult, J ˆ 9.6 Hz, 1H), 1.58 (mult, 2H), 1.26 (mult, 2H), 1.11 (d of
mult, J ˆ 9.6 Hz, 1H). IR: 3353 (br, OH), 2952, 2866, 1762 (st,
6
1
Compound 24 was prepared in 25% yield from cyclohexanone. H
13
NMR: d 2.05 (br s, 4H), 1.53 (br s, 6H); C NMR: d 151 (t,
FC ˆ 280 Hz, CF ), 88.1 (JFC ˆ 18.7 Hz), 26.5, 26.2, 24.4. IR: 2939,
859, 1759 (C=CF
�
1
J
2
C=CF ), 1448, 1248, 1065, 1048, 1018, 811, 774, 744 cm . Spectral
2
�
1
2
2
), 1453, 1268, 1213, 738 cm [29].
data for 16 agree with those cited previously [7].

S. Elsheimer et al. / Journal of Fluorine Chemistry 102 (2000) 3±9
9
[
14] C. Wakselman, in: M. Hudlicky, A. Pavlath (Eds.), The Chemistry of
Organic Fluorine Compounds II, American Chemical Society,
Washington, DC, 1995, p. 446.
References
[
[
1] R.E. Banks, Organofluorine Compounds and their Industrial
Applications, Ellis Horwood, Chichester, UK, 1979.
2] R.L. Powell, in: M. Hudlicky, A. Pavlath (Eds.), The Chemistry of
Organic Fluorine Compounds II, American Chemical Society,
Washington, DC, 1995, p. 1089.
[15] F. Tellier, J.-M. Duffault, M. Baudry, R. Sauv eà tre, J. Fluorine Chem.
91 (1996) 79.
[16] W.H. Gumprecht, in: M. Hudlicky, A. Pavlath (Eds.), The Chemistry
of Organic Fluorine Compounds II, American Chemical Society,
Washington, DC, 1995, p. 422.
[
[
3] M.L. Robin, in: M. Hudlicky, A. Pavlath (Eds.), The Chemistry of
Organic Fluorine Compounds II, American Chemical Society,
Washington, DC, 1995, p. 1099.
[17] S. Matsubara, H. Matsuda, T. Hamatani, M. Schlosser, Tetrahedron
44 (1988) 2855.
[18] C.H. DePuy, C.A. Bishop, J. Am. Chem. Soc. 82 (1960) 2535.
[19] R.A. Rossi, A.B. Pierini, S.M. Palacios, Advances in Free Radical
Chemistry, JAI Press, 1990, pp. 193±252.
4] N. Suprenant, T. Nunno, M. Kravett, M. Breton, Halogenated organic
containing wastes: treatment technologies, Noyes Data Corp., Park
Ridge, NJ, 1988.
[20] B. Bigot, D. Roux, L. Salem, J. Am. Chem. Soc. 103 (1981) 5271.
[21] E. Wolthius, B. Bosenbroek, G. Dewall, E. Geels, A. Leegwater, J.
Org. Chem. 28 (1963) 148.
[
[
5] M.J. Molina, F.S. Rowland, Nature 810 (1974) 249.
6] S. Elsheimer, M. Michael, A. Landavazo, D.K. Slattery, J. Weeks, J.
Org. Chem. 53 (1988) 6151.
[22] E. Tobler, D.J. Foster, J. Org. Chem. 29 (1964) 2839.
[23] W.R. Dolbier Jr., K.S. Medinger, A. Greenberg, J.F. Liebman,
Tetrahedron 38 (1982) 2415.
[
[
[
7] S. Elsheimer, D.K. Slattery, M. Michael, J. Weeks, K. Topoleski, J.
Org. Chem. 54 (1989) 3992.
8] J. Gonzalez, C.J. Foti, S. Elsheimer, J. Org. Chem. 56 (1991)
[24] D. Seyferth, R.M. Simon, D.J. Sepelak, H.A. Klein, J. Am. Chem.
Soc. 105 (1983) 4634.
4
323.
9] J. Gonzalez, M.J. Foti, S. Elsheimer, Organic Syntheses 72 (1995)
25.
10] S. Elsheimer, C.J. Foti, M.D. Bartberger, J. Org. Chem. 61 (1996)
252.
[25] C. Hu, F. Qing, C. Shen, J. Chem. Soc. Perkin Trans. I (1993) 335.
[26] S. Elsheimer, W.R. Dolbier Jr., M. Murla, K. Seppelt, G. Paprott, J.
Org. Chem. 49 (1984) 205.
2
[
6
[27] L.F. Fieser, M.J. Haddadin, Can. J. Chem. 43 (1965) 1599.
[28] M.S. Newman, H.M. Dali, W.M. Hung, J. Org. Chem. 40 (1975) 262
and references therein.
[
[
[
11] D.J. Burton, L.J. Kehoe, J. Org. Chem. 35 (1970) 1339.
12] R. Keese, E.P. Krebs, Angew. Chem. Int. Ed. Eng. 10 (1971) 262.
13] R.D. Chambers, Fluorine in Organic Chemistry, Wiley/Interscience,
London, 1973, p. 99.
[29] D.J. Burton, D.G. Naae, Synth. Comm. 3 (1973) 197.
[30] D.K. Slattery, M.S. Thesis, Univ. Central Florida, 1989.