Syntheses of selectively fluorinated cyclodecenones: The first deployment of the neutral oxy-Cope rearrangement in organofluorine chemistry

Article abstract of DOI:10.1039/b311261f

Melallated haloalkenes were used to open epoxides in moderate to good yield. The homoallylic alcohols obtained underwent Swern oxidation to afford three γ,γ-difluorinated β,γ-enones, which reacted with either vinyllithium, 2-lithio-2H-dihydropyran or another metallated haloalkene to afford substituted trans-1,2-divinylcyclohexanols of different degrees of stability. These intermediates underwent neutral thermal oxy-Cope rearrangements when heated in xylene in Ace tubes. The first-formed enols ketonised without loss of HF to afford a range of cyclodecenones in moderate to good yield; X-ray crystallography was used extensively for product characterisation. All substrates rearranged more rapidly than a cisltrans mixture of 1,2-divinylcyclohexanols.

Full text of DOI:10.1039/b311261f

View Article Online / Journal Homepage / Table of Contents for this issue
Syntheses of selectively fluorinated cyclodecenones:
the first deployment of the neutral oxy-Cope rearrangement in
organofluorine chemistry
Gianluca DiMartino,a Michael B. Hursthouse,b Mark E. Light,b Jonathan M. Percy,*c
Neil S. Spencera and Malcolm Tolleya
a School of Chemical Sciences, University of Birmingham, Edgbaston, Birmingham,
UK B15 2TT
b EPSRC National X-Ray Crystallography Service, Department of Chemistry,
University of Southampton, Highfield, Southampton, UK SO17 1BJ
c Department of Chemistry, University of Leicester, University Road, Leicester, UK LE1 7RH.
E-mail: jmp29@le.ac.uk
Received 16th September 2003, Accepted 30th October 2003
First published as an Advance Article on the web 17th November 2003
Metallated haloalkenes were used to open epoxides in moderate to good yield. The homoallylic alcohols obtained
underwent Swern oxidation to afford three γ,γ-difluorinated β,γ-enones, which reacted with either vinyllithium,
2-lithio-2H-dihydropyran or another metallated haloalkene to afford substituted trans-1,2-divinylcyclohexanols of
different degrees of stability. These intermediates underwent neutral thermal oxy-Cope rearrangements when heated
in xylene in Ace® tubes. The first-formed enols ketonised without loss of HF to afford a range of cyclodecenones in
moderate to good yield; X-ray crystallography was used extensively for product characterisation. All substrates
rearranged more rapidly than a cis/trans mixture of 1,2-divinylcyclohexanols.
potassium alkoxide then allows rearrangement at temperatures
as low as Ϫ78 ЊC, although temperatures at and above ambient
are more common. The reaction leads directly to the formation
of a reactive enolate which can be trapped, adding complexity
Introduction
In 1964, Berson and Jones recognized that the thermal (oxy-
Cope) rearrangement of a 1,5-diene 1, carrying an hydroxyl
group at C-3, resulted in the formation of an enol 2 which
or a convenient functionality for subsequent highly selective
underwent irreversible tautomerisation to a δ,ε-unsaturated
reactions. The anionic oxy-Cope has indeed become a tool of
carbonyl compound 3 (Scheme 1) driving the reaction from left
major strategic power6 for the construction of complex cyclic
to right.1
systems.7–12
Although the conditions of the anionic oxy-Cope are milder,
the methodology cannot be applied to certain classes of sub-
strates. The alkoxide must be present as a loose ion pair (or may
be more dissociated still) for the C3–C4 bond to be weakened
significantly, so high nucleophilicity/basicity at oxygen is an
inevitable consequence, and may lead to undesirable side
reactions.
Scheme 1 Minimal framework for neutral oxy-Cope rearrangement.
Our interest in the oxy-Cope rearrangement arose from the
seminal observations of Dolbier and co-workers. Even a single
The oxy-Cope rearrangement differs from the Cope
rearrangement in several ways. Firstly, the oxy-Cope substrate
can be prepared easily either by addition of a vinylmetal
reagent to a β,γ-unsaturated carbonyl compound, or by the
1,2-addition of an allylmetal to an α,β-unsaturated carbonyl
compound. Secondly, the rapid tautomerisation of the enol
species to a carbonyl compound effectively removes the oxy-Cope
product from the equilibrium and thus the rearrangement is
irreversible.2 This driving force can overcome the development
of strain which impedes Cope rearrangements,3 so, importantly
for natural product synthesis, medium rings can be constructed
easily. Furthermore, the product of the oxy-Cope rearrange-
ment contains two remote and useful functional groups which
may be manipulated or transformed further. The reaction can
require high temperatures; although the overall driving force is
significant, strong bonds must be broken in the rate determin-
ing step, and retro-ene reactions can compete, particularly with
simple acyclic substrates. Hydroxyl group protection can block
the fragmentation pathway,4 but a simple change of reaction
conditions affords the highest level of control.
fluorine atom affects the equilibrium position of the Cope
rearrangement shown in Scheme 2,13 though with a modest
effect on ∆G‡ compared to the isotopically labeled parent
system described by Doering and co-workers.14
Progressive terminal difluorination of 1,5-hexadiene lowers
∆H‡ and ∆G‡ for the reaction, while ∆S‡ increases. The ∆S‡
values are consistent with passage through concerted chair
transition states and the rearrangements of 4 and 5 are also
effectively irreversible, the thermodynamic advantage of the
products arising from the well known destabilising effect of CF2
centres upon alkenes.15 More complex situations arise with sys-
tems that are constrained to adopt boat conformations16 to
bring the diene termini together;17 kinetic and computational
insights are being developed to elucidate the role of biradical
reaction pathways under these constraints.18,19
The oxy-Cope rearrangement involving fluorinated sub-
strates has not, to our knowledge, been reported anywhere
else.20 We wished to know if we could use our alkenylmetal
chemistry21–23 to construct precursors to the oxy-Cope
rearrangement easily and to determine qualitatively if the
thermal neutral oxy-Cope rearrangement would be accelerated
by the presence of one or two terminal CF2 groups.24–27 Because
Spectacular rate enhancements are available if the rearrange-
ment is carried out via the conjugate base;5 generation of the
T h i s j o u r n a l i s © T h e R o y a l S o c i e t y o f C h e m i s t r y 2 0 0 3
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4423

View Article Online
Scheme 2 Cope rearrangements reported by Doering,14 and Dolbier18 and co-workers.
difluoroalkenes are well known electrophiles and their exposure
to potassium alkoxides would be expected to result in nucleo-
philic displacement of a fluoride ion,28,29 we concentrated
exclusively on the thermal neutral rearrangements.
As fluorinated cyclodecenones were unknown in the liter-
ature, we chose to investigate the ring expansion reaction of
cyclohexyl templates; Scheme 3 shows the overall route. Nucleo-
philic ring opening of cyclohexene oxide, followed by Swern
oxidation and the addition of a second vinylmetal would deliver
divinylcyclohexanols suitable for rearrangement. Issues would
arise with the relatively high basicity of the difluorovinyllithium
reagents, the potential sensitivity of the β,γ-unsaturated ketone
intermediate either during isolation or the addition of the
second vinylmetal and the potential retro-ene side reaction. All
of these issues are addressed in the manuscript.
Scheme
4
Reagents and conditions: i, n-BuLi, THF, Ϫ78 ЊC;
ii, BF3.OEt2; iii, cyclohexene oxide; iv, DMSO, oxalyl chloride, DCM,
then Et3N, Ϫ78 ЊC; v, vinyllithium, THF, Ϫ78 ЊC then NH4Cl; vi, ∆,
xylene, Ace® tube (Table 1).
microanalytically pure. The ketones could be stored success-
fully without any trace of isomerisation. Formation of the
dienolate or dienol followed by γ-protonation would move the
alkenyl group into conjugation with the ketonic carbonyl while
rehybridising the CF2 centre from sp2 to sp3. We stirred 8c with
triethylamine in CD2Cl2 for one week but detected no change
in the 19F NMR spectrum, so these β,γ-unsaturated ketone
intermediates are more robust than suspected.
Commercial vinylmagnesium bromide gave unacceptable
results in subsequent attempts at vinylation and we were
unable to obtain good yields of divinylcyclohexanols using the
Gadwood method37 for the preparation in situ of vinyllithium
from vinyl bromide. However, transmetallation of vinyltributyl-
tin with n-butyllithium followed by the addition of 8a–8c at
Ϫ78 ЊC delivered 9a–9c in good to moderate yields. These
intermediates were very difficult to purify fully; traces of
tetraalkyltin compounds adhered through column chromato-
graphy but clean 19F NMR spectra were obtained. These
suggested strongly that only one of the two possible diastereo-
isomers had been formed (as a racemic modification). The sense
of attack predicted (on 11 in Scheme 5) would lead to the trans-
dienols 9a–c based upon the known tendency of 2-methyl-
Scheme
3
Proposed route to fluorinated cyclodecenones using
metallated fluoroalkenes and the neutral oxy-Cope rearrangement.
Results and discussion
Fluorinated alkenylmetals 6a and 6b were generated according
to precedent directly from HCFC-133a30 and HFC-134a31,32
respectively; in the presence of freshly-distilled BF3.etherate,33
both ring opened cyclohexene oxide in good yield to afford the
known homoallyl alcohols 7a (69%) and 7b (76%) (Scheme 4).
We generated known 6c from the corresponding stannane34
and used the same electrophile/Lewis acid combination to
afford 7c (64%). The use of the tin chemistry is undesirable but
gave cleaner products than our original procedure. Coe et al.
have described cyclisation reactions of homoallylic alcohols
related to 7b;35 the stereochemistry of 7b and the presence of
the boron halide presumably prevent the cyclisation here.
Swern oxidation, selected because it is known to be mild, was
carried out using a standard procedure36 to deliver ketones
8a–8c in moderate to good yield (Table 1).
The former compounds were shown to be pure by capillary
GC and characterised to HRMS while 8c was obtained
Scheme 5 Trajectory of nucleophilic attack upon vinylcyclohexanol
predicted by Ashby.38
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4424

View Article Online
Table 1 Synthesis of first generation cyclodecenones
Alcohol
Yield(%)
Ketone
Yield(%)
Dienol
Yield(%)a
Enone
T/ЊC
t/hours
Yield(%)
7a
7b
7c
69
76
64
8a
8b
8c
83
72
52
9a
9b
9c
78
76
52
10a
10b
10c
125
125
125
8
46
32
67
72
68
a Estimated yield; see text.
Table 2 Syntheses of cyclodecenones incorporating dihydropyranyl units
Ketone
Dienol
Yield(%)a
Enone
T/ЊC
t/hours
Yield(%)
8a
8b
8c
13a
13b
13c
97
78
100
14a
14b
14c
150
140
155
3
17
9
65
84
62
a Estimated yield; see text.
cyclohexanone to undergo equatorial attack with
phenyllithium.38
graphic analysis after vapour diffusion recrystallisation from
dichloromethane (hexane precipitant).
The first rearrangements were carried out in xylene in Ace®
tubes at 125 ЊC (also Table 1), with the periodic withdrawal of
aliquots for monitoring by 19F NMR. In all cases, the terminal
Fig. 1 shows the X-ray crystal structure of 14b; the alkene
geometry can be seen clearly (the corresponding 3JH–F coupling
is 34.8 Hz, supporting our assignment for 10b), together with
the cis-junction between the tetrahydropyranyl and cyclo-
decenone rings.
2
vinylic 19F signals (Ϫ88 to Ϫ122 ppm, JF–F 40–80 Hz) dis-
appeared as the AB systems (2JF–F ∼250 Hz) arising from the
rehybridised CF2 centre developed; the fluorine atoms are
diastereotopic because the cyclodecenone ring is chiral. All
reactions were driven to completion before isolation of the
crystalline products 10a–c by evaporation of the xylene and
column chromatography. The purity of the products was shown
by GC but we were only able to obtain satisfactory micro-
analysis in one instance (10c) despite repeated attempts. It is
likely that accurate combustion analysis was prevented by the
presence of traces of tetrabutyltin which could not be fully
removed. All three cyclodecenones invariably crystallised as fine
needles preventing the collection of any X-ray data.
The alkene stereochemistry can be assigned in the case of
10b; in the 1H NMR spectrum, the alkenyl proton shows a large
(33.7 Hz) coupling consistent only with the transoid H–C᎐C–F
᎐
dihedral (the Z-alkene). The same geometry was inferred for
10a and 10c. There was modest broadening in the 19F NMR
spectrum and in the methylene region of the 1H NMR spectrum
of 10a, presumably because of fluxional movements of the ring.
We did not carry out VT NMR studies as there seemed little
information to gain in these instances.
Fig. 1 X-Ray crystal structure of 14b.
One of the fluorine atoms shows a large 3JH–F splitting to the
H-10 proton (31.6 Hz in 14b, 30.9 Hz in 14c) consistent with the
close to antiperiplanar dihedral relationship in both the crystal
structure (175Њ) and the solution conformer. The alkene shows
some twisting out of plane (4.9Њ) which is not unexpected for a
transoid alkene in a 10-membered ring. The structure for 14a
showed essentially the same features.
We were also able to add metallated dihydropyran 12 to
8a–8c to afford three very unstable dienols 13a–13c (Scheme 6).
Extensive 2D NMR analysis was undertaken to allow the
1
assignment of the rather complex, though well-dispersed, H
NMR spectrum of 14b. In the COSY spectrum (Fig. 2), almost
all the possible correlations are visible. The alkene proton H-7 is
clearly coupled to both H-6 protons and connectivity can be
demonstrated around the ring into the H-3 methylene protons.
The chemical shifts of the two H-3 protons differ by almost 1
ppm which is surprising. One possible explanation could be a
transannular interaction with the alkenyl group which select-
ively and strongly deshields one proton in the methylene pair.
The two H-13 protons next to the oxygen of the tetrahydro-
pyranyl ring provide a convenient place to begin tracing the
connectivity through the fused ring.
Confirmation of the assignment and support for the congru-
ence between solid state and solution conformations came from
an NOE experiment (GOESY40). Examination of the crystal
structure suggests that the H-7 alkene proton lies close to one
of each of the H-5 (at 2.55 Å) and H-3 (at 2.57 Å) methylene
protons, which are mutually close (at 2.56 Å). The observed
transannular transfer of magnetisation from H-7 to H-3 is
Scheme 6 Reagents and conditions: i, t-BuLi, THP, 0 ЊC; ii, THF, cool
to Ϫ78 ЊC then 8a–8c; iii, ∆, xylene, Ace® tube (Table 2).
Dihydropyran was metallated in tetrahydropyran using the
method of Meyers.39 The adducts showed very clean 19F NMR
spectra, consistent with the formation of a single (racemic)
diastereoisomer but resisted all attempts at purification,
decomposing rapidly on silica gel. However, the crude products
rearranged smoothly in xylene in Ace® tubes at 150 ЊC to afford
crystalline products 14a–14c in good yield (Table 2).
All afforded satisfactory microanalyses and in two cases,
sufficiently good crystals could be grown for X-ray crystallo-
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4425

View Article Online
Fig. 2 500 MHz 1H–1H gradient COSY NMR spectrum for 14b.
therefore to be expected and the H-7/H-5 and H-5/H-3 magnet-
isation transfers are consistent with these predictions. Sharp
NMR spectra were obtained, consistent with lower degrees of
ring mobility for 14a–c.
These results were interesting because they showed that we
could introduce further ring architecture and additional func-
tionality via the second vinylmetal reagent. They also con-
firmed our ideas about the sense of stereoselection in the enone
addition reaction; the most obvious explanation for the various
stereochemical outcomes in 14a–14c involves the rearrange-
ment of a trans-dienol through a chair transition state 15. How-
ever, 16, which also leads to the correct alkene geometry cannot
be excluded.
Scheme 7 Reagents and conditions: i, 180 ЊC, PhMe.41
The results differ significantly from those described by
Barriault and co-workers (Scheme 7).41 Retro-ene reaction (a
significant side reaction for simple acyclic species) dominates
here. Vigorous heating of 17 in a sealed tube in toluene afforded
18 alone, and none of the expected 20 (a strain-relieving
transannular-ene product from first-formed oxy-Cope educt
19). Our outcome presumably arises from the considerably
greater ease with which these systems (9,13) rearrange so
that the fragmentation in Scheme 7 which leads to 18 does not
compete with the neutral oxy-Cope.
Further variation in the second alkenylmetal fragment was
sought through the reactions of 6a with 8b and 8c, to afford 21b
(46%) and 21c (62%), and 6b with 8a–8c to afford 22a (57%
estimated yield), 22b (62% estimated yield) and 22c (56%)
(Scheme 8).
Scheme 8 Reagents and conditions: i, 6a (for 21b,c) or 6b (for 22a–c),
THF, Ϫ78 ЊC; ii, ∆, xylene, Ace® tube (see Table 3).
These were very difficult compounds to purify fully, though
we were satisfied as to their identity from multinuclear NMR.
Although we could detect only a single diastereoisomer for
21b and 21c, separable diastereoisomers were obtained for 22a
(2.3 : 1) and 22c (1.8 : 1). We have not been able to propose the
identity of the major stereoisomer by comparing 13C and 19F
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4426

View Article Online
Table 3 Preparation of more highly fluorinated cyclodecenones
Nucleophile
Ketone
Dienol
Yield(%)a
Enone
T/ЊC
t/hours
Yield(%)
6a
6a
6b
6b
6b
8b
8c
8a
8b
8c
21b
21c
22a
22b
22c
46
62
57
62
56
23b
23c
24a
24b
24c
110
150
150
100
100
40
2
1.75
47
30
69
71
73
54
66
a Estimated yield.
NMR chemical shifts, nor can we explain why a second stereo-
isomer should be formed in these cases alone. The NMR
spectra of 21a and 22b require comment; for 22b, most of the
13C and 1H NMR spectra were badly overlapped, making
the identification of discrete multiplets impossible, but with
integration of the 19F NMR spectra, the two diastereoisomers
could be identified clearly with discrete signals for 5 of the 6
different fluorine atoms in each diastereoisomer. The signals
from the sixth fluorines in each diastereoisomer were over-
lapped so that the signal appeared as a doublet of triplets. We
also note that a signal from the other internal fluorine in the
minor diastereoisomer (Ϫ176.5 ppm, ddt, 111.6, 34.0, 5.0 Hz)
contains a surprisingly large coupling to a proton. One 34.0 Hz
splitting arises from a cis-coupling to another alkenyl fluorine
3
but the third splitting must arise from a JH–F coupling to the
allylic proton in 22b. The origin of this splitting is far from
clear, as similar interactions in 8b result in much smaller coup-
ling constants (7.6 Hz in 8b). However, in 22a, which has a
simpler 1H NMR spectrum, the putative 3JH–F coupling can be
identified in the 19F (31.8 Hz) and 1H NMR (31.6 Hz) spectra,
suggesting strongly that it represents a genuinely large allylic
H–F coupling. The putative 3JH–F coupling also appears clearly
in the 19F NMR spectrum of 7b, but is obscured by overlap in
the 1H NMR spectrum.
All five compounds underwent oxy-Cope rearrangement suc-
cessfully (Table 3) to afford crystalline solids from which we
obtained X-ray crystal structures as unambiguous proof of
structure. Only Z-configured alkenes were obtained, although it
is possible that we simply did not detect a second product in the
rather complex 19F NMR spectra of the crude material.
We were concerned that the first formed enols from the oxy-
Cope would lose HF to form the conjugated enones. We saw no
evidence of such products but we were careful to neutralise the
tube contents before working with them. When mutually super-
imposed, the crystal structures for 23c and 24c show a high
degree of congruence but differ significantly in the location of
the C–Cl and C–F bonds (Fig. 3a).
Fig. 3 (a) Superimposed X-ray crystal structures for 23c and 24c
(fluorine atoms in green, chlorine atom in yellow). (b) Superimposed
X-ray crystal structures for 23b and 24a (fluorine atoms in green,
chlorine atoms in yellow, hydrogens hidden for clarity).
The same issue arose with 23b and 24a (Fig. 3b). In both
cases, the Cl–C–C–O dihedral angle is considerably closer to
90Њ (79.1Њ) than the H–C–C–O angle in the fluorocongener
(64.3Њ). The expansion of this angle causes some distortion to
the dihedral angles around the adjacent methylene groups.
From studies of 2-halocyclohexanones,42,43 different halogen
stereoelectronic effects have been noted. The interaction
between the C–X σ bond and the C᎐O π* orbital favours axial
᎐
disposition of the C–Cl bond but is much weaker for the C–F
bond because of the opposition between the high electro-
negativity of the fluorine atom and the Xϩ character required
(Fig. 4).
The approach of the Cl–C–C–O dihedral angle towards 90Њ
in these systems may reflect this preference, at least in the solid
Fig. 4 Stereoelectronic effects of the conformations of α-halocarbonyl
compounds.
cis/trans mixture following the procedure of Holt.45 Choosing
150 ЊC as the rearrangement temperature, we found that 11a,
14a and 24a all rearranged completely within 3 hours, whereas
25 was recovered unchanged.
state. The antiparallel arrangement between the C–F and C᎐O
᎐
bonds in 24a and 24c allows dipolar minimisation as in the
calculated minimum energy conformation of N-methyl fluoro-
acetamide.44 Unfortunately, the signals for the H-2 protons in
1
the H NMR spectra of 23b and 23c are too complex for the
solution Cl–C–C᎐O dihedral to be estimated.
᎐
To make qualitative reactivity comparisons between the
various systems, we prepared 1,2-divinylcyclohexanol 25 as a
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4427

View Article Online
We were able to form 26 but only after heating for 24 hours at
225 ЊC, in contrast to the literature observation.46 Indeed all the
fluorinated systems described in this manuscript rearrange
more rapidly than 25, consistent with the effects of CF2 centres
upon the Cope rearrangement and indicating that the thermal
neutral oxy-Cope rearrangement could be an extremely valu-
able tool for the construction of cyclic molecules which are
selectively fluorinated to different degrees. Furthermore, the
addition of fluorinated and non-fluorinated vinylmetals to
γ,γ-difluorinated vinylcyclohexenones appears to be a flexible,
reliable and robust method of precursor synthesis. Further
work will attempt to probe the importance of diradical
character in these rearrangements.
cation of this work.20 Other data appear after the preparative
details for the compound.
trans-2-(1Ј-Chloro-2Ј,2Ј-difluoroethenyl)-cyclohexan-1-ol 7a
Alcohol 7a was prepared according to our published
procedure.30
trans-2-(1Ј,2Ј,2Ј-Trifluoroethenyl)-cyclohexan-1-ol 7b
A 2-necked round bottomed flask was fitted with a Rotaflo tap
and suba seal. The flask was evacuated through the Rotaflo tap,
and cooled to Ϫ78 ЊC; THF (2.5 ml) was added to the evacu-
ated flask which was sealed. 1,2,2,2-Tetrafluoroethane (50 ml,
2.1 mmol [Vm @ 20 ЊC = 24043 ml mmolϪ1 vs. @ 0 ЊC (STP) =
22402 ml molϪ1.]) was used, then n-butyllithium (1.7 ml of a
1.8 M solution in hexane, 3.0 mmol) was added and the mixture
was left to stir for 45 minutes at Ϫ78 ЊC. Work-up in the usual
manner afforded an orange oil (0.29 g). Purification by column
chromatography (10% diethyl ether in petroleum ether) gave 7b
(0.20 g, 76%) as a yellow oil; Rf (10% diethyl ether in petroleum
Experimental
NMR spectra were recorded on a Bruker AC-300 spectrometer.
19F-NMR (300 MHz) spectra were referenced to fluorotrichloro-
methane as the internal standard. 1H-NMR (300 MHz) and 13C-
NMR (75 MHz) spectra were referenced to residual chloroform
at 7.26 ppm. 13C-NMR spectra were recorded using the JMOD or
ether) 0.10; νmax (film)/cmϪ1 3404 (OH), 1794, 1729 (C᎐C);
᎐
1
PENDANT pulse sequences. 500 MHz H-COSY and GOESY
δH (300 MHz, CDCl3) 3.56–3.40 (1H, m, H-1), 3.30 (1H, s,
OH ), 2.28–1.87 (2H, m, H-2, H-6), 1.80–1.55 (3H, m, –CH2),
1.52–1.33 (1H, m, –CH2), 1.31–1.05 (3H, m, –CH2); δC (75
NMR spectra were recorded on a Bruker DRX-500 spectrometer.
400 MHz 19F, 1H-COSY and HSQC NMR spectra were recorded
on a Bruker AMX-400 spectrometer. J values are reported in Hz.
Low and high resolution mass analyses were recorded on a VG
ProSpec mass spectrometer, Kratos Profile mass spectrometer or
a VG ZabSpec mass spectrometer. Chemical ionisation (CIϩ)
methods used ammonia as the reagent gas. A Micromass LCT
mass spectrometer was also used for both low resolution
(ES-TOF) mass spectra (using methanol mobile phase) and
HRMS measurements (using a lockmass incorporated into the
mobile phase). All mass data containing Cl refer to 35Cl isotope.
Precoated aluminium-backed silica plates were supplied by
1
2
MHz, CDCl3) 153.9 (dd, JC–F 272.4, JC–F 47.5), 129.4 (ddd,
1JC–F 235.1, JC–F 52.0, JC–F 14.1), 69.0, 43.8 (d, JC–F 20.9),
34.2, 25.1, 25.0, 24.0; δF (282 MHz, CDCl3) Ϫ105.2 (1F, dd,
2J 89.0, 3Jcis 31.8), Ϫ124.9 (1F, dd, 3Jtrans 113.2, 2J 89.0), Ϫ185.0
(1F, d, Jtrans 113.2, Jcis 31.8, JH–F 31.8); m/z (CI) 198 (80%,
M ϩ [NH4]ϩ), 134 (100%). The data were in agreement with
those reported by Dubuffet et al.33
2
2
2
3
3
3
2-[1Ј-(N,N-Diethylcarbamoyloxy)-2Ј,2Ј-difluoroethenyl]-cyclo-
hexan-1-ol 7c
E. Merck, A. G. Darmstadt, Germany. (silica gel 60 F254
,
A
solution of 1-(N,N-diethyl-carbamoyloxy)-2,2-difluoro-
thickness 0.2 mm, Art.no. 5554). Anisaldehyde, potassium
permanganate staining or ultraviolet light were employed for
visualisation. Column chromatography was performed using
silica gel (E. Merk, A. G. Kieselgel, Art. 9385). Infra red spectra
were obtained from a Perkin Elmer 1600 series FTIR spectro-
meter, in the region 4000–500 cmϪ1. Elemental analyses were
performed at the University of North London. X-ray crystallo-
graphic analyses were performed, by the EPSRC X-ray crystal-
lography service, at the University of Southampton.
Tetrahydrofuran was dried by refluxing with sodium metal and
benzophenone under dry nitrogen, until a deep purple colour
persisted. The solvent was then collected by syringe as required.
Diethylether, hexane, toluene and DCM were dried by refluxing
with calcium hydride under dry nitrogen then distilled and col-
lected by syringe as required. HCFC-133a and HFC-134a were
supplied by Fluorochem and used as received. Alcohol 7a was
prepared according to our published procedure.30 1-(N,N-diethyl-
carbamoyloxy)-2,2,2-trifluoroethane and 1-(N,N-diethyl-carb-
amoyloxy)-2,2-difluoroethenyl tributylstannane were prepared
according to our published procedure.34 Cyclohexene oxide,
oxalyl chloride, dimethylsulfoxide, triethylamine and boron tri-
fluoride etherate were purchased from the Aldrich Chemical
Company and were freshly distilled before the reactions. t-Butyl-
lithium and n-butyllithium were titrated using 1,3-diphenyl-
2-propanone-p-toluene sulfonylhydrazone according to the
procedure of Lipton and co-workers.47 Sealed tube reactions were
performed in Ace® pressure tubes (8648B). Light petroleum
refers to the fraction boiling in the range 40–60 ЊC.
ethenyl tributylstannane34 (0.61 g, 2 mmol) in dry THF (6 ml)
was cooled to Ϫ78 ЊC. n-Butyllithium (1.21 ml of a 1.78 M
solution in hexane) was added to the solution over 10 minutes
and the mixture became yellow during this time. The suspen-
sion was stirred at Ϫ78 ЊC for 45 minutes, then boron trifluoride
etherate (0.50 ml, 4 mmol) was added in one portion, followed
by the addition of cyclohexene oxide (0.20 ml, 2.0 mmol). The
solution was stirred for a further two hours at Ϫ78 ЊC before
being allowed to warm slowly. At Ϫ30 ЊC, saturated aqueous
ammonium chloride solution (25 ml) was added. The aqueous
layer was extracted with ether (3 × 15 ml) and the combined
organic extracts were washed with saturated sodium bicarb-
onate solution (2 × 15 ml), dried with magnesium sulfate and
evaporated in vacuo to leave a yellow oil. Purification by flash
chromatography (15% ethyl acetate in hexane) yielded the alco-
hol 7c (0.354 g, 64%) as a pale yellow oil; Rf (15% ethyl acetate
3
in hexane) 0.35; δH (300 MHz, CDCl3) 4.25 (1H, d, JH–H 2.9,
H-1), 3.40–3.20 (5H, m, NCH2CH3, OH ), 2.30–2.15 (1H, m,
H-2), 2.12–1.98 (1H, m, –CH2), 1.82–1.61 (3H, m, –CH2), 1.45–
1.10 (10H, m, –CH2, NCH2CH3); δC (75 MHz, CDCl3) 155.2
(dd, 1JC–F 288.8, 1JC–F 282.6), 154.5, 112.1 (dd, 2JC–F 43.5, 2JC–F
14.1), 70.0, 68.0, 42.0, 39.0, 28.0, 26.0, 24.0 18.0; δF (282 MHz,
CDCl3) Ϫ96.8 (1F, d, 2JF–F 59.8), Ϫ110.2 (1F, d, 2JF–F 59.8); m/z
(CI) 278 (100%, M ϩ H)ϩ, 260 (7%, M Ϫ OH), 100 (30%,
CONEt2) in agreement with literature data.34
2-(1Ј-Chloro-2Ј,2Ј-difluoroethenyl)-cyclohexan-1-one 8a
A solution of dimethyl sulfoxide (5.8 ml, 28 mmol) in DCM
(25 ml, 4.8M) was added dropwise to a stirred solution of
oxalyl chloride (1.21 ml, 12.7 mmol) in DCM (14 ml) at Ϫ78 ЊC.
The mixture was stirred for 30 minutes at Ϫ78 ЊC then 7a
(1.05 g, 5.86 mmol) was added over ten minutes at Ϫ78 ЊC as a
solution in DMSO/DCM (6 ml, 1 : 1.9). The reaction mixture
was stirred for a further 60 minutes at Ϫ78 ЊC. Triethylamine
Crystallographic experimental†
Crystallographic data for 14a were published in our communi-
† CCDC reference numbers 216343 and 219860–219864. See http://
www.rsc.org/suppdata/ob/b3/b311261f/ for crystallographic data in .cif
or other electronic format.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4428

View Article Online
(4.06 ml, 29 mmol) was added slowly and the yellow solution
was left stirring at Ϫ78 ЊC for one hour further and allowed to
warm to Ϫ30 ЊC. The reaction was quenched at this temper-
ature after 30 minutes with saturated aqueous bicarbonate
solution (50 ml). The aqueous layer was extracted with di-
chloromethane (3 × 10 ml). The combined organic extracts
were washed with brine (25 ml), dried (MgSO4), filtered and
concentrated in vacuo to leave a yellow oil (1.15 g). Purification
by flash chromatography (10% diethyl ether in petroleum ether)
yielded 8a (0.945 g, 83%, 99% by GC) as a yellow oil; Rf
filtered and concentrated in vacuo. Column chromatography
(10% diethyl ether in light petroleum) afforded dienol 9a
(assumed trans) (0.151 g, 78%) as a clear oil; Rf (10% diethyl
ether in light petroleum) 0.35; δH (300 MHz, CDCl3) 5.81 (1H,
dd, 3JH–H 16.9, 3JH–H 10.7, H-1Ј), 5.32 (1H, d, 3JH–H 16.9, H-2Ј),
5.12 (1H, d, 3JH–H 10.7, H-2Ј), 2.52 (1H, s, OH ), 2.16–1.99 (1H,
m, H-2), 1.90–1.20 (8H, m, –CH2); δF (282 MHz, CDCl3) Ϫ88.7
(1F, d, 2JF–F 43.2), Ϫ91.2 (1F, d, 2JF–F 43.2); m/z (CI) 240 (18%,
M ϩ [NH4]ϩ), 222 (90%, Mϩ). The compound was not charac-
terised more fully because it contained traces of tetrabutyltin
which could not be removed.
(10% diethyl ether in petroleum ether) 0.5; νmax (film)/cmϪ1 1748
3
(C᎐C), 1717 (C᎐O); δ (300 MHz, CDCl3) 3.40 (1H, dd, JH–H
᎐
᎐
H
3
3
11.8, JH–H 5.5, H-2), 2.53 (1H, d, JH–H 14.3, H-6), 2.40–2.20
(1H, m, H-6), 2.18–1.92 (4H, m, –CH2), 1.80–1.61 (2H, m,
–CH2); δC (75 MHz, CDCl3) 206.1, 154.5 (t, JC–F 287.7), 90.7
(dd, JC–F 36.5, JC–F 38.0), 50.8, 41.7, 31.8, 27.1, 24.6; δF (282
MHz, CDCl3) Ϫ87.3 (1F, d, JF–F 42.6), Ϫ92.3 (1F, d,
trans-1-Ethenyl-2-(1Ј,2Ј,2Ј-trifluoroethenyl)-cyclohexan-1-ol 9b
1
Dienol 9b was prepared using the same procedure as for 9a
from 8b (0.178 g, 1.0 mmol) and 1-(tributylstannyl)ethene
(0.316 g, 1.0 mmol). Work up in the usual manner afforded a
yellow oil (0.29 g) which was purified by column chromato-
graphy (10% diethyl ether in light petroleum) to afford dienol 9b
(assumed trans) (0.14 g, 68%) as a colourless oil; Rf, (10% di-
ethyl ether in petroleum ether) 0.1; νmax (film)/cmϪ1 3500 (OH),
2
2
2
2JF–F 42.6); [HRMS (CI, [M ϩ NH4]ϩ) Found 194.03122. Calc.
for C8H9ClOF2 194.03099]; m/z 194 (15 %, Mϩ), 159 (50 %,
M Ϫ Cl), 39 (100%).
1786, 1756, 1716 (C᎐C); δ (300 MHz, CDCl3) 5.82 (1H, dd,
᎐
2-(1Ј,2Ј,2Ј-Trifluoroethenyl)-cyclohexan-1-one 8b
H
3JH–H 16.7, 3JH–H 10.7, H-1Љ), 5.31 (1H, d, 3JH–H 16.7, H-2Љ), 5.11
Ketone 8b was prepared from 7b, using the procedure for 8a on
the same scale. Flash chromatography (10% ethyl ether in pet-
roleum ether) yielded 8b (0.756 g, 72%, 98% by GC) as a yellow
oil; Rf (10% ethyl ether in petroleum ether) 0.25; νmax (film)/cmϪ1
3
(1H, d, JH–H 10.7, H-2Љ), 2.42–2.30 (1H, m, H-2), 2.12–1.90
(1H, m, OH ), 1.89–1.10 (8H, m, –CH2); δC (75 MHz, CDCl3)
1
1
2
153.9 (ddd, JC–F 287.6, JC–F 272.4, JC–F 48.0), 144.0, 130.2
1
2
2
(ddd, JC–F 236.8, JC–F 52.0, JC–F 14.7), 113.0, 58.0, 42.8 (dd,
2JC–F 18.6, JC–F 2.8), 38.5, 26.0, 24.5, 21.0; δF (282 MHz,
1797 (C᎐C), 1721 (C᎐O); δH (300 MHz; CDCl3) 3.30–3.10 (1H,
3
᎐
᎐
CDCl3) Ϫ105.5 (1F, dd, 2JF–F 85.2, 3J cFi–sF 31.8), Ϫ122.2 (1F, dd,
m, H-2), 2.45–2.35 (1H, m, H-6a), 2.30–2.20 (1H, m, H-6b),
2.15–1.95 (2H, m, H-3), 1.95–1.50 (4H, m, –CH2); δC (75 MHz,
CDCl3) 205.0, 154.1 (ddd, 1JC–F 287.6, 273.5, 2JC–F 45.8), 126.5
trans
trans
2JF–F 85.2, J F–F 112.5, JF–H 3.8), Ϫ179.1 (1F, dt, J F–F 112.5,
3J cFi–sF31.8); m/z (CI) 224 (13%, Mϩ[NH4]ϩ), 206 (90%, Mϩ). The
compound was not characterised more fully because it
contained traces of tetrabutyltin which could not be removed.
3
4
3
1
2
2
2
(ddd, JC–F 236.2, JC–F 52.6, JC–F 16.9), 40.8 (d, JC–F 20.9),
40.5, 30.3, 29.8, 27.0; δF (282 MHz, CDCl3) Ϫ104.1 (1F, dd,
cis
trans
2JF–F 85.2, J F–F 33.7), Ϫ123.2 (1F, ddd, J F–F 114.0, JF–F
3
3
2
trans
cis
4
3
3
3
85.2, JF–H 2.5), Ϫ180.4 (1F, ddd, J F–F 114.0, J F–F 33.7, JF–H
7.6); [HRMS (CI, M ϩ [NH4]ϩ) Found 196.0950. Calc. for
C8H13F3NO 196.0949, found 196.0950]; m/z (CI) 196 (100%,
M ϩ [NH4]ϩ), 178 (20, Mϩ), 158 (50, M Ϫ HF).
trans-1-Ethenyl-2-(1Ј-[N, N-diethylcarbamoyloxy]-2Ј-difluoro-
ethenyl)-cyclohexan-1-ol 9c
Dienol 9c was prepared from 8c (0.275 g, 1.0 mmol) and
1-(tributylstannyl)ethene (0.316 g, 1.0 mmol) as for 9c except
for the fact that the temperature was maintained at Ϫ78 ЊC for
four hours after the addition of n-butyllithium to the stannane.
The reaction was quenched at Ϫ78 ЊC with saturated ammo-
nium chloride solution (20 ml). Work up in the usual manner
followed by column chromatography (10% diethyl ether in light
petroleum) afforded dienol 9c (assumed trans) (0.206 g, 68%) as
a colourless oil; Rf (10% diethyl ether in light petroleum) 0.3;
νmax (film)/cmϪ1 3414 (OH), 1763 (C᎐O), 1703 (C᎐C); δ (300
2-[1Ј-(N,N-Diethylcarbamoyloxy)-2Ј,2Ј-difluoroethenyl]-cyclo-
hexan-1-one 8c
Ketone 8c was prepared from 7c, using the procedure for 8a on
the same scale. Flash chromatography (10% ethyl ether in pet-
roleum ether) yielded 8c (0.838 g, 52%) as a yellow oil; Rf (10%
ethyl ether in petroleum ether) 0.2; (Found: C, 56.70; H, 7.00;
N, 5.00. Calc. For C13H19O3F2N: C, 56.73; H, 6.91; N 5.09%);
νmax (film)/cmϪ1 1777 (C᎐C), 1724 (C᎐O); δ (300 MHz; CDCl )
᎐
᎐
H
᎐
᎐
H
3
3
3
MHz, CDCl3) 5.89 (1H, dd, JH–H 17.3, JH–H 10.7, H-1Љ) 5.31
3.40–3.28 (5H, m, H-2, 2 × NCH2), 2.58–2.45 (1H, m, H-6),
2.40–2.22 (2H, m, –CH2), 2.15–1.82 (2H, m, –CH2), 1.80–1.56
(3H, m, –CH2), 1.20–1.05 (6H, m, CH3); δC (75 MHz; CDCl3)
205.9, 152.2 (dd, 1JC–F 289.4, 1JC–F 281.4), 152.9, 109.2 (dd, 2JC–F
44.6, 2JC–F 15.2), 50.2, 41.9, 41.8, 29.0, 26.0, 24.6, 13.8; δF (282
3
3
(1H, d, JH–H 17.3, H-2Љ), 5.09 (1H, d, JH–H 10.7, H-2Љ), 3.61
(1H, s, OH ), 3.40–3.19 (4H, m, NCH2), 2.48–2.37 (1H, m, H-2),
1.88–1.02 (8H, envelope, –CH2); δC (75 MHz, CDCl3) 155.5 (t,
1JC–F 282.0), 155.0, 144.8, 112.5, 112.4, 74.6, 44.9, 42.5, 37.6,
29.8, 26.2, 21.2, 13.8; δF (282 MHz, CDCl3) Ϫ97.7 (1F, d, 2JF–F
56.6), Ϫ105.6 (1F, d, 2JF–F 56.6); [HRMS ES, Found 326.1539.
Calc. for C15H23O3F2NNa 326.1544]; m/z 304 (18%, [M ϩ H]ϩ),
286 (20%, M Ϫ OH).
2
2
MHz; CDCl3) Ϫ95.8 (1F, d, JF–F 61.0), Ϫ108.2 (1F, d, JF–F
61.0); m/z (CI) 275 (4%, Mϩ), 260 (M Ϫ Me), 100 (100%,
CONEt2).
trans-1-Ethenyl-2-(1Ј-chloro-2Ј,2Ј-difluoroethenyl)-cyclohexan-
1-ol 9a
5-Chloro-4,4-difluoro-cyclodec-5Z-en-1-one 10a
n-Butyllithium (0.53 ml of a 1.9 M solution in hexane, 1.0
mmol) was added dropwise to a stirred solution of 1-(tributyl-
stannyl)ethene (0.316 g, 1.0 mmol) in THF (0.5 ml) at Ϫ78 ЊC.
The mixture was left stirring for one hour at the same temper-
ature then 8a (0.194 g, 1.0 mmol) was added. The solution
was allowed to warm to Ϫ40 ЊC after one hour and left for
30 minutes further at this temperature. The mixture was then
quenched with saturated aqueous ammonium chloride solution
(20 ml). The phases were separated and the aqueous phase was
extracted with diethyl ether (3 × 10 ml). The combined organic
extracts were washed with saturated sodium bicarbonate solu-
tion (2 × 10 ml), brine (2 × 10 ml), dried (magnesium sulfate),
A solution of dienol 9a (0.222 g, 1.0 mmol) in xylene (5 ml) was
sealed in an Ace® tube and heated to 125 ЊC in an oil bath. The
reaction was followed by 19F NMR of aliquots until the starting
material was consumed completely (8 hours). Concentration
and column chromatography (10% ether in light petroleum)
afforded a white solid which was recrystallised from dichloro-
methane/hexane to afford 10a (0.149 g, 67%) as colourless
needles; mp 76–77 ЊC; Rf (10% ether in light petroleum) 0.3;
νmax (Nujol mull)/cmϪ1 1715 (C᎐O); δH (300 MHz, CDCl3) 6.11
᎐
(1H, t, 3JH–H 7.3, H-6), 3.20–1.10 (12H, envelope); δC (75 MHz,
CDCl3) 207.3, 131.9 (t, 3JC–F 5.1), 129.2 (t, 2JC–F 31.1), 119.7 (t,
2
1JC–F 247.3), 33.4 (t, JC–F 28.3), 56.4, 45.0, 37.1, 28.2, 22.3;
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4429

View Article Online
δF (282 MHz, CDCl3) Ϫ97.8 (1F, br d, 2JF–F 233.9), Ϫ106.1 (1F,
br d, 2JF–F 233.9); [HRMS (CI, [M ϩ H]ϩ) Found: 240.096787.
Calc. for C10H17O35ClF2N 240.096674]; m/z 240 (100%,
M ϩ [NH4]ϩ). Satisfactory microanalysis could not be obtained
for this compound.
be detected by 19F NMR (assumed trans). δF (282 MHz, CDCl3)
cis
2
3
2
Ϫ105.9 (1F, dd, JF–F 83.9, J F–F 32.4), Ϫ122.6 (1F, ddd, JF–F
83.9, 3J tFra–nFs 111.3, 4JF–H 3.8), Ϫ180.1 (1F, dtd, 3JF–F 111.3, 3JF–F
4
32.4, JF–H 5.1). The dienol was used for the subsequent step
without further purification.
trans-2-[1Ј-(N,N-Diethylcarbamoyloxy)-2Ј,2Ј-difluoroethenyl]-
1-(2Љ,3Љ-dihydropyranyl)-cyclohexan-1-ol 13c
4,4,5-Trifluoro-cyclodec-5Z-ene-1-one 10b
From dienol 9b (0.206 g, 1.0 mmol) at 125 ЊC (46 hours).
Concentration and column chromatography (10% ether in light
petroleum) gave a white solid which recrystallised from
dichloromethane/hexane to afford 10b (0.148 g, 72%) as fine
colourless needles; mp 62–63 ЊC; Rf (15% ether in light petrol-
eum) 0.2; νmax (Nujol mull)/cmϪ1 1719 (C᎐O); δ (300 MHz,
Dienol 13c was prepared from 2,3-dihydropyran (0.18 ml,
2.0 mmol) and 8c (0.275 g, 1.0 mmol) as an unstable yellow oil
(0.274g, 100% estimated yield). Only one diastereoisomer could
be detected by 19F NMR (assumed trans). δF (282 MHz, CDCl3)
Ϫ97.1 (1F, d, 2JF–F 55.9), Ϫ84.2 (1F, d, 2JF–F 55.9).
᎐
H
CDCl3) 5.21 (1H, dt, 3JH–F 33.7, 3JH–H 7.7, H-6), 3.10–2.75 (1H,
m, –CH2), 2.73–1.50 (10H, envelope, –CH2), 1.50–1.12 (1H, m,
–CH2); δC (75 MHz, CDCl3) 208.0, 150.2 (dt, 1JC–F 261.7, 2JC–F
8-Chloro-9,9-difluoro-14-oxa-cis-bicyclo-[8.4.0]-tetradeca-7Z-
en-2-one 14a
1
2
2
32.2,), 117.1 (td, JC–F 244.7, JC–F 39.0), 112.0 (dt, JC–F 9.6,
A solution of dienol 13a (0.278 g, 1.0 mmol) in xylene (5 ml)
was sealed in an Ace® tube and heated to 150 ЊC in an oil bath.
The reaction was followed by 19F NMR of aliquots until the
starting material was consumed completely (3 hours). Con-
centration and column chromatography (10% ether in light
petroleum) gave a colourless solid which recrystallised from
dichloromethane/hexane to afford 14a (0.181 g, 65%) as
colourless rhombi; mp 137–138 ЊC; (Found: C, 56.20; H, 6.18.
Calc. for C13H17ClF2O2: C, 56.02; H, 6.15%); Rf (10% ether in
3JC–F 3.4), 44.0, 36.9, 33.6 (td, JC–F 27.7, JC–F 2.3), 28.5, 23.5,
22.1; δF (282 MHz, CDCl3) Ϫ104.8 (1F, dd, 2JF–F 253.0, 3J 24.2),
Ϫ112.2(1F, dd, 2JF–F 253.0, 3JF–F 31.9), Ϫ123.8(1F, dd, 3JF–F 31.9,
3JF–F 24.2); [HRMS (CI, [M ϩ H]ϩ) Found: 224.127287. Calc. for
C10H17F3NO 224.126224]; m/z 224 (100%, [M ϩ NH4]ϩ). Satis-
factory microanalysis could not be obtained for this compound.
2
3
5-(N,N-Diethylcarbamoyloxy)-4,4-difluoro-cyclodec-5Z-en-1-one
10c
light petroleum) 0.1; νmax (Nujol mull)/cmϪ1 1710 (C᎐O);
᎐
δH (500 MHz, CDCl3); 6.25 (1H, d, 3J 10.7, H-7), 4.16 (1H, dd,
2JH–H 11.5, 3JH–H 5.5, H-13a), 3.73–3.68 (1H, m, H-1), 3.48 (1H,
dt, 2JH–H 11.5, 3JH–H 2.7, H-13b), 3.10 (1H, dd, 2JH–H 19.1, 3JH–H
10.3, H-3), 3.09–2.85 (1H, m, H-10), 2.48–2.28 (2H, m, H-6,
H-12), 2.23–1.80 (4H, m, H-3, H-12, H-4, H-6), 1.79–1.51 (3H,
m, H-11, H-4, H-5), 1.50–1.32 (2H, m, H-5, H-11); δC (126
MHz, CDCl3) 207.5, 133.1, 129.3 (t, 2JC–F 30.5), 120.1 (dd, 1JC–F
From dienol 9c (0.303 g, 1.0 mmol) at 125 ЊC (32 hours).
Concentration and column chromatography (10% ether in light
petroleum) gave a colourless solid which recrystallised from
dichloromethane/hexane to afford 10c (0.206 g, 68%) as fine
colourless needles; mp 83–85 ЊC; Rf (10% ether in light petrol-
eum); 0.20; (Found: C, 59.68; H, 7.87; N, 4.30. Calc. for
C15H23F2NO3: C, 59.39; H, 7.64; N, 4.62%); νmax (Nujol mull)/
1
2
2
cmϪ1 1731 (C᎐O); δH (300 MHz, CDCl3) 5.70 (1H, t, 3JH–H 7.7,
257.7, JC–F 247.0), 83.2, 69.5, 40.3 (dd, JC–F 27.7, JC–F 20.9),
38.5, 29.1, 28.5, 23.1, 22.0, 20.5; δF (282 MHz, CDCl3)
᎐
H-6), 3.58–3.20 (4H, m, NCH2), 3.00–2.30 (4H, envelope, H-2,
H-10), 2.30–1.50 (8H, envelope, –CH2), 1.40–1.05 (6H, m,
2
3
2
Ϫ95.2(1F, dd, t, JF–F 239.7, JF–H 3.8), Ϫ102.8 (1F, dd, JF–F
3
2
239.7, JF–H 31.8); m/z (CI) 296 (10%, M ϩ [NH4]ϩ), 279 (8%,
–CH3); δC (75 MHz, CDCl3) 208.0, 153.2, 139.9 (t, JC–F 31.1),
[M ϩ H]ϩ), 258 (20, M Ϫ HF), 84 (100). The 1H and 13C NMR
spectrum were assigned fully by COSY, GOESY and HSQC
experiments.
123.2, 118.8 (t, 1JF–F 246.4), 42.1, 34.1 (t, 2JC–F 28.3), 45.8, 37.6,
29.1, 26.8, 22.9, 14.0; δF (282 MHz, CDCl3) Ϫ104.8 (1F, br d,
2JF–F 242.2), Ϫ111.2 (1F, br dd, 2JF–F 242.2, 3JF–H 28.0); [HRMS
(CI, [M ϩ H]ϩ) Found: 304.172661. Calc. for (C15H24O3F2N)
304.172425]; m/z (EI) 303 (15%, Mϩ), 100 (CONEt2, 100).
8,9,9-Trifluoro-14-oxa-cis-bicyclo-[8.4.0]-tetradeca-7Z-en-2-one
14b20
2-(1Ј-Chloro-2Ј,2Ј-difluoroethenyl)-1-(2Љ,3Љ-dihydropyranyl)-
cyclohexan-1-ol 13a
Was prepared from dienol 14b (0.262 g, 1.0 mmol) at 140 ЊC
(17 hours). Concentration and column chromatography (10%
ether in light petroleum) gave a white solid which recrystallised
from dichloromethane/hexane to afford 14b (0.220 g, 84%)
as colourless rhombi; mp 93–94 ЊC; Rf (10% ether in light pet-
roleum) 0.20; (Found: C, 59.68; H, 6.61. Calc. for C13H17O2F3:
t-BuLi (1.22 ml of a 1.64 M solution in n-pentane) was added
dropwise to a solution of 2,3-dihydropyran (0.18 ml, 2.0 mmol)
in dry tetrahydropyran (THP, 0.35 ml) at 0 ЊC. The mixture was
stirred for 1 h then the solution was cooled to Ϫ78 ЊC. THF
(0.25 ml) was added and the mixture was stirred for 5 minutes
further. Ketone 8a (0.275 g, 1.0 mmol) was added dropwise
over 5 minutes; the mixture was stirred at Ϫ78 ЊC for one hour
further and quenched with water (50 ml) The aqueous layer
was extracted with ethyl ether (3 × 10 ml). The combined
organic extracts were washed with saturated aqueous bicarbon-
ate solution (3 × 10 ml), brine (3 × 10 ml), dried (MgSO4),
filtered and concentrated in vacuo to afford dienol 13a as an
unstable yellow oil (0.267 g, 97% estimated yield). Only one
diastereoisomer could be detected by 19F NMR (assumed
C, 59.54; H, 6.49%); νmax (Nujol mull)/cmϪ1 1712 (C᎐O);
᎐
3
3
δH (500 MHz, CDCl3); 5.32* (1H, ddq, JH–F 34.8, JH–H
12.2, J,4JH–F 3.4, H-7), 4.16* (1H, br dd, JH–H 11.8, JH–H 5.5,
3
2
3
H-13), 3.80–3.65 (1H, m, H-1), 3.48 (1H, ddt, 3JH–H 12.7, 2JH–H
11.8, 3JH–H, JH–H 2.9, H-13), 3.10 (1H, dd, 2JH–H 19.1, 3JH–H 10.5,
4
H-3), 2.88 (1H, dddd, 3JH–F 31.6, 3JH–F 10.8, 3JH–H 5.5, 3JH–H 2.9,
H-10), 2.38–2.30 (2H, m, H-6, H-12), 2.19–2.05 (3H, m, H-3,
2
H-12, H-4), 2.05–1.95 (1H, m, H-6), 1.85* (1H, d sept., JH–H
3
4
2
3
11.5, JH–H, JH–H 2.9, H-5), 1.73* (1H, ttd, JH–H JH–H 13.7,
3JH–H 5.4, 4JH–F 1.0, H-11), 1.63–1.55 (1H, m, H-4), 1.40 (1H, d,
2JH–H 14.2, H-11), 1.35–1.23 (1H, m, H-5); δC (126 MHz, CDCl3)
207.5, 150.5 (ddd, 1JC–F 260.5, 2JC–F 38.4, 2JC–F 27.7), 117.6 (ddd,
1JC–F 252.0, 1JC–F 244.7, 2JC–F 37.9), 113.4–113.0 (m), 83.2, 69.5,
40.9 (dd, 2JC–F 27.1, 2JC–F 21.5), 38.5, 29.1, 28.5, 23.1, 22.0, 20.5;
2
trans). δF (282 MHz, CDCl3) Ϫ88.9 (1F, d, JF–F 39.4), Ϫ92.2
(1F, d, 2JF–F 39.4). The dienol was used for the subsequent step
without further purification.
2
3
δF (282 MHz, CDCl3 Ϫ101.9 (1F, dd, JF–F 255.6, JF–F 26.7),
trans-2-(1Ј,2Ј,2Ј-Trifluoroethenyl)-1-(2Љ,3Љ-dihydropyranyl)-
cyclohexan-1-ol 13b
2
3
3
Ϫ109.5 (1F, dd, JF–F 255.6, JF–F 31.6), Ϫ124.8 (1F, t, JF–H
34.8); m/z (CI) 280 (100%, [M ϩ NH4]ϩ). The 1H and 13C NMR
spectra were assigned fully by COSY, GOESY and HSQC
experiments. Gaussian and Lorentzian broadening was used to
resolve the signals indicated* fully.
Dienol 13b was prepared from 2,3-dihydropyran (0.409 ml,
4.5 mmol) and 8b (0.267 g, 1.5 mmol) as an unstable yellow oil
(0.208g, 78% estimated yield). Only one diastereoisomer could
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4430

View Article Online
Crystallographic data for 14b:† C13H17F3O2, crystal size
0.3 × 0.2 × 0.2 mm, M = 262.27, monoclinic, a = 10.1391(5), b =
8.2152(4), c = 14.6853(5) Å, β = 96.703(3) deg, U = 1214.85(9)
Å3, T = 150(2) K, space group P21/c, Z = 4, µ(Mo–Kα) = 0.125
mmϪ1, 9829 reflections measured, 2130 unique (Rint = 0.0605)
which were used in all calculations. Final R indices [F 2>2σ(F 2)]
R1 = 0.0352, wR2 = 0.0858; R indices (all data) R1 = 0.0503,
wR2 = 0.0933.
in petroleum ether) 0.2; νmax (film)/cmϪ1 3428 (OH), 1763
(C᎐O), 1719 (C᎐C); δ (300 MHz, CDCl3) 3.40–3.23 (4H, m,
᎐
᎐
H
NCH2), 2.90 (1H, s, OH ), 2.50 (1H, s, H-2), 2.21–1.29 (8H, m,
–CH2), 1.25–1.05 (6H, m, NCH2CH3); δC (75 MHz, CDCl3)
1
1
155.0 (t, JC–F 284.3), 153.7 (d, JC–F 292.7), 152.2, 112.3 (dd,
2
2
2
2J
40.1, J
15.3), 99.0 (dd, J
33.9, J
16.4), 73.4,
C–F
C–F
C–F
C–F
41.8, 40.6, 33.1, 25.4, 21.8, 20.7, 13.2; δF (282 MHz, CDCl3)
Ϫ80.5 (1F, d, 2JF–F 40.7), Ϫ84.2 (1F, d, 2JF–F 40.7), Ϫ95.4 (1F, d,
2
2JF–F 54.7), Ϫ107.5 (1F, d, JF–F 54.7); m/z (ES) 396.1 (100%,
[M ϩ Na]ϩ).
8-(N,N-Diethylcarbamoyloxy)-9,9-difluoro-14-oxa-bicyclo-
[8.4.0]-tetradeca-7-en-2-one 14c
cis- and trans-1-(1Ј,2Ј,2Ј-Trifluoroethenyl)-2-(1Љ-chloro-2Љ,2Љ-
difluoroethenyl)-cyclohexan-1-ol 22a
Was prepared from dienol 13c (0.359 g, 1.0 mmol) at 155 ЊC
(9 hours). Concentration and column chromatography (10%
ether in light petroleum) gave a white solid which recrystallised
from dichloromethane/hexane to afford 14c (0.222 g, 62%) as
colourless rhombi; mp 139–141 ЊC; Rf (10% ether in light
petroleum) 0.3; (Found: C, 60.09; H, 7.75; N, 3.93. Calc. for
C18H27O4F2N: C, 60.20; H, 7.50; N 3.90%); νmax (Nujol mull)/
cmϪ1 1738, 1707 (C᎐O); δ (500 MHz, CDCl3); 5.80 (1H, dd,
1,2,2-Trifluoro-1-lithioethene 6b (1.5 mmol) was generated
according to the procedure used for 7b, and ketone 8a (0.292 g,
1.5 mmol) was added dropwise over 5 minutes. The reaction
was stirred at Ϫ78 ЊC for three hours and quenched by the
addition of saturated aqueous ammonium chloride solution
(20 ml). The aqueous layer was extracted with diethyl ether
(3 × 10 ml). The combined organic extracts were washed with
saturated sodium bicarbonate solution (2 × 15 ml) brine (2 ×
15 ml), dried (MgSO4), filtered and concentrated in vacuo to
leave a yellow oil. Purification by column chromatography (10%
diethyl ether in petroleum ether) afforded 22a (0.24 g, 57% total
yield) as a mixture for diastereoisomers (2.3 : 1) as a yellow
oil. Major diastereoisomer; Rf (10% diethyl ether in petrol-
eum ether) 0.2 (0.165 g, 40%); νmax (film)/cmϪ1 3432 (OH),
᎐
H
3JH–H 11.0, JH–H 2.6, H-7), 4.08 (1H, dd, JH–H 10.7, JH–H
5.5, H-13), 3.71–3.67 (1H, m, H-1), 3.50–3.28 (5H, m, H-13,
2 × NCH2), 3.17–3.00 (1H, dd, 2JH–H 19.1, 3JH–H 9.9, H-3), 2.78
3
2
3
3
2
(1H, d, JF–H 30.9, H-10), 2.33 (1H, d, JH–H 14.0, H-6), 2.20–
1.89 (4H, m, H-3, H-12, H-4), 1.86–1.48 (4H, m, H-11, H-4,
H-5, H-6), 1.42–1.28 (2H, m, H-11, H-5), 1.25–1.02 (6H, m;
2 × –CH3); δC (126 MHz, CDCl3) 207.5, 152.5, 141.0 (dd,
2JC–F 33.9, 2JC–F 27.7), 123.5, 119.0 (dd, 1JC–F 255.4, 1JC–F 245.3),
2
2
1788 (C᎐C), 1725 (C᎐C); δ (300 MHz, CDCl3) 2.82 (1H, d,
᎐
᎐
83.2, 69.5, 42.0, 40.3 (dd, JC–F 28.3, JC–F 20.3), 38.5, 29.5,
26.5, 23.0, 22.0, 20.2, 13.8; δF (282 MHz, CDCl3) Ϫ92.4 (1F, d,
H
3
3JH–H 12.5, H-2), 2.45 (1H, s, OH ), 2.10 (1H, d, JH–H 12.5,
2JF–F 246.0), Ϫ108.4 (1F, dd, JF–F 246.0, JF–H 31.8); [HRMS
(CI, [M ϩ H]ϩ) Found: 360.199311. Calc. for C18H28O4F2N:
360.198640]; m/z (CI) 360 (7%, [M ϩ H]ϩ), 279 (12), 242 (13),
2
3
H-3ax), 1.90–1.52 (5H, m, –CH2), 1.48–1.19 (2H, m, –CH2);
1
1
δC (75 MHz, CDCl3) 154.2 (t, JC–F 289.3), 154.1 (ddd, JC–F
1
2
1
2
288.2, JC–F 282.0, JC–F 47.5), 131.8 (ddd, JC–F 240.7, JC–F
36.7, JC–F 14.7), 92.4 (dd, JC–F 41.2, JC–F 17.5), 73.7 (d, JC–F
22.0), 41.5, 37.1, 29.5, 25.5, 20.2; δF (282 MHz, CDCl3) Ϫ86.1
2
2
2
2
217 (10), 173 (14), 100 (CONEt2, 100). The H and 13C NMR
1
spectra were assigned fully by COSY, GOESY and HSQC
experiments.
2
2
(1F, d, JF–F 39.4), Ϫ90.7 (1F, d, JF–F 39.4), Ϫ100.2 (1F, dd,
2JF–F 84.6, 3J cFi–sF 35.0), Ϫ115.4 (1F, dd, 3J tFr–aFns 113.9, 2JF–F 84.6),
trans
cis
3
3
4
Ϫ178.7 (1F, ddd, J F–F 113.9 J F–F 35.0, JF–H 7.6); m/z
(CI) 294 (100%, M ϩ [NH4]ϩ), 240 (15, M Ϫ HCl), 220 (12,
M Ϫ HCl, HF); minor diastereoisomer Rf (10% diethyl ether in
petroleum ether) 0.1 (0.069 g, 0.24 mmol, 17%); νmax (film)/cmϪ1
trans-1-(1Ј-Chloro-2Ј,2Ј-difluoroethenyl)-2-(1Љ,2Љ,2Љ-trifluoro-
ethenyl)-cyclohexan-1-ol 21b
Was prepared from 6a (1.5 mmol), generated as for 7a and
ketone 8b (0.267 g, 1.5 mmol). After addition, the mixture was
left stirring at Ϫ78 ЊC one hour, allowed to warm to Ϫ30 ЊC for
30 minutes then quenched by the addition of saturated aqueous
ammonium chloride solution (20 ml). Work up in the usual
manner followed by column chromatography (5% diethyl ether
in petroleum ether) afforded 21b (assumed trans) (0.191g, 46%)
as a yellow oil; Rf (5% diethyl ether in petroleum ether) 0.1;
3464, 1724 (C᎐C); δ (300 MHz, CDCl3) 2.95–2.80 (1H, m,
᎐
H
H-2), 2.35–2.18 (1H, m, OH ), 2.09–1.05 (8H, m, –CH2);
2
δF (282 MHz, CDCl3) Ϫ85.3 (1F, d, J 36.5), Ϫ90.1 (1F, d,
2
3
2J 36.5), Ϫ100.2 (1F, dd, J 83.9, Jcis 35.0), Ϫ115.5 (1F, dd,
2J 83.9, Jtrans 113.7), Ϫ178.7 (1F, ddd, Jtrans 113.7, Jcis 35.0,
4JF–H 7.6); m/z (CI) 294 (35%, M ϩ [NH4]ϩ), 183 (30), 165 (52),
154 (73). The dienol was used for the subsequent step without
further characterisation.
3
3
3
3
3
δH (300 MHz, CDCl3) 2.78 (1H, br dd, JH–F 31.6, JH–H 12.5,
H-2), 2.28 (1H, s, OH ), 2.20–1.50 (6H, m, –CH2), 1.48–1.15
1
(2H, m, –CH2); δC (75 MHz, CDCl3) 153.9 (ddd, JC–F 326.6,
1,2-Di-(1Ј,2Ј,2Ј-trifluoroethenyl)-cyclohexan-1-ol 22b
2
1
1
1JC–F 275.8, JC–F 47.5), 153.8 (dd, JC–F 293.3, JC–F 287.7),
1
2
2
2
Was prepared as for 22a from 6b (1.5 mmol) and ketone 8b
(0.267 g, 1.5 mmol). Work up in the usual manner followed by
column chromatography (10% diethyl ether in petroleum ether)
afforded 22b (0.242 g 62% estimated yield) as an inseparable
mixture of diastereoisomers (1.9 : 1, see text) Rf (10% diethyl
ether in petroleum ether) 0.23; δH (300 MHz, CDCl3) 3.75–3.65
(1H, m, OH ), 2.91–2.55 (3H, m, –CH2), 2.42–1.15 (5H, m,
–CH2); δC (75 MHz, CDCl3) 158.1–149.8 (m), 133.9–
127.8 (m), 73.2–71.8 (m), 44.0–41.0 (m), 35.8, 24.5, 23.0, 19.5;
δF (282 MHz, CDCl3) major diastereoisomer); Ϫ99.7 (dd,
130.2 (ddd, JC–F 235.7, JC–F 51.6, JC–F 15.8), 97.3 (dd, JC–F
27.7, 2JC–F 15.8), 75.5, 40.5 (d, 2JC–F 18.1), 35.8, 34.7, 34.2, 20.6;
2
δF (282 MHz, CDCl3) Ϫ79.9 (1F, d, JF–F 43.2), Ϫ85.4 (1F, d,
2JF–F 43.2), Ϫ102.8 (1F, dd, 2JF–F 80.1, 3J cFi–sF 31.8), Ϫ120.6 (1F,
trans
trans
cis
3
2
3
3
dd, J F–F 111.9, J 80.1), Ϫ181.2 (1F, dtd, J F–F 111.9, J F–F
31.8, JF–H 31.8, JF–H 5.1); [HRMS (EI, Mϩ) Found:
276.034783. Calc. for C10H1035ClF5O 276.034034]; m/z (ES) 276
(75%, Mϩ), 256 (65, M Ϫ HF), 236 (60, M Ϫ H2F2), 221 (100,
M Ϫ HF, Cl).
3
4
cis
cis
3
2
3
2JF–F 83.6, J F–F 34.9), Ϫ102.8 (dd, JF–F 79.3, J F–F 32.5),
trans-1-(1Ј-Chloro-2Ј,2Ј-difluoroethenyl)-2-[1Љ-(N,N-di-
ethylcarbamoyloxy)-2Љ,2Љ-difluoroethenyl)-cyclohexan-1-ol 21c
trans
trans
3
2
3
Ϫ115.1 (dd, J F–F 114.1, JF–F 83.6), Ϫ120.9 (ddt, J F–F 111.9,
2JF–F 80.1, JH–F 4.0); minor diastereoisomer Ϫ99.4 (dd, JF–F
4
2
Was prepared from 6a (1.5 mmol) and ketone 8c (0.412 g,
1.5 mmol) which were stirred at Ϫ78 ЊC for three hours before
being quenched with saturated aqueous ammonium chloride
solution (20 ml). Work up in the usual manner followed by
column chromatography (15% diethyl ether in petroleum ether)
afforded 21c (0.347 g, 62%) as a yellow oil; Rf (15% diethyl ether
78.2, 3J cFi–sF 34.0), Ϫ121.4 (ddt, 3J Ftr–aFns 111.6, 2JF–F 78.2, 4JH–F 5.0),
cis
trans
2
3
3
Ϫ103.2 (dd, JF–F 77.6, J F–F 32.0), Ϫ113.1 (dd, J F–F 111.3,
2JF–F 77.6); both diastereoisomers Ϫ180.8 (app. dt, 2JF–F 113,6,
3J cFi–sF 32.0); m/z (CI) 240 (60%, M ϩ 1-HF), 149 (100%). The
dienol was used for the subsequent step without further
characterisation.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4431

View Article Online
cis- and trans-1-(1Ј,2Ј,2Ј-Trifluoroethenyl)-2-[1Ј-(N,N-diethyl-
carbamoyloxy)-2Ј,2Ј difluoroethenyl]-cyclohexan-1-ol 22c
colourless blocks; mp 99–102 ЊC; (Found: C, 48.44; H, 5.62; N,
3.58. Calc. For C15H20ClF4NO3: C, 48.20; H, 5.39; N, 3.75%); Rf
(10% ether in light petroleum) 0.2; νmax (Nujol mull)/cmϪ1 1722
Was prepared as for 22a from 6b (1.5 mmol) and ketone 8c
(0.412 g, 1.5 mmol). Work up in the usual manner followed by
column chromatography (10% diethyl ether in petroleum ether)
afforded 22c (0.300 g, 56% total yield) as a mixture of
diastereoisomers (1.8 : 1), as a yellow oil. The two diastereo-
isomers were separated and characterised: major diastereo-
isomer (assumed trans) (0.193 g, 36%); Rf (10% diethyl ether in
petroleum ether) 0.25: δH (300 MHz, CDCl3) 4.10 (1H, s, OH ),
(C᎐O); δ (300 MHz, CDCl3) 6.02–5.88 (1H, m, H-6), 4.52–
᎐
H
4.38 (1H, m, H-2), 3.47–3.19 (4H, m, 2 × NCH2), 3.00–2.75
(1H, m, H-10), 2.74–2.47 (1H, m, H-10), 2.29–2.01 (2H, m,
H-7), 1.99–1.62 (3H, m, –CH2), 1.52–1.30 (1H, m, –CH2); 1.28–
1.10 (6H, m, 2 × –CH3); δC (126 MHz, CDCl3) 194.5, 151.4,
136.3–135.6 (m), 127.9, 115.8–109.0 (m), 57.6–56.0 (m), 42.4,
42.0, 37.9, 26.6, 26.0, 20.9, 13.9, 13.2; δF (282 MHz, CDCl3)
Ϫ108.3 (1F, d, 2JF–F 271.2), Ϫ119.2 (1F, d, 2JF–F 290.0); Ϫ123.9
3
3.42–3.20 (4H, m, 2 × NCH2), 2.62 (1H, d, JH–H 12.1, H-2),
2
2
(1F, d, JF–F 271.2), Ϫ126.8 (d, JF–F 290.0); m/z 396.5 (100%,
2.14–2.08 (1H, m, H-3), 1.98–1.30 (6H, m, –CH2), 1.20–1.08
(6H, m, 2 × –CH3); δC (75 MHz, CDCl3) 159.5–150.1 (m), 134.2
[M ϩ Na]ϩ).
Crystallographic data for 23c:† C15H20ClF4NO3, crystal size
0.4 × 0.3 × 0.2 mm, M = 373.77, monoclinic, a = 11.1981(5), b =
11.8463(7), c = 13.1682(6) Å, β = 94.524(3) deg, U = 1741.40(15)
Å3, T = 150(2) K, space group P21/n, Z = 4, µ(Mo–Kα) =
1
2
2
2
(ddd, JC–F 245.8, JC–F 35.6, JC–F 15.8), 111.1 (dd, JC–F 40.7,
2JC–F 15.3), 71.0 (2JC–F 19.2), 42.0, 57.5, 39.5, 28.0, 26.5,
24.1, 13.5; δF (282 MHz, CDCl3) Ϫ94.05 (1F, d, 2J 53.4), Ϫ107.3
cis
2
2
3
(1F, d, JF–F 53.4), Ϫ101.2 (1F, dd, JF–F 76.3, J F–F 35.6),
Ϫ111.5 (1F, dd, 3J tFra–nFs 111.9 2JF–F 76.3), Ϫ168.6 (1F, br d, 3J tFra–nFs
111.9 (cis coupling invisible due to signal breadth); [HRMS
(CI, [M ϩ H]ϩ) Found: 358.14518. Calc. for [C15H20F5NO3]
358.14416; m/z (CI) 358 (37%, [M ϩ H]ϩ), 340 (100%,
M Ϫ OH): and minor diastereoisomer (0.107 g, 20%); Rf (10%
diethyl ether in petroleum ether) 0.10: δH (300 MHz, CDCl3)
4.55 (1H, s, OH ), 3.42–3.20 (4H, m, 2 × NCH2), 2.85–2.75 (1H,
m, H-2), 1.90–1.25 (8H, m, –CH2), 1.20–1.09 (6H, m, 2 ×
NCH2CH3); δC (75 MHz, CDCl3) 159.5–150.1 (m), 134.2, 111.1,
55.5, 65.6, 42.3, 34.7, 25.5, 22.2, 19.9, 13.5; δF (282 MHz,
0.272 mmϪ1, 14497 reflections measured, 3067 unique (Rint
=
0.0583) which were used in all calculations. Final R indices
[F 2 > 2σ(F 2)] R1 = 0.0416, wR2 = 0.1052; R indices (all data)
R1 = 0.0643, wR2 = 0.1154.
5-Chloro-2,3,3,4,4-pentafluoro-cyclodec-5Z-en-1-one 24a
Was prepared from dienol 22a (0.276 g, 1.0 mmol) at 150 ЊC
(1.75 hours). Concentration and column chromatography (10%
ether in light petroleum) gave a white solid which recrystallised
from dichloromethane/hexane to afford 24a (0.201 g, 73%) as
colourless blocks; mp 92–98 ЊC; Rf (10% ether in light petrol-
eum) 0.1; δH (300 MHz, CDCl3) 6.55–6.40 (1H, m, H-6), 4.88
(1H, dtd, 2JH–F 46.7, 3JH–F 9.2, 3JH–FC 9.2, 4JH–F 4.4, H-2), 3.15–
2.98 (1H, m, H-10), 2.60–2.45 (1H, m, H-10), 2.38–2.20 (2H, m,
H-7), 2.10–1.18 (4H, m, –CH2); δC (75 MHz, CDCl3) 199.0,
2
2
CDCl3) Ϫ95.05 (1F, d, JF–F 52.1), Ϫ100.6 (1F, dd, JF–F 84.5,
cis
3J F–F 33.7), Ϫ105.9 (1F, ddd, JF–F 52.1, JF–H 19.1, JF–H, 3.1),
2
4
5
Ϫ115.6 (1F, dd, 3J tFra–nFs 113.2, 2JF–F 84.5), Ϫ176.8 (1F, ddd, 3J tFra–nFs
cis
113.2, J F–F 33.7, JF–H 19.1); [HRMS (CI, [M ϩ H]ϩ) Found:
358.143378. Calc. for C15H20F5NO3 358.14416]; m/z (CI) 358
(10%, [M ϩ H]ϩ), 340 (100%, M Ϫ OH).
3
4
1
2
129.9–128.0 (m), 90.2 (dt, JC–F 199.5, JC–F 24.4), 42.5, 38.0,
29.5, 20.5; distinct resonances for ring methylene carbons were
2
not observed; δF (282 MHz, CDCl3) Ϫ105.8 (1F, br d, JF–F
2-Chloro-3,3,4,4,5-pentafluoro-cyclodec-5Z-en-1-one 23b
291.1), Ϫ107.4 (1F, d, 2JF–F 291.1); Ϫ107.8 (1F, d, 2JF–F 258.8),
2
A solution of dienol 21b (0.276 g, 1.0 mmol) in xylene (5 ml)
was sealed in an Ace® tube and heated to 110 ЊC in an oil bath.
The reaction was followed by 19F NMR of aliquots until the
starting material was consumed completely (40 hours). Concen-
tration and column chromatography (10% ether in petroleum
ether) then recrystallisation afforded 23b (0.191 g, 69%) as
colorless blocks: mp 82–83 ЊC; νmax (Nujol mull)/cmϪ1 1721
Ϫ111.0 (1F, d, JF–F 258.8), (Ϫ199.1)–(Ϫ199.7) (1F, m); m/z
(CI) 294 (2.5%, [M ϩ NH4]ϩ), 167 (100%). Neither satis-
factory microanalysis nor HRMS could be obtained for this
compound.
Crystallographic data for 24a:† C10H10ClF5O, crystal size
0.2 × 0.2 × 0.3 mm, M = 276.63, orthorhombic, a = 7.979(1),
b = 9.742(1), c = 14.053(2) Å, U = 1092.4(2) Å3, T = 150(2) K,
space group P2(1)2(1)2(1), Z = 4, µ(Mo–Kα) = 0.399 mmϪ1
,
(C᎐O); δ (300 MHz, CDCl3) 5.67–5.30 (1H, m, H-6), 4.60–
᎐
H
4.29 (1H, m, H-2), 3.14–2.72 (1H, m, H-10), 2.70–2.23 (2H, m,
H-10, –CH2), 2.20–1.63 (4H, m, –CH2), 1.61–1.18 (1H, m,
6657 reflections measured, 2190 unique (Rint
= 0.0334)
which were used in all calculations. Final R indices [F 2
>
2
2σ(F 2)] R1 = 0.0283, wR2 = 0.0598; R indices (all data)
R1 = 0.0334, wR2 = 0.0621.
–CH2); δC (75 MHz, CDCl3) 196.0, 117.2, 61.1 (t, JC–F 20.3),
56.8, 23.7, 23.5, 20.7; δF (376 MHz, PhMe-d8, 213K) Ϫ106.4
2
3
3
2
(1F, dt, JF–F 256.5, JF–F, JF–H 20.3), Ϫ118.2 (1F, br d, JF–F
2
3
266.5); Ϫ121.8 (1F, d, JF–F 256.5), Ϫ123.2 (br t, JF–F 27.3),
Ϫ124.8 (1F, dd, 2JF–F 266.5, 3JF–F 19.0); [HRMS (CI, [M ϩ H]ϩ)
Found: 276.035289. Calc. for C10H10ClF5O 276.034034];
m/z 276 (28%, Mϩ). Satisfactory microanalysis could not be
obtained for this compound.
2,3,3,4,4,5-Hexafluoro-cyclodec-5Z-en-1-one 24b
Was prepared from dienol 22b (0.260 g, 1.0 mmol) as mixture of
two diastereoisomers (1.9 : 1 ratio) at 100 ЊC (47 hours). Con-
centration and column chromatography (10% ether in light
petroleum) gave a white solid which recrystallised from di-
chloromethane/hexane to afford 24b (0.140 g, 54%) as colorless
blocks; mp 46–49 ЊC; Rf (10% ether in light petroleum) 0.1;
Crystallographic data for 23b:† C10H10ClF5O, crystal size
0.3 × 0.3 × 0.2 mm, M = 276.63, monoclinic, a = 8.7576(4),
b = 13.7337(10), c = 9.8743(4) Å, β = 106.530(2) deg, U =
1138.54(11) Å3, T = 150(2) K, space group Pc, Z = 4, µ(Mo–Kα)
νmax (Nujol mull)/cmϪ1 1724 (C᎐O); δH (500 MHz, CDCl3) 5.57
᎐
(1H, ddd, JH–F 32.7, 3J 7.4, 2.9, H-6), 4.90 (1H, app. ddt,
3
= 0.383 mmϪ1, 12683 reflections measured, 3664 unique (Rint
=
3JH–F 46.5, 3JH–F 9.0, 3JH–F 4.8, 4JH–F 4.8 H-2), 3.20–2.80 (1H, m,
0.0378) which were used in all calculations. Final R indices
[F 2 > 2σ(F 2)] R1 = 0.0272, wR2 = 0.0594; R indices (all data)
R1 = 0.0272, wR2 = 0.0604.
3
H-10a), 2.50–2.37 (1H, m, H-7a), 2.30 (1H, dd, J 18.9, 9.2,
H-9a), 2.16–1.98 (2H, m, H-7b, H-10b), 1.96–1.84 (1H, m,
H-8a), 1.77–1.68 (1H, m, H-9b), 1.46–1.32 (1H, m, H-8b);
2
1
δC (75 MHz, CDCl3) 199.2 (d, JC–F 26.4), 146.1 (ddd, JC–F
2-Chloro-3,3,4,4-tetrafluoro-5-(N,N-diethylcarbamoyloxy)-
cyclodec-5Z-en-1-one 23c
2
2
1
262.2, JC–F 33.9, JC–F 26.6), 116.8–116.1 (m), 90.2 (dt, JC–F
2
201.8, JC–F 26.6); 57.5, 38.4, 23.8, 20.4; δF (282 MHz, CDCl3)
2
2
Was prepared from dienol 21c (0.373 g, 1.0 mmol) at 150 ЊC
(2 hours). Concentration and column chromatography (10%
ether in light petroleum) gave a white solid which recrystallised
from dichloromethane/hexane to afford 23c (0.265 g, 71%)
Ϫ113.4 (1F, br d, JF–F 287.4), Ϫ116.3 (1F, br d, JF–F 287.4);
2 2
Ϫ120.1 (1F, br d, JF–F 277.2), Ϫ124.1 (1F, d, JF–F 277.2),
(Ϫ127.1)–(Ϫ127.5) (1F, m), (Ϫ205.6)–(Ϫ206.1) (1F, m);
[HRMS (CI [M ϩ NH4]ϩ) Found: 278.096651. Calc. for
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4432

View Article Online
C10H14F6NO 278.097959]; m/z (CI) 278 (50%, M ϩ [NH4]ϩ).
114.0, 73.6, 53.1, 29.1, 27.0, 25.1, 23.3; m/z (CI) 152 (5%, Mϩ),
151 (7%, M Ϫ 1).
The 1H NMR spectrum was assigned partially by COSY analy-
1
sis and by comparison with the fully-assigned H NMR spec-
trum for 14b. Satisfactory microanalysis could not be obtained
for this compound.
Cyclodec-5-en-1-one 26
A solution of dienols 25 (0.152 g, 1.0 mmol) in xylene (5 ml)
was sealed in an Ace® tube and heated to 225 ЊC in an oil bath.
The reaction was followed by H NMR of aliquots until the
Crystallographic data for 24b:† C10H10F6O, crystal size 0.3 ×
0.3 × 0.2 mm, M = 260.18, triclinic, a = 5.996(1), b = 8.007(2),
c = 11.386(2) Å, α = 80.90(3) deg, β = 87.03(3) deg, γ = 78.66(3),
U = 529.10(18) Å3, T = 150(2) K, space group P-1, Z = 2,
µ(Mo–Kα) = 0.175 mmϪ1, 6510 reflections measured, 2053
unique (Rint = 0.0578) which were used in all calculations. Final
R indices [F 2 > 2σ(F 2)] R1 = 0.0425, wR2 = 0.1189; R indices
(all data) R1 = 0.0493, wR2 = 0.1248.
1
starting material was consumed completely (24 hours). Concen-
tration and column chromatography (10% ether in petroleum
ether) afforded 124 (0.074 g, 49%) as colourless oil; νmax (film)/
cmϪ1 1708 (C᎐O); δ (300 MHz, CDCl3) 5.39–5.20 (1H,
᎐
H
m, H-6), 5.18–4.98 (1H, m, H-5), 2.58–1.08 (14H, –CH2),
δC (75 MHz, CDCl3) 212.9, 134.4, 131.2, 45.5, 43.1, 34.1, 33.3,
28.7, 27.9, 22.1; m/z 170 (18%, M ϩ NH4ϩ), 153 (25%, M ϩ 1),
identical with the data reported by Marvell and Whalley.46
5-(N,N-Diethylcarbamoyloxy)-2,3,3,4,4-pentafluoro-cyclodec-
5Z-en-1-one 24c
Was prepared from dienol 22c (0.357 g, 1.0 mmol) at 100 ЊC
(30 hours). Concentration and column chromatography (10%
ether in light petroleum) gave a white solid which recrystallised
from dichloromethane/hexane to afford 24c (0.234 g, 66%) as
colourless rods; mp 105 ЊC; Rf (10% ether in light petroleum)
0.15; (Found: C, 50.40; H, 5.60; N, 3.80; Calc. For C15H19-
NO3F2: C, 50.40; H, 5.60; N, 3.90%); νmax (Nujol mull)/cmϪ1
Acknowledgements
We thank the Engineering and Physical Sciences Research
Council of the UK for project grants GR/K88293 (studentship
to G. D.) and 98XR20 (X-ray crystallographic support).
Notes and references
1731 (C᎐O); δ (300 MHz, CDCl3) 6.10–5.98 (1H, m, H-6),
᎐
H
2
3
3
4.88 (1H, ddd, JH–F 47.1, JH–F 9.2, JH–F 4.8, H-2), 3.48–3.20
(4H, m, NCH2), 3.20–3.02 (1H, m, H-10), 2.40–1.52 (6H, m,
–CH2), 1.50–1.30 (1H, m, –CH2), 1.27 (3H, t, J 7.0, CH3),
1.15 (3H, t, J 7.0, CH3); δC (75 MHz, CDCl3) 199.0, 151.8,
128.3, 90.4 (dd, JC–F 202.3, JC–F 28.2), 41.9, 37.9, 27.1, 26.2,
1 J. A. Berson and M. Jones, J. Am. Chem. Soc., 1964, 86, 5019–5020.
2 For an extensive review of oxy-Cope and related rearrangements,
see: H. Frauenrath, in Stereoselective Synthesis, ed. G. Helmchen,
R. W. Hoffmann, J. Mulzer, and E. Schaumann, Stuttgart, 1996,
vol. 6, pp. 3547––3688.
3 P. Heimbach, Angew. Chem., Int. Ed. Engl., 1964, 3, 702–703.
4 D. L. J. Clive, P. W. M. Sgarbi, X. He, S. Y. Sun, J. H. Zhang and
L. G. Ou, Can. J. Chem., 2003, 81, 811–824.
5 D. A. Evans and A. M. Golob, J. Am. Chem. Soc., 1975, 97, 4765–
4766.
6 L. A. Paquette, Tetrahedron, 1997, 53, 13971–14020.
7 L. Gentric, I. Hanna and L. Ricard, Org. Lett., 2003, 5, 1139–1142.
8 J. M. Yu, T. Wang, X. Y. Z. Wearing, J. Ma and J. M. Cook, J. Org.
Chem., 2003, 68, 5852–5859.
9 L. A. Paquette, R. Guevel, S. Sakamoto, I. H. Kim and J. Crawford,
J. Org. Chem., 2003, 68, 6096–6107.
10 J. Yang, Y. O. Long and L. A. Paquette, J. Am. Chem. Soc., 2003,
125, 1567–1574.
11 H. Miyaoka, T. Baba, H. Mitome and Y. Yamada, Tetrahedron Lett.,
2001, 42, 9233–9236.
12 J. T. Njardarson and J. L. Wood, Org. Lett., 2001, 3, 2431–2434.
13 W. R. Dolbier, A. C. Alty and O. Phanstiel, J. Am. Chem. Soc., 1987,
109, 3046–3050.
3
3
1
2
2
20.3, 13.5; δF (282 MHz, CDCl3) Ϫ112.1 (1F, d, JF–F 286.1),
Ϫ115.3 (1F, d, 2JF–F 286.1); Ϫ118.1 (1F, d, 2JF–F 267.0), Ϫ123.3
2
3
(1F, dt, JF–F 267.0, JF–F 19.0), (Ϫ204.4)–(Ϫ204.8) (1F, m);
[HRMS (CI, [M ϩ H]ϩ) Found: 357.137280. Calc. for
C15H20F5NO3 357.136335]; m/z 358 (100%, M ϩ [NH4]ϩ), 375
(90%, [M ϩ H]ϩ).
Crystallographic data for 24c:† C15H20F5NO3, crystal size
0.3 × 0.3 × 0.1 mm, M = 357.32, monoclinic, a = 11.1132(2),
b = 11.0996(2), c = 13.8507(3) Å, β = 102.8902(12) deg, U =
1665.46(6) Å3, T = 150(2) K, space group P21/n, Z = 4, µ(Mo–
Kα) = 0.134 mmϪ1, 29759 reflections measured, 3396 unique
(Rint = 0.0446) which were used in all calculations. Final R
indices [F 2 > 2σ(F 2)] R1 = 0.0344, wR2 = 0.0800; R indices (all
data) R1 = 0.0454, wR2 = 0.0854.
14 W. E. Doering, V. G. Toscano and G. H. Beasley, Tetrahedron, 1971,
27, 5299–5306.
cis- and trans-1,2-Divinyl-cyclohexan-1-ols 25
15 S. Y. Wang and W. T. Borden, J. Am. Chem. Soc., 1989, 111, 7282–
7283.
Vinylmagnesium bromide (1.0 ml of a 1 M solution in THF,
1.0 mmol) was added dropwise over 10 minutes to a solution of
2-chlorocyclohexanone (0.132 g, 1.0 mmol) in dry THF (0.5 ml)
at 0 ЊC. The mixture was stirred for two hours, then further
vinylmagnesium bromide (1.0 ml of a 1 M solution in THF,
1.0 mmol) was added and the mixture was heated to reflux
(63 ЊC),45 stirred for three hours at this temperature and
quenched carefully with saturated aqueous ammonium chloride
solution (20 ml). The aqueous layer was extracted with diethyl
ether (3 × 10 ml) and the combined organic extracts were
washed with brine (2 × 15 ml), dried (MgSO4), filtered and
concentrated in vacuo. Purification by column chromatography
(10% diethyl ether in petroleum ether) afforded 25 as a mixture
of two diastereoisomers (0.1 g, 4.1 : 1, 73%). Major diastereo-
isomer: δH (300 MHz, CDCl3) 5.99–5.80 (2H, m, H-1Ј, H-1Љ),
5.28–4.92 (4H, m, H-2Ј, H-2Љ), 2.50–2.35 (1H, m, OH ), 2.18–
2.02 (1H, m, H-2), 2.90–1.38 (7H, m, –CH2), 1.37–1.15 (1H,
m, –CH2); δC (75 MHz, CDCl3) 146.0, 138.9, 116.5, 112.2,
73.5, 42.0, 39.1, 25.3, 22.8; m/z (CI) 152 (12%, Mϩ), 151 (14%,
M Ϫ 1); Minor diastereoisomer δH (300 MHz, CDCl3) 6.30–
6.16 (1H, m, H-1Ј), 5.69–5.32 (1H, m, H-1Љ), 5.33–4.90 (4H, m,
H-2Ј-2Љ), 2.82–2.70 (1H, m, OH ), 2.45–2.20 (1H, m, H-2), 2.15–
1.15 (8H, m, –CH2); δC (75 MHz, CDCl3) 140.0, 138.8, 117.0,
16 K. J. Shea and R. B. Phillips, J. Am. Chem. Soc., 1980, 102, 3156–
3162.
17 W. R. Dolbier and K. W. Palmer, J. Am. Chem. Soc., 1993, 115,
9349–9350.
18 W. R. Dolbier, K. W. Palmer, F. Tian, P. Fiedorow, A. Zaganiaczyk
and H. Koroniak, Collect. Czech. Chem. Commun., 2002, 67, 1517–
1532.
19 F. Haeffner, K. N. Houk, S. M. Schulze and J. K. Lee, J. Org. Chem.,
2003, 68, 2310–2316 describes recent treatments of anionic
sigmatropic rearrangements using DFT methods in which biradical
pathways appear significant. For an experimental observation of
biradical character in the Cope rearrangement of a perhalogenated
hexadiene, see : N. Y. Jing and D. M. Lemal, J. Am. Chem. Soc.,
1993, 115, 8481–8482.
20 G. Dimartino, T. Gelbrich, M. B. Hursthouse, M. E. Light,
J. M. Percy and N. S. Spencer, Chem. Commun., 1999, 2535–2536
describes the preliminary publication of these results.
21 L. R. Cox, G. A. DeBoos, J. J. Fullbrook, J. M. Percy, N. S. Spencer
and M. Tolley, Org. Lett., 2003, 5, 337–339.
22 G. A. DeBoos, J. J. Fullbrook and J. M. Percy, Org. Lett., 2001, 3,
2859–2861.
23 G. A. DeBoos, J. J. Fullbrook, W. M. Owton, J. M. Percy and
A. C. Thomas, Synlett, 2000, 963–966.
24 S. T. Patel, J. M. Percy and R. D. Wilkes, Tetrahedron, 1995, 51,
11327–11336.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4433

View Article Online
25 S. T. Patel, J. M. Percy and R. D. Wilkes, J. Org. Chem., 1996, 61,
166–173.
26 M. J. Broadhurst, S. J. Brown, J. M. Percy and M. E. Prime, J. Chem.
Soc., Perkin Trans. 1, 2000, 3217–3226.
27 G. Dimartino and J. M. Percy, Chem. Commun., 2000, 2339–
2340.
36 A. J. Mancuso, S.-L. Huang and D. Swern, J. Org. Chem., 1978, 43,
2480–2482.
37 R. C. Gadwood and R. M. Lett, J. Org. Chem., 1982, 47, 2268–2275.
38 E. C. Ashby and J. T. Laemmle, Chem. Rev., 1975, 75, 521–546.
39 M. Shimano and A. I. Meyers, Tetrahedron Lett., 1994, 35, 7727–
7730.
28 W. Yuan, R. J. Berman and M. H. Gelb, J. Am. Chem. Soc., 1987,
109, 8071–8081.
29 F. Tellier, M. Audouin, M. Baudry and R. Sauvêtre, Tetrahedron
Lett., 1998, 39, 5041–5044.
40 J. Stonehouse, P. Adell, J. Keeler and A. J. Shaka, J. Am. Chem. Soc.,
1994, 116, 6037–6038.
41 J. M. Warrington, G. P. A. Yap and L. Barriault, Org. Lett., 2000, 2,
663–665.
30 J. M. Bainbridge, S. J. Brown, P. N. Ewing, R. R. Gibson and J. M.
Percy, J. Chem. Soc., Perkin Trans. 1, 1998, 2541–2545.
31 J. Burdon, P. L. Coe, I. B. Haslock and R. L. Powell, Chem.
Commun., 1996, 49–50.
32 J. M. Bainbridge, S. Corr, M. Kanai and J. M. Percy, Tetrahedron
Lett., 2000, 41, 971–974.
42 E. A. Basso, C. Kaiser, R. Rittner and J. B. Lambert, J. Org. Chem.,
1993, 58, 7865–7869.
43 E. L. Eliel, S. H. Wilen and L. N. Mander, Stereochemistry of
Organic Compounds, Wiley-Interscience, Chichester, 1994, 732–
734.
44 J. W. Banks, A. S. Batsanov, J. A. K. Howard, D. O’Hagan, H. S.
Rzepa and S. Martin-Santamaria, J. Chem. Soc., Perkin Trans. 2,
1999, 2409–2411.
33 T. Dubuffet, R. Sauvêtre and J. F. Normant, J. Organomet. Chem.,
1988, 341, 11–18.
34 J. A. Howarth, W. M. Owton, J. M. Percy and M. H. Rock,
Tetrahedron, 1995, 51, 10289–10302.
35 P. L. Coe, J. Burdon and I. B. Haslock, J. Fluorine Chem., 2000, 102,
43–50.
45 D. A. Holt, Tetrahedron Lett., 1981, 22, 2243–2246.
46 E. N. Marvell and W. Whalley, Tetrahedron Lett., 1970, 509–512.
47 M. F. Lipton, C. M. Sorensen, A. C. Sadler and R. H. Shapiro,
J. Organomet. Chem., 1980, 186, 155–158.
O r g . B i o m o l . C h e m . , 2 0 0 3 , 1, 4 4 2 3 – 4 4 3 4
4434